recoll/src/doc/user/usermanual.html

10756 lines
521 KiB
HTML

<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"
"http://www.w3.org/TR/html4/loose.dtd">
<html>
<head>
<meta name="generator" content=
"HTML Tidy for HTML5 for Linux version 5.6.0">
<meta http-equiv="Content-Type" content=
"text/html; charset=utf-8">
<title>Recoll user manual</title>
<link rel="stylesheet" type="text/css" href="docbook-xsl.css">
<meta name="generator" content="DocBook XSL Stylesheets V1.79.1">
<meta name="description" content=
"Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.3 or any later version published by the Free Software Foundation; with no Invariant Sections, no Front-Cover Texts, and no Back-Cover Texts. A copy of the license can be found at the following location: GNU web site. This document introduces full text search notions and describes the installation and use of the Recoll application. This version describes Recoll 1.29.">
</head>
<body bgcolor="white" text="black" link="#0000FF" vlink="#840084"
alink="#0000FF">
<div lang="en" class="book">
<div class="titlepage">
<div>
<div>
<h1 class="title"><a name="idm1" id="idm1"></a>Recoll
user manual</h1>
</div>
<div>
<div class="author">
<h3 class="author"><span class=
"firstname">Jean-Francois</span> <span class=
"surname">Dockes</span></h3>
<div class="affiliation">
<div class="address">
<p><code class="email">&lt;<a class="email" href=
"mailto:jfd@recoll.org">jfd@recoll.org</a>&gt;</code></p>
</div>
</div>
</div>
</div>
<div>
<p class="copyright">Copyright © 2005-2020 Jean-Francois
Dockes</p>
</div>
<div>
<div class="abstract">
<p><code class="literal">Permission is granted to copy,
distribute and/or modify this document under the terms
of the GNU Free Documentation License, Version 1.3 or
any later version published by the Free Software
Foundation; with no Invariant Sections, no Front-Cover
Texts, and no Back-Cover Texts. A copy of the license
can be found at the following location: <a class=
"ulink" href="http://www.gnu.org/licenses/fdl.html"
target="_top">GNU web site</a>.</code></p>
<p>This document introduces full text search notions
and describes the installation and use of the
<span class="application">Recoll</span> application.
This version describes <span class=
"application">Recoll</span> 1.29.</p>
</div>
</div>
</div>
<hr>
</div>
<div class="toc">
<p><b>Table of Contents</b></p>
<dl class="toc">
<dt><span class="chapter">1. <a href=
"#RCL.INTRODUCTION">Introduction</a></span></dt>
<dd>
<dl>
<dt><span class="sect1">1.1. <a href=
"#RCL.INTRODUCTION.TRYIT">Giving it a
try</a></span></dt>
<dt><span class="sect1">1.2. <a href=
"#RCL.INTRODUCTION.SEARCH">Full text
search</a></span></dt>
<dt><span class="sect1">1.3. <a href=
"#RCL.INTRODUCTION.RECOLL">Recoll
overview</a></span></dt>
</dl>
</dd>
<dt><span class="chapter">2. <a href=
"#RCL.INDEXING">Indexing</a></span></dt>
<dd>
<dl>
<dt><span class="sect1">2.1. <a href=
"#RCL.INDEXING.INTRODUCTION">Introduction</a></span></dt>
<dd>
<dl>
<dt><span class="sect2">2.1.1. <a href=
"#RCL.INDEXING.INTRODUCTION.MODES">Indexing
modes</a></span></dt>
<dt><span class="sect2">2.1.2. <a href=
"#RCL.INDEXING.INTRODUCTION.CONFIG">Configurations,
multiple indexes</a></span></dt>
<dt><span class="sect2">2.1.3. <a href=
"#idm235">Document types</a></span></dt>
<dt><span class="sect2">2.1.4. <a href=
"#idm284">Indexing failures</a></span></dt>
<dt><span class="sect2">2.1.5. <a href=
"#idm294">Recovery</a></span></dt>
</dl>
</dd>
<dt><span class="sect1">2.2. <a href=
"#RCL.INDEXING.STORAGE">Index storage</a></span></dt>
<dd>
<dl>
<dt><span class="sect2">2.2.1. <a href=
"#RCL.INDEXING.STORAGE.FORMAT"><span class=
"application">Xapian</span> index
formats</a></span></dt>
<dt><span class="sect2">2.2.2. <a href=
"#RCL.INDEXING.STORAGE.SECURITY">Security
aspects</a></span></dt>
<dt><span class="sect2">2.2.3. <a href=
"#RCL.INDEXING.STORAGE.BIG">Special considerations
for big indexes</a></span></dt>
</dl>
</dd>
<dt><span class="sect1">2.3. <a href=
"#RCL.INDEXING.CONFIG">Index
configuration</a></span></dt>
<dd>
<dl>
<dt><span class="sect2">2.3.1. <a href=
"#RCL.INDEXING.CONFIG.MULTIPLE">Multiple
indexes</a></span></dt>
<dt><span class="sect2">2.3.2. <a href=
"#RCL.INDEXING.CONFIG.SENS">Index case and
diacritics sensitivity</a></span></dt>
<dt><span class="sect2">2.3.3. <a href=
"#RCL.INDEXING.CONFIG.THREADS">Indexing threads
configuration (<span class=
"application">Unix</span>-like
systems)</a></span></dt>
<dt><span class="sect2">2.3.4. <a href=
"#RCL.INDEXING.CONFIG.GUI">The index configuration
GUI</a></span></dt>
</dl>
</dd>
<dt><span class="sect1">2.4. <a href=
"#RCL.INDEXING.REMOVABLE">Removable
volumes</a></span></dt>
<dt><span class="sect1">2.5. <a href=
"#RCL.INDEXING.WebQUEUE"><span class=
"application">Unix</span>-like systems: indexing
visited Web pages</a></span></dt>
<dt><span class="sect1">2.6. <a href=
"#RCL.INDEXING.EXTATTR"><span class=
"application">Unix</span>-like systems: using extended
attributes</a></span></dt>
<dt><span class="sect1">2.7. <a href=
"#RCL.INDEXING.EXTTAGS"><span class=
"application">Unix</span>-like systems: importing
external tags</a></span></dt>
<dt><span class="sect1">2.8. <a href=
"#RCL.INDEXING.PDF">The PDF input
handler</a></span></dt>
<dd>
<dl>
<dt><span class="sect2">2.8.1. <a href=
"#RCL.INDEXING.PDF.XMP">XMP fields
extraction</a></span></dt>
<dt><span class="sect2">2.8.2. <a href=
"#RCL.INDEXING.PDF.ATTACH">PDF attachment
indexing</a></span></dt>
</dl>
</dd>
<dt><span class="sect1">2.9. <a href=
"#RCL.INDEXING.OCR">Recoll and OCR</a></span></dt>
<dt><span class="sect1">2.10. <a href=
"#RCL.INDEXING.PERIODIC">Periodic
indexing</a></span></dt>
<dt><span class="sect1">2.11. <a href=
"#RCL.INDEXING.MONITOR"><span class=
"application">Unix</span>-like systems: real time
indexing</a></span></dt>
</dl>
</dd>
<dt><span class="chapter">3. <a href=
"#RCL.SEARCH">Searching</a></span></dt>
<dd>
<dl>
<dt><span class="sect1">3.1. <a href=
"#RCL.SEARCH.INTRODUCTION">Introduction</a></span></dt>
<dt><span class="sect1">3.2. <a href=
"#RCL.SEARCH.GUI">Searching with the Qt graphical user
interface</a></span></dt>
<dd>
<dl>
<dt><span class="sect2">3.2.1. <a href=
"#RCL.SEARCH.GUI.SIMPLE">Simple
search</a></span></dt>
<dt><span class="sect2">3.2.2. <a href=
"#RCL.SEARCH.GUI.RESLIST">The result
list</a></span></dt>
<dt><span class="sect2">3.2.3. <a href=
"#RCL.SEARCH.GUI.RESTABLE">The result
table</a></span></dt>
<dt><span class="sect2">3.2.4. <a href=
"#RCL.SEARCH.GUI.RUNSCRIPT"><span class=
"application">Unix</span>-like systems: running
arbitrary commands on result files</a></span></dt>
<dt><span class="sect2">3.2.5. <a href=
"#RCL.SEARCH.GUI.THUMBNAILS"><span class=
"application">Unix</span>-like systems: displaying
thumbnails</a></span></dt>
<dt><span class="sect2">3.2.6. <a href=
"#RCL.SEARCH.GUI.PREVIEW">The preview
window</a></span></dt>
<dt><span class="sect2">3.2.7. <a href=
"#RCL.SEARCH.GUI.FRAGBUTS">The Query Fragments
window</a></span></dt>
<dt><span class="sect2">3.2.8. <a href=
"#RCL.SEARCH.GUI.COMPLEX">Complex/advanced
search</a></span></dt>
<dt><span class="sect2">3.2.9. <a href=
"#RCL.SEARCH.GUI.TERMEXPLORER">The term explorer
tool</a></span></dt>
<dt><span class="sect2">3.2.10. <a href=
"#RCL.SEARCH.GUI.MULTIDB">Multiple
indexes</a></span></dt>
<dt><span class="sect2">3.2.11. <a href=
"#RCL.SEARCH.GUI.HISTORY">Document
history</a></span></dt>
<dt><span class="sect2">3.2.12. <a href=
"#RCL.SEARCH.GUI.SORT">Sorting search results and
collapsing duplicates</a></span></dt>
<dt><span class="sect2">3.2.13. <a href=
"#RCL.SEARCH.GUI.SHORTCUTS">Keyboard
shortcuts</a></span></dt>
<dt><span class="sect2">3.2.14. <a href=
"#RCL.SEARCH.GUI.TIPS">Search tips</a></span></dt>
<dt><span class="sect2">3.2.15. <a href=
"#RCL.SEARCH.SAVING">Saving and restoring queries
(1.21 and later)</a></span></dt>
<dt><span class="sect2">3.2.16. <a href=
"#RCL.SEARCH.GUI.CUSTOM">Customizing the search
interface</a></span></dt>
</dl>
</dd>
<dt><span class="sect1">3.3. <a href=
"#RCL.SEARCH.KIO">Searching with the KDE KIO
slave</a></span></dt>
<dd>
<dl>
<dt><span class="sect2">3.3.1. <a href=
"#RCL.SEARCH.KIO.INTRO">What's this</a></span></dt>
<dt><span class="sect2">3.3.2. <a href=
"#RCL.SEARCH.KIO.SEARCHABLEDOCS">Searchable
documents</a></span></dt>
</dl>
</dd>
<dt><span class="sect1">3.4. <a href=
"#RCL.SEARCH.COMMANDLINE">Searching on the command
line</a></span></dt>
<dt><span class="sect1">3.5. <a href=
"#RCL.SEARCH.LANG">The query language</a></span></dt>
<dd>
<dl>
<dt><span class="sect2">3.5.1. <a href=
"#RCL.SEARCH.LANG.RANGES">Range
clauses</a></span></dt>
<dt><span class="sect2">3.5.2. <a href=
"#RCL.SEARCH.LANG.MODIFIERS">Modifiers</a></span></dt>
</dl>
</dd>
<dt><span class="sect1">3.6. <a href=
"#RCL.SEARCH.ANCHORWILD">Anchored searches and
wildcards</a></span></dt>
<dd>
<dl>
<dt><span class="sect2">3.6.1. <a href=
"#RCL.SEARCH.WILDCARDS">More about
wildcards</a></span></dt>
<dt><span class="sect2">3.6.2. <a href=
"#RCL.SEARCH.ANCHOR">Anchored
searches</a></span></dt>
</dl>
</dd>
<dt><span class="sect1">3.7. <a href=
"#RCL.SEARCH.SYNONYMS">Using Synonyms
(1.22)</a></span></dt>
<dt><span class="sect1">3.8. <a href=
"#RCL.SEARCH.PTRANS">Path translations</a></span></dt>
<dt><span class="sect1">3.9. <a href=
"#RCL.SEARCH.CASEDIAC">Search case and diacritics
sensitivity</a></span></dt>
<dt><span class="sect1">3.10. <a href=
"#RCL.SEARCH.DESKTOP">Desktop
integration</a></span></dt>
<dd>
<dl>
<dt><span class="sect2">3.10.1. <a href=
"#RCL.SEARCH.SHORTCUT">Hotkeying
recoll</a></span></dt>
<dt><span class="sect2">3.10.2. <a href=
"#RCL.KICKER-APPLET">The KDE Kicker Recoll
applet</a></span></dt>
</dl>
</dd>
</dl>
</dd>
<dt><span class="chapter">4. <a href=
"#RCL.PROGRAM">Programming interface</a></span></dt>
<dd>
<dl>
<dt><span class="sect1">4.1. <a href=
"#RCL.PROGRAM.FILTERS">Writing a document input
handler</a></span></dt>
<dd>
<dl>
<dt><span class="sect2">4.1.1. <a href=
"#RCL.PROGRAM.FILTERS.SIMPLE">Simple input
handlers</a></span></dt>
<dt><span class="sect2">4.1.2. <a href=
"#RCL.PROGRAM.FILTERS.MULTIPLE">"Multiple"
handlers</a></span></dt>
<dt><span class="sect2">4.1.3. <a href=
"#RCL.PROGRAM.FILTERS.ASSOCIATION">Telling
<span class="application">Recoll</span> about the
handler</a></span></dt>
<dt><span class="sect2">4.1.4. <a href=
"#RCL.PROGRAM.FILTERS.HTML">Input handler
output</a></span></dt>
<dt><span class="sect2">4.1.5. <a href=
"#RCL.PROGRAM.FILTERS.PAGES">Page
numbers</a></span></dt>
</dl>
</dd>
<dt><span class="sect1">4.2. <a href=
"#RCL.PROGRAM.FIELDS">Field data
processing</a></span></dt>
<dt><span class="sect1">4.3. <a href=
"#RCL.PROGRAM.PYTHONAPI">Python API</a></span></dt>
<dd>
<dl>
<dt><span class="sect2">4.3.1. <a href=
"#RCL.PROGRAM.PYTHONAPI.INTRO">Introduction</a></span></dt>
<dt><span class="sect2">4.3.2. <a href=
"#RCL.PROGRAM.PYTHONAPI.ELEMENTS">Interface
elements</a></span></dt>
<dt><span class="sect2">4.3.3. <a href=
"#RCL.PROGRAM.PYTHONAPI.LOG">Log messages for
Python scripts</a></span></dt>
<dt><span class="sect2">4.3.4. <a href=
"#RCL.PROGRAM.PYTHONAPI.SEARCH">Python search
interface</a></span></dt>
<dt><span class="sect2">4.3.5. <a href=
"#RCL.PROGRAM.PYTHONAPI.UPDATE">Creating Python
external indexers</a></span></dt>
<dt><span class="sect2">4.3.6. <a href=
"#RCL.PROGRAM.PYTHONAPI.COMPAT">Package
compatibility with the previous
version</a></span></dt>
</dl>
</dd>
</dl>
</dd>
<dt><span class="chapter">5. <a href=
"#RCL.INSTALL">Installation and
configuration</a></span></dt>
<dd>
<dl>
<dt><span class="sect1">5.1. <a href=
"#RCL.INSTALL.BINARY">Installing a binary
copy</a></span></dt>
<dt><span class="sect1">5.2. <a href=
"#RCL.INSTALL.EXTERNAL">Supporting
packages</a></span></dt>
<dt><span class="sect1">5.3. <a href=
"#RCL.INSTALL.BUILDING">Building from
source</a></span></dt>
<dd>
<dl>
<dt><span class="sect2">5.3.1. <a href=
"#RCL.INSTALL.BUILDING.PREREQS">Prerequisites</a></span></dt>
<dt><span class="sect2">5.3.2. <a href=
"#RCL.INSTALL.BUILDING.BUILDING">Building</a></span></dt>
<dt><span class="sect2">5.3.3. <a href=
"#RCL.INSTALL.BUILDING.INSTALL">Installing</a></span></dt>
<dt><span class="sect2">5.3.4. <a href=
"#RCL.INSTALL.BUILDING.PYTHON">Python API
package</a></span></dt>
<dt><span class="sect2">5.3.5. <a href=
"#RCL.INSTALL.BUILDING.SOLARIS">Building on
Solaris</a></span></dt>
</dl>
</dd>
<dt><span class="sect1">5.4. <a href=
"#RCL.INSTALL.CONFIG">Configuration
overview</a></span></dt>
<dd>
<dl>
<dt><span class="sect2">5.4.1. <a href=
"#RCL.INSTALL.CONFIG.ENVIR">Environment
variables</a></span></dt>
<dt><span class="sect2">5.4.2. <a href=
"#RCL.INSTALL.CONFIG.RECOLLCONF">Recoll main
configuration file, recoll.conf</a></span></dt>
<dt><span class="sect2">5.4.3. <a href=
"#RCL.INSTALL.CONFIG.FIELDS">The fields
file</a></span></dt>
<dt><span class="sect2">5.4.4. <a href=
"#RCL.INSTALL.CONFIG.MIMEMAP">The mimemap
file</a></span></dt>
<dt><span class="sect2">5.4.5. <a href=
"#RCL.INSTALL.CONFIG.MIMECONF">The mimeconf
file</a></span></dt>
<dt><span class="sect2">5.4.6. <a href=
"#RCL.INSTALL.CONFIG.MIMEVIEW">The mimeview
file</a></span></dt>
<dt><span class="sect2">5.4.7. <a href=
"#RCL.INSTALL.CONFIG.PTRANS">The <code class=
"filename">ptrans</code> file</a></span></dt>
<dt><span class="sect2">5.4.8. <a href=
"#RCL.INSTALL.CONFIG.EXAMPLES">Examples of
configuration adjustments</a></span></dt>
</dl>
</dd>
</dl>
</dd>
</dl>
</div>
<div class="list-of-tables">
<p><b>List of Tables</b></p>
<dl>
<dt>3.1. <a href="#idm1465">Keyboard shortcuts</a></dt>
</dl>
</div>
<div class="chapter">
<div class="titlepage">
<div>
<div>
<h1 class="title"><a name="RCL.INTRODUCTION" id=
"RCL.INTRODUCTION"></a>Chapter&nbsp;1.&nbsp;Introduction</h1>
</div>
</div>
</div>
<p>This document introduces full text search notions and
describes the installation and use of the <span class=
"application">Recoll</span> application. It is updated for
<span class="application">Recoll</span> 1.29.</p>
<p><span class="application">Recoll</span> was for a long
time dedicated to Unix-like systems. It was only lately
(2015) ported to <span class="application">MS-Windows</span>.
Many references in this manual, especially file locations,
are specific to Unix, and not valid on <span class=
"application">Windows</span>, where some described features
are also not available. The manual will be progressively
updated. Until this happens, on <span class=
"application">Windows</span>, most references to shared files
can be translated by looking under the Recoll installation
directory (Typically <code class="filename">C:/Program Files
(x86)/Recoll</code>, esp. anything referenced in <code class=
"filename">/usr/share</code> in this document will be found
int the <code class="filename">Share</code> subdirectory).
The user configuration is stored by default under
<code class="filename">AppData/Local/Recoll</code> inside the
user directory, along with the index itself.</p>
<div class="sect1">
<div class="titlepage">
<div>
<div>
<h2 class="title" style="clear: both"><a name=
"RCL.INTRODUCTION.TRYIT" id=
"RCL.INTRODUCTION.TRYIT"></a>1.1.&nbsp;Giving it a
try</h2>
</div>
</div>
</div>
<p>If you do not like reading manuals (who does?) but wish
to give <span class="application">Recoll</span> a try, just
<a class="link" href="#RCL.INSTALL.BINARY" title=
"5.1.&nbsp;Installing a binary copy">install</a> the
application and start the <span class=
"command"><strong>recoll</strong></span> graphical user
interface (GUI), which will ask permission to index your
home directory, allowing you to search immediately after
indexing completes.</p>
<p>Do not do this if your home directory contains a huge
number of documents and you do not want to wait or are very
short on disk space. In this case, you may first want to
customize the <a class="link" href="#RCL.INDEXING.CONFIG"
title="2.3.&nbsp;Index configuration">configuration</a> to
restrict the indexed area (shortcut: from the <span class=
"command"><strong>recoll</strong></span> GUI go to:
<span class="guimenu">Preferences</span><span class=
"guimenuitem">Indexing configuration</span>, then adjust
the <span class="guilabel">Top directories</span>
section).</p>
<p>On <span class="application">Unix</span>-like systems,
you may need to install the appropriate <a class="link"
href="#RCL.INSTALL.EXTERNAL" title=
"5.2.&nbsp;Supporting packages">supporting applications</a>
for document types that need them (for example <span class=
"application">antiword</span> for <span class=
"application">Microsoft Word</span> files). The
<span class="application">Recoll</span> for <span class=
"application">Windows</span> package is self-contained and
includes most useful auxiliary programs.</p>
</div>
<div class="sect1">
<div class="titlepage">
<div>
<div>
<h2 class="title" style="clear: both"><a name=
"RCL.INTRODUCTION.SEARCH" id=
"RCL.INTRODUCTION.SEARCH"></a>1.2.&nbsp;Full text
search</h2>
</div>
</div>
</div>
<p><span class="application">Recoll</span> is a full text
search application, which means that it finds your data by
content rather than by external attributes (like the file
name). You specify words (terms) which should or should not
appear in the text you are looking for, and receive in
return a list of matching documents, ordered so that the
most <span class="emphasis"><em>relevant</em></span>
documents will appear first.</p>
<p>You do not need to remember in what file or email
message you stored a given piece of information. You just
ask for related terms, and the tool will return a list of
documents where these terms are prominent, in a similar way
to Internet search engines.</p>
<p>Full text search applications try to determine which
documents are most relevant to the search terms you
provide. Computer algorithms for determining relevance can
be very complex, and in general are inferior to the power
of the human mind to rapidly determine relevance. The
quality of relevance guessing is probably the most
important aspect when evaluating a search application.
<span class="application">Recoll</span> relies on the
<span class="application">Xapian</span> probabilistic
information retrieval library to determine relevance.</p>
<p>In many cases, you are looking for all the forms of a
word, including plurals, different tenses for a verb, or
terms derived from the same root or <span class=
"emphasis"><em>stem</em></span> (example: <em class=
"replaceable"><code>floor, floors, floored,
flooring...</code></em>). Queries are usually automatically
expanded to all such related terms (words that reduce to
the same stem). This can be prevented for searching for a
specific form.</p>
<p>Stemming, by itself, does not accommodate for
misspellings or phonetic searches. A full text search
application may also support this form of approximation.
For example, a search for <em class=
"replaceable"><code>aliterattion</code></em> returning no
result might propose <em class=
"replaceable"><code>alliteration, alteration, alterations,
or altercation</code></em> as possible replacement terms.
<span class="application">Recoll</span> bases its
suggestions on the actual index contents, so that
suggestions may be made for words which would not appear in
a standard dictionary.</p>
</div>
<div class="sect1">
<div class="titlepage">
<div>
<div>
<h2 class="title" style="clear: both"><a name=
"RCL.INTRODUCTION.RECOLL" id=
"RCL.INTRODUCTION.RECOLL"></a>1.3.&nbsp;Recoll
overview</h2>
</div>
</div>
</div>
<p><span class="application">Recoll</span> uses the
<a class="ulink" href="http://www.xapian.org" target=
"_top"><span class="application">Xapian</span></a>
information retrieval library as its storage and retrieval
engine. <span class="application">Xapian</span> is a very
mature package using <a class="ulink" href=
"http://www.xapian.org/docs/intro_ir.html" target="_top">a
sophisticated probabilistic ranking model</a>.</p>
<p>The <span class="application">Xapian</span> library
manages an index database which describes where terms
appear in your document files. It efficiently processes the
complex queries which are produced by the <span class=
"application">Recoll</span> query expansion mechanism, and
is in charge of the all-important relevance computation
task.</p>
<p><span class="application">Recoll</span> provides the
mechanisms and interface to get data into and out of the
index. This includes translating the many possible document
formats into pure text, handling term variations (using
<span class="application">Xapian</span> stemmers), and
spelling approximations (using the <span class=
"application">aspell</span> speller), interpreting user
queries and presenting results.</p>
<p>In a shorter way, <span class=
"application">Recoll</span> does the dirty footwork,
<span class="application">Xapian</span> deals with the
intelligent parts of the process.</p>
<p>The <span class="application">Xapian</span> index can be
big (roughly the size of the original document set), but it
is not a document archive. <span class=
"application">Recoll</span> can only display documents that
still exist at the place from which they were indexed.</p>
<p><span class="application">Recoll</span> stores all
internal data in <span class="application">Unicode
UTF-8</span> format, and it can index many types of files
with different character sets, encodings, and languages
into the same index. It can process documents embedded
inside other documents (for example a PDF document stored
inside a Zip archive sent as an email attachment...), down
to an arbitrary depth.</p>
<p>Stemming is the process by which <span class=
"application">Recoll</span> reduces words to their radicals
so that searching does not depend, for example, on a word
being singular or plural (floor, floors), or on a verb
tense (flooring, floored). Because the mechanisms used for
stemming depend on the specific grammatical rules for each
language, there is a separate <span class=
"application">Xapian</span> stemmer module for most common
languages where stemming makes sense.</p>
<p><span class="application">Recoll</span> stores the
unstemmed versions of terms in the main index and uses
auxiliary databases for term expansion (one for each
stemming language), which means that you can switch
stemming languages between searches, or add a language
without needing a full reindex.</p>
<p>Storing documents written in different languages in the
same index is possible, and commonly done. In this
situation, you can specify several stemming languages for
the index.</p>
<p><span class="application">Recoll</span> currently makes
no attempt at automatic language recognition, which means
that the stemmer will sometimes be applied to terms from
other languages with potentially strange results. In
practise, even if this introduces possibilities of
confusion, this approach has been proven quite useful, and
it is much less cumbersome than separating your documents
according to what language they are written in.</p>
<p>By default, <span class="application">Recoll</span>
strips most accents and diacritics from terms, and converts
them to lower case before either storing them in the index
or searching for them. As a consequence, it is impossible
to search for a particular capitalization of a term
(<code class="literal">US</code> / <code class=
"literal">us</code>), or to discriminate two terms based on
diacritics (<code class="literal">sake</code> /
<code class="literal">saké</code>, <code class=
"literal">mate</code> / <code class=
"literal">maté</code>).</p>
<p><span class="application">Recoll</span> can optionally
store the raw terms, without accent stripping or case
conversion. In this configuration, default searches will
behave as before, but it is possible to perform searches
sensitive to case and diacritics. This is described in more
detail in the section about <a class="link" href=
"#RCL.INDEXING.CONFIG.SENS" title=
"2.3.2.&nbsp;Index case and diacritics sensitivity">index
case and diacritics sensitivity</a>.</p>
<p><span class="application">Recoll</span> uses many
parameters to define exactly what to index, and how to
classify and decode the source documents. These are kept in
<a class="link" href="#RCL.INDEXING.CONFIG" title=
"2.3.&nbsp;Index configuration">configuration files</a>. A
default configuration is copied into a standard location
(usually something like <code class=
"filename">/usr/share/recoll/examples</code>) during
installation. The default values set by the configuration
files in this directory may be overridden by values set
inside your personal configuration. With the default
configuration, <span class="application">Recoll</span> will
index your home directory with generic parameters. Most
common parameters can be set by using configuration menus
in the <span class="command"><strong>recoll</strong></span>
GUI. Some less common parameters can only be set by editing
the text files (the new values will be preserved by the
GUI).</p>
<p>The <a class="link" href="#RCL.INDEXING.PERIODIC.EXEC"
title="Running the indexer">indexing process</a> is started
automatically (after asking permission), the first time you
execute the <span class=
"command"><strong>recoll</strong></span> GUI. Indexing can
also be performed by executing the <span class=
"command"><strong>recollindex</strong></span> command.
<span class="application">Recoll</span> indexing is
multithreaded by default when appropriate hardware
resources are available, and can perform in parallel
multiple tasks for text extraction, segmentation and index
updates.</p>
<p><a class="link" href="#RCL.SEARCH" title=
"Chapter&nbsp;3.&nbsp;Searching">Searches</a> are usually
performed inside the <span class=
"command"><strong>recoll</strong></span> GUI, which has
many options to help you find what you are looking for.
However, there are other ways to query the index:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style="list-style-type: disc;">
<li class="listitem">
<p>A <a class="link" href="#RCL.SEARCH.COMMANDLINE"
title=
"3.4.&nbsp;Searching on the command line">command
line interface</a>.</p>
</li>
<li class="listitem">
<p>A <a class="link" href="#RCL.PROGRAM.PYTHONAPI"
title="4.3.&nbsp;Python API"><span class=
"application">Python</span> programming
interface</a></p>
</li>
<li class="listitem">
<p>A <a class="link" href="#RCL.SEARCH.KIO" title=
"3.3.&nbsp;Searching with the KDE KIO slave"><span class="application">
KDE</span> KIO slave module</a>.</p>
</li>
<li class="listitem">
<p>A Ubuntu Unity <a class="ulink" href=
"https://www.lesbonscomptes.com/recoll/pages/download.html"
target="_top">Scope</a> module.</p>
</li>
<li class="listitem">
<p>A Gnome Shell <a class="ulink" href=
"https://www.lesbonscomptes.com/recoll/pages/download.html"
target="_top">Search Provider</a>.</p>
</li>
<li class="listitem">
<p>A <a class="ulink" href=
"https://framagit.org/medoc92/recollwebui" target=
"_top">Web interface</a>.</p>
</li>
</ul>
</div>
</div>
</div>
<div class="chapter">
<div class="titlepage">
<div>
<div>
<h1 class="title"><a name="RCL.INDEXING" id=
"RCL.INDEXING"></a>Chapter&nbsp;2.&nbsp;Indexing</h1>
</div>
</div>
</div>
<div class="sect1">
<div class="titlepage">
<div>
<div>
<h2 class="title" style="clear: both"><a name=
"RCL.INDEXING.INTRODUCTION" id=
"RCL.INDEXING.INTRODUCTION"></a>2.1.&nbsp;Introduction</h2>
</div>
</div>
</div>
<p>Indexing is the process by which the set of documents is
analyzed and the data entered into the database.
<span class="application">Recoll</span> indexing is
normally incremental: documents will only be processed if
they have been modified since the last run. On the first
execution, all documents will need processing. A full index
build can be forced later by specifying an option to the
indexing command (<span class=
"command"><strong>recollindex</strong></span> <code class=
"option">-z</code> or <code class="option">-Z</code>).</p>
<p><span class=
"command"><strong>recollindex</strong></span> skips files
which caused an error during a previous pass. This is a
performance optimization, and the command line option
<code class="option">-k</code> can be set to retry failed
files, for example after updating an input handler.</p>
<p>The following sections give an overview of different
aspects of the indexing processes and configuration, with
links to detailed sections.</p>
<p>Depending on your data, temporary files may be needed
during indexing, some of them possibly quite big. You can
use the <code class="envar">RECOLL_TMPDIR</code> or
<code class="envar">TMPDIR</code> environment variables to
determine where they are created (the default is to use
<code class="filename">/tmp</code>). Using <code class=
"envar">TMPDIR</code> has the nice property that it may
also be taken into account by auxiliary commands executed
by <span class=
"command"><strong>recollindex</strong></span>.</p>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.INDEXING.INTRODUCTION.MODES" id=
"RCL.INDEXING.INTRODUCTION.MODES"></a>2.1.1.&nbsp;Indexing
modes</h3>
</div>
</div>
</div>
<p><span class="application">Recoll</span> indexing can
be performed along two main modes:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style=
"list-style-type: disc;">
<li class="listitem">
<p><b><a class="link" href="#RCL.INDEXING.PERIODIC"
title="2.10.&nbsp;Periodic indexing">Periodic (or
batch) indexing</a> .&nbsp;</b><span class=
"command"><strong>recollindex</strong></span> is
executed at discrete times. On <span class=
"application">Unix</span>-like systems, the typical
usage is to have a nightly run <a class="link"
href="#RCL.INDEXING.PERIODIC.AUTOMAT" title=
"Linux: using cron to automate indexing">programmed</a>
into your <span class=
"command"><strong>cron</strong></span> file. On
<span class="application">Windows</span>, this is
the only mode available, and the Windows Task
Scheduler can be used to run indexing. In both
cases, the GUI includes an easy interface to the
system batch scheduler.</p>
</li>
<li class="listitem">
<p><b><a class="link" href="#RCL.INDEXING.MONITOR"
title=
"2.11.&nbsp;Unix-like systems: real time indexing">Real
time indexing</a> .&nbsp;</b>(Only available on
<span class="application">Unix</span>-like
systems). <span class=
"command"><strong>recollindex</strong></span> runs
permanently as a daemon and uses a file system
alteration monitor (e.g. <span class=
"application">inotify</span>) to detect file
changes. New or updated files are indexed at once.
Monitoring a big file system tree can consume
significant system resources.</p>
</li>
</ul>
</div>
<div class="simplesect">
<div class="titlepage">
<div>
<div>
<h4 class="title"><a name="idm190" id=
"idm190"></a><span class=
"application">Unix</span>-like systems: choosing
an indexing mode</h4>
</div>
</div>
</div>
<p>The choice between the two methods is mostly a
matter of preference, and they can be combined by
setting up multiple indexes (ie: use periodic indexing
on a big documentation directory, and real time
indexing on a small home directory), or, with
<span class="application">Recoll</span> 1.24 and newer,
by <a class="link" href="#RCL.INDEXING.MONITOR" title=
"2.11.&nbsp;Unix-like systems: real time indexing">configuring
the index so that only a subset of the tree will be
monitored.</a></p>
<p>The choice of method and the parameters used can be
configured from the <span class=
"command"><strong>recoll</strong></span> GUI:
<span class="guimenu">Preferences</span><span class=
"guimenuitem">Indexing schedule</span> dialog.</p>
</div>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.INDEXING.INTRODUCTION.CONFIG" id=
"RCL.INDEXING.INTRODUCTION.CONFIG"></a>2.1.2.&nbsp;Configurations,
multiple indexes</h3>
</div>
</div>
</div>
<p><span class="application">Recoll</span> supports
defining multiple indexes, each defined by its own
configuration directory. A configuration directory
contains <a class="link" href="#RCL.INDEXING.CONFIG"
title="2.3.&nbsp;Index configuration">several files</a>
which describe what should be indexed and how.</p>
<p>When <span class=
"command"><strong>recoll</strong></span> or <span class=
"command"><strong>recollindex</strong></span> is first
executed, it creates a default configuration directory.
This configuration is the one used for indexing and
querying when no specific configuration is specified. It
is located in <code class=
"filename">$HOME/.recoll/</code> for <span class=
"application">Unix</span>-like systems and <code class=
"filename">%LOCALAPPDATA%\Recoll</code> on <span class=
"application">Windows</span> (typically <code class=
"filename">C:\Users\[me]\Appdata\Local\Recoll</code>).</p>
<p>All configuration parameters have defaults, defined in
system-wide files. Without further customisation, the
default configuration will process your complete home
directory, with a reasonable set of defaults. It can be
adjusted to process a different area of the file system,
select files in different ways, and many other
things.</p>
<p>In some cases, it may be useful to create additional
configuration directories, for example, to separate
personal and shared indexes, or to take advantage of the
organization of your data to improve search
precision.</p>
<p>In order to do this, you would create an empty
directory in a location of your choice, and then instruct
<span class="command"><strong>recoll</strong></span> or
<span class="command"><strong>recollindex</strong></span>
to use it by setting either a command line option
(<code class="literal">-c</code> <em class=
"replaceable"><code>/some/directory</code></em>), or an
environment variable (<code class=
"envar">RECOLL_CONFDIR</code>=<em class=
"replaceable"><code>/some/directory</code></em>). Any
modification performed by the commands (e.g.
configuration customisation or searches by <span class=
"command"><strong>recoll</strong></span> or index
creation by <span class=
"command"><strong>recollindex</strong></span>) would then
apply to the new directory and not to the default
one.</p>
<p>Once multiple indexes are created, you can use each of
them separately by setting the <code class=
"literal">-c</code> option or the <code class=
"envar">RECOLL_CONFDIR</code> environment variable when
starting a command, to select the desired index.</p>
<p>It is also possible to instruct one configuration to
query one or several other indexes in addition to its
own, by using the <span class="guimenuitem">External
index</span> function in the <span class=
"command"><strong>recoll</strong></span> GUI, or some
other functions in the command line and programming
tools.</p>
<p>A plausible usage scenario for the multiple index
feature would be for a system administrator to set up a
central index for shared data, that you choose to search
or not in addition to your personal data. Of course,
there are other possibilities. for example, there are
many cases where you know the subset of files that should
be searched, and where narrowing the search can improve
the results. You can achieve approximately the same
effect with the directory filter in advanced search, but
multiple indexes may have better performance and may be
worth the trouble in some cases.</p>
<p>A more advanced use case would be to use multiple
index to improve indexing performance, by updating
several indexes in parallel (using multiple CPU cores and
disks, or possibly several machines), and then merging
them, or querying them in parallel.</p>
<p>See the section about <a class="link" href=
"#RCL.INDEXING.CONFIG.MULTIPLE" title=
"2.3.1.&nbsp;Multiple indexes">configuring multiple
indexes</a> for more detail</p>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name="idm235" id=
"idm235"></a>2.1.3.&nbsp;Document types</h3>
</div>
</div>
</div>
<p><span class="application">Recoll</span> knows about
quite a few different document types. The parameters for
document types recognition and processing are set in
<a class="link" href="#RCL.INDEXING.CONFIG" title=
"2.3.&nbsp;Index configuration">configuration
files</a>.</p>
<p>Most file types, like HTML or word processing files,
only hold one document. Some file types, like email
folders or zip archives, can hold many individually
indexed documents, which may themselves be compound ones.
Such hierarchies can go quite deep, and <span class=
"application">Recoll</span> can process, for example, a
<span class="application">LibreOffice</span> document
stored as an attachment to an email message inside an
email folder archived in a zip file...</p>
<p><span class=
"command"><strong>recollindex</strong></span> processes
plain text, HTML, OpenDocument (Open/LibreOffice), email
formats, and a few others internally.</p>
<p>Other file types (ie: postscript, pdf, ms-word, rtf
...) need external applications for preprocessing. The
list is in the <a class="link" href=
"#RCL.INSTALL.EXTERNAL" title=
"5.2.&nbsp;Supporting packages">installation</a> section.
After every indexing operation, <span class=
"application">Recoll</span> updates a list of commands
that would be needed for indexing existing files types.
This list can be displayed by selecting the menu option
<span class="guimenu">File</span><span class=
"guimenuitem">Show Missing Helpers</span> in the
<span class="command"><strong>recoll</strong></span> GUI.
It is stored in the <code class="filename">missing</code>
text file inside the configuration directory.</p>
<p>After installing a missing handler, you may need to
tell <span class=
"command"><strong>recollindex</strong></span> to retry
the failed files, by adding option <code class=
"literal">-k</code> to the command line, or by using the
GUI <span class="guimenu">File</span><span class=
"guimenuitem">Special indexing</span> menu. This is
because <span class=
"command"><strong>recollindex</strong></span>, in its
default operation mode, will not retry files which caused
an error during an earlier pass. In special cases, it may
be useful to reset the data for a category of files
before indexing. See the <span class=
"command"><strong>recollindex</strong></span> manual
page. If your index is not too big, it may be simpler to
just reset it.</p>
<p>By default, <span class="application">Recoll</span>
will try to index any file type that it has a way to
read. This is sometimes not desirable, and there are ways
to either exclude some types, or on the contrary define a
positive list of types to be indexed. In the latter case,
any type not in the list will be ignored.</p>
<p>Excluding files by name can be done by adding wildcard
name patterns to the <a class="link" href=
"#RCL.INSTALL.CONFIG.RECOLLCONF.SKIPPEDNAMES">skippedNames</a>
list, which can be done from the GUI Index configuration
menu. Excluding by type can be done by setting the
<a class="link" href=
"#RCL.INSTALL.CONFIG.RECOLLCONF.EXCLUDEDMIMETYPES">excludedmimetypes</a>
list in the configuration file (1.20 and later). This can
be redefined for subdirectories.</p>
<p>You can also define an exclusive list of MIME types to
be indexed (no others will be indexed), by setting the
<a class="link" href=
"#RCL.INSTALL.CONFIG.RECOLLCONF.INDEXEDMIMETYPES">indexedmimetypes</a>
configuration variable. Example:</p>
<pre class="programlisting">
indexedmimetypes = text/html application/pdf
</pre>
<p>It is possible to redefine this parameter for
subdirectories. Example:</p>
<pre class="programlisting">
[/path/to/my/dir]
indexedmimetypes = application/pdf
</pre>
<p>(When using sections like this, don't forget that they
remain in effect until the end of the file or another
section indicator).</p>
<p><code class="literal">excludedmimetypes</code> or
<code class="literal">indexedmimetypes</code>, can be set
either by editing the <a class="link" href=
"#RCL.INSTALL.CONFIG.RECOLLCONF" title=
"5.4.2.&nbsp;Recoll main configuration file, recoll.conf">
configuration file (<code class=
"filename">recoll.conf</code>)</a> for the index, or by
using the GUI index configuration tool.</p>
<div class="note" style=
"margin-left: 0.5in; margin-right: 0.5in;">
<h3 class="title">Note about MIME types</h3>
<p>When editing the <code class=
"literal">indexedmimetypes</code> or <code class=
"literal">excludedmimetypes</code> lists, you should
use the MIME values listed in the <code class=
"filename">mimemap</code> file or in Recoll result
lists in preference to <code class="literal">file
-i</code> output: there are a number of differences.
The <code class="literal">file -i</code> output should
only be used for files without extensions, or for which
the extension is not listed in <code class=
"filename">mimemap</code></p>
</div>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name="idm284" id=
"idm284"></a>2.1.4.&nbsp;Indexing failures</h3>
</div>
</div>
</div>
<p>Indexing may fail for some documents, for a number of
reasons: a helper program may be missing, the document
may be corrupt, we may fail to uncompress a file because
no file system space is available, etc.</p>
<p>The <span class="application">Recoll</span> indexer in
versions 1.21 and later does not retry failed files by
default, because some indexing failures can be quite
costly (for example failing to uncompress a big file
because of insufficient disk space). Retrying will only
occur if an explicit option (<code class=
"option">-k</code>) is set on the <span class=
"command"><strong>recollindex</strong></span> command
line, or if a script executed when <span class=
"command"><strong>recollindex</strong></span> starts up
says so. The script is defined by a configuration
variable (<code class=
"literal">checkneedretryindexscript</code>), and makes a
rather lame attempt at deciding if a helper command may
have been installed, by checking if any of the common
<code class="filename">bin</code> directories have
changed.</p>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name="idm294" id=
"idm294"></a>2.1.5.&nbsp;Recovery</h3>
</div>
</div>
</div>
<p>In the rare case where the index becomes corrupted
(which can signal itself by weird search results or
crashes), the index files need to be erased before
restarting a clean indexing pass. Just delete the
<code class="filename">xapiandb</code> directory (see
<a class="link" href="#RCL.INDEXING.STORAGE" title=
"2.2.&nbsp;Index storage">next section</a>), or,
alternatively, start the next <span class=
"command"><strong>recollindex</strong></span> with the
<code class="option">-z</code> option, which will reset
the database before indexing. The difference between the
two methods is that the second will not change the
current index format, which may be undesirable if a newer
format is supported by the <span class=
"application">Xapian</span> version.</p>
</div>
</div>
<div class="sect1">
<div class="titlepage">
<div>
<div>
<h2 class="title" style="clear: both"><a name=
"RCL.INDEXING.STORAGE" id=
"RCL.INDEXING.STORAGE"></a>2.2.&nbsp;Index
storage</h2>
</div>
</div>
</div>
<p>The default location for the index data is the
<code class="filename">xapiandb</code> subdirectory of the
<span class="application">Recoll</span> configuration
directory, typically <code class=
"filename">$HOME/.recoll/xapiandb/</code>. This can be
changed via two different methods (with different
purposes):</p>
<div class="orderedlist">
<ol class="orderedlist" type="1">
<li class="listitem">
<p>For a given configuration directory, you can
specify a non-default storage location for the index
by setting the <code class="varname">dbdir</code>
parameter in the configuration file (see the
<a class="link" href="#RCL.INSTALL.CONFIG.RECOLLCONF"
title=
"5.4.2.&nbsp;Recoll main configuration file, recoll.conf">
configuration section</a>). This method would mainly
be of use if you wanted to keep the configuration
directory in its default location, but desired
another location for the index, typically out of disk
occupation or performance concerns.</p>
</li>
<li class="listitem">
<p>You can specify a different configuration
directory by setting the <code class=
"envar">RECOLL_CONFDIR</code> environment variable,
or using the <code class="option">-c</code> option to
the <span class="application">Recoll</span> commands.
This method would typically be used to index
different areas of the file system to different
indexes. For example, if you were to issue the
following command:</p>
<pre class=
"programlisting">recoll -c ~/.indexes-email</pre>
<p>Then <span class="application">Recoll</span> would
use configuration files stored in <code class=
"filename">~/.indexes-email/</code> and, (unless
specified otherwise in <code class=
"filename">recoll.conf</code>) would look for the
index in <code class=
"filename">~/.indexes-email/xapiandb/</code>.</p>
<p>Using multiple configuration directories and
<a class="link" href="#RCL.INSTALL.CONFIG.RECOLLCONF"
title=
"5.4.2.&nbsp;Recoll main configuration file, recoll.conf">
configuration options</a> allows you to tailor
multiple configurations and indexes to handle
whatever subset of the available data you wish to
make searchable.</p>
</li>
</ol>
</div>
<p>The size of the index is determined by the size of the
set of documents, but the ratio can vary a lot. For a
typical mixed set of documents, the index size will often
be close to the data set size. In specific cases (a set of
compressed mbox files for example), the index can become
much bigger than the documents. It may also be much smaller
if the documents contain a lot of images or other
non-indexed data (an extreme example being a set of mp3
files where only the tags would be indexed).</p>
<p>Of course, images, sound and video do not increase the
index size, which means that in most cases, the space used
by the index will be negligible compared to the total
amount of data on the computer.</p>
<p>The index data directory (<code class=
"filename">xapiandb</code>) only contains data that can be
completely rebuilt by an index run (as long as the original
documents exist), and it can always be destroyed
safely.</p>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.INDEXING.STORAGE.FORMAT" id=
"RCL.INDEXING.STORAGE.FORMAT"></a>2.2.1.&nbsp;<span class="application">Xapian</span>
index formats</h3>
</div>
</div>
</div>
<p><span class="application">Xapian</span> versions
usually support several formats for index storage. A
given major <span class="application">Xapian</span>
version will have a current format, used to create new
indexes, and will also support the format from the
previous major version.</p>
<p><span class="application">Xapian</span> will not
convert automatically an existing index from the older
format to the newer one. If you want to upgrade to the
new format, or if a very old index needs to be converted
because its format is not supported any more, you will
have to explicitly delete the old index (typically
<code class="filename">~/.recoll/xapiandb</code>), then
run a normal indexing command. Using <span class=
"command"><strong>recollindex</strong></span> option
<code class="option">-z</code> would not work in this
situation.</p>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.INDEXING.STORAGE.SECURITY" id=
"RCL.INDEXING.STORAGE.SECURITY"></a>2.2.2.&nbsp;Security
aspects</h3>
</div>
</div>
</div>
<p>The <span class="application">Recoll</span> index does
not hold complete copies of the indexed documents (it
almost does after version 1.24). But it does hold enough
data to allow for an almost complete reconstruction. If
confidential data is indexed, access to the database
directory should be restricted.</p>
<p><span class="application">Recoll</span> will create
the configuration directory with a mode of 0700 (access
by owner only). As the index data directory is by default
a sub-directory of the configuration directory, this
should result in appropriate protection.</p>
<p>If you use another setup, you should think of the kind
of protection you need for your index, set the directory
and files access modes appropriately, and also maybe
adjust the <code class="literal">umask</code> used during
index updates.</p>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.INDEXING.STORAGE.BIG" id=
"RCL.INDEXING.STORAGE.BIG"></a>2.2.3.&nbsp;Special
considerations for big indexes</h3>
</div>
</div>
</div>
<p>This only needs concern you if your index is going to
be bigger than around 5 GBytes. Beyond 10 GBytes, it
becomes a serious issue. Most people have much smaller
indexes. For reference, 5 GBytes would be around 2000
bibles, a lot of text. If you have a huge text dataset
(remember: images don't count, the text content of PDFs
is typically less than 5% of the file size), read on.</p>
<p>The amount of writing performed by Xapian during index
creation is not linear with the index size (it is
somewhere between linear and quadratic). For big indexes
this becomes a performance issue, and may even be an SSD
disk wear issue.</p>
<p>The problem can be mitigated by observing the
following rules:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style=
"list-style-type: disc;">
<li class="listitem">
<p>Partition the data set and create several
indexes of reasonable size rather than a huge one.
These indexes can then be queried in parallel
(using the <span class="application">Recoll</span>
external indexes facility), or merged using
<span class=
"command"><strong>xapian-compact</strong></span>.</p>
</li>
<li class="listitem">
<p>Have a lot of RAM available and set the
<code class="literal">idxflushmb</code>
<span class="application">Recoll</span>
configuration parameter as high as you can without
swapping (experimentation will be needed). 200
would be a minimum in this context.</p>
</li>
<li class="listitem">
<p>Use Xapian 1.4.10 or newer, as this version
brought a significant improvement in the amount of
writes.</p>
</li>
</ul>
</div>
</div>
</div>
<div class="sect1">
<div class="titlepage">
<div>
<div>
<h2 class="title" style="clear: both"><a name=
"RCL.INDEXING.CONFIG" id=
"RCL.INDEXING.CONFIG"></a>2.3.&nbsp;Index
configuration</h2>
</div>
</div>
</div>
<p>Variables stored inside the <a class="link" href=
"#RCL.INSTALL.CONFIG" title=
"5.4.&nbsp;Configuration overview"><span class=
"application">Recoll</span> configuration files</a> control
which areas of the file system are indexed, and how files
are processed. The values can be set by editing the text
files. Most of the more commonly used ones can also be
adjusted by using the <a class="link" href=
"#RCL.INDEXING.CONFIG.GUI" title=
"2.3.4.&nbsp;The index configuration GUI">dialogs in the
<span class="command"><strong>recoll</strong></span>
GUI</a>.</p>
<p>The first time you start <span class=
"command"><strong>recoll</strong></span>, you will be asked
whether or not you would like it to build the index. If you
want to adjust the configuration before indexing, just
click <span class="guilabel">Cancel</span> at this point,
which will get you into the configuration interface. If you
exit at this point, <code class="filename">recoll</code>
will have created a default configuration directory with
empty configuration files, which you can then edit.</p>
<p>The configuration is documented inside the <a class=
"link" href="#RCL.INSTALL.CONFIG" title=
"5.4.&nbsp;Configuration overview">installation chapter</a>
of this document, or in the <a class="ulink" href=
"https://www.lesbonscomptes.com/recoll/manpages/recoll.conf.5.html"
target="_top"><span class="citerefentry"><span class=
"refentrytitle">recoll.conf</span>(5)</span></a> manual
page. Both documents are automatically generated from the
comments inside the configuration file.</p>
<p>The most immediately useful variable is probably
<a class="link" href=
"#RCL.INSTALL.CONFIG.RECOLLCONF.TOPDIRS"><code class=
"varname">topdirs</code></a>, which lists the subtrees and
files to be indexed.</p>
<p>The applications needed to index file types other than
text, HTML or email (ie: pdf, postscript, ms-word...) are
described in the <a class="link" href=
"#RCL.INSTALL.EXTERNAL" title=
"5.2.&nbsp;Supporting packages">external packages
section</a>.</p>
<p>There are two incompatible types of Recoll indexes,
depending on the treatment of character case and
diacritics. A <a class="link" href=
"#RCL.INDEXING.CONFIG.SENS" title=
"2.3.2.&nbsp;Index case and diacritics sensitivity">further
section</a> describes the two types in more detail. The
default type is appropriate in most cases.</p>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.INDEXING.CONFIG.MULTIPLE" id=
"RCL.INDEXING.CONFIG.MULTIPLE"></a>2.3.1.&nbsp;Multiple
indexes</h3>
</div>
</div>
</div>
<p>Multiple <span class="application">Recoll</span>
indexes can be created by using several configuration
directories which are typically set to index different
areas of the file system.</p>
<p>A specific index can be selected by setting the
<code class="envar">RECOLL_CONFDIR</code> environment
variable or giving the <code class="option">-c</code>
option to <span class=
"command"><strong>recoll</strong></span> and <span class=
"command"><strong>recollindex</strong></span>.</p>
<p>The <span class=
"command"><strong>recollindex</strong></span> program,
used for creating or updating indexes, always works on a
single index. The different configurations are entirely
independent (no parameters are ever shared between
configurations when indexing).</p>
<p>All the search interfaces (<span class=
"command"><strong>recoll</strong></span>, <span class=
"command"><strong>recollq</strong></span>, the Python
API, etc.) operate with a main configuration, from which
both configuration and index data are used, and can also
query data from multiple additional indexes. Only the
index data from the latter is used, their configuration
parameters are ignored. This implies that some parameters
should be consistent among index configurations which are
to be used together.</p>
<p>When searching, the current main index (defined by
<code class="envar">RECOLL_CONFDIR</code> or <code class=
"option">-c</code>) is always active. If this is
undesirable, you can set up your base configuration to
index an empty directory.</p>
<p>Index configuration parameters can be set either by
using a text editor on the files, or, for most
parameters, by using the <a class="link" href=
"#RCL.INDEXING.CONFIG.GUI" title=
"2.3.4.&nbsp;The index configuration GUI"><span class=
"command"><strong>recoll</strong></span> index
configuration GUI</a>. In the latter case, the
configuration directory for which parameters are modified
is the one which was selected by <code class=
"envar">RECOLL_CONFDIR</code> or the <code class=
"option">-c</code> parameter, and there is no way to
switch configurations within the GUI.</p>
<p>See the <a class="link" href=
"#RCL.INSTALL.CONFIG.RECOLLCONF" title=
"5.4.2.&nbsp;Recoll main configuration file, recoll.conf">
configuration section</a> for a detailed description of
the parameters</p>
<p>Some configuration parameters must be consistent among
a set of multiple indexes used together for searches.
Most importantly, all indexes to be queried concurrently
must have the same option concerning character case and
diacritics stripping, but there are other constraints.
Most of the relevant parameters affect the <a class=
"link" href="#RCL.INSTALL.CONFIG.RECOLLCONF.TERMS" title=
"Parameters affecting how we generate terms and organize the index">
term generation</a>.</p>
<p>Using multiple configurations implies a small level of
command line or file manager usage. The user must
explicitly create additional configuration directories,
the GUI will not do it. This is to avoid mistakenly
creating additional directories when an argument is
mistyped. Also, the GUI or the indexer must be launched
with a specific option or environment to work on the
right configuration.</p>
<div class="simplesect">
<div class="titlepage">
<div>
<div>
<h4 class="title"><a name="idm415" id=
"idm415"></a>In practise: creating and using an
additional index</h4>
</div>
</div>
</div>
<p>Initially creating the configuration and index:</p>
<pre class="programlisting">
mkdir <em class=
"replaceable"><code>/path/to/my/new/config</code></em></pre>
<p>Configuring the new index can be done from the
<span class="command"><strong>recoll</strong></span>
GUI, launched from the command line to pass the
<code class="literal">-c</code> option (you could
create a desktop file to do it for you), and then using
the <a class="link" href="#RCL.INDEXING.CONFIG.GUI"
title="2.3.4.&nbsp;The index configuration GUI">GUI
index configuration tool</a> to set up the index.</p>
<pre class="programlisting">
recoll -c <em class=
"replaceable"><code>/path/to/my/new/config</code></em></pre>
<p>Alternatively, you can just start a text editor on
the main configuration file:</p>
<pre class="programlisting">
<em class="replaceable"><code>someEditor</code></em> <em class=
"replaceable"><code>/path/to/my/new/config</code></em>/<a class=
"link" href="#RCL.INSTALL.CONFIG.RECOLLCONF" title=
"5.4.2.&nbsp;Recoll main configuration file, recoll.conf"><code class="filename">recoll.conf</code></a>
</pre>
<p>Creating and updating the index can be done from the
command line:</p>
<pre class="programlisting">recollindex -c <em class=
"replaceable"><code>/path/to/my/new/config</code></em>
</pre>
<p>or from the File menu of a GUI launched with the
same option (<span class=
"command"><strong>recoll</strong></span>, see
above).</p>
<p>The same GUI would also let you set up batch
indexing for the new index. Real time indexing can only
be set up from the GUI for the default index (the menu
entry will be inactive if the GUI was started with a
non-default <code class="literal">-c</code>
option).</p>
<p>The new index can be queried alone with</p>
<pre class="programlisting">
recoll -c <em class=
"replaceable"><code>/path/to/my/new/config</code></em></pre>
<p>Or, in parallel with the default index, by starting
<span class="command"><strong>recoll</strong></span>
without a <code class="literal">-c</code> option, and
using the <span class="guimenu">Preferences</span>
<span class="guimenuitem">External Index Dialog</span>
menu.</p>
</div>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.INDEXING.CONFIG.SENS" id=
"RCL.INDEXING.CONFIG.SENS"></a>2.3.2.&nbsp;Index
case and diacritics sensitivity</h3>
</div>
</div>
</div>
<p>As of <span class="application">Recoll</span> version
1.18 you have a choice of building an index with terms
stripped of character case and diacritics, or one with
raw terms. For a source term of <code class=
"literal">Résumé</code>, the former will store
<code class="literal">resume</code>, the latter
<code class="literal">Résumé</code>.</p>
<p>Each type of index allows performing searches
insensitive to case and diacritics: with a raw index, the
user entry will be expanded to match all case and
diacritics variations present in the index. With a
stripped index, the search term will be stripped before
searching.</p>
<p>A raw index allows using case and diacritics to
discriminate between terms, e.g., returning different
results when searching for <code class=
"literal">US</code> and <code class="literal">us</code>
or <code class="literal">resume</code> and <code class=
"literal">résumé</code>. Read the <a class="link" href=
"#RCL.SEARCH.CASEDIAC" title=
"3.9.&nbsp;Search case and diacritics sensitivity">section
about search case and diacritics sensitivity</a> for more
details.</p>
<p>The type of index to be created is controlled by the
<code class="literal">indexStripChars</code>
configuration variable which can only be changed by
editing the configuration file. Any change implies an
index reset (not automated by <span class=
"application">Recoll</span>), and all indexes in a search
must be set in the same way (again, not checked by
<span class="application">Recoll</span>).</p>
<p><span class="application">Recoll</span> creates a
stripped index by default if <code class=
"literal">indexStripChars</code> is not set.</p>
<p>As a cost for added capability, a raw index will be
slightly bigger than a stripped one (around 10%). Also,
searches will be more complex, so probably slightly
slower, and the feature is relatively little used, so
that a certain amount of weirdness cannot be
excluded.</p>
<p>One of the most adverse consequence of using a raw
index is that some phrase and proximity searches may
become impossible: because each term needs to be
expanded, and all combinations searched for, the
multiplicative expansion may become unmanageable.</p>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.INDEXING.CONFIG.THREADS" id=
"RCL.INDEXING.CONFIG.THREADS"></a>2.3.3.&nbsp;Indexing
threads configuration (<span class=
"application">Unix</span>-like systems)</h3>
</div>
</div>
</div>
<p>The <span class="application">Recoll</span> indexing
process <span class=
"command"><strong>recollindex</strong></span> can use
multiple threads to speed up indexing on multiprocessor
systems. The work done to index files is divided in
several stages and some of the stages can be executed by
multiple threads. The stages are:</p>
<div class="orderedlist">
<ol class="orderedlist" type="1">
<li class="listitem">
<p>File system walking: this is always performed by
the main thread.</p>
</li>
<li class="listitem">
<p>File conversion and data extraction.</p>
</li>
<li class="listitem">
<p>Text processing (splitting, stemming, etc.).</p>
</li>
<li class="listitem">
<p><span class="application">Xapian</span> index
update.</p>
</li>
</ol>
</div>
<p>You can also read a <a class="ulink" href=
"http://www.recoll.org/pages/idxthreads/threadingRecoll.html"
target="_top">longer document</a> about the
transformation of <span class="application">Recoll</span>
indexing to multithreading.</p>
<p>The threads configuration is controlled by two
configuration file parameters.</p>
<div class="variablelist">
<dl class="variablelist">
<dt><span class="term"><code class=
"varname">thrQSizes</code></span></dt>
<dd>
<p>This variable defines the job input queues
configuration. There are three possible queues for
stages 2, 3 and 4, and this parameter should give
the queue depth for each stage (three integer
values). If a value of -1 is used for a given
stage, no queue is used, and the thread will go on
performing the next stage. In practise, deep queues
have not been shown to increase performance. A
value of 0 for the first queue tells <span class=
"application">Recoll</span> to perform
autoconfiguration (no need for anything else in
this case, thrTCounts is not used) - this is the
default configuration.</p>
</dd>
<dt><span class="term"><code class=
"varname">thrTCounts</code></span></dt>
<dd>
<p>This defines the number of threads used for each
stage. If a value of -1 is used for one of the
queue depths, the corresponding thread count is
ignored. It makes no sense to use a value other
than 1 for the last stage because updating the
<span class="application">Xapian</span> index is
necessarily single-threaded (and protected by a
mutex).</p>
</dd>
</dl>
</div>
<div class="note" style=
"margin-left: 0.5in; margin-right: 0.5in;">
<h3 class="title">Note</h3>
<p>If the first value in <code class=
"varname">thrQSizes</code> is 0, <code class=
"varname">thrTCounts</code> is ignored.</p>
</div>
<p>The following example would use three queues (of depth
2), and 4 threads for converting source documents, 2 for
processing their text, and one to update the index. This
was tested to be the best configuration on the test
system (quadri-processor with multiple disks).</p>
<pre class="programlisting">
thrQSizes = 2 2 2
thrTCounts = 4 2 1
</pre>
<p>The following example would use a single queue, and
the complete processing for each document would be
performed by a single thread (several documents will
still be processed in parallel in most cases). The
threads will use mutual exclusion when entering the index
update stage. In practise the performance would be close
to the precedent case in general, but worse in certain
cases (e.g. a Zip archive would be performed purely
sequentially), so the previous approach is preferred.
YMMV... The 2 last values for thrTCounts are ignored.</p>
<pre class="programlisting">
thrQSizes = 2 -1 -1
thrTCounts = 6 1 1
</pre>
<p>The following example would disable multithreading.
Indexing will be performed by a single thread.</p>
<pre class="programlisting">
thrQSizes = -1 -1 -1
</pre>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name="RCL.INDEXING.CONFIG.GUI"
id="RCL.INDEXING.CONFIG.GUI"></a>2.3.4.&nbsp;The
index configuration GUI</h3>
</div>
</div>
</div>
<p>Most parameters for a given index configuration can be
set from a <span class=
"command"><strong>recoll</strong></span> GUI running on
this configuration (either as default, or by setting
<code class="envar">RECOLL_CONFDIR</code> or the
<code class="option">-c</code> option.)</p>
<p>The interface is started from the <span class=
"guimenu">Preferences</span><span class=
"guimenuitem">Index Configuration</span> menu entry. It
is divided in four tabs, <span class="guilabel">Global
parameters</span>, <span class="guilabel">Local
parameters</span>, <span class="guilabel">Web
history</span> (which is explained in the next section)
and <span class="guilabel">Search parameters</span>.</p>
<p>The <span class="guilabel">Global parameters</span>
tab allows setting global variables, like the lists of
top directories, skipped paths, or stemming
languages.</p>
<p>The <span class="guilabel">Local parameters</span> tab
allows setting variables that can be redefined for
subdirectories. This second tab has an initially empty
list of customisation directories, to which you can add.
The variables are then set for the currently selected
directory (or at the top level if the empty line is
selected).</p>
<p>The <span class="guilabel">Search parameters</span>
section defines parameters which are used at query time,
but are global to an index and affect all search tools,
not only the GUI.</p>
<p>The meaning for most entries in the interface is
self-evident and documented by a <code class=
"literal">ToolTip</code> popup on the text label. For
more detail, you will need to refer to the <a class=
"link" href="#RCL.INSTALL.CONFIG" title=
"5.4.&nbsp;Configuration overview">configuration
section</a> of this guide.</p>
<p>The configuration tool normally respects the comments
and most of the formatting inside the configuration file,
so that it is quite possible to use it on hand-edited
files, which you might nevertheless want to backup
first...</p>
</div>
</div>
<div class="sect1">
<div class="titlepage">
<div>
<div>
<h2 class="title" style="clear: both"><a name=
"RCL.INDEXING.REMOVABLE" id=
"RCL.INDEXING.REMOVABLE"></a>2.4.&nbsp;Removable
volumes</h2>
</div>
</div>
</div>
<p><span class="application">Recoll</span> used to have no
support for indexing removable volumes (portable disks, USB
keys, etc.). Recent versions have improved the situation
and support indexing removable volumes in two different
ways:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style="list-style-type: disc;">
<li class="listitem">
<p>By indexing the volume in the main, fixed, index,
and ensuring that the volume data is not purged if
the indexing runs while the volume is mounted. (since
<span class="application">Recoll</span> 1.25.2).</p>
</li>
<li class="listitem">
<p>By storing a volume index on the volume itself
(since <span class="application">Recoll</span>
1.24).</p>
</li>
</ul>
</div>
<div class="simplesect">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.INDEXING.REMOVABLE.MAIN" id=
"RCL.INDEXING.REMOVABLE.MAIN"></a>Indexing
removable volumes in the main index</h3>
</div>
</div>
</div>
<p>As of version 1.25.2, <span class=
"application">Recoll</span> provides a simple way to
ensure that the index data for an absent volume will not
be purged. Two conditions must be met:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style=
"list-style-type: disc;">
<li class="listitem">
<p>The volume mount point must be a member of the
<code class="literal">topdirs</code> list.</p>
</li>
<li class="listitem">
<p>The mount directory must be empty (when the
volume is not mounted).</p>
</li>
</ul>
</div>
<p>If <span class=
"command"><strong>recollindex</strong></span> finds that
one of the <code class="literal">topdirs</code> is empty
when starting up, any existing data for the tree will be
preserved by the indexing pass (no purge for this
area).</p>
</div>
<div class="simplesect">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.INDEXING.REMOVABLE.SELF" id=
"RCL.INDEXING.REMOVABLE.SELF"></a>Self contained
volumes</h3>
</div>
</div>
</div>
<p>As of <span class="application">Recoll</span> 1.24, it
has become possible to build self-contained datasets
including a <span class="application">Recoll</span>
configuration directory and index together with the
indexed documents, and to move such a dataset around (for
example copying it to an USB drive), without having to
adjust the configuration for querying the index.</p>
<div class="note" style=
"margin-left: 0.5in; margin-right: 0.5in;">
<h3 class="title">Note</h3>
<p>This is a query-time feature only. The index must
only be updated in its original location. If an update
is necessary in a different location, the index must be
reset.</p>
</div>
<p>The principle of operation is that the configuration
stores the location of the original configuration
directory, which must reside on the movable volume. If
the volume is later mounted elsewhere, <span class=
"application">Recoll</span> adjusts the paths stored
inside the index by the difference between the original
and current locations of the configuration directory.</p>
<p>To make a long story short, here follows a script to
create a <span class="application">Recoll</span>
configuration and index under a given directory (given as
single parameter). The resulting data set (files + recoll
directory) can later to be moved to a CDROM or thumb
drive. Longer explanations come after the script.</p>
<pre class="programlisting">#!/bin/sh
fatal()
{
echo $*;exit 1
}
usage()
{
fatal "Usage: init-recoll-volume.sh &lt;top-directory&gt;"
}
test $# = 1 || usage
topdir=$1
test -d "$topdir" || fatal $topdir should be a directory
confdir="$topdir/recoll-config"
test ! -d "$confdir" || fatal $confdir should not exist
mkdir "$confdir"
cd "$topdir"
topdir=`pwd`
cd "$confdir"
confdir=`pwd`
(echo topdirs = '"'$topdir'"'; \
echo orgidxconfdir = $topdir/recoll-config) &gt; "$confdir/recoll.conf"
recollindex -c "$confdir"
</pre>
<p>The examples below will assume that you have a dataset
under <code class="filename">/home/me/mydata/</code>,
with the index configuration and data stored inside
<code class=
"filename">/home/me/mydata/recoll-confdir</code>.</p>
<p>In order to be able to run queries after the dataset
has been moved, you must ensure the following:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style=
"list-style-type: disc;">
<li class="listitem">
<p>The main configuration file must define the
<a class="link" href=
"#RCL.INSTALL.CONFIG.RECOLLCONF.ORGIDXCONFDIR">orgidxconfdir</a>
variable to be the original location of the
configuration directory (<code class=
"filename">orgidxconfdir=/home/me/mydata/recoll-confdir</code>
must be set inside <code class=
"filename">/home/me/mydata/recoll-confdir/recoll.conf</code>
in the example above).</p>
</li>
<li class="listitem">
<p>The configuration directory must exist with the
documents, somewhere under the directory which will
be moved. E.g. if you are moving <code class=
"filename">/home/me/mydata</code> around, the
configuration directory must exist somewhere below
this point, for example <code class=
"filename">/home/me/mydata/recoll-confdir</code>,
or <code class=
"filename">/home/me/mydata/sub/recoll-confdir</code>.</p>
</li>
<li class="listitem">
<p>You should keep the default locations for the
index elements which are relative to the
configuration directory by default (principally
<code class="literal">dbdir</code>). Only the paths
referring to the documents themselves (e.g.
<code class="literal">topdirs</code> values) should
be absolute (in general, they are only used when
indexing anyway).</p>
</li>
</ul>
</div>
<p>Only the first point needs an explicit user action,
the <span class="application">Recoll</span> defaults are
compatible with the third one, and the second is
natural.</p>
<p>If, after the move, the configuration directory needs
to be copied out of the dataset (for example because the
thumb drive is too slow), you can set the <a class="link"
href=
"#RCL.INSTALL.CONFIG.RECOLLCONF.CURIDXCONFDIR">curidxconfdir</a>,
variable inside the copied configuration to define the
location of the moved one. For example if <code class=
"filename">/home/me/mydata</code> is now mounted onto
<code class="filename">/media/me/somelabel</code>, but
the configuration directory and index has been copied to
<code class="filename">/tmp/tempconfig</code>, you would
set <code class="literal">curidxconfdir</code> to
<code class=
"filename">/media/me/somelabel/recoll-confdir</code>
inside <code class=
"filename">/tmp/tempconfig/recoll.conf</code>.
<code class="literal">orgidxconfdir</code> would still be
<code class=
"filename">/home/me/mydata/recoll-confdir</code> in the
original and the copy.</p>
<p>If you are regularly copying the configuration out of
the dataset, it will be useful to write a script to
automate the procedure. This can't really be done inside
<span class="application">Recoll</span> because there are
probably many possible variants. One example would be to
copy the configuration to make it writable, but keep the
index data on the medium because it is too big - in this
case, the script would also need to set <code class=
"literal">dbdir</code> in the copied configuration.</p>
<p>The same set of modifications (<span class=
"application">Recoll</span> 1.24) has also made it
possible to run queries from a readonly configuration
directory (with slightly reduced function of course, such
as not recording the query history).</p>
</div>
</div>
<div class="sect1">
<div class="titlepage">
<div>
<div>
<h2 class="title" style="clear: both"><a name=
"RCL.INDEXING.WebQUEUE" id=
"RCL.INDEXING.WebQUEUE"></a>2.5.&nbsp;<span class=
"application">Unix</span>-like systems: indexing
visited Web pages</h2>
</div>
</div>
</div>
<p>With the help of a <span class=
"application">Firefox</span> extension, <span class=
"application">Recoll</span> can index the Internet pages
that you visit. The extension has a long history: it was
initially designed for the <span class=
"application">Beagle</span> indexer, then adapted to
<span class="application">Recoll</span> and the
<span class="application">Firefox</span> <span class=
"application">XUL</span> API. The current version of the
extension is located in the <a class="ulink" href=
"https://addons.mozilla.org/en-US/firefox/addon/recoll-we/"
target="_top">Mozilla add-ons repository</a> uses the
<span class="application">WebExtensions</span> API, and
works with current <span class="application">Firefox</span>
versions.</p>
<p>The extension works by copying visited Web pages to an
indexing queue directory, which <span class=
"application">Recoll</span> then processes, storing the
data into a local cache, then indexing it, then removing
the file from the queue.</p>
<div class="note" style=
"margin-left: 0.5in; margin-right: 0.5in;">
<h3 class="title">The local cache is not an archive</h3>
<p>As mentioned above, a copy of the indexed Web pages is
retained by Recoll in a local cache (from which data is
fetched for previews, or when resetting the index). The
cache is not changed by an index reset, just read for
indexing. The cache has a maximum size, which can be
adjusted from the <span class="guilabel">Index
configuration</span> / <span class="guilabel">Web
history</span> panel (<code class=
"literal">webcachemaxmbs</code> parameter in <code class=
"filename">recoll.conf</code>). Once the maximum size is
reached, old pages are erased to make room for new ones.
The pages which you want to keep indefinitely need to be
explicitly archived elsewhere. Using a very high value
for the cache size can avoid data erasure, but see the
above 'Howto' page for more details and gotchas.</p>
</div>
<p>The visited Web pages indexing feature can be enabled on
the <span class="application">Recoll</span> side from the
GUI <span class="guilabel">Index configuration</span>
panel, or by editing the configuration file (set
<code class="varname">processwebqueue</code> to 1).</p>
<p>The <span class="application">Recoll</span> GUI has a
tool to list and edit the contents of the Web cache.
(<span class="guimenu">Tools</span><span class=
"guimenuitem">Webcache editor</span>)</p>
<p>The <span class=
"command"><strong>recollindex</strong></span> command has
two options to help manage the Web cache:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style="list-style-type: disc;">
<li class="listitem"><code class=
"option">--webcache-compact</code> will recover the
space from erased entries. It may need to use twice the
disk space currently needed for the Web cache.</li>
<li class="listitem"><code class=
"option">--webcache-burst <em class=
"replaceable"><code>destdir</code></em></code> will
extract all current entries into pairs of metadata and
data files created inside <em class=
"replaceable"><code>destdir</code></em></li>
</ul>
</div>
<p>You can find more details on Web indexing, its usage and
configuration in a <a class="ulink" href=
"https://www.lesbonscomptes.com/recoll/faqsandhowtos/IndexWebHistory"
target="_top">Recoll 'Howto' entry</a>.</p>
</div>
<div class="sect1">
<div class="titlepage">
<div>
<div>
<h2 class="title" style="clear: both"><a name=
"RCL.INDEXING.EXTATTR" id=
"RCL.INDEXING.EXTATTR"></a>2.6.&nbsp;<span class=
"application">Unix</span>-like systems: using
extended attributes</h2>
</div>
</div>
</div>
<p>User extended attributes are named pieces of information
that most modern file systems can attach to any file.</p>
<p><span class="application">Recoll</span> processes
extended attributes as document fields by default.</p>
<p>A <a class="ulink" href=
"http://www.freedesktop.org/wiki/CommonExtendedAttributes"
target="_top">freedesktop standard</a> defines a few
special attributes, which are handled as such by
<span class="application">Recoll</span>:</p>
<div class="variablelist">
<dl class="variablelist">
<dt><span class="term">mime_type</span></dt>
<dd>
<p>If set, this overrides any other determination of
the file MIME type.</p>
</dd>
<dt><span class="term">charset</span></dt>
<dd>
<p>If set, this defines the file character set
(mostly useful for plain text files).</p>
</dd>
</dl>
</div>
<p>By default, other attributes are handled as <span class=
"application">Recoll</span> fields of the same name.</p>
<p>On Linux, the <code class="literal">user</code> prefix
is removed from the name.</p>
<p>The name translation can be configured more precisely
inside the <a class="link" href=
"#RCL.INSTALL.CONFIG.FIELDS" title=
"5.4.3.&nbsp;The fields file"><code class=
"filename">fields</code> configuration file</a>.</p>
</div>
<div class="sect1">
<div class="titlepage">
<div>
<div>
<h2 class="title" style="clear: both"><a name=
"RCL.INDEXING.EXTTAGS" id=
"RCL.INDEXING.EXTTAGS"></a>2.7.&nbsp;<span class=
"application">Unix</span>-like systems: importing
external tags</h2>
</div>
</div>
</div>
<p>During indexing, it is possible to import metadata for
each file by executing commands. This allows, for example,
extracting tag data from an external application and
storing it in a field for indexing.</p>
<p>See the <a class="link" href=
"#RCL.INSTALL.CONFIG.RECOLLCONF.METADATACMDS">section about
the <code class="literal">metadatacmds</code> field</a> in
the main configuration chapter for a description of the
configuration syntax.</p>
<p>For example, if you would want <span class=
"application">Recoll</span> to use tags managed by
<span class="application">tmsu</span> in a field named
<em class="replaceable"><code>tags</code></em>, you would
add the following to the configuration file:</p>
<pre class="programlisting">[/some/area/of/the/fs]
metadatacmds = ; <em class=
"replaceable"><code>tags</code></em> = tmsu tags %f
</pre>
<div class="note" style=
"margin-left: 0.5in; margin-right: 0.5in;">
<h3 class="title">Note</h3>
<p>Depending on the <span class="application">tmsu</span>
version, you may need/want to add options like
<code class="literal">--database=/some/db</code>.</p>
</div>
<p>You may want to restrict this processing to a subset of
the directory tree, because it may slow down indexing a bit
(<code class="literal">[some/area/of/the/fs]</code>).</p>
<p>Note the initial semi-colon after the equal sign.</p>
<p>In the example above, the output of <span class=
"command"><strong>tmsu</strong></span> is used to set a
field named <em class="replaceable"><code>tags</code></em>.
The field name is arbitrary and could be <em class=
"replaceable"><code>tmsu</code></em> or <em class=
"replaceable"><code>myfield</code></em> just the same, but
<em class="replaceable"><code>tags</code></em> is an alias
for the standard <span class="application">Recoll</span>
<code class="literal">keywords</code> field, and the
<span class="command"><strong>tmsu</strong></span> output
will just augment its contents. This will avoid the need to
extend the <a class="link" href="#RCL.PROGRAM.FIELDS"
title="4.2.&nbsp;Field data processing">field
configuration</a>.</p>
<p>Once re-indexing is performed (you will need to force
the file reindexing, <span class=
"application">Recoll</span> will not detect the need by
itself), you will be able to search from the query
language, through any of its aliases: <em class=
"replaceable"><code>tags:some/alternate/values</code></em>
or <em class=
"replaceable"><code>tags:all,these,values</code></em> (the
compact field search syntax is supported for recoll 1.20
and later. For older versions, you would need to repeat the
<em class="replaceable"><code>tags:</code></em> specifier
for each term, e.g. <em class=
"replaceable"><code>tags:some</code></em> <code class=
"literal">OR</code> <em class=
"replaceable"><code>tags:alternate</code></em>).</p>
<p>Tags changes will not be detected by the indexer if the
file itself did not change. One possible workaround would
be to update the file <code class="literal">ctime</code>
when you modify the tags, which would be consistent with
how extended attributes function. A pair of <span class=
"command"><strong>chmod</strong></span> commands could
accomplish this, or a <code class="literal">touch -a</code>
. Alternatively, just couple the tag update with a
<code class="literal">recollindex -e -i</code> <em class=
"replaceable"><code>/path/to/the/file</code></em>.</p>
</div>
<div class="sect1">
<div class="titlepage">
<div>
<div>
<h2 class="title" style="clear: both"><a name=
"RCL.INDEXING.PDF" id=
"RCL.INDEXING.PDF"></a>2.8.&nbsp;The PDF input
handler</h2>
</div>
</div>
</div>
<p>The PDF format is very important for scientific and
technical documentation, and document archival. It has
extensive facilities for storing metadata along with the
document, and these facilities are actually used in the
real world.</p>
<p>In consequence, the <span class=
"command"><strong>rclpdf.py</strong></span> PDF input
handler has more complex capabilities than most others, and
it is also more configurable. Specifically, <span class=
"command"><strong>rclpdf.py</strong></span> has the
following features:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style="list-style-type: disc;">
<li class="listitem">
<p>It can be configured to extract specific metadata
tags from an XMP packet.</p>
</li>
<li class="listitem">
<p>It can extract PDF attachments.</p>
</li>
<li class="listitem">
<p>It can automatically perform OCR if the document
text is empty. This is done by executing an external
program and is now described in a <a class="link"
href="#RCL.INDEXING.OCR" title=
"2.9.&nbsp;Recoll and OCR">separate section</a>,
because the OCR framework can also be used with
non-PDF image files.</p>
</li>
</ul>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name="RCL.INDEXING.PDF.XMP"
id="RCL.INDEXING.PDF.XMP"></a>2.8.1.&nbsp;XMP
fields extraction</h3>
</div>
</div>
</div>
<p>The <code class="filename">rclpdf.py</code> script in
<span class="application">Recoll</span> version 1.23.2
and later can extract XMP metadata fields by executing
the <span class="command"><strong>pdfinfo</strong></span>
command (usually found with <span class=
"application">poppler-utils</span>). This is controlled
by the <a class="link" href=
"#RCL.INSTALL.CONFIG.RECOLLCONF.PDFEXTRAMETA">pdfextrameta</a>
configuration variable, which specifies which tags to
extract and, possibly, how to rename them.</p>
<p>The <a class="link" href=
"#RCL.INSTALL.CONFIG.RECOLLCONF.PDFEXTRAMETAFIX">pdfextrametafix</a>
variable can be used to designate a file with Python code
to edit the metadata fields (available for <span class=
"application">Recoll</span> 1.23.3 and later. 1.23.2 has
equivalent code inside the handler script). Example:</p>
<pre class="programlisting">import sys
import re
class MetaFixer(object):
def __init__(self):
pass
def metafix(self, nm, txt):
if nm == 'bibtex:pages':
txt = re.sub(r'--', '-', txt)
elif nm == 'someothername':
# do something else
pass
elif nm == 'stillanother':
# etc.
pass
return txt
def wrapup(self, metaheaders):
pass
</pre>
<p>If the 'metafix()' method is defined, it is called for
each metadata field. A new MetaFixer object is created
for each PDF document (so the object can keep state for,
for example, eliminating duplicate values). If the
'wrapup()' method is defined, it is called at the end of
XMP fields processing with the whole metadata as
parameter, as an array of '(nm, val)' pairs, allowing an
alternate approach for editing or adding/deleting
fields.</p>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name="RCL.INDEXING.PDF.ATTACH"
id="RCL.INDEXING.PDF.ATTACH"></a>2.8.2.&nbsp;PDF
attachment indexing</h3>
</div>
</div>
</div>
<p>If <span class="application">pdftk</span> is
installed, and if the the <a class="link" href=
"#RCL.INSTALL.CONFIG.RECOLLCONF.PDFATTACH">pdfattach</a>
configuration variable is set, the PDF input handler will
try to extract PDF attachments for indexing as
sub-documents of the PDF file. This is disabled by
default, because it slows down PDF indexing a bit even if
not one attachment is ever found (PDF attachments are
uncommon in my experience).</p>
</div>
</div>
<div class="sect1">
<div class="titlepage">
<div>
<div>
<h2 class="title" style="clear: both"><a name=
"RCL.INDEXING.OCR" id=
"RCL.INDEXING.OCR"></a>2.9.&nbsp;Recoll and OCR</h2>
</div>
</div>
</div>
<p>This is new in <span class="application">Recoll</span>
1.26.5. Older versions had a more limited, non-caching
capability to execute an external OCR program in the PDF
handler. The new function has the following features:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style="list-style-type: disc;">
<li class="listitem">
<p>The OCR output is cached, stored as separate
files. The caching is ultimately based on a hash
value of the original file contents, so that it is
immune to file renames. A first path-based layer
ensures fast operation for unchanged (unmoved files),
and the data hash (which is still orders of magnitude
faster than OCR) is only re-computed if the file has
moved. OCR is only performed if the file was not
previously processed or if it changed.</p>
</li>
<li class="listitem">
<p>The support for a specific program is implemented
in a simple Python module. It should be
straightforward to add support for any OCR engine
with a capability to run from the command line.</p>
</li>
<li class="listitem">
<p>Modules initially exist for <span class=
"application">tesseract</span> (Linux and Windows),
and <span class="application">ABBYY FineReader</span>
(Linux, tested with version 11). ABBYY FineReader is
a commercial closed source program, but it sometimes
perform better than tesseract.</p>
</li>
<li class="listitem">
<p>The OCR is currently only called from the PDF
handler, but there should be no problem using it for
other image types.</p>
</li>
</ul>
</div>
<p>To enable this feature, you need to install one of the
supported OCR applications (<span class=
"application">tesseract</span> or <span class=
"application">ABBYY</span>), enable OCR in the PDF handler,
and tell <span class="application">Recoll</span> where the
appropriate command resides. The last parts are done by
setting configuration variables. See the <a class="link"
href="#RCL.INSTALL.CONFIG.RECOLLCONF.OCR" title=
"Parameters for OCR processing">relevant section</a>. All
parameters can be localized in subdirectories through the
usual main configuration mechanism (path sections).</p>
</div>
<div class="sect1">
<div class="titlepage">
<div>
<div>
<h2 class="title" style="clear: both"><a name=
"RCL.INDEXING.PERIODIC" id=
"RCL.INDEXING.PERIODIC"></a>2.10.&nbsp;Periodic
indexing</h2>
</div>
</div>
</div>
<div class="simplesect">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.INDEXING.PERIODIC.EXEC" id=
"RCL.INDEXING.PERIODIC.EXEC"></a>Running the
indexer</h3>
</div>
</div>
</div>
<p>The <span class=
"command"><strong>recollindex</strong></span> program
performs index updates. You can start it either from the
command line or from the <span class=
"guimenu">File</span> menu in the <span class=
"command"><strong>recoll</strong></span> GUI program.
When started from the GUI, the indexing will run on the
same configuration <span class=
"command"><strong>recoll</strong></span> was started on.
When started from the command line, <span class=
"command"><strong>recollindex</strong></span> will use
the <code class="envar">RECOLL_CONFDIR</code> variable or
accept a <code class="option">-c</code> <em class=
"replaceable"><code>confdir</code></em> option to specify
a non-default configuration directory.</p>
<p>If the <span class=
"command"><strong>recoll</strong></span> program finds no
index when it starts, it will automatically start
indexing (except if canceled).</p>
<p>The GUI <span class="guimenu">File</span> menu has
entries to start or stop the current indexing operation.
When indexing is not currently running, you have a choice
between <span class="guimenuitem">Update Index</span> or
<span class="guimenuitem">Rebuild Index</span>. The first
choice only processes changed files, the second one
erases the index before starting so that all files are
processed.</p>
<p>On Linux and Windows, the GUI can be used to manage
the indexing operation. Stopping the indexer can be done
from the <span class=
"command"><strong>recoll</strong></span> GUI <span class=
"guimenu">File</span><span class="guimenuitem">Stop
Indexing</span> menu entry.</p>
<p>On Linux, the <span class=
"command"><strong>recollindex</strong></span> indexing
process can be interrupted by sending an interrupt
(<span class="keysym">Ctrl-C</span>, SIGINT) or terminate
(SIGTERM) signal.</p>
<p>When stopped, some time may elapse before <span class=
"command"><strong>recollindex</strong></span> exits,
because it needs to properly flush and close the
index.</p>
<p>After an interruption, the index will be somewhat
inconsistent because some operations which are normally
performed at the end of the indexing pass will have been
skipped (for example, the stemming and spelling databases
will be inexistent or out of date). You just need to
restart indexing at a later time to restore consistency.
The indexing will restart at the interruption point (the
full file tree will be traversed, but files that were
indexed up to the interruption and for which the index is
still up to date will not need to be reindexed).</p>
</div>
<div class="simplesect">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.INDEXING.PERIODIC.CMDLINE" id=
"RCL.INDEXING.PERIODIC.CMDLINE"></a>recollindex
command line</h3>
</div>
</div>
</div>
<p><span class=
"command"><strong>recollindex</strong></span> has many
options which are listed in its <a class="ulink" href=
"https://www.lesbonscomptes.com/recoll/manpages/recollindex.1.html"
target="_top">manual page</a>. Only a few will be
described here.</p>
<p>Option <code class="option">-z</code> will reset the
index when starting. This is almost the same as
destroying the index files (the nuance is that the
<span class="application">Xapian</span> format version
will not be changed).</p>
<p>Option <code class="option">-Z</code> will force the
update of all documents without resetting the index
first. This will not have the "clean start" aspect of
<code class="option">-z</code>, but the advantage is that
the index will remain available for querying while it is
rebuilt, which can be a significant advantage if it is
very big (some installations need days for a full index
rebuild).</p>
<p>Option <code class="option">-k</code> will force
retrying files which previously failed to be indexed, for
example because of a missing helper program.</p>
<p>Of special interest also, maybe, are the <code class=
"option">-i</code> and <code class="option">-f</code>
options. <code class="option">-i</code> allows indexing
an explicit list of files (given as command line
parameters or read on <code class=
"literal">stdin</code>). <code class="option">-f</code>
tells <span class=
"command"><strong>recollindex</strong></span> to ignore
file selection parameters from the configuration.
Together, these options allow building a custom file
selection process for some area of the file system, by
adding the top directory to the <code class=
"varname">skippedPaths</code> list and using an
appropriate file selection method to build the file list
to be fed to <span class=
"command"><strong>recollindex</strong></span>
<code class="option">-if</code>. Trivial example:</p>
<pre class="programlisting">
find . -name indexable.txt -print | recollindex -if
</pre>
<p><span class=
"command"><strong>recollindex</strong></span>
<code class="option">-i</code> will not descend into
subdirectories specified as parameters, but just add them
as index entries. It is up to the external file selection
method to build the complete file list.</p>
</div>
<div class="simplesect">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.INDEXING.PERIODIC.AUTOMAT" id=
"RCL.INDEXING.PERIODIC.AUTOMAT"></a>Linux: using
<span class="command"><strong>cron</strong></span>
to automate indexing</h3>
</div>
</div>
</div>
<p>The most common way to set up indexing is to have a
cron task execute it every night. For example the
following <code class="filename">crontab</code> entry
would do it every day at 3:30AM (supposing <span class=
"command"><strong>recollindex</strong></span> is in your
PATH):</p>
<pre class="screen">
30 3 * * * recollindex &gt; /some/tmp/dir/recolltrace 2&gt;&amp;1
</pre>
<p>Or, using <span class=
"command"><strong>anacron</strong></span>:</p>
<pre class="screen">
1 15 su mylogin -c "recollindex recollindex &gt; /tmp/rcltraceme 2&gt;&amp;1"
</pre>
<p>The <span class="application">Recoll</span> GUI has
dialogs to manage <code class="filename">crontab</code>
entries for <span class=
"command"><strong>recollindex</strong></span>. You can
reach them from the <span class=
"guimenu">Preferences</span><span class=
"guimenuitem">Indexing Schedule</span> menu. They only
work with the good old <span class=
"command"><strong>cron</strong></span>, and do not give
access to all features of <span class=
"command"><strong>cron</strong></span> scheduling.
Entries created via the tool are marked with a
<code class="literal">RCLCRON_RCLINDEX=</code> marker so
that the tool knows which entries belong to it. As a side
effect, this sets an environment variable for the
process, but it's not actually used, this is just a
marker.</p>
<p>The usual command to edit your <code class=
"filename">crontab</code> is <span class=
"command"><strong>crontab</strong></span> <code class=
"option">-e</code> (which will usually start the
<span class="command"><strong>vi</strong></span> editor
to edit the file). You may have more sophisticated tools
available on your system.</p>
<p>Please be aware that there may be differences between
your usual interactive command line environment and the
one seen by crontab commands. Especially the PATH
variable may be of concern. Please check the crontab
manual pages about possible issues.</p>
</div>
</div>
<div class="sect1">
<div class="titlepage">
<div>
<div>
<h2 class="title" style="clear: both"><a name=
"RCL.INDEXING.MONITOR" id=
"RCL.INDEXING.MONITOR"></a>2.11.&nbsp;<span class=
"application">Unix</span>-like systems: real time
indexing</h2>
</div>
</div>
</div>
<p>Real time monitoring/indexing is performed by starting
the <span class=
"command"><strong>recollindex</strong></span> <code class=
"option">-m</code> command. With this option, <span class=
"command"><strong>recollindex</strong></span> will detach
from the terminal and become a daemon, permanently
monitoring file changes and updating the index.</p>
<p>In this situation, the <span class=
"command"><strong>recoll</strong></span> GUI <span class=
"guimenu">File</span> menu makes two operations available:
<span class="guimenuitem">Stop</span> and <span class=
"guimenuitem">Trigger incremental pass</span>.</p>
<p><span class="guimenuitem">Trigger incremental
pass</span> has the same effect as restarting the indexer,
and will cause a complete walk of the indexed area,
processing the changed files, then switch to monitoring.
This is only marginally useful, maybe in cases where the
indexer is configured to delay updates, or to force an
immediate rebuild of the stemming and phonetic data, which
are only processed at intervals by the real time
indexer.</p>
<p>While it is convenient that data is indexed in real
time, repeated indexing can generate a significant load on
the system when files such as email folders change. Also,
monitoring large file trees by itself significantly taxes
system resources. You probably do not want to enable it if
your system is short on resources. Periodic indexing is
adequate in most cases.</p>
<p>As of <span class="application">Recoll</span> 1.24, you
can set the <a class="link" href=
"#RCL.INSTALL.CONFIG.RECOLLCONF.MONITORDIRS">monitordirs</a>
configuration variable to specify that only a subset of
your indexed files will be monitored for instant indexing.
In this situation, an incremental pass on the full tree can
be triggered by either restarting the indexer, or just
running <span class=
"command"><strong>recollindex</strong></span>, which will
notify the running process. The <span class=
"command"><strong>recoll</strong></span> GUI also has a
menu entry for this.</p>
<div class="simplesect">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.INDEXING.MONITOR.START.SYSTEMD" id=
"RCL.INDEXING.MONITOR.START.SYSTEMD"></a>Automatic
daemon start with systemd</h3>
</div>
</div>
</div>
<p>The installation contains two example files (in
<code class="filename">share/recoll/examples</code>) for
starting the indexing daemon with <span class=
"application">systemd</span>.</p>
<p><code class="filename">recollindex-user.service</code>
would be used for starting <span class=
"command"><strong>recollindex</strong></span> as a user
service, and can be installed with the following
commands:</p>
<pre class=
"programlisting">systemctl --user link /usr/share/recoll/examples/recollindex-user.service
systemctl --user enable --now recollindex-user.service</pre>
<p>The indexer will start when the user logs in and run
while there is a session open for them.</p>
<p><code class=
"filename">recollindex-system.service</code> would be
used for starting the indexer at boot time, running as a
specific user. It can be useful when running the text
search as a shared service (e.g. when users access it
through the WEB UI). You will need to edit it to replace
the @SOMEUSER@ value with something which makes sense in
your case, then install it as a regular <span class=
"application">systemd</span> system service. Of course,
if you want to run several such units, you will also need
to rename the installed file.</p>
</div>
<div class="simplesect">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.INDEXING.MONITOR.START" id=
"RCL.INDEXING.MONITOR.START"></a>Automatic daemon
start from the desktop session</h3>
</div>
</div>
</div>
<p>Under <span class="application">KDE</span>,
<span class="application">Gnome</span> and some other
desktop environments, the daemon can automatically
started when you log in, by creating a desktop file
inside the <code class=
"filename">~/.config/autostart</code> directory. This can
be done for you by the <span class=
"application">Recoll</span> GUI. Use the <span class=
"guimenu">Preferences-&gt;Indexing Schedule</span>
menu.</p>
<p>With older <span class="application">X11</span>
setups, starting the daemon is normally performed as part
of the user session script.</p>
<p>The <code class="filename">rclmon.sh</code> script can
be used to easily start and stop the daemon. It can be
found in the <code class="filename">examples</code>
directory (typically <code class=
"filename">/usr/local/[share/]recoll/examples</code>).</p>
<p>For example, a good old <span class=
"application">xdm</span>-based session could have a
<code class="filename">.xsession</code> script with the
following lines at the end:</p>
<pre class="programlisting">recollconf=$HOME/.recoll-home
recolldata=/usr/local/share/recoll
RECOLL_CONFDIR=$recollconf $recolldata/examples/rclmon.sh start
fvwm
</pre>
<p>The indexing daemon gets started, then the window
manager, for which the session waits.</p>
<p>By default the indexing daemon will monitor the state
of the X11 session, and exit when it finishes, it is not
necessary to kill it explicitly. (The <span class=
"application">X11</span> server monitoring can be
disabled with option <code class="option">-x</code> to
<span class=
"command"><strong>recollindex</strong></span>).</p>
<p>If you use the daemon completely out of an
<span class="application">X11</span> session, you need to
add option <code class="option">-x</code> to disable
<span class="application">X11</span> session monitoring
(else the daemon will not start).</p>
</div>
<div class="simplesect">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.INDEXING.MONITOR.DETAILS" id=
"RCL.INDEXING.MONITOR.DETAILS"></a>Miscellaneous
details</h3>
</div>
</div>
</div>
<p>By default, the messages from the indexing daemon will
be sent to the same file as those from the interactive
commands (<code class="literal">logfilename</code>). You
may want to change this by setting the <code class=
"varname">daemlogfilename</code> and <code class=
"varname">daemloglevel</code> configuration parameters.
Also the log file will only be truncated when the daemon
starts. If the daemon runs permanently, the log file may
grow quite big, depending on the log level.</p>
<p><b>Increasing resources for inotify.&nbsp;</b>On Linux
systems, monitoring a big tree may need increasing the
resources available to inotify, which are normally
defined in <code class=
"filename">/etc/sysctl.conf</code>.</p>
<pre class="programlisting">
### inotify
#
# cat /proc/sys/fs/inotify/max_queued_events - 16384
# cat /proc/sys/fs/inotify/max_user_instances - 128
# cat /proc/sys/fs/inotify/max_user_watches - 16384
#
# -- Change to:
#
fs.inotify.max_queued_events=32768
fs.inotify.max_user_instances=256
fs.inotify.max_user_watches=32768
</pre>
<p>Especially, you will need to trim your tree or adjust
the <code class="literal">max_user_watches</code> value
if indexing exits with a message about errno <code class=
"literal">ENOSPC</code> (28) from <code class=
"function">inotify_add_watch</code>.</p>
<p><b>Slowing down the reindexing rate for fast changing
files.&nbsp;</b>When using the real time monitor, it may
happen that some files need to be indexed, but change so
often that they impose an excessive load for the system.
<span class="application">Recoll</span> provides a
configuration option to specify the minimum time before
which a file, specified by a wildcard pattern, cannot be
reindexed. See the <code class=
"varname">mondelaypatterns</code> parameter in the
<a class="link" href=
"#RCL.INSTALL.CONFIG.RECOLLCONF.MISC" title=
"Miscellaneous parameters">configuration section</a>.</p>
</div>
</div>
</div>
<div class="chapter">
<div class="titlepage">
<div>
<div>
<h1 class="title"><a name="RCL.SEARCH" id=
"RCL.SEARCH"></a>Chapter&nbsp;3.&nbsp;Searching</h1>
</div>
</div>
</div>
<div class="sect1">
<div class="titlepage">
<div>
<div>
<h2 class="title" style="clear: both"><a name=
"RCL.SEARCH.INTRODUCTION" id=
"RCL.SEARCH.INTRODUCTION"></a>3.1.&nbsp;Introduction</h2>
</div>
</div>
</div>
<p>Getting answers to specific queries is of course the
whole point of <span class="application">Recoll</span>. The
multiple provided interfaces always understand simple
queries made of one or several words, and return
appropriate results in most cases.</p>
<p>In order to make the most of <span class=
"application">Recoll</span> though, it may be worthwhile to
understand how it processes your input. Five different
modes exist:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style="list-style-type: disc;">
<li class="listitem">
<p>In <code class="literal">All Terms</code> mode,
<span class="application">Recoll</span> looks for
documents containing all your input terms.</p>
</li>
<li class="listitem">
<p><code class="literal">Query Language</code> mode
behaves like <code class="literal">All Terms</code>
in the absence of special input, but it can also do
much more. This is the best mode for getting the most
of <span class="application">Recoll</span>.</p>
</li>
<li class="listitem">
<p>In <code class="literal">Any Term</code> mode,
<span class="application">Recoll</span> looks for
documents containing any your input terms, preferring
those which contain more.</p>
</li>
<li class="listitem">
<p>In <code class="literal">File Name</code> mode,
<span class="application">Recoll</span> will only
match file names, not content. Using a small subset
of the index allows things like left-hand wildcards
without performance issues, and may sometimes be
useful.</p>
</li>
<li class="listitem">
<p>The GUI <code class="literal">Advanced
Search</code> mode is actually not more powerful than
the query language, but it helps you build complex
queries without having to remember the language, and
avoids any interpretation ambiguity, as it bypasses
the user input parser.</p>
</li>
</ul>
</div>
<p>These five input modes are supported by the different
user interfaces which are described in the following
sections.</p>
</div>
<div class="sect1">
<div class="titlepage">
<div>
<div>
<h2 class="title" style="clear: both"><a name=
"RCL.SEARCH.GUI" id=
"RCL.SEARCH.GUI"></a>3.2.&nbsp;Searching with the Qt
graphical user interface</h2>
</div>
</div>
</div>
<p>The <span class="command"><strong>recoll</strong></span>
program provides the main user interface for searching. It
is based on the <span class="application">Qt</span>
library.</p>
<p><span class="command"><strong>recoll</strong></span> has
two search interfaces:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style="list-style-type: disc;">
<li class="listitem">
<p>Simple search (the default, on the main screen)
has a single entry field where you can enter multiple
words.</p>
</li>
<li class="listitem">
<p>Advanced search (a panel accessed through the
<span class="guilabel">Tools</span> menu or the
toolbox bar icon) has multiple entry fields, which
you may use to build a logical condition, with
additional filtering on file type, location in the
file system, modification date, and size.</p>
</li>
</ul>
</div>
<p>In most cases, you can enter the terms as you think
them, even if they contain embedded punctuation or other
non-textual characters (e.g. <span class=
"application">Recoll</span> can handle things like email
addresses).</p>
<p>The main case where you should enter text differently
from how it is printed is for east-asian languages
(Chinese, Japanese, Korean). Words composed of single or
multiple characters should be entered separated by white
space in this case (they would typically be printed without
white space).</p>
<p>Some searches can be quite complex, and you may want to
re-use them later, perhaps with some tweaking. <span class=
"application">Recoll</span> can save and restore searches.
See <a class="link" href="#RCL.SEARCH.SAVING" title=
"3.2.15.&nbsp;Saving and restoring queries (1.21 and later)">
Saving and restoring queries</a>.</p>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name="RCL.SEARCH.GUI.SIMPLE"
id="RCL.SEARCH.GUI.SIMPLE"></a>3.2.1.&nbsp;Simple
search</h3>
</div>
</div>
</div>
<div class="procedure">
<ol class="procedure" type="1">
<li class="step">
<p>Start the <span class=
"command"><strong>recoll</strong></span>
program.</p>
</li>
<li class="step">
<p>Possibly choose a search mode: <span class=
"guilabel">Any term</span>, <span class=
"guilabel">All terms</span>, <span class=
"guilabel">File name</span> or <span class=
"guilabel">Query language</span>.</p>
</li>
<li class="step">
<p>Enter search term(s) in the text field at the
top of the window.</p>
</li>
<li class="step">
<p>Click the <span class="guilabel">Search</span>
button or hit the <span class=
"keycap"><strong>Enter</strong></span> key to start
the search.</p>
</li>
</ol>
</div>
<p>The initial default search mode is <span class=
"guilabel">Query language</span>. Without special
directives, this will look for documents containing all
of the search terms (the ones with more terms will get
better scores), just like the <span class="guilabel">All
terms</span> mode. <span class="guilabel">Any term</span>
will search for documents where at least one of the terms
appear. <span class="guilabel">File name</span> will
exclusively look for file names, not contents</p>
<p>All search modes allow terms to be expanded with
wildcards characters (<code class="literal">*</code>,
<code class="literal">?</code>, <code class=
"literal">[]</code>). See the <a class="link" href=
"#RCL.SEARCH.WILDCARDS" title=
"3.6.1.&nbsp;More about wildcards">section about
wildcards</a> for more details.</p>
<p>In all modes except <span class="guilabel">File
name</span>, you can search for exact phrases (adjacent
words in a given order) by enclosing the input inside
double quotes. Ex: <code class="literal">"virtual
reality"</code>.</p>
<p>The <span class="guilabel">Query Language</span>
features are described in <a class="link" href=
"#RCL.SEARCH.LANG" title="3.5.&nbsp;The query language">a
separate section</a>.</p>
<p>When using a stripped index (the default), character
case has no influence on search, except that you can
disable stem expansion for any term by capitalizing it.
Ie: a search for <code class="literal">floor</code> will
also normally look for <code class=
"literal">flooring</code>, <code class=
"literal">floored</code>, etc., but a search for
<code class="literal">Floor</code> will only look for
<code class="literal">floor</code>, in any character
case. Stemming can also be disabled globally in the
preferences. When using a raw index, <a class="link"
href="#RCL.SEARCH.CASEDIAC" title=
"3.9.&nbsp;Search case and diacritics sensitivity">the
rules are a bit more complicated</a>.</p>
<p><span class="application">Recoll</span> remembers the
last few searches that you performed. You can directly
access the search history by clicking the clock button on
the right of the search entry, while the latter is empty.
Otherwise, the history is used for entry completion (see
next). Only the search texts are remembered, not the mode
(all/any/file name).</p>
<p>While text is entered in the search area, <span class=
"command"><strong>recoll</strong></span> will display
possible completions, filtered from the history and the
index search terms. This can be disabled with a GUI
Preferences option.</p>
<p>Double-clicking on a word in the result list or a
preview window will insert it into the simple search
entry field.</p>
<p>You can cut and paste any text into an <span class=
"guilabel">All terms</span> or <span class="guilabel">Any
term</span> search field, punctuation, newlines and all -
except for wildcard characters (single <code class=
"literal">?</code> characters are ok). <span class=
"application">Recoll</span> will process it and produce a
meaningful search. This is what most differentiates this
mode from the <span class="guilabel">Query
Language</span> mode, where you have to care about the
syntax.</p>
<p>You can use the <a class="link" href=
"#RCL.SEARCH.GUI.COMPLEX" title=
"3.2.8.&nbsp;Complex/advanced search"><span class=
"guimenu">Tools</span><span class=
"guimenuitem">Advanced search</span></a> dialog for more
complex searches.</p>
<p>The <span class="guilabel">File name</span> search
mode will specifically look for file names. The point of
having a separate file name search is that wild card
expansion can be performed more efficiently on a small
subset of the index (allowing wild cards on the left of
terms without excessive cost). Things to know:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style=
"list-style-type: disc;">
<li class="listitem">
<p>White space in the entry should match white
space in the file name, and is not treated
specially.</p>
</li>
<li class="listitem">
<p>The search is insensitive to character case and
accents, independently of the type of index.</p>
</li>
<li class="listitem">
<p>An entry without any wild card character and not
capitalized will be prepended and appended with '*'
(ie: <em class="replaceable"><code>etc</code></em>
-&gt; <em class=
"replaceable"><code>*etc*</code></em>, but
<em class="replaceable"><code>Etc</code></em> -&gt;
<em class="replaceable"><code>etc</code></em>).</p>
</li>
<li class="listitem">
<p>If you have a big index (many files),
excessively generic fragments may result in
inefficient searches.</p>
</li>
</ul>
</div>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name="RCL.SEARCH.GUI.RESLIST"
id="RCL.SEARCH.GUI.RESLIST"></a>3.2.2.&nbsp;The
result list</h3>
</div>
</div>
</div>
<p>After starting a search, a list of results will
instantly be displayed in the main window.</p>
<p>By default, the document list is presented in order of
relevance (how well the system estimates that the
document matches the query). You can sort the result by
ascending or descending date by using the vertical arrows
in the toolbar.</p>
<p>Clicking the <code class="literal">Preview</code> link
for an entry will open an internal preview window for the
document. Further <code class="literal">Preview</code>
clicks for the same search will open tabs in the existing
preview window. You can use <span class=
"keycap"><strong>Shift</strong></span>+Click to force the
creation of another preview window, which may be useful
to view the documents side by side. (You can also browse
successive results in a single preview window by typing
<span class=
"keycap"><strong>Shift</strong></span>+<span class=
"keycap"><strong>ArrowUp/Down</strong></span> in the
window).</p>
<p>Clicking the <code class="literal">Open</code> link
will start an external viewer for the document. By
default, <span class="application">Recoll</span> lets the
desktop choose the appropriate application for most
document types. See <a class="link" href=
"#RCL.SEARCH.GUI.RESLIST.APPLICATIONS" title=
"Customising the applications">further</a> for
customizing the applications.</p>
<p>You can click on the <code class="literal">Query
details</code> link at the top of the results page to see
the query actually performed, after stem expansion and
other processing.</p>
<p>Double-clicking on any word inside the result list or
a preview window will insert it into the simple search
text.</p>
<p>The result list is divided into pages (the size of
which you can change in the preferences). Use the arrow
buttons in the toolbar or the links at the bottom of the
page to browse the results.</p>
<p>The <code class="literal">Preview</code> and
<code class="literal">Open</code> edit links may not be
present for all entries, meaning that <span class=
"application">Recoll</span> has no configured way to
preview a given file type (which was indexed by name
only), or no configured external editor for the file
type. This can sometimes be adjusted simply by tweaking
the <a class="link" href="#RCL.INSTALL.CONFIG.MIMEMAP"
title="5.4.4.&nbsp;The mimemap file"><code class=
"filename">mimemap</code></a> and <a class="link" href=
"#RCL.INSTALL.CONFIG.MIMEVIEW" title=
"5.4.6.&nbsp;The mimeview file"><code class=
"filename">mimeview</code></a> configuration files (the
latter can be modified with the user preferences
dialog).</p>
<p>The format of the result list entries is entirely
configurable by using the preference dialog to <a class=
"link" href="#RCL.SEARCH.GUI.CUSTOM.RESLIST" title=
"The result list format">edit an HTML fragment</a>.</p>
<div class="simplesect">
<div class="titlepage">
<div>
<div>
<h4 class="title"><a name=
"RCL.SEARCH.GUI.RESLIST.APPLICATIONS" id=
"RCL.SEARCH.GUI.RESLIST.APPLICATIONS"></a>Customising
the applications</h4>
</div>
</div>
</div>
<p>By default <span class="application">Recoll</span>
lets the desktop choose what application should be used
to open a given document, with exceptions.</p>
<p>The details of this behaviour can be customized with
the <span class="guimenu">Preferences</span>
<span class="guimenuitem">GUI configuration</span>
<span class="guimenuitem">User interface</span>
<span class="guimenuitem">Choose editor
applications</span> dialog or by editing the <a class=
"link" href="#RCL.INSTALL.CONFIG.MIMEVIEW" title=
"5.4.6.&nbsp;The mimeview file"><code class=
"filename">mimeview</code> configuration file.</a></p>
<p>When <span class="guilabel">Use desktop
preferences</span>, at the top of the dialog, is
checked, the desktop default is generally used, but
there is a small default list of exceptions, for MIME
types where the <span class="application">Recoll</span>
choice should override the desktop one. These are
applications which are well integrated with
<span class="application">Recoll</span>, for example,
on Linux, <span class="application">evince</span> for
viewing PDF and Postscript files because of its support
for opening the document at a specific page and passing
a search string as an argument. You can add or remove
document types to the exceptions by using the
dialog.</p>
<p>If you prefer to completely customize the choice of
applications, you can uncheck <span class=
"guilabel">Use desktop preferences</span>, in which
case the <span class="application">Recoll</span>
predefined applications will be used, and can be
changed for each document type. This is probably not
the most convenient approach in most cases.</p>
<p>In all cases, the applications choice dialog accepts
multiple selections of MIME types in the top section,
and lets you define how they are processed in the
bottom one. In most cases, you will be using
<code class="literal">%f</code> as a place holder to be
replaced by the file name in the application command
line.</p>
<p>You may also change the choice of applications by
editing the <a class="link" href=
"#RCL.INSTALL.CONFIG.MIMEVIEW" title=
"5.4.6.&nbsp;The mimeview file"><code class=
"filename">mimeview</code></a> configuration file if
you find this more convenient.</p>
<p>Under <span class="application">Unix</span>-like
systems, each result list entry also has a right-click
menu with an <span class="guilabel">Open With</span>
entry. This lets you choose an application from the
list of those which registered with the desktop for the
document MIME type, on a case by case basis.</p>
</div>
<div class="sect3">
<div class="titlepage">
<div>
<div>
<h4 class="title"><a name=
"RCL.SEARCH.GUI.RESLIST.SUGGS" id=
"RCL.SEARCH.GUI.RESLIST.SUGGS"></a>No results:
the spelling suggestions</h4>
</div>
</div>
</div>
<p>When a search yields no result, and if the
<span class="application">aspell</span> dictionary is
configured, <span class="application">Recoll</span>
will try to check for misspellings among the query
terms, and will propose lists of replacements. Clicking
on one of the suggestions will replace the word and
restart the search. You can hold any of the modifier
keys (Ctrl, Shift, etc.) while clicking if you would
rather stay on the suggestion screen because several
terms need replacement.</p>
</div>
<div class="sect3">
<div class="titlepage">
<div>
<div>
<h4 class="title"><a name=
"RCL.SEARCH.GUI.RESULTLIST.MENU" id=
"RCL.SEARCH.GUI.RESULTLIST.MENU"></a>The result
list right-click menu</h4>
</div>
</div>
</div>
<p>Apart from the preview and edit links, you can
display a pop-up menu by right-clicking over a
paragraph in the result list. This menu has the
following entries:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style=
"list-style-type: disc;">
<li class="listitem">
<p><span class="guilabel">Preview</span></p>
</li>
<li class="listitem">
<p><span class="guilabel">Open</span></p>
</li>
<li class="listitem">
<p><span class="guilabel">Open With</span></p>
</li>
<li class="listitem">
<p><span class="guilabel">Run Script</span></p>
</li>
<li class="listitem">
<p><span class="guilabel">Copy File
Name</span></p>
</li>
<li class="listitem">
<p><span class="guilabel">Copy Url</span></p>
</li>
<li class="listitem">
<p><span class="guilabel">Save to File</span></p>
</li>
<li class="listitem">
<p><span class="guilabel">Find similar</span></p>
</li>
<li class="listitem">
<p><span class="guilabel">Preview Parent
document</span></p>
</li>
<li class="listitem">
<p><span class="guilabel">Open Parent
document</span></p>
</li>
<li class="listitem">
<p><span class="guilabel">Open Snippets
Window</span></p>
</li>
</ul>
</div>
<p>The <span class="guilabel">Preview</span> and
<span class="guilabel">Open</span> entries do the same
thing as the corresponding links.</p>
<p><span class="guilabel">Open With</span>
(<span class="application">Unix</span>-like systems)
lets you open the document with one of the applications
claiming to be able to handle its MIME type (the
information comes from the <code class=
"literal">.desktop</code> files in <code class=
"filename">/usr/share/applications</code>).</p>
<p><span class="guilabel">Run Script</span>
(<span class="application">Unix</span>-like systems)
allows starting an arbitrary command on the result
file. It will only appear for results which are
top-level files. See <a class="link" href=
"#RCL.SEARCH.GUI.RUNSCRIPT" title=
"3.2.4.&nbsp;Unix-like systems: running arbitrary commands on result files">
further</a> for a more detailed description.</p>
<p>The <span class="guilabel">Copy File Name</span> and
<span class="guilabel">Copy Url</span> copy the
relevant data to the clipboard, for later pasting.</p>
<p><span class="guilabel">Save to File</span> allows
saving the contents of a result document to a chosen
file. This entry will only appear if the document does
not correspond to an existing file, but is a
subdocument inside such a file (ie: an email
attachment). It is especially useful to extract
attachments with no associated editor.</p>
<p>The <span class="guilabel">Open/Preview Parent
document</span> entries allow working with the higher
level document (e.g. the email message an attachment
comes from). <span class="application">Recoll</span> is
sometimes not totally accurate as to what it can or
can't do in this area. For example the <span class=
"guilabel">Parent</span> entry will also appear for an
email which is part of an mbox folder file, but you
can't actually visualize the mbox (there will be an
error dialog if you try).</p>
<p>If the document is a top-level file, <span class=
"guilabel">Open Parent</span> will start the default
file manager on the enclosing filesystem directory.</p>
<p>The <span class="guilabel">Find similar</span> entry
will select a number of relevant term from the current
document and enter them into the simple search field.
You can then start a simple search, with a good chance
of finding documents related to the current result. I
can't remember a single instance where this function
was actually useful to me...</p>
<p><a name="RCL.SEARCH.GUI.RESULTLIST.MENU.SNIPPETS"
id="RCL.SEARCH.GUI.RESULTLIST.MENU.SNIPPETS"></a>The
<span class="guilabel">Open Snippets Window</span>
entry will only appear for documents which support page
breaks (typically PDF, Postscript, DVI). The snippets
window lists extracts from the document, taken around
search terms occurrences, along with the corresponding
page number, as links which can be used to start the
native viewer on the appropriate page. If the viewer
supports it, its search function will also be primed
with one of the search terms.</p>
</div>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name="RCL.SEARCH.GUI.RESTABLE"
id="RCL.SEARCH.GUI.RESTABLE"></a>3.2.3.&nbsp;The
result table</h3>
</div>
</div>
</div>
<p>As an alternative to the result list, the results can
also be displayed in spreadsheet-like fashion. You can
switch to this presentation by clicking the table-like
icon in the toolbar (this is a toggle, click again to
restore the list).</p>
<p>Clicking on the column headers will allow sorting by
the values in the column. You can click again to invert
the order, and use the header right-click menu to reset
sorting to the default relevance order (you can also use
the sort-by-date arrows to do this).</p>
<p>Both the list and the table display the same
underlying results. The sort order set from the table is
still active if you switch back to the list mode. You can
click twice on a date sort arrow to reset it from
there.</p>
<p>The header right-click menu allows adding or deleting
columns. The columns can be resized, and their order can
be changed (by dragging). All the changes are recorded
when you quit <span class=
"command"><strong>recoll</strong></span></p>
<p>Hovering over a table row will update the detail area
at the bottom of the window with the corresponding
values. You can click the row to freeze the display. The
bottom area is equivalent to a result list paragraph,
with links for starting a preview or a native
application, and an equivalent right-click menu. Typing
<span class="keycap"><strong>Esc</strong></span> (the
Escape key) will unfreeze the display.</p>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.SEARCH.GUI.RUNSCRIPT" id=
"RCL.SEARCH.GUI.RUNSCRIPT"></a>3.2.4.&nbsp;<span class="application">Unix</span>-like
systems: running arbitrary commands on result
files</h3>
</div>
</div>
</div>
<p>Apart from the <span class="guilabel">Open</span> and
<span class="guilabel">Open With</span> operations, which
allow starting an application on a result document (or a
temporary copy), based on its MIME type, it is also
possible to run arbitrary commands on results which are
top-level files, using the <span class="guilabel">Run
Script</span> entry in the results pop-up menu.</p>
<p>The commands which will appear in the <span class=
"guilabel">Run Script</span> submenu must be defined by
<code class="literal">.desktop</code> files inside the
<code class="filename">scripts</code> subdirectory of the
current configuration directory.</p>
<p>Here follows an example of a <code class=
"literal">.desktop</code> file, which could be named for
example, <code class=
"filename">~/.recoll/scripts/myscript.desktop</code> (the
exact file name inside the directory is irrelevant):</p>
<pre class="programlisting">
[Desktop Entry]
Type=Application
Name=MyFirstScript
Exec=/home/me/bin/tryscript %F
MimeType=*/*
</pre>
<p>The <code class="literal">Name</code> attribute
defines the label which will appear inside the
<span class="guilabel">Run Script</span> menu. The
<code class="literal">Exec</code> attribute defines the
program to be run, which does not need to actually be a
script, of course. The <code class=
"literal">MimeType</code> attribute is not used, but
needs to exist.</p>
<p>The commands defined this way can also be used from
links inside the <a class="link" href=
"#RCL.SEARCH.GUI.CUSTOM.RESLIST.PARA" title=
"The paragraph format">result paragraph</a>.</p>
<p>As an example, it might make sense to write a script
which would move the document to the trash and purge it
from the <span class="application">Recoll</span>
index.</p>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.SEARCH.GUI.THUMBNAILS" id=
"RCL.SEARCH.GUI.THUMBNAILS"></a>3.2.5.&nbsp;<span class="application">Unix</span>-like
systems: displaying thumbnails</h3>
</div>
</div>
</div>
<p>The default format for the result list entries and the
detail area of the result table display an icon for each
result document. The icon is either a generic one
determined from the MIME type, or a thumbnail of the
document appearance. Thumbnails are only displayed if
found in the standard <span class=
"application">freedesktop</span> location, where they
would typically have been created by a file manager.</p>
<p>Recoll has no capability to create thumbnails. A
relatively simple trick is to use the <span class=
"guilabel">Open parent document/folder</span> entry in
the result list popup menu. This should open a file
manager window on the containing directory, which should
in turn create the thumbnails (depending on your
settings). Restarting the search should then display the
thumbnails.</p>
<p>There are also <a class="ulink" href=
"https://www.lesbonscomptes.com/recoll/faqsandhowtos/ResultsThumbnails.html"
target="_top">some pointers about thumbnail
generation</a> in the <span class=
"application">Recoll</span> FAQ.</p>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name="RCL.SEARCH.GUI.PREVIEW"
id="RCL.SEARCH.GUI.PREVIEW"></a>3.2.6.&nbsp;The
preview window</h3>
</div>
</div>
</div>
<p>The preview window opens when you first click a
<code class="literal">Preview</code> link inside the
result list.</p>
<p>Subsequent preview requests for a given search open
new tabs in the existing window (except if you hold the
<span class="keycap"><strong>Shift</strong></span> key
while clicking which will open a new window for side by
side viewing).</p>
<p>Starting another search and requesting a preview will
create a new preview window. The old one stays open until
you close it.</p>
<p>You can close a preview tab by typing <span class=
"keycap"><strong>Ctrl-W</strong></span> (<span class=
"keycap"><strong>Ctrl</strong></span> + <span class=
"keycap"><strong>W</strong></span>) in the window.
Closing the last tab, or using the window manager button
in the top of the frame will also close the window.</p>
<p>You can display successive or previous documents from
the result list inside a preview tab by typing
<span class=
"keycap"><strong>Shift</strong></span>+<span class=
"keycap"><strong>Down</strong></span> or <span class=
"keycap"><strong>Shift</strong></span>+<span class=
"keycap"><strong>Up</strong></span> (<span class=
"keycap"><strong>Down</strong></span> and <span class=
"keycap"><strong>Up</strong></span> are the arrow
keys).</p>
<p>A right-click menu in the text area allows switching
between displaying the main text or the contents of
fields associated to the document (ie: author, abtract,
etc.). This is especially useful in cases where the term
match did not occur in the main text but in one of the
fields. In the case of images, you can switch between
three displays: the image itself, the image metadata as
extracted by <span class=
"command"><strong>exiftool</strong></span> and the
fields, which is the metadata stored in the index.</p>
<p>You can print the current preview window contents by
typing <span class=
"keycap"><strong>Ctrl-P</strong></span> (<span class=
"keycap"><strong>Ctrl</strong></span> + <span class=
"keycap"><strong>P</strong></span>) in the window
text.</p>
<div class="sect3">
<div class="titlepage">
<div>
<div>
<h4 class="title"><a name=
"RCL.SEARCH.GUI.PREVIEW.SEARCH" id=
"RCL.SEARCH.GUI.PREVIEW.SEARCH"></a>Searching
inside the preview</h4>
</div>
</div>
</div>
<p>The preview window has an internal search
capability, mostly controlled by the panel at the
bottom of the window, which works in two modes: as a
classical editor incremental search, where we look for
the text entered in the entry zone, or as a way to walk
the matches between the document and the <span class=
"application">Recoll</span> query that found it.</p>
<div class="variablelist">
<dl class="variablelist">
<dt><span class="term">Incremental text
search</span></dt>
<dd>
<p>The preview tabs have an internal incremental
search function. You initiate the search either
by typing a <span class=
"keycap"><strong>/</strong></span> (slash) or
<span class=
"keycap"><strong>CTL-F</strong></span> inside the
text area or by clicking into the <span class=
"guilabel">Search for:</span> text field and
entering the search string. You can then use the
<span class="guilabel">Next</span> and
<span class="guilabel">Previous</span> buttons to
find the next/previous occurrence. You can also
type <span class=
"keycap"><strong>F3</strong></span> inside the
text area to get to the next occurrence.</p>
<p>If you have a search string entered and you
use Ctrl-Up/Ctrl-Down to browse the results, the
search is initiated for each successive document.
If the string is found, the cursor will be
positioned at the first occurrence of the search
string.</p>
</dd>
<dt><span class="term">Walking the match
lists</span></dt>
<dd>
<p>If the entry area is empty when you click the
<span class="guilabel">Next</span> or
<span class="guilabel">Previous</span> buttons,
the editor will be scrolled to show the next
match to any search term (the next highlighted
zone). If you select a search group from the
dropdown list and click <span class=
"guilabel">Next</span> or <span class=
"guilabel">Previous</span>, the match list for
this group will be walked. This is not the same
as a text search, because the occurrences will
include non-exact matches (as caused by stemming
or wildcards). The search will revert to the text
mode as soon as you edit the entry area.</p>
</dd>
</dl>
</div>
</div>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name="RCL.SEARCH.GUI.FRAGBUTS"
id="RCL.SEARCH.GUI.FRAGBUTS"></a>3.2.7.&nbsp;The
Query Fragments window</h3>
</div>
</div>
</div>
<p>Selecting the <span class="guimenu">Tools</span>
<span class="guimenuitem">Query Fragments</span> menu
entry will open a window with radio- and check-buttons
which can be used to activate query language fragments
for filtering the current query. This can be useful if
you have frequent reusable selectors, for example,
filtering on alternate directories, or searching just one
category of files, not covered by the standard category
selectors.</p>
<p>The contents of the window are entirely customizable,
and defined by the contents of the <code class=
"filename">fragbuts.xml</code> file inside the
configuration directory. The sample file distributed with
<span class="application">Recoll</span> (which you should
be able to find under <code class=
"filename">/usr/share/recoll/examples/fragbuts.xml</code>),
contains an example which filters the results from the
Web history.</p>
<p>Here follows an example:</p>
<pre class="programlisting">
&lt;?xml version="1.0" encoding="UTF-8"?&gt;
&lt;fragbuts version="1.0"&gt;
&lt;radiobuttons&gt;
&lt;!-- Actually useful: toggle Web queue results inclusion --&gt;
&lt;fragbut&gt;
&lt;label&gt;Include Web Results&lt;/label&gt;
&lt;frag&gt;&lt;/frag&gt;
&lt;/fragbut&gt;
&lt;fragbut&gt;
&lt;label&gt;Exclude Web Results&lt;/label&gt;
&lt;frag&gt;-rclbes:BGL&lt;/frag&gt;
&lt;/fragbut&gt;
&lt;fragbut&gt;
&lt;label&gt;Only Web Results&lt;/label&gt;
&lt;frag&gt;rclbes:BGL&lt;/frag&gt;
&lt;/fragbut&gt;
&lt;/radiobuttons&gt;
&lt;buttons&gt;
&lt;fragbut&gt;
&lt;label&gt;Example: Year 2010&lt;/label&gt;
&lt;frag&gt;date:2010-01-01/2010-12-31&lt;/frag&gt;
&lt;/fragbut&gt;
&lt;fragbut&gt;
&lt;label&gt;Example: c++ files&lt;/label&gt;
&lt;frag&gt;ext:cpp OR ext:cxx&lt;/frag&gt;
&lt;/fragbut&gt;
&lt;fragbut&gt;
&lt;label&gt;Example: My Great Directory&lt;/label&gt;
&lt;frag&gt;dir:/my/great/directory&lt;/frag&gt;
&lt;/fragbut&gt;
&lt;/buttons&gt;
&lt;/fragbuts&gt;
</pre>
<p>Each <code class="literal">radiobuttons</code> or
<code class="literal">buttons</code> section defines a
line of checkbuttons or radiobuttons inside the window.
Any number of buttons can be selected, but the
radiobuttons in a line are exclusive.</p>
<p>Each <code class="literal">fragbut</code> section
defines the label for a button, and the Query Language
fragment which will be added (as an AND filter) before
performing the query if the button is active.</p>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name="RCL.SEARCH.GUI.COMPLEX"
id=
"RCL.SEARCH.GUI.COMPLEX"></a>3.2.8.&nbsp;Complex/advanced
search</h3>
</div>
</div>
</div>
<p>The advanced search dialog helps you build more
complex queries without memorizing the search language
constructs. It can be opened through the <span class=
"guilabel">Tools</span> menu or through the main
toolbar.</p>
<p><span class="application">Recoll</span> keeps a
history of searches. See <a class="link" href=
"#RCL.SEARCH.GUI.COMPLEX.HISTORY" title=
"Advanced search history">Advanced search
history</a>.</p>
<p>The dialog has two tabs:</p>
<div class="orderedlist">
<ol class="orderedlist" type="1">
<li class="listitem">
<p>The first tab lets you specify terms to search
for, and permits specifying multiple clauses which
are combined to build the search.</p>
</li>
<li class="listitem">
<p>The second tab lets filter the results according
to file size, date of modification, MIME type, or
location.</p>
</li>
</ol>
</div>
<p>Click on the <span class="guilabel">Start
Search</span> button in the advanced search dialog, or
type <span class="keycap"><strong>Enter</strong></span>
in any text field to start the search. The button in the
main window always performs a simple search.</p>
<p>Click on the <code class="literal">Show query
details</code> link at the top of the result page to see
the query expansion.</p>
<div class="sect3">
<div class="titlepage">
<div>
<div>
<h4 class="title"><a name=
"RCL.SEARCH.GUI.COMPLEX.TERMS" id=
"RCL.SEARCH.GUI.COMPLEX.TERMS"></a>Advanced
search: the "find" tab</h4>
</div>
</div>
</div>
<p>This part of the dialog lets you constructc a query
by combining multiple clauses of different types. Each
entry field is configurable for the following
modes:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style=
"list-style-type: disc;">
<li class="listitem">
<p>All terms.</p>
</li>
<li class="listitem">
<p>Any term.</p>
</li>
<li class="listitem">
<p>None of the terms.</p>
</li>
<li class="listitem">
<p>Phrase (exact terms in order within an
adjustable window).</p>
</li>
<li class="listitem">
<p>Proximity (terms in any order within an
adjustable window).</p>
</li>
<li class="listitem">
<p>Filename search.</p>
</li>
</ul>
</div>
<p>Additional entry fields can be created by clicking
the <span class="guilabel">Add clause</span>
button.</p>
<p>When searching, the non-empty clauses will be
combined either with an AND or an OR conjunction,
depending on the choice made on the left (<span class=
"guilabel">All clauses</span> or <span class=
"guilabel">Any clause</span>).</p>
<p>Entries of all types except "Phrase" and "Near"
accept a mix of single words and phrases enclosed in
double quotes. Stemming and wildcard expansion will be
performed as for simple search.</p>
<p><b>Phrases and Proximity searches.&nbsp;</b>These
two clauses work in similar ways, with the difference
that proximity searches do not impose an order on the
words. In both cases, an adjustable number (slack) of
non-matched words may be accepted between the searched
ones (use the counter on the left to adjust this
count). For phrases, the default count is zero (exact
match). For proximity it is ten (meaning that two
search terms, would be matched if found within a window
of twelve words). Examples: a phrase search for
<code class="literal">quick fox</code> with a slack of
0 will match <code class="literal">quick fox</code> but
not <code class="literal">quick brown fox</code>. With
a slack of 1 it will match the latter, but not
<code class="literal">fox quick</code>. A proximity
search for <code class="literal">quick fox</code> with
the default slack will match the latter, and also
<code class="literal">a fox is a cunning and quick
animal</code>.</p>
</div>
<div class="sect3">
<div class="titlepage">
<div>
<div>
<h4 class="title"><a name=
"RCL.SEARCH.GUI.COMPLEX.FILTER" id=
"RCL.SEARCH.GUI.COMPLEX.FILTER"></a>Advanced
search: the "filter" tab</h4>
</div>
</div>
</div>
<p>This part of the dialog has several sections which
allow filtering the results of a search according to a
number of criteria</p>
<div class="itemizedlist">
<ul class="itemizedlist" style=
"list-style-type: disc;">
<li class="listitem">
<p>The first section allows filtering by dates of
last modification. You can specify both a minimum
and a maximum date. The initial values are set
according to the oldest and newest documents
found in the index.</p>
</li>
<li class="listitem">
<p>The next section allows filtering the results
by file size. There are two entries for minimum
and maximum size. Enter decimal numbers. You can
use suffix multipliers: <code class=
"literal">k/K</code>, <code class=
"literal">m/M</code>, <code class=
"literal">g/G</code>, <code class=
"literal">t/T</code> for 1E3, 1E6, 1E9, 1E12
respectively.</p>
</li>
<li class="listitem">
<p>The next section allows filtering the results
by their MIME types, or MIME categories (ie:
media/text/message/etc.).</p>
<p>You can transfer the types between two boxes,
to define which will be included or excluded by
the search.</p>
<p>The state of the file type selection can be
saved as the default (the file type filter will
not be activated at program start-up, but the
lists will be in the restored state).</p>
</li>
<li class="listitem">
<p>The bottom section allows restricting the
search results to a sub-tree of the indexed area.
You can use the <span class=
"guilabel">Invert</span> checkbox to search for
files not in the sub-tree instead. If you use
directory filtering often and on big subsets of
the file system, you may think of setting up
multiple indexes instead, as the performance may
be better.</p>
<p>You can use relative/partial paths for
filtering. Ie, entering <code class=
"literal">dirA/dirB</code> would match either
<code class=
"filename">/dir1/dirA/dirB/myfile1</code> or
<code class=
"filename">/dir2/dirA/dirB/someother/myfile2</code>.</p>
</li>
</ul>
</div>
</div>
<div class="sect3">
<div class="titlepage">
<div>
<div>
<h4 class="title"><a name=
"RCL.SEARCH.GUI.COMPLEX.HISTORY" id=
"RCL.SEARCH.GUI.COMPLEX.HISTORY"></a>Advanced
search history</h4>
</div>
</div>
</div>
<p>The advanced search tool memorizes the last 100
searches performed. You can walk the saved searches by
using the up and down arrow keys while the keyboard
focus belongs to the advanced search dialog.</p>
<p>The complex search history can be erased, along with
the one for simple search, by selecting the
<span class="guimenu">File</span><span class=
"guimenuitem">Erase Search History</span> menu
entry.</p>
</div>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.SEARCH.GUI.TERMEXPLORER" id=
"RCL.SEARCH.GUI.TERMEXPLORER"></a>3.2.9.&nbsp;The
term explorer tool</h3>
</div>
</div>
</div>
<p><span class="application">Recoll</span> automatically
manages the expansion of search terms to their
derivatives (ie: plural/singular, verb inflections). But
there are other cases where the exact search term is not
known. For example, you may not remember the exact
spelling, or only know the beginning of the name.</p>
<p>The search will only propose replacement terms with
spelling variations when no matching document were found.
In some cases, both proper spellings and mispellings are
present in the index, and it may be interesting to look
for them explicitly.</p>
<p>The term explorer tool (started from the toolbar icon
or from the <span class="guilabel">Term explorer</span>
entry of the <span class="guilabel">Tools</span> menu)
can be used to search the full index terms list. It has
three modes of operations:</p>
<div class="variablelist">
<dl class="variablelist">
<dt><span class="term">Wildcard</span></dt>
<dd>
<p>In this mode of operation, you can enter a
search string with shell-like wildcards (*, ?, []).
ie: <em class="replaceable"><code>xapi*</code></em>
would display all index terms beginning with
<em class="replaceable"><code>xapi</code></em>.
(More about wildcards <a class="link" href=
"#RCL.SEARCH.WILDCARDS" title=
"3.6.1.&nbsp;More about wildcards">here</a> ).</p>
</dd>
<dt><span class="term">Regular expression</span></dt>
<dd>
<p>This mode will accept a regular expression as
input. Example: <em class=
"replaceable"><code>word[0-9]+</code></em>. The
expression is implicitly anchored at the beginning.
Ie: <em class="replaceable"><code>press</code></em>
will match <em class=
"replaceable"><code>pression</code></em> but not
<em class=
"replaceable"><code>expression</code></em>. You can
use <em class=
"replaceable"><code>.*press</code></em> to match
the latter, but be aware that this will cause a
full index term list scan, which can be quite
long.</p>
</dd>
<dt><span class="term">Stem expansion</span></dt>
<dd>
<p>This mode will perform the usual stem expansion
normally done as part user input processing. As
such it is probably mostly useful to demonstrate
the process.</p>
</dd>
<dt><span class="term">Spelling/Phonetic</span></dt>
<dd>
<p>In this mode, you enter the term as you think it
is spelled, and <span class=
"application">Recoll</span> will do its best to
find index terms that sound like your entry. This
mode uses the <span class=
"application">Aspell</span> spelling application,
which must be installed on your system for things
to work (if your documents contain non-ascii
characters, <span class="application">Recoll</span>
needs an aspell version newer than 0.60 for UTF-8
support). The language which is used to build the
dictionary out of the index terms (which is done at
the end of an indexing pass) is the one defined by
your NLS environment. Weird things will probably
happen if languages are mixed up.</p>
</dd>
<dt><span class="term">Show index
statistics</span></dt>
<dd>
<p>This will print a long list of boring numbers
about the index</p>
</dd>
<dt><span class="term">List files which could not be
indexed</span></dt>
<dd>
<p>This will show the files which caused errors,
usually because <span class=
"command"><strong>recollindex</strong></span> could
not translate their format into text.</p>
</dd>
</dl>
</div>
<p>Note that in cases where <span class=
"application">Recoll</span> does not know the beginning
of the string to search for (ie a wildcard expression
like <em class="replaceable"><code>*coll</code></em>),
the expansion can take quite a long time because the full
index term list will have to be processed. The expansion
is currently limited at 10000 results for wildcards and
regular expressions. It is possible to change the limit
in the configuration file.</p>
<p>Double-clicking on a term in the result list will
insert it into the simple search entry field. You can
also cut/paste between the result list and any entry
field (the end of lines will be taken care of).</p>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name="RCL.SEARCH.GUI.MULTIDB"
id=
"RCL.SEARCH.GUI.MULTIDB"></a>3.2.10.&nbsp;Multiple
indexes</h3>
</div>
</div>
</div>
<p>See the section describing <a class="link" href=
"#RCL.INDEXING.CONFIG.MULTIPLE" title=
"2.3.1.&nbsp;Multiple indexes">the use of multiple
indexes</a> for generalities. Only the aspects concerning
the <span class="command"><strong>recoll</strong></span>
GUI are described here.</p>
<p>A <span class="command"><strong>recoll</strong></span>
program instance is always associated with a specific
index, which is the one to be updated when requested from
the <span class="guimenu">File</span> menu, but it can
use any number of <span class="application">Recoll</span>
indexes for searching. The external indexes can be
selected through the <span class="guilabel">external
indexes</span> tab in the preferences dialog.</p>
<p>Index selection is performed in two phases. A set of
all usable indexes must first be defined, and then the
subset of indexes to be used for searching. These
parameters are retained across program executions (there
are kept separately for each <span class=
"application">Recoll</span> configuration). The set of
all indexes is usually quite stable, while the active
ones might typically be adjusted quite frequently.</p>
<p>The main index (defined by <code class=
"envar">RECOLL_CONFDIR</code>) is always active. If this
is undesirable, you can set up your base configuration to
index an empty directory.</p>
<p>When adding a new index to the set, you can select
either a <span class="application">Recoll</span>
configuration directory, or directly a <span class=
"application">Xapian</span> index directory. In the first
case, the <span class="application">Xapian</span> index
directory will be obtained from the selected
configuration.</p>
<p>As building the set of all indexes can be a little
tedious when done through the user interface, you can use
the <code class="envar">RECOLL_EXTRA_DBS</code>
environment variable to provide an initial set. This
might typically be set up by a system administrator so
that every user does not have to do it. The variable
should define a colon-separated list of index
directories, ie:</p>
<pre class=
"screen">export RECOLL_EXTRA_DBS=/some/place/xapiandb:/some/other/db</pre>
<p>Another environment variable, <code class=
"envar">RECOLL_ACTIVE_EXTRA_DBS</code> allows adding to
the active list of indexes. This variable was suggested
and implemented by a <span class=
"application">Recoll</span> user. It is mostly useful if
you use scripts to mount external volumes with
<span class="application">Recoll</span> indexes. By using
<code class="envar">RECOLL_EXTRA_DBS</code> and
<code class="envar">RECOLL_ACTIVE_EXTRA_DBS</code>, you
can add and activate the index for the mounted volume
when starting <span class=
"command"><strong>recoll</strong></span>. Unreachable
indexes will automatically be deactivated when starting
up.</p>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name="RCL.SEARCH.GUI.HISTORY"
id=
"RCL.SEARCH.GUI.HISTORY"></a>3.2.11.&nbsp;Document
history</h3>
</div>
</div>
</div>
<p>Documents that you actually view (with the internal
preview or an external tool) are entered into the
document history, which is remembered.</p>
<p>You can display the history list by using the
<span class="guilabel">Tools/</span><span class=
"guilabel">Doc History</span> menu entry.</p>
<p>You can erase the document history by using the
<span class="guilabel">Erase document history</span>
entry in the <span class="guimenu">File</span> menu.</p>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name="RCL.SEARCH.GUI.SORT" id=
"RCL.SEARCH.GUI.SORT"></a>3.2.12.&nbsp;Sorting
search results and collapsing duplicates</h3>
</div>
</div>
</div>
<p>The documents in a result list are normally sorted in
order of relevance. It is possible to specify a different
sort order, either by using the vertical arrows in the
GUI toolbox to sort by date, or switching to the result
table display and clicking on any header. The sort order
chosen inside the result table remains active if you
switch back to the result list, until you click one of
the vertical arrows, until both are unchecked (you are
back to sort by relevance).</p>
<p>Sort parameters are remembered between program
invocations, but result sorting is normally always
inactive when the program starts. It is possible to keep
the sorting activation state between program invocations
by checking the <span class="guilabel">Remember sort
activation state</span> option in the preferences.</p>
<p>It is also possible to hide duplicate entries inside
the result list (documents with the exact same contents
as the displayed one). The test of identity is based on
an MD5 hash of the document container, not only of the
text contents (so that ie, a text document with an image
added will not be a duplicate of the text only).
Duplicates hiding is controlled by an entry in the
<span class="guilabel">GUI configuration</span> dialog,
and is off by default.</p>
<p>When a result document does have undisplayed
duplicates, a <code class="literal">Dups</code> link will
be shown with the result list entry. Clicking the link
will display the paths (URLs + ipaths) for the duplicate
entries.</p>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.SEARCH.GUI.SHORTCUTS" id=
"RCL.SEARCH.GUI.SHORTCUTS"></a>3.2.13.&nbsp;Keyboard
shortcuts</h3>
</div>
</div>
</div>
<p>A number of common actions within the graphical
interface can be triggered through keyboard shortcuts. As
of <span class="application">Recoll</span> 1.29, many of
the shortcut values can be customised from a screen in
the GUI preferences. Most shortcuts are specific to a
given context (e.g. within a preview window, within the
result table).</p>
<div class="table">
<a name="idm1465" id="idm1465"></a>
<p class="title"><b>Table&nbsp;3.1.&nbsp;Keyboard
shortcuts</b></p>
<div class="table-contents">
<table class="table" summary="Keyboard shortcuts"
border="1">
<colgroup>
<col align="left" class="c1">
<col align="left" class="c2">
</colgroup>
<thead>
<tr>
<th align="left">Description</th>
<th align="left">Default value</th>
</tr>
</thead>
<tbody>
<tr>
<td colspan="2" align="left"><span class=
"command"><strong>Context: almost
everywhere</strong></span></td>
</tr>
<tr>
<td align="left">Program exit</td>
<td align="left">Ctrl+Q</td>
</tr>
<tr>
<td colspan="2" align="left"><span class=
"command"><strong>Context: advanced
search</strong></span></td>
</tr>
<tr>
<td align="left">Load the next entry from the
search history</td>
<td align="left">Up</td>
</tr>
<tr>
<td align="left">Load the previous entry from
the search history</td>
<td align="left">Down</td>
</tr>
<tr>
<td colspan="2" align="left"><span class=
"command"><strong>Context: main
window</strong></span></td>
</tr>
<tr>
<td align="left">Clear search. This will move
the keyboard cursor to the simple search entry
and erase the current text</td>
<td align="left">Ctrl+S</td>
</tr>
<tr>
<td align="left">Move the keyboard cursor to
the search entry area without erasing the
current text</td>
<td align="left">Ctrl+L</td>
</tr>
<tr>
<td align="left">Move the keyboard cursor to
the search entry area without erasing the
current text</td>
<td align="left">Ctrl+Shift+S</td>
</tr>
<tr>
<td align="left">Toggle displaying the current
results as a table or as a list</td>
<td align="left">Ctrl+T</td>
</tr>
<tr>
<td colspan="2" align="left"><span class=
"command"><strong>Context: main window, when
showing the results as a
table</strong></span></td>
</tr>
<tr>
<td align="left">Move the keyboard cursor to
currently the selected row in the table, or to
the first one if none is selected</td>
<td align="left">Ctrl+R</td>
</tr>
<tr>
<td align="left">Jump to row 0-9 or a-z in the
table</td>
<td align="left">Ctrl+[0-9] or
Ctrl+Shift+[a-z]</td>
</tr>
<tr>
<td align="left">Cancel the current
selection</td>
<td align="left">Esc</td>
</tr>
<tr>
<td colspan="2" align="left"><span class=
"command"><strong>Context: preview
window</strong></span></td>
</tr>
<tr>
<td align="left">Close the preview window</td>
<td align="left">Esc</td>
</tr>
<tr>
<td align="left">Close the current tab</td>
<td align="left">Ctrl+W</td>
</tr>
<tr>
<td align="left">Open a print dialog for the
current tab contents</td>
<td align="left">Ctrl+P</td>
</tr>
<tr>
<td align="left">Load the next result from the
list to the current tab</td>
<td align="left">Shift+Down</td>
</tr>
<tr>
<td align="left">Load the previous result from
the list to the current tab</td>
<td align="left">Shift+Up</td>
</tr>
<tr>
<td colspan="2" align="left"><span class=
"command"><strong>Context: result
table</strong></span></td>
</tr>
<tr>
<td align="left">Copy the text contained in the
selected document to the clipboard</td>
<td align="left">Ctrl+G</td>
</tr>
<tr>
<td align="left">Open the current document and
exit Recoll</td>
<td align="left">Ctrl+Shift+O</td>
</tr>
<tr>
<td align="left">Open the current document</td>
<td align="left">Ctrl+O</td>
</tr>
<tr>
<td align="left">Show a full preview for the
current document</td>
<td align="left">Ctrl+D</td>
</tr>
<tr>
<td align="left">Toggle showing the column
names</td>
<td align="left">Ctrl+H</td>
</tr>
<tr>
<td align="left">Show a snippets (keyword in
context) list for the current document</td>
<td align="left">Ctrl+E</td>
</tr>
<tr>
<td align="left">Toggle showing the row
letters/numbers</td>
<td align="left">Ctrl+V</td>
</tr>
<tr>
<td colspan="2" align="left"><span class=
"command"><strong>Context: snippets
window</strong></span></td>
</tr>
<tr>
<td align="left">Close the snippets window</td>
<td align="left">Esc</td>
</tr>
<tr>
<td align="left">Find in the snippets list
(method #1)</td>
<td align="left">Ctrl+F</td>
</tr>
<tr>
<td align="left">Find in the snippets list
(method #2)</td>
<td align="left">/</td>
</tr>
<tr>
<td align="left">Find the next instance of the
search term</td>
<td align="left">F3</td>
</tr>
<tr>
<td align="left">Find the previous instance of
the search term</td>
<td align="left">Shift+F3</td>
</tr>
</tbody>
</table>
</div>
</div><br class="table-break">
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name="RCL.SEARCH.GUI.TIPS" id=
"RCL.SEARCH.GUI.TIPS"></a>3.2.14.&nbsp;Search
tips</h3>
</div>
</div>
</div>
<div class="sect3">
<div class="titlepage">
<div>
<div>
<h4 class="title"><a name=
"RCL.SEARCH.GUI.TIPS.TERMS" id=
"RCL.SEARCH.GUI.TIPS.TERMS"></a>Terms and search
expansion</h4>
</div>
</div>
</div>
<p><b>Term completion.&nbsp;</b>While typing into the
simple search entry, a popup menu will appear and show
completions for the current string. Values preceded by
a clock icon come from the history, those preceded by a
magnifier icon come from the index terms. This can be
disabled in the preferences.</p>
<p><b>Picking up new terms from result or preview
text.&nbsp;</b>Double-clicking on a word in the result
list or in a preview window will copy it to the simple
search entry field.</p>
<p><b>Wildcards.&nbsp;</b>Wildcards can be used inside
search terms in all forms of searches. <a class="link"
href="#RCL.SEARCH.WILDCARDS" title=
"3.6.1.&nbsp;More about wildcards">More about
wildcards</a>.</p>
<p><b>Automatic suffixes.&nbsp;</b>Words like
<code class="literal">odt</code> or <code class=
"literal">ods</code> can be automatically turned into
query language <code class="literal">ext:xxx</code>
clauses. This can be enabled in the <span class=
"guilabel">Search preferences</span> panel in the
GUI.</p>
<p><b>Disabling stem expansion.&nbsp;</b>Entering a
capitalized word in any search field will prevent stem
expansion (no search for <code class=
"literal">gardening</code> if you enter <code class=
"literal">Garden</code> instead of <code class=
"literal">garden</code>). This is the only case where
character case should make a difference for a
<span class="application">Recoll</span> search. You can
also disable stem expansion or change the stemming
language in the preferences.</p>
<p><b>Finding related documents.&nbsp;</b>Selecting the
<span class="guilabel">Find similar documents</span>
entry in the result list paragraph right-click menu
will select a set of "interesting" terms from the
current result, and insert them into the simple search
entry field. You can then possibly edit the list and
start a search to find documents which may be
apparented to the current result.</p>
<p><b>File names.&nbsp;</b>File names are added as
terms during indexing, and you can specify them as
ordinary terms in normal search fields (<span class=
"application">Recoll</span> used to index all
directories in the file path as terms. This has been
abandoned as it did not seem really useful).
Alternatively, you can use the specific file name
search which will <span class=
"emphasis"><em>only</em></span> look for file names,
and may be faster than the generic search especially
when using wildcards.</p>
</div>
<div class="sect3">
<div class="titlepage">
<div>
<div>
<h4 class="title"><a name=
"RCL.SEARCH.GUI.TIPS.PHRASES" id=
"RCL.SEARCH.GUI.TIPS.PHRASES"></a>Working with
phrases and proximity</h4>
</div>
</div>
</div>
<p><b>Phrases searches.&nbsp;</b>A phrase can be looked
for by enclosing a number of terms in double quotes.
Example: <code class="literal">"user manual"</code>
will look only for occurrences of <code class=
"literal">user</code> immediately followed by
<code class="literal">manual</code>. You can use the
<span class="guilabel">"Phrase"</span> field of the
advanced search dialog to the same effect. Phrases can
be entered along simple terms in all simple or advanced
search entry fields, except <span class=
"guilabel">"Phrase"</span>.</p>
<p><b>Proximity searches.&nbsp;</b>A proximity search
differs from a phrase search in that it does not impose
an order on the terms. Proximity searches can be
entered by specifying the <span class=
"guilabel">"Proximity"</span> type in the advanced
search, or by postfixing a phrase search with a 'p'.
Example: "user manual"p would also match "manual user".
Also see <a class="link" href=
"#RCL.SEARCH.LANG.MODIFIERS" title=
"3.5.2.&nbsp;Modifiers">the modifier section</a> from
the query language documentation.</p>
<p><b>AutoPhrases.&nbsp;</b>This option can be set in
the preferences dialog. If it is set, a phrase will be
automatically built and added to simple searches when
looking for <code class="literal">Any terms</code>.
This will not change radically the results, but will
give a relevance boost to the results where the search
terms appear as a phrase. Ie: searching for
<code class="literal">virtual reality</code> will still
find all documents where either <code class=
"literal">virtual</code> or <code class=
"literal">reality</code> or both appear, but those
which contain <code class="literal">virtual
reality</code> should appear sooner in the list.</p>
<p>Phrase searches can slow down a query if most of the
terms in the phrase are common. If the <code class=
"varname">autophrase</code> option is on, very common
terms will be removed from the automatically
constructed phrase. The removal threshold can be
adjusted from the search preferences.</p>
<p><b>Phrases and abbreviations.&nbsp;</b>Dotted
abbreviations like <code class="literal">I.B.M.</code>
are also automatically indexed as a word without the
dots: <code class="literal">IBM</code>. Searching for
the word inside a phrase (ie: <code class=
"literal">"the IBM company"</code>) will only match the
dotted abrreviation if you increase the phrase slack
(using the advanced search panel control, or the
<code class="literal">o</code> query language
modifier). Literal occurrences of the word will be
matched normally.</p>
</div>
<div class="sect3">
<div class="titlepage">
<div>
<div>
<h4 class="title"><a name=
"RCL.SEARCH.GUI.TIPS.MISC" id=
"RCL.SEARCH.GUI.TIPS.MISC"></a>Others</h4>
</div>
</div>
</div>
<p><b>Using fields.&nbsp;</b>You can use the <a class=
"link" href="#RCL.SEARCH.LANG" title=
"3.5.&nbsp;The query language">query language</a> and
field specifications to only search certain parts of
documents. This can be especially helpful with email,
for example only searching emails from a specific
originator: <code class="literal">search tips
from:helpfulgui</code></p>
<p><b>Adjusting the result table columns.&nbsp;</b>When
displaying results in table mode, you can use a right
click on the table headers to activate a pop-up menu
which will let you adjust what columns are displayed.
You can drag the column headers to adjust their order.
You can click them to sort by the field displayed in
the column. You can also save the result list in CSV
format.</p>
<p><b>Changing the GUI geometry.&nbsp;</b>It is
possible to configure the GUI in wide form factor by
dragging the toolbars to one of the sides (their
location is remembered between sessions), and moving
the category filters to a menu (can be set in the
<span class="guimenu">Preferences</span><span class=
"guimenuitem">GUI configuration</span><span class=
"guimenuitem">User interface</span> panel).</p>
<p><b>Query explanation.&nbsp;</b>You can get an exact
description of what the query looked for, including
stem expansion, and Boolean operators used, by clicking
on the result list header.</p>
<p><b>Advanced search history.&nbsp;</b>You can display
any of the last 100 complex searches performed by using
the up and down arrow keys while the advanced search
panel is active.</p>
<p><b>Forced opening of a preview window.&nbsp;</b>You
can use <span class=
"keycap"><strong>Shift</strong></span>+Click on a
result list <code class="literal">Preview</code> link
to force the creation of a preview window instead of a
new tab in the existing one.</p>
</div>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name="RCL.SEARCH.SAVING" id=
"RCL.SEARCH.SAVING"></a>3.2.15.&nbsp;Saving and
restoring queries (1.21 and later)</h3>
</div>
</div>
</div>
<p>Both simple and advanced query dialogs save recent
history, but the amount is limited: old queries will
eventually be forgotten. Also, important queries may be
difficult to find among others. This is why both types of
queries can also be explicitly saved to files, from the
GUI menus: <span class="guimenu">File</span>
<span class="guimenuitem">Save last query / Load last
query</span></p>
<p>The default location for saved queries is a
subdirectory of the current configuration directory, but
saved queries are ordinary files and can be written or
moved anywhere.</p>
<p>Some of the saved query parameters are part of the
preferences (e.g. <code class="literal">autophrase</code>
or the active external indexes), and may differ when the
query is loaded from the time it was saved. In this case,
<span class="application">Recoll</span> will warn of the
differences, but will not change the user
preferences.</p>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name="RCL.SEARCH.GUI.CUSTOM"
id=
"RCL.SEARCH.GUI.CUSTOM"></a>3.2.16.&nbsp;Customizing
the search interface</h3>
</div>
</div>
</div>
<p>You can customize some aspects of the search interface
by using the <span class="guimenu">GUI
configuration</span> entry in the <span class=
"guimenu">Preferences</span> menu.</p>
<p>There are several tabs in the dialog, dealing with the
interface itself, the parameters used for searching and
returning results, and what indexes are searched.</p>
<p><a name="RCL.SEARCH.GUI.CUSTOM.UI" id=
"RCL.SEARCH.GUI.CUSTOM.UI"></a><b>User interface
parameters:&nbsp;</b></p>
<div class="itemizedlist">
<ul class="itemizedlist" style=
"list-style-type: disc;">
<li class="listitem">
<p><span class="guilabel">Highlight color for query
terms</span>: Terms from the user query are
highlighted in the result list samples and the
preview window. The color can be chosen here. Any
Qt color string should work (ie <code class=
"literal">red</code>, <code class=
"literal">#ff0000</code>). The default is
<code class="literal">blue</code>.</p>
</li>
<li class="listitem">
<p><span class="guilabel">Style sheet</span>: The
name of a <span class="application">Qt</span> style
sheet text file which is applied to the whole
Recoll application on startup. The default value is
empty, but there is a skeleton style sheet
(<code class="filename">recoll.qss</code>) inside
the <code class=
"filename">/usr/share/recoll/examples</code>
directory. Using a style sheet, you can change most
<span class=
"command"><strong>recoll</strong></span> graphical
parameters: colors, fonts, etc. See the sample file
for a few simple examples.</p>
<p>You should be aware that parameters (e.g.: the
background color) set inside the <span class=
"application">Recoll</span> GUI style sheet will
override global system preferences, with possible
strange side effects: for example if you set the
foreground to a light color and the background to a
dark one in the desktop preferences, but only the
background is set inside the <span class=
"application">Recoll</span> style sheet, and it is
light too, then text will appear light-on-light
inside the <span class="application">Recoll</span>
GUI.</p>
</li>
<li class="listitem">
<p><span class="guilabel">Maximum text size
highlighted for preview</span> Inserting highlights
on search term inside the text before inserting it
in the preview window involves quite a lot of
processing, and can be disabled over the given text
size to speed up loading.</p>
</li>
<li class="listitem">
<p><span class="guilabel">Prefer HTML to plain text
for preview</span> if set, Recoll will display HTML
as such inside the preview window. If this causes
problems with the Qt HTML display, you can uncheck
it to display the plain text version instead.</p>
</li>
<li class="listitem">
<p><span class="guilabel">Activate links in
preview</span> if set, Recoll will turn HTTP links
found inside plain text into proper HTML anchors,
and clicking a link inside a preview window will
start the default browser on the link target.</p>
</li>
<li class="listitem">
<p><span class="guilabel">Plain text to HTML line
style</span>: when displaying plain text inside the
preview window, <span class=
"application">Recoll</span> tries to preserve some
of the original text line breaks and indentation.
It can either use PRE HTML tags, which will well
preserve the indentation but will force horizontal
scrolling for long lines, or use BR tags to break
at the original line breaks, which will let the
editor introduce other line breaks according to the
window width, but will lose some of the original
indentation. The third option has been available in
recent releases and is probably now the best one:
use PRE tags with line wrapping.</p>
</li>
<li class="listitem">
<p><span class="guilabel">Choose editor
application</span>: this opens a dialog which
allows you to select the application to be used to
open each MIME type. The default is to use the
<span class=
"command"><strong>xdg-open</strong></span> utility,
but you can use this dialog to override it, setting
exceptions for MIME types that will still be opened
according to <span class=
"application">Recoll</span> preferences. This is
useful for passing parameters like page numbers or
search strings to applications that support them
(e.g. <span class="application">evince</span>).
This cannot be done with <span class=
"command"><strong>xdg-open</strong></span> which
only supports passing one parameter.</p>
</li>
<li class="listitem">
<p><span class="guilabel">Disable Qt autocompletion
in search entry</span>: this will disable the
completion popup. Il will only appear, and display
the full history, either if you enter only white
space in the search area, or if you click the clock
button on the right of the area.</p>
</li>
<li class="listitem">
<p><span class="guilabel">Document filter choice
style</span>: this will let you choose if the
document categories are displayed as a list or a
set of buttons, or a menu.</p>
</li>
<li class="listitem">
<p><span class="guilabel">Start with simple search
mode</span>: this lets you choose the value of the
simple search type on program startup. Either a
fixed value (e.g. <code class="literal">Query
Language</code>, or the value in use when the
program last exited.</p>
</li>
<li class="listitem">
<p><span class="guilabel">Start with advanced
search dialog open</span> : If you use this dialog
frequently, checking the entries will get it to
open when recoll starts.</p>
</li>
<li class="listitem">
<p><span class="guilabel">Remember sort activation
state</span> if set, Recoll will remember the sort
tool stat between invocations. It normally starts
with sorting disabled.</p>
</li>
</ul>
</div>
<p><a name="RCL.SEARCH.GUI.CUSTOM.RL" id=
"RCL.SEARCH.GUI.CUSTOM.RL"></a><b>Result list
parameters:&nbsp;</b></p>
<div class="itemizedlist">
<ul class="itemizedlist" style=
"list-style-type: disc;">
<li class="listitem">
<p><span class="guilabel">Number of results in a
result page</span></p>
</li>
<li class="listitem">
<p><span class="guilabel">Result list font</span>:
There is quite a lot of information shown in the
result list, and you may want to customize the font
and/or font size. The rest of the fonts used by
<span class="application">Recoll</span> are
determined by your generic Qt config (try the
<span class=
"command"><strong>qtconfig</strong></span>
command).</p>
</li>
<li class="listitem">
<p><a name="RCL.SEARCH.GUI.CUSTOM.RESULTPARA" id=
"RCL.SEARCH.GUI.CUSTOM.RESULTPARA"></a><span class=
"guilabel">Edit result list paragraph format
string</span>: allows you to change the
presentation of each result list entry. See the
<a class="link" href=
"#RCL.SEARCH.GUI.CUSTOM.RESLIST" title=
"The result list format">result list customisation
section</a>.</p>
</li>
<li class="listitem">
<p><a name="RCL.SEARCH.GUI.CUSTOM.RESULTHEAD" id=
"RCL.SEARCH.GUI.CUSTOM.RESULTHEAD"></a><span class=
"guilabel">Edit result page HTML header
insert</span>: allows you to define text inserted
at the end of the result page HTML header. More
detail in the <a class="link" href=
"#RCL.SEARCH.GUI.CUSTOM.RESLIST" title=
"The result list format">result list customisation
section</a>.</p>
</li>
<li class="listitem">
<p><span class="guilabel">Date format</span>:
allows specifying the format used for displaying
dates inside the result list. This should be
specified as an strftime() string (man
strftime).</p>
</li>
<li class="listitem">
<p><a name="RCL.SEARCH.GUI.CUSTOM.ABSSEP" id=
"RCL.SEARCH.GUI.CUSTOM.ABSSEP"></a><span class=
"guilabel">Abstract snippet separator</span>: for
synthetic abstracts built from index data, which
are usually made of several snippets from different
parts of the document, this defines the snippet
separator, an ellipsis by default.</p>
</li>
</ul>
</div>
<p><a name="RCL.SEARCH.GUI.CUSTOM.SEARCH" id=
"RCL.SEARCH.GUI.CUSTOM.SEARCH"></a><b>Search
parameters:&nbsp;</b></p>
<div class="itemizedlist">
<ul class="itemizedlist" style=
"list-style-type: disc;">
<li class="listitem">
<p><span class="guilabel">Hide duplicate
results</span>: decides if result list entries are
shown for identical documents found in different
places.</p>
</li>
<li class="listitem">
<p><span class="guilabel">Stemming language</span>:
stemming obviously depends on the document's
language. This listbox will let you chose among the
stemming databases which were built during indexing
(this is set in the <a class="link" href=
"#RCL.INSTALL.CONFIG.RECOLLCONF" title=
"5.4.2.&nbsp;Recoll main configuration file, recoll.conf">
main configuration file</a>), or later added with
<span class="command"><strong>recollindex
-s</strong></span> (See the recollindex manual).
Stemming languages which are dynamically added will
be deleted at the next indexing pass unless they
are also added in the configuration file.</p>
</li>
<li class="listitem">
<p><span class="guilabel">Automatically add phrase
to simple searches</span>: a phrase will be
automatically built and added to simple searches
when looking for <code class="literal">Any
terms</code>. This will give a relevance boost to
the results where the search terms appear as a
phrase (consecutive and in order).</p>
</li>
<li class="listitem">
<p><span class="guilabel">Autophrase term frequency
threshold percentage</span>: very frequent terms
should not be included in automatic phrase searches
for performance reasons. The parameter defines the
cutoff percentage (percentage of the documents
where the term appears).</p>
</li>
<li class="listitem">
<p><span class="guilabel">Replace abstracts from
documents</span>: this decides if we should
synthesize and display an abstract in place of an
explicit abstract found within the document
itself.</p>
</li>
<li class="listitem">
<p><span class="guilabel">Dynamically build
abstracts</span>: this decides if <span class=
"application">Recoll</span> tries to build document
abstracts (lists of <span class=
"emphasis"><em>snippets</em></span>) when
displaying the result list. Abstracts are
constructed by taking context from the document
information, around the search terms.</p>
</li>
<li class="listitem">
<p><span class="guilabel">Synthetic abstract
size</span>: adjust to taste...</p>
</li>
<li class="listitem">
<p><span class="guilabel">Synthetic abstract
context words</span>: how many words should be
displayed around each term occurrence.</p>
</li>
<li class="listitem">
<p><span class="guilabel">Query language magic file
name suffixes</span>: a list of words which
automatically get turned into <code class=
"literal">ext:xxx</code> file name suffix clauses
when starting a query language query (e.g.:
<code class="literal">doc xls xlsx...</code>). This
will save some typing for people who use file types
a lot when querying.</p>
</li>
</ul>
</div>
<p><a name="RCL.SEARCH.GUI.CUSTOM.EXTRADB" id=
"RCL.SEARCH.GUI.CUSTOM.EXTRADB"></a><b>External
indexes:&nbsp;</b>This panel will let you browse for
additional indexes that you may want to search. External
indexes are designated by their database directory (ie:
<code class=
"filename">/home/someothergui/.recoll/xapiandb</code>,
<code class=
"filename">/usr/local/recollglobal/xapiandb</code>).</p>
<p>Once entered, the indexes will appear in the
<span class="guilabel">External indexes</span> list, and
you can chose which ones you want to use at any moment by
checking or unchecking their entries.</p>
<p>Your main database (the one the current configuration
indexes to), is always implicitly active. If this is not
desirable, you can set up your configuration so that it
indexes, for example, an empty directory. An alternative
indexer may also need to implement a way of purging the
index from stale data,</p>
<div class="sect3">
<div class="titlepage">
<div>
<div>
<h4 class="title"><a name=
"RCL.SEARCH.GUI.CUSTOM.RESLIST" id=
"RCL.SEARCH.GUI.CUSTOM.RESLIST"></a>The result
list format</h4>
</div>
</div>
</div>
<p>Recoll normally uses a full function HTML processor
to display the result list and the <a class="link"
href="#RCL.SEARCH.GUI.RESULTLIST.MENU.SNIPPETS">snippets
window</a>. Depending on the version, this may be based
on either Qt WebKit or Qt WebEngine. It is then
possible to completely customise the result list with
full support for CSS and Javascript.</p>
<p>It is also possible to build <span class=
"application">Recoll</span> to use a simpler Qt
QTextBrowser widget to display the HTML, which may be
necessary if the ones above are not ported on the
system, or to reduce the application size and
dependencies. There are limits to what you can do in
this case, but it is still possible to decide what data
each result will contain, and how it will be
displayed.</p>
<p>The result list presentation can be customized by
adjusting two elements:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style=
"list-style-type: disc;">
<li class="listitem">
<p>The paragraph format</p>
</li>
<li class="listitem">
<p>HTML code inside the header section. For
versions 1.21 and later, this is also used for
the <a class="link" href=
"#RCL.SEARCH.GUI.RESULTLIST.MENU.SNIPPETS">snippets
window</a>.</p>
</li>
</ul>
</div>
<p>The paragraph format and the header fragment can be
edited from the <span class="guilabel">Result
list</span> tab of the <span class="guilabel">GUI
configuration</span>.</p>
<p>The header fragment is used both for the result list
and the snippets window. The snippets list is a table
and has a <code class="literal">snippets</code> class
attribute. Each paragraph in the result list is a
table, with class <code class="literal">respar</code>,
but this can be changed by editing the paragraph
format.</p>
<p>There are a few examples on the <a class="ulink"
href="http://www.recoll.org/pages/custom.html" target=
"_top">page about customising the result list</a> on
the <span class="application">Recoll</span> web
site.</p>
<div class="sect4">
<div class="titlepage">
<div>
<div>
<h5 class="title"><a name=
"RCL.SEARCH.GUI.CUSTOM.RESLIST.PARA" id=
"RCL.SEARCH.GUI.CUSTOM.RESLIST.PARA"></a>The
paragraph format</h5>
</div>
</div>
</div>
<p>This is an arbitrary HTML string where the
following printf-like <code class="literal">%</code>
substitutions will be performed:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style=
"list-style-type: disc;">
<li class="listitem">
<p><b>%A.&nbsp;</b>Abstract</p>
</li>
<li class="listitem">
<p><b>%D.&nbsp;</b>Date</p>
</li>
<li class="listitem">
<p><b>%I.&nbsp;</b>Icon image name. This is
normally determined from the MIME type. The
associations are defined inside the <a class=
"link" href="#RCL.INSTALL.CONFIG.MIMECONF"
title=
"5.4.5.&nbsp;The mimeconf file"><code class=
"filename">mimeconf</code> configuration
file</a>. If a thumbnail for the file is found
at the standard Freedesktop location, this will
be displayed instead.</p>
</li>
<li class="listitem">
<p><b>%K.&nbsp;</b>Keywords (if any)</p>
</li>
<li class="listitem">
<p><b>%L.&nbsp;</b>Precooked Preview, Edit, and
possibly Snippets links</p>
</li>
<li class="listitem">
<p><b>%M.&nbsp;</b>MIME type</p>
</li>
<li class="listitem">
<p><b>%N.&nbsp;</b>result Number inside the
result page</p>
</li>
<li class="listitem">
<p><b>%P.&nbsp;</b>Parent folder Url. In the
case of an embedded document, this is the
parent folder for the top level container
file.</p>
</li>
<li class="listitem">
<p><b>%R.&nbsp;</b>Relevance percentage</p>
</li>
<li class="listitem">
<p><b>%S.&nbsp;</b>Size information</p>
</li>
<li class="listitem">
<p><b>%T.&nbsp;</b>Title or Filename if not
set.</p>
</li>
<li class="listitem">
<p><b>%t.&nbsp;</b>Title or empty.</p>
</li>
<li class="listitem">
<p><b>%(filename).&nbsp;</b>File name.</p>
</li>
<li class="listitem">
<p><b>%U.&nbsp;</b>Url</p>
</li>
</ul>
</div>
<p>The format of the Preview, Edit, and Snippets
links is <code class="literal">&lt;a
href="P%N"&gt;</code>, <code class="literal">&lt;a
href="E%N"&gt;</code> and <code class="literal">&lt;a
href="A%N"&gt;</code> where <em class=
"replaceable"><code>docnum</code></em> (%N) expands
to the document number inside the result page).</p>
<p>A link target defined as <code class=
"literal">"F%N"</code> will open the document
corresponding to the <code class="literal">%P</code>
parent folder expansion, usually creating a file
manager window on the folder where the container file
resides. E.g.:</p>
<pre class=
"programlisting">&lt;a href="F%N"&gt;%P&lt;/a&gt;</pre>
<p>A link target defined as <code class=
"literal">R%N|<em class=
"replaceable"><code>scriptname</code></em></code>
will run the corresponding script on the result file
(if the document is embedded, the script will be
started on the top-level parent). See the <a class=
"link" href="#RCL.SEARCH.GUI.RUNSCRIPT" title=
"3.2.4.&nbsp;Unix-like systems: running arbitrary commands on result files">
section about defining scripts</a>.</p>
<p>In addition to the predefined values above, all
strings like <code class=
"literal">%(fieldname)</code> will be replaced by the
value of the field named <code class=
"literal">fieldname</code> for this document. Only
stored fields can be accessed in this way, the value
of indexed but not stored fields is not known at this
point in the search process (see <a class="link"
href="#RCL.PROGRAM.FIELDS" title=
"4.2.&nbsp;Field data processing">field
configuration</a>). There are currently very few
fields stored by default, apart from the values above
(only <code class="literal">author</code> and
<code class="literal">filename</code>), so this
feature will need some custom local configuration to
be useful. An example candidate would be the
<code class="literal">recipient</code> field which is
generated by the message input handlers.</p>
<p>The default value for the paragraph format string
is:</p>
<pre class="screen">
"&lt;table class=\"respar\"&gt;\n"
"&lt;tr&gt;\n"
"&lt;td&gt;&lt;a href='%U'&gt;&lt;img src='%I' width='64'&gt;&lt;/a&gt;&lt;/td&gt;\n"
"&lt;td&gt;%L &amp;nbsp;&lt;i&gt;%S&lt;/i&gt; &amp;nbsp;&amp;nbsp;&lt;b&gt;%T&lt;/b&gt;&lt;br&gt;\n"
"&lt;span style='white-space:nowrap'&gt;&lt;i&gt;%M&lt;/i&gt;&amp;nbsp;%D&lt;/span&gt;&amp;nbsp;&amp;nbsp;&amp;nbsp; &lt;i&gt;%U&lt;/i&gt;&amp;nbsp;%i&lt;br&gt;\n"
"%A %K&lt;/td&gt;\n"
"&lt;/tr&gt;&lt;/table&gt;\n"
</pre>
<p>You may, for example, try the following for a more
web-like experience:</p>
<pre class="screen">
&lt;u&gt;&lt;b&gt;&lt;a href="P%N"&gt;%T&lt;/a&gt;&lt;/b&gt;&lt;/u&gt;&lt;br&gt;
%A&lt;font color=#008000&gt;%U - %S&lt;/font&gt; - %L
</pre>
<p>Note that the P%N link in the above paragraph
makes the title a preview link. Or the clean
looking:</p>
<pre class="screen">
&lt;img src="%I" align="left"&gt;%L &lt;font color="#900000"&gt;%R&lt;/font&gt;
&amp;nbsp;&amp;nbsp;&lt;b&gt;%T&amp;&lt;/b&gt;&lt;br&gt;%S&amp;nbsp;
&lt;font color="#808080"&gt;&lt;i&gt;%U&lt;/i&gt;&lt;/font&gt;
&lt;table bgcolor="#e0e0e0"&gt;
&lt;tr&gt;&lt;td&gt;&lt;div&gt;%A&lt;/div&gt;&lt;/td&gt;&lt;/tr&gt;
&lt;/table&gt;%K
</pre>
<p>These samples, and some others are <a class=
"ulink" href=
"http://www.recoll.org/pages/custom.html" target=
"_top">on the web site, with pictures to show how
they look.</a></p>
<p>It is also possible to <a class="link" href=
"#RCL.SEARCH.GUI.CUSTOM.ABSSEP">define the value of
the snippet separator inside the abstract
section</a>.</p>
</div>
</div>
</div>
</div>
<div class="sect1">
<div class="titlepage">
<div>
<div>
<h2 class="title" style="clear: both"><a name=
"RCL.SEARCH.KIO" id=
"RCL.SEARCH.KIO"></a>3.3.&nbsp;Searching with the KDE
KIO slave</h2>
</div>
</div>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name="RCL.SEARCH.KIO.INTRO"
id="RCL.SEARCH.KIO.INTRO"></a>3.3.1.&nbsp;What's
this</h3>
</div>
</div>
</div>
<p>The <span class="application">Recoll</span> KIO slave
allows performing a <span class=
"application">Recoll</span> search by entering an
appropriate URL in a KDE open dialog, or with an
HTML-based interface displayed in <span class=
"command"><strong>Konqueror</strong></span>.</p>
<p>The HTML-based interface is similar to the Qt-based
interface, but slightly less powerful for now. Its
advantage is that you can perform your search while
staying fully within the KDE framework: drag and drop
from the result list works normally and you have your
normal choice of applications for opening files.</p>
<p>The alternative interface uses a directory view of
search results. Due to limitations in the current KIO
slave interface, it is currently not obviously useful (to
me).</p>
<p>The interface is described in more detail inside a
help file which you can access by entering <code class=
"filename">recoll:/</code> inside the <span class=
"command"><strong>konqueror</strong></span> URL line
(this works only if the recoll KIO slave has been
previously installed).</p>
<p>The instructions for building this module are located
in the source tree. See: <code class=
"filename">kde/kio/recoll/00README.txt</code>. Some Linux
distributions do package the kio-recoll module, so check
before diving into the build process, maybe it's already
out there ready for one-click installation.</p>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.SEARCH.KIO.SEARCHABLEDOCS" id=
"RCL.SEARCH.KIO.SEARCHABLEDOCS"></a>3.3.2.&nbsp;Searchable
documents</h3>
</div>
</div>
</div>
<p>As a sample application, the <span class=
"application">Recoll</span> KIO slave could allow
preparing a set of HTML documents (for example a manual)
so that they become their own search interface inside
<span class=
"command"><strong>konqueror</strong></span>.</p>
<p>This can be done by either explicitly inserting
<code class="literal">&lt;a
href="recoll://..."&gt;</code> links around some document
areas, or automatically by adding a very small
<span class="application">javascript</span> program to
the documents, like the following example, which would
initiate a search by double-clicking any term:</p>
<pre class=
"programlisting">&lt;script language="JavaScript"&gt;
function recollsearch() {
var t = document.getSelection();
window.location.href = 'recoll://search/query?qtp=a&amp;p=0&amp;q=' +
encodeURIComponent(t);
}
&lt;/script&gt;
....
&lt;body ondblclick="recollsearch()"&gt;
</pre>
</div>
</div>
<div class="sect1">
<div class="titlepage">
<div>
<div>
<h2 class="title" style="clear: both"><a name=
"RCL.SEARCH.COMMANDLINE" id=
"RCL.SEARCH.COMMANDLINE"></a>3.4.&nbsp;Searching on
the command line</h2>
</div>
</div>
</div>
<p>There are several ways to obtain search results as a
text stream, without a graphical interface:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style="list-style-type: disc;">
<li class="listitem">
<p>By passing option <code class="option">-t</code>
to the <span class=
"command"><strong>recoll</strong></span> program, or
by calling it as <span class=
"command"><strong>recollq</strong></span> (through a
link).</p>
</li>
<li class="listitem">
<p>By using the <span class=
"command"><strong>recollq</strong></span>
program.</p>
</li>
<li class="listitem">
<p>By writing a custom <span class=
"application">Python</span> program, using the
<a class="link" href="#RCL.PROGRAM.PYTHONAPI" title=
"4.3.&nbsp;Python API">Recoll Python API</a>.</p>
</li>
</ul>
</div>
<p>The first two methods work in the same way and
accept/need the same arguments (except for the additional
<code class="option">-t</code> to <span class=
"command"><strong>recoll</strong></span>). The query to be
executed is specified as command line arguments.</p>
<p><span class="command"><strong>recollq</strong></span> is
not always built by default. You can use the <code class=
"filename">Makefile</code> in the <code class=
"filename">query</code> directory to build it. This is a
very simple program, and if you can program a little c++,
you may find it useful to taylor its output format to your
needs. Apart from being easily customised, <span class=
"command"><strong>recollq</strong></span> is only really
useful on systems where the Qt libraries are not available,
else it is redundant with <code class="literal">recoll
-t</code>.</p>
<p><span class="command"><strong>recollq</strong></span>
has a <a class="ulink" href=
"https://www.lesbonscomptes.com/recoll/manpages/recollq.1.html"
target="_top">man page</a>. The Usage string follows:</p>
<pre class="programlisting">
recollq: usage:
-P: Show the date span for all the documents present in the index
[-o|-a|-f] [-q] &lt;query string&gt;
Runs a recoll query and displays result lines.
Default: will interpret the argument(s) as a xesam query string
Query elements:
* Implicit AND, exclusion, field spec: t1 -t2 title:t3
* OR has priority: t1 OR t2 t3 OR t4 means (t1 OR t2) AND (t3 OR t4)
* Phrase: "t1 t2" (needs additional quoting on cmd line)
-o Emulate the GUI simple search in ANY TERM mode
-a Emulate the GUI simple search in ALL TERMS mode
-f Emulate the GUI simple search in filename mode
-q is just ignored (compatibility with the recoll GUI command line)
Common options:
-c &lt;configdir&gt; : specify config directory, overriding $RECOLL_CONFDIR
-d also dump file contents
-n [first-]&lt;cnt&gt; define the result slice. The default value for [first]
is 0. Without the option, the default max count is 2000.
Use n=0 for no limit
-b : basic. Just output urls, no mime types or titles
-Q : no result lines, just the processed query and result count
-m : dump the whole document meta[] array for each result
-A : output the document abstracts
-S fld : sort by field &lt;fld&gt;
-D : sort descending
-s stemlang : set stemming language to use (must exist in index...)
Use -s "" to turn off stem expansion
-T &lt;synonyms file&gt;: use the parameter (Thesaurus) for word expansion
-i &lt;dbdir&gt; : additional index, several can be given
-e use url encoding (%xx) for urls
-F &lt;field name list&gt; : output exactly these fields for each result.
The field values are encoded in base64, output in one line and
separated by one space character. This is the recommended format
for use by other programs. Use a normal query with option -m to
see the field names. Use -F '' to output all fields, but you probably
also want option -N in this case
-N : with -F, print the (plain text) field names before the field values
</pre>
<p>Sample execution:</p>
<pre class="programlisting">
recollq 'ilur -nautique mime:text/html'
Recoll query: ((((ilur:(wqf=11) OR ilurs) AND_NOT (nautique:(wqf=11) OR nautiques OR nautiqu OR nautiquement)) FILTER Ttext/html))
4 results
text/html [file:///Users/uncrypted-dockes/projets/bateaux/ilur/comptes.html] [comptes.html] 18593 bytes
text/html [file:///Users/uncrypted-dockes/projets/nautique/webnautique/articles/ilur1/index.html] [Constructio...
text/html [file:///Users/uncrypted-dockes/projets/pagepers/index.html] [psxtcl/writemime/recoll]...
text/html [file:///Users/uncrypted-dockes/projets/bateaux/ilur/factEtCie/recu-chasse-maree....
</pre>
</div>
<div class="sect1">
<div class="titlepage">
<div>
<div>
<h2 class="title" style="clear: both"><a name=
"RCL.SEARCH.LANG" id=
"RCL.SEARCH.LANG"></a>3.5.&nbsp;The query
language</h2>
</div>
</div>
</div>
<p>The query language processor is activated in the GUI
simple search entry when the search mode selector is set to
<span class="guilabel">Query Language</span>. It can also
be used with the KIO slave or the command line search. It
broadly has the same capabilities as the complex search
interface in the GUI.</p>
<p>The language was based on the now defunct <a class=
"ulink" href=
"http://www.xesam.org/main/XesamUserSearchLanguage95"
target="_top">Xesam</a> user search language
specification.</p>
<p>If the results of a query language search puzzle you and
you doubt what has been actually searched for, you can use
the GUI <code class="literal">Show Query</code> link at the
top of the result list to check the exact query which was
finally executed by Xapian.</p>
<p>Here follows a sample request that we are going to
explain:</p>
<pre class="programlisting">
author:"john doe" Beatles OR Lennon Live OR Unplugged -potatoes
</pre>
<p>This would search for all documents with <em class=
"replaceable"><code>John Doe</code></em> appearing as a
phrase in the author field (exactly what this is would
depend on the document type, ie: the <code class=
"literal">From:</code> header, for an email message), and
containing either <em class=
"replaceable"><code>beatles</code></em> or <em class=
"replaceable"><code>lennon</code></em> and either
<em class="replaceable"><code>live</code></em> or
<em class="replaceable"><code>unplugged</code></em> but not
<em class="replaceable"><code>potatoes</code></em> (in any
part of the document).</p>
<p>An element is composed of an optional field
specification, and a value, separated by a colon (the field
separator is the last colon in the element). Examples:
<em class="replaceable"><code>Eugenie</code></em>,
<em class="replaceable"><code>author:balzac</code></em>,
<em class="replaceable"><code>dc:title:grandet</code></em>
<em class="replaceable"><code>dc:title:"eugenie
grandet"</code></em></p>
<p>The colon, if present, means "contains". Xesam defines
other relations, which are mostly unsupported for now
(except in special cases, described further down).</p>
<p>All elements in the search entry are normally combined
with an implicit AND. It is possible to specify that
elements be OR'ed instead, as in <em class=
"replaceable"><code>Beatles</code></em> <code class=
"literal">OR</code> <em class=
"replaceable"><code>Lennon</code></em>. The <code class=
"literal">OR</code> must be entered literally (capitals),
and it has priority over the AND associations: <em class=
"replaceable"><code>word1</code></em> <em class=
"replaceable"><code>word2</code></em> <code class=
"literal">OR</code> <em class=
"replaceable"><code>word3</code></em> means <em class=
"replaceable"><code>word1</code></em> AND (<em class=
"replaceable"><code>word2</code></em> <code class=
"literal">OR</code> <em class=
"replaceable"><code>word3</code></em>) not (<em class=
"replaceable"><code>word1</code></em> AND <em class=
"replaceable"><code>word2</code></em>) <code class=
"literal">OR</code> <em class=
"replaceable"><code>word3</code></em>.</p>
<p><span class="application">Recoll</span> versions 1.21
and later, allow using parentheses to group elements, which
will sometimes make things clearer, and may allow
expressing combinations which would have been difficult
otherwise.</p>
<p>An element preceded by a <code class="literal">-</code>
specifies a term that should <span class=
"emphasis"><em>not</em></span> appear.</p>
<p>As usual, words inside quotes define a phrase (the order
of words is significant), so that <em class=
"replaceable"><code>title:"prejudice pride"</code></em> is
not the same as <em class=
"replaceable"><code>title:prejudice
title:pride</code></em>, and is unlikely to find a
result.</p>
<p>Words inside phrases and capitalized words are not
stem-expanded. Wildcards may be used anywhere inside a
term. Specifying a wild-card on the left of a term can
produce a very slow search (or even an incorrect one if the
expansion is truncated because of excessive size). Also see
<a class="link" href="#RCL.SEARCH.WILDCARDS" title=
"3.6.1.&nbsp;More about wildcards">More about
wildcards</a>.</p>
<p>To save you some typing, recent <span class=
"application">Recoll</span> versions (1.20 and later)
interpret a comma-separated list of terms for a field as an
AND list inside the field. Use slash characters ('/') for
an OR list. No white space is allowed. So</p>
<pre class="programlisting">author:john,lennon</pre>
<p>will search for documents with <code class=
"literal">john</code> and <code class=
"literal">lennon</code> inside the <code class=
"literal">author</code> field (in any order), and</p>
<pre class="programlisting">author:john/ringo</pre>
<p>would search for <code class="literal">john</code> or
<code class="literal">ringo</code>. This behaviour only
happens for field queries (input without a field, comma- or
slash- separated input will produce a phrase search). You
can use a <code class="literal">text</code> field name to
search the main text this way.</p>
<p>Modifiers can be set on a double-quote value, for
example to specify a proximity search (unordered). See
<a class="link" href="#RCL.SEARCH.LANG.MODIFIERS" title=
"3.5.2.&nbsp;Modifiers">the modifier section</a>. No space
must separate the final double-quote and the modifiers
value, e.g. <em class="replaceable"><code>"two
one"po10</code></em></p>
<p><span class="application">Recoll</span> currently
manages the following default fields:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style="list-style-type: disc;">
<li class="listitem">
<p><code class="literal">title</code>, <code class=
"literal">subject</code> or <code class=
"literal">caption</code> are synonyms which specify
data to be searched for in the document title or
subject.</p>
</li>
<li class="listitem">
<p><code class="literal">author</code> or
<code class="literal">from</code> for searching the
documents originators.</p>
</li>
<li class="listitem">
<p><code class="literal">recipient</code> or
<code class="literal">to</code> for searching the
documents recipients.</p>
</li>
<li class="listitem">
<p><code class="literal">keyword</code> for searching
the document-specified keywords (few documents
actually have any).</p>
</li>
<li class="listitem">
<p><code class="literal">filename</code> for the
document's file name. This is not necessarily set for
all documents: internal documents contained inside a
compound one (for example an EPUB section) do not
inherit the container file name any more, this was
replaced by an explicit field (see next).
Sub-documents can still have a specific <code class=
"literal">filename</code>, if it is implied by the
document format, for example the attachment file name
for an email attachment.</p>
</li>
<li class="listitem">
<p><code class="literal">containerfilename</code>.
This is set for all documents, both top-level and
contained sub-documents, and is always the name of
the filesystem directory entry which contains the
data. The terms from this field can only be matched
by an explicit field specification (as opposed to
terms from <code class="literal">filename</code>
which are also indexed as general document content).
This avoids getting matches for all the sub-documents
when searching for the container file name.</p>
</li>
<li class="listitem">
<p><code class="literal">ext</code> specifies the
file name extension (Ex: <code class=
"literal">ext:html</code>).</p>
</li>
<li class="listitem">
<p><code class="literal">rclmd5</code> the MD5
checksum for the document. This is used for
displaying the duplicates of a search result (when
querying with the option to collapse duplicate
results). Incidentally, this could be used to find
the duplicates of any given file by computing its MD5
checksum and executing a query with just the
<code class="literal">rclmd5</code> value.</p>
</li>
</ul>
</div>
<p><span class="application">Recoll</span> 1.20 and later
have a way to specify aliases for the field names, which
will save typing, for example by aliasing <code class=
"literal">filename</code> to <em class=
"replaceable"><code>fn</code></em> or <code class=
"literal">containerfilename</code> to <em class=
"replaceable"><code>cfn</code></em>. See the <a class=
"link" href="#RCL.INSTALL.CONFIG.FIELDS" title=
"5.4.3.&nbsp;The fields file">section about the
<code class="filename">fields</code> file</a>.</p>
<p>The document input handlers used while indexing have the
possibility to create other fields with arbitrary names,
and aliases may be defined in the configuration, so that
the exact field search possibilities may be different for
you if someone took care of the customisation.</p>
<p>The field syntax also supports a few field-like, but
special, criteria:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style="list-style-type: disc;">
<li class="listitem">
<p><code class="literal">dir</code> for filtering the
results on file location (Ex: <code class=
"literal">dir:/home/me/somedir</code>). <code class=
"literal">-dir</code> also works to find results not
in the specified directory (release &gt;= 1.15.8).
Tilde expansion will be performed as usual (except
for a bug in versions 1.19 to 1.19.11p1). Wildcards
will be expanded, but please <a class="link" href=
"#RCL.SEARCH.WILDCARDS.PATH" title=
"Wildcards and path filtering">have a look</a> at an
important limitation of wildcards in path
filters.</p>
<p>Relative paths also make sense, for example,
<code class="literal">dir:share/doc</code> would
match either <code class=
"filename">/usr/share/doc</code> or <code class=
"filename">/usr/local/share/doc</code></p>
<p>Several <code class="literal">dir</code> clauses
can be specified, both positive and negative. For
example the following makes sense:</p>
<pre class="programlisting">
dir:recoll dir:src -dir:utils -dir:common
</pre>
<p>This would select results which have both
<code class="filename">recoll</code> and <code class=
"filename">src</code> in the path (in any order), and
which have not either <code class=
"filename">utils</code> or <code class=
"filename">common</code>.</p>
<p>You can also use <code class="literal">OR</code>
conjunctions with <code class="literal">dir:</code>
clauses.</p>
<p>A special aspect of <code class=
"literal">dir</code> clauses is that the values in
the index are not transcoded to UTF-8, and never
lower-cased or unaccented, but stored as binary. This
means that you need to enter the values in the exact
lower or upper case, and that searches for names with
diacritics may sometimes be impossible because of
character set conversion issues. Non-ASCII UNIX file
paths are an unending source of trouble and are best
avoided.</p>
<p>You need to use double-quotes around the path
value if it contains space characters.</p>
</li>
<li class="listitem">
<p><code class="literal">size</code> for filtering
the results on file size. Example: <code class=
"literal">size&lt;10000</code>. You can use
<code class="literal">&lt;</code>, <code class=
"literal">&gt;</code> or <code class=
"literal">=</code> as operators. You can specify a
range like the following: <code class=
"literal">size&gt;100 size&lt;1000</code>. The usual
<code class="literal">k/K, m/M, g/G, t/T</code> can
be used as (decimal) multipliers. Ex: <code class=
"literal">size&gt;1k</code> to search for files
bigger than 1000 bytes.</p>
</li>
<li class="listitem">
<p><code class="literal">date</code> for searching or
filtering on dates. The syntax for the argument is
based on the ISO8601 standard for dates and time
intervals. Only dates are supported, no times. The
general syntax is 2 elements separated by a
<code class="literal">/</code> character. Each
element can be a date or a period of time. Periods
are specified as <code class=
"literal">P</code><em class=
"replaceable"><code>n</code></em><code class=
"literal">Y</code><em class=
"replaceable"><code>n</code></em><code class=
"literal">M</code><em class=
"replaceable"><code>n</code></em><code class=
"literal">D</code>. The <em class=
"replaceable"><code>n</code></em> numbers are the
respective numbers of years, months or days, any of
which may be missing. Dates are specified as
<em class=
"replaceable"><code>YYYY</code></em>-<em class=
"replaceable"><code>MM</code></em>-<em class=
"replaceable"><code>DD</code></em>. The days and
months parts may be missing. If the <code class=
"literal">/</code> is present but an element is
missing, the missing element is interpreted as the
lowest or highest date in the index. Examples:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style=
"list-style-type: circle;">
<li class="listitem">
<p><code class=
"literal">2001-03-01/2002-05-01</code> the
basic syntax for an interval of dates.</p>
</li>
<li class="listitem">
<p><code class=
"literal">2001-03-01/P1Y2M</code> the same
specified with a period.</p>
</li>
<li class="listitem">
<p><code class="literal">2001/</code> from the
beginning of 2001 to the latest date in the
index.</p>
</li>
<li class="listitem">
<p><code class="literal">2001</code> the whole
year of 2001</p>
</li>
<li class="listitem">
<p><code class="literal">P2D/</code> means 2
days ago up to now if there are no documents
with dates in the future.</p>
</li>
<li class="listitem">
<p><code class="literal">/2003</code> all
documents from 2003 or older.</p>
</li>
</ul>
</div>
<p>Periods can also be specified with small letters
(ie: p2y).</p>
</li>
<li class="listitem">
<p><code class="literal">mime</code> or <code class=
"literal">format</code> for specifying the MIME type.
These clauses are processed besides the normal
Boolean logic of the search. Multiple values will be
OR'ed (instead of the normal AND). You can specify
types to be excluded, with the usual <code class=
"literal">-</code>, and use wildcards. Example:
<em class="replaceable"><code>mime:text/*
-mime:text/plain</code></em> Specifying an explicit
boolean operator before a <code class=
"literal">mime</code> specification is not supported
and will produce strange results.</p>
</li>
<li class="listitem">
<p><code class="literal">type</code> or <code class=
"literal">rclcat</code> for specifying the category
(as in text/media/presentation/etc.). The
classification of MIME types in categories is defined
in the <span class="application">Recoll</span>
configuration (<code class=
"filename">mimeconf</code>), and can be modified or
extended. The default category names are those which
permit filtering results in the main GUI screen.
Categories are OR'ed like MIME types above, and can
be negated with <code class="literal">-</code>.</p>
</li>
<li class="listitem">
<p><code class="literal">issub</code> for specifying
that only standalone (<code class=
"literal">issub:0</code>) or only embedded
(<code class="literal">issub:1</code>) documents
should be returned as results.</p>
</li>
</ul>
</div>
<div class="note" style=
"margin-left: 0.5in; margin-right: 0.5in;">
<h3 class="title">Note</h3>
<p><code class="literal">mime</code>, <code class=
"literal">rclcat</code>, <code class=
"literal">size</code>, <code class="literal">issub</code>
and <code class="literal">date</code> criteria always
affect the whole query (they are applied as a final
filter), even if set with other terms inside a
parenthese.</p>
</div>
<div class="note" style=
"margin-left: 0.5in; margin-right: 0.5in;">
<h3 class="title">Note</h3>
<p><code class="literal">mime</code> (or the equivalent
<code class="literal">rclcat</code>) is the <span class=
"emphasis"><em>only</em></span> field with an
<code class="literal">OR</code> default. You do need to
use <code class="literal">OR</code> with <code class=
"literal">ext</code> terms for example.</p>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name="RCL.SEARCH.LANG.RANGES"
id="RCL.SEARCH.LANG.RANGES"></a>3.5.1.&nbsp;Range
clauses</h3>
</div>
</div>
</div>
<p><span class="application">Recoll</span> 1.24 and later
support range clauses on fields which have been
configured to support it. No default field uses them
currently, so this paragraph is only interesting if you
modified the fields configuration and possibly use a
custom input handler.</p>
<p>A range clause looks like one of the following:</p>
<pre class="programlisting"><em class=
"replaceable"><code>myfield</code></em>:<em class=
"replaceable"><code>small</code></em>..<em class=
"replaceable"><code>big</code></em>
<em class="replaceable"><code>myfield</code></em>:<em class=
"replaceable"><code>small</code></em>..
<em class="replaceable"><code>myfield</code></em>:..<em class=
"replaceable"><code>big</code></em>
</pre>
<p>The nature of the clause is indicated by the two dots
<code class="literal">..</code>, and the effect is to
filter the results for which the <em class=
"replaceable"><code>myfield</code></em> value is in the
possibly open-ended interval.</p>
<p>See the section about the <a class="link" href=
"#RCL.INSTALL.CONFIG.FIELDS" title=
"5.4.3.&nbsp;The fields file"><code class=
"filename">fields</code> configuration file</a> for the
details of configuring a field for range searches (list
them in the [values] section).</p>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.SEARCH.LANG.MODIFIERS" id=
"RCL.SEARCH.LANG.MODIFIERS"></a>3.5.2.&nbsp;Modifiers</h3>
</div>
</div>
</div>
<p>Some characters are recognized as search modifiers
when found immediately after the closing double quote of
a phrase, as in <code class="literal">"some
term"modifierchars</code>. The actual "phrase" can be a
single term of course. Supported modifiers:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style=
"list-style-type: disc;">
<li class="listitem">
<p><code class="literal">l</code> can be used to
turn off stemming (mostly makes sense with
<code class="literal">p</code> because stemming is
off by default for phrases).</p>
</li>
<li class="listitem">
<p><code class="literal">s</code> can be used to
turn off synonym expansion, if a synonyms file is
in place (only for <span class=
"application">Recoll</span> 1.22 and later).</p>
</li>
<li class="listitem">
<p><code class="literal">o</code> can be used to
specify a "slack" for phrase and proximity
searches: the number of additional terms that may
be found between the specified ones. If
<code class="literal">o</code> is followed by an
integer number, this is the slack, else the default
is 10.</p>
</li>
<li class="listitem">
<p><code class="literal">p</code> can be used to
turn the default phrase search into a proximity one
(unordered). Example: <code class="literal">"order
any in"p</code></p>
</li>
<li class="listitem">
<p><code class="literal">C</code> will turn on case
sensitivity (if the index supports it).</p>
</li>
<li class="listitem">
<p><code class="literal">D</code> will turn on
diacritics sensitivity (if the index supports
it).</p>
</li>
<li class="listitem">
<p>A weight can be specified for a query element by
specifying a decimal value at the start of the
modifiers. Example: <code class=
"literal">"Important"2.5</code>.</p>
</li>
</ul>
</div>
</div>
</div>
<div class="sect1">
<div class="titlepage">
<div>
<div>
<h2 class="title" style="clear: both"><a name=
"RCL.SEARCH.ANCHORWILD" id=
"RCL.SEARCH.ANCHORWILD"></a>3.6.&nbsp;Anchored
searches and wildcards</h2>
</div>
</div>
</div>
<p>Some special characters are interpreted by <span class=
"application">Recoll</span> in search strings to expand or
specialize the search. Wildcards expand a root term in
controlled ways. Anchor characters can restrict a search to
succeed only if the match is found at or near the beginning
of the document or one of its fields.</p>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name="RCL.SEARCH.WILDCARDS"
id="RCL.SEARCH.WILDCARDS"></a>3.6.1.&nbsp;More
about wildcards</h3>
</div>
</div>
</div>
<p>All words entered in <span class=
"application">Recoll</span> search fields will be
processed for wildcard expansion before the request is
finally executed.</p>
<p>The wildcard characters are:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style=
"list-style-type: disc;">
<li class="listitem">
<p><code class="literal">*</code> which matches 0
or more characters.</p>
</li>
<li class="listitem">
<p><code class="literal">?</code> which matches a
single character.</p>
</li>
<li class="listitem">
<p><code class="literal">[]</code> which allow
defining sets of characters to be matched (ex:
<code class="literal">[</code><strong class=
"userinput"><code>abc</code></strong><code class=
"literal">]</code> matches a single character which
may be 'a' or 'b' or 'c', <code class=
"literal">[</code><strong class=
"userinput"><code>0-9</code></strong><code class=
"literal">]</code> matches any number.</p>
</li>
</ul>
</div>
<p>You should be aware of a few things when using
wildcards.</p>
<div class="itemizedlist">
<ul class="itemizedlist" style=
"list-style-type: disc;">
<li class="listitem">
<p>Using a wildcard character at the beginning of a
word can make for a slow search because
<span class="application">Recoll</span> will have
to scan the whole index term list to find the
matches. However, this is much less a problem for
field searches, and queries like <em class=
"replaceable"><code>author:*@domain.com</code></em>
can sometimes be very useful.</p>
</li>
<li class="listitem">
<p>For <span class="application">Recoll</span>
version 18 only, when working with a raw index
(preserving character case and diacritics), the
literal part of a wildcard expression will be
matched exactly for case and diacritics. This is
not true any more for versions 19 and later.</p>
</li>
<li class="listitem">
<p>Using a <code class="literal">*</code> at the
end of a word can produce more matches than you
would think, and strange search results. You can
use the <a class="link" href=
"#RCL.SEARCH.GUI.TERMEXPLORER" title=
"3.2.9.&nbsp;The term explorer tool">term
explorer</a> tool to check what completions exist
for a given term. You can also see exactly what
search was performed by clicking on the link at the
top of the result list. In general, for natural
language terms, stem expansion will produce better
results than an ending <code class=
"literal">*</code> (stem expansion is turned off
when any wildcard character appears in the
term).</p>
</li>
</ul>
</div>
<div class="sect3">
<div class="titlepage">
<div>
<div>
<h4 class="title"><a name=
"RCL.SEARCH.WILDCARDS.PATH" id=
"RCL.SEARCH.WILDCARDS.PATH"></a>Wildcards and
path filtering</h4>
</div>
</div>
</div>
<p>Due to the way that <span class=
"application">Recoll</span> processes wildcards inside
<code class="literal">dir</code> path filtering
clauses, they will have a multiplicative effect on the
query size. A clause containing wildcards in several
paths elements, like, for example, <code class=
"literal">dir:</code><em class=
"replaceable"><code>/home/me/*/*/docdir</code></em>,
will almost certainly fail if your indexed tree is of
any realistic size.</p>
<p>Depending on the case, you may be able to work
around the issue by specifying the paths elements more
narrowly, with a constant prefix, or by using 2
separate <code class="literal">dir:</code> clauses
instead of multiple wildcards, as in <code class=
"literal">dir:</code><em class=
"replaceable"><code>/home/me</code></em> <code class=
"literal">dir:</code><em class=
"replaceable"><code>docdir</code></em>. The latter
query is not equivalent to the initial one because it
does not specify a number of directory levels, but
that's the best we can do (and it may be actually more
useful in some cases).</p>
</div>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name="RCL.SEARCH.ANCHOR" id=
"RCL.SEARCH.ANCHOR"></a>3.6.2.&nbsp;Anchored
searches</h3>
</div>
</div>
</div>
<p>Two characters are used to specify that a search hit
should occur at the beginning or at the end of the text.
<code class="literal">^</code> at the beginning of a term
or phrase constrains the search to happen at the start,
<code class="literal">$</code> at the end force it to
happen at the end.</p>
<p>As this function is implemented as a phrase search it
is possible to specify a maximum distance at which the
hit should occur, either through the controls of the
advanced search panel, or using the query language, for
example, as in:</p>
<pre class="programlisting">"^someterm"o10</pre>
<p>which would force <code class=
"literal">someterm</code> to be found within 10 terms of
the start of the text. This can be combined with a field
search as in <code class=
"literal">somefield:"^someterm"o10</code> or <code class=
"literal">somefield:someterm$</code>.</p>
<p>This feature can also be used with an actual phrase
search, but in this case, the distance applies to the
whole phrase and anchor, so that, for example,
<code class="literal">bla bla my unexpected term</code>
at the beginning of the text would be a match for
<code class="literal">"^my term"o5</code>.</p>
<p>Anchored searches can be very useful for searches
inside somewhat structured documents like scientific
articles, in case explicit metadata has not been supplied
(a most frequent case), for example for looking for
matches inside the abstract or the list of authors (which
occur at the top of the document).</p>
</div>
</div>
<div class="sect1">
<div class="titlepage">
<div>
<div>
<h2 class="title" style="clear: both"><a name=
"RCL.SEARCH.SYNONYMS" id=
"RCL.SEARCH.SYNONYMS"></a>3.7.&nbsp;Using Synonyms
(1.22)</h2>
</div>
</div>
</div>
<p><b>Term synonyms and text search:&nbsp;</b>in general,
there are two main ways to use term synonyms for searching
text:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style="list-style-type: disc;">
<li class="listitem">
<p>At index creation time, they can be used to alter
the indexed terms, either increasing or decreasing
their number, by expanding the original terms to all
synonyms, or by reducing all synonym terms to a
canonical one.</p>
</li>
<li class="listitem">
<p>At query time, they can be used to match texts
containing terms which are synonyms of the ones
specified by the user, either by expanding the query
for all synonyms, or by reducing the user entry to
canonical terms (the latter only works if the
corresponding processing has been performed while
creating the index).</p>
</li>
</ul>
</div>
<p><span class="application">Recoll</span> only uses
synonyms at query time. A user query term which part of a
synonym group will be optionally expanded into an
<code class="literal">OR</code> query for all terms in the
group.</p>
<p>Synonym groups are defined inside ordinary text files.
Each line in the file defines a group.</p>
<p>Example:</p>
<pre class="programlisting">
hi hello "good morning"
# not sure about "au revoir" though. Is this english ?
bye goodbye "see you" \
"au revoir"
</pre>
<p>As usual, lines beginning with a <code class=
"literal">#</code> are comments, empty lines are ignored,
and lines can be continued by ending them with a
backslash.</p>
<p>Multi-word synonyms are supported, but be aware that
these will generate phrase queries, which may degrade
performance and will disable stemming expansion for the
phrase terms.</p>
<p>The contents of the synonyms file must be casefolded
(not only lowercased), because this is what expected at the
point in the query processing where it is used. There are a
few cases where this makes a difference, for example,
German sharp s should be expressed as <code class=
"literal">ss</code>, Greek final sigma as sigma. For
reference, Python3 has an easy way to casefold words
(str.casefold()).</p>
<p>The synonyms file can be specified in the <span class=
"guilabel">Search parameters</span> tab of the <span class=
"guilabel">GUI configuration</span> <span class=
"guilabel">Preferences</span> menu entry, or as an option
for command-line searches.</p>
<p>Once the file is defined, the use of synonyms can be
enabled or disabled directly from the <span class=
"guilabel">Preferences</span> menu.</p>
<p>The synonyms are searched for matches with user terms
after the latter are stem-expanded, but the contents of the
synonyms file itself is not subjected to stem expansion.
This means that a match will not be found if the form
present in the synonyms file is not present anywhere in the
document set (same with accents when using a raw
index).</p>
<p>The synonyms function is probably not going to help you
find your letters to Mr. Smith. It is best used for
domain-specific searches. For example, it was initially
suggested by a user performing searches among historical
documents: the synonyms file would contains nicknames and
aliases for each of the persons of interest.</p>
</div>
<div class="sect1">
<div class="titlepage">
<div>
<div>
<h2 class="title" style="clear: both"><a name=
"RCL.SEARCH.PTRANS" id=
"RCL.SEARCH.PTRANS"></a>3.8.&nbsp;Path
translations</h2>
</div>
</div>
</div>
<p>In some cases, the document paths stored inside the
index do not match the actual ones, so that document
previews and accesses will fail. This can occur in a number
of circumstances:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style="list-style-type: disc;">
<li class="listitem">
<p>When using multiple indexes it is a relatively
common occurrence that some will actually reside on a
remote volume, for example mounted via NFS. In this
case, the paths used to access the documents on the
local machine are not necessarily the same than the
ones used while indexing on the remote machine. For
example, <code class="filename">/home/me</code> may
have been used as a <code class=
"literal">topdirs</code> elements while indexing, but
the directory might be mounted as <code class=
"filename">/net/server/home/me</code> on the local
machine.</p>
</li>
<li class="listitem">
<p>The case may also occur with removable disks. It
is perfectly possible to configure an index to live
with the documents on the removable disk, but it may
happen that the disk is not mounted at the same place
so that the documents paths from the index are
invalid.</p>
</li>
<li class="listitem">
<p>As a last example, one could imagine that a big
directory has been moved, but that it is currently
inconvenient to run the indexer.</p>
</li>
</ul>
</div>
<p><span class="application">Recoll</span> has a facility
for rewriting access paths when extracting the data from
the index. The translations can be defined for the main
index and for any additional query index.</p>
<p>The path translation facility will be useful whenever
the documents paths seen by the indexer are not the same as
the ones which should be used at query time.</p>
<p>In the above NFS example, <span class=
"application">Recoll</span> could be instructed to rewrite
any <code class="filename">file:///home/me</code> URL from
the index to <code class=
"filename">file:///net/server/home/me</code>, allowing
accesses from the client.</p>
<p>The translations are defined in the <a class="link"
href="#RCL.INSTALL.CONFIG.PTRANS" title=
"5.4.7.&nbsp;The ptrans file"><code class=
"filename">ptrans</code></a> configuration file, which can
be edited by hand or from the GUI external indexes
configuration dialog: <span class=
"guimenu">Preferences</span><span class=
"guimenuitem">External index dialog</span>, then click the
<span class="guilabel">Paths translations</span> button on
the right below the index list.</p>
<div class="note" style=
"margin-left: 0.5in; margin-right: 0.5in;">
<h3 class="title">Note</h3>
<p>Due to a current bug, the GUI must be restarted after
changing the <code class="filename">ptrans</code> values
(even when they were changed from the GUI).</p>
</div>
</div>
<div class="sect1">
<div class="titlepage">
<div>
<div>
<h2 class="title" style="clear: both"><a name=
"RCL.SEARCH.CASEDIAC" id=
"RCL.SEARCH.CASEDIAC"></a>3.9.&nbsp;Search case and
diacritics sensitivity</h2>
</div>
</div>
</div>
<p>For <span class="application">Recoll</span> versions
1.18 and later, and <span class="emphasis"><em>when working
with a raw index</em></span> (not the default), searches
can be sensitive to character case and diacritics. How this
happens is controlled by configuration variables and what
search data is entered.</p>
<p>The general default is that searches entered without
upper-case or accented characters are insensitive to case
and diacritics. An entry of <code class=
"literal">resume</code> will match any of <code class=
"literal">Resume</code>, <code class=
"literal">RESUME</code>, <code class=
"literal">résumé</code>, <code class=
"literal">Résumé</code> etc.</p>
<p>Two configuration variables can automate switching on
sensitivity (they were documented but actually did nothing
until <span class="application">Recoll</span> 1.22):</p>
<div class="variablelist">
<dl class="variablelist">
<dt><span class="term">autodiacsens</span></dt>
<dd>
<p>If this is set, search sensitivity to diacritics
will be turned on as soon as an accented character
exists in a search term. When the variable is set to
true, <code class="literal">resume</code> will start
a diacritics-unsensitive search, but <code class=
"literal">résumé</code> will be matched exactly. The
default value is <span class=
"emphasis"><em>false</em></span>.</p>
</dd>
<dt><span class="term">autocasesens</span></dt>
<dd>
<p>If this is set, search sensitivity to character
case will be turned on as soon as an upper-case
character exists in a search term <span class=
"emphasis"><em>except for the first one</em></span>.
When the variable is set to true, <code class=
"literal">us</code> or <code class=
"literal">Us</code> will start a
diacritics-unsensitive search, but <code class=
"literal">US</code> will be matched exactly. The
default value is <span class=
"emphasis"><em>true</em></span> (contrary to
<code class="literal">autodiacsens</code>).</p>
</dd>
</dl>
</div>
<p>As in the past, capitalizing the first letter of a word
will turn off its stem expansion and have no effect on
case-sensitivity.</p>
<p>You can also explicitly activate case and diacritics
sensitivity by using modifiers with the query language.
<code class="literal">C</code> will make the term
case-sensitive, and <code class="literal">D</code> will
make it diacritics-sensitive. Examples:</p>
<pre class="programlisting">
"us"C
</pre>
<p>will search for the term <code class="literal">us</code>
exactly (<code class="literal">Us</code> will not be a
match).</p>
<pre class="programlisting">
"resume"D
</pre>
<p>will search for the term <code class=
"literal">resume</code> exactly (<code class=
"literal">résumé</code> will not be a match).</p>
<p>When either case or diacritics sensitivity is activated,
stem expansion is turned off. Having both does not make
much sense.</p>
</div>
<div class="sect1">
<div class="titlepage">
<div>
<div>
<h2 class="title" style="clear: both"><a name=
"RCL.SEARCH.DESKTOP" id=
"RCL.SEARCH.DESKTOP"></a>3.10.&nbsp;Desktop
integration</h2>
</div>
</div>
</div>
<p>Being independent of the desktop type has its drawbacks:
<span class="application">Recoll</span> desktop integration
is minimal. However there are a few tools available:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style="list-style-type: disc;">
<li class="listitem">
<p>Users of recent Ubuntu-derived distributions, or
any other Gnome desktop systems (e.g. Fedora) can
install the <a class="ulink" href=
"https://www.lesbonscomptes.com/recoll/pages/download.html#gssp"
target="_top">Recoll GSSP</a> (Gnome Shell Search
Provider).</p>
</li>
<li class="listitem">
<p>The <span class="application">KDE</span> KIO Slave
was described in a <a class="link" href=
"#RCL.SEARCH.KIO" title=
"3.3.&nbsp;Searching with the KDE KIO slave">previous
section</a>. It can provide search results inside
<span class=
"command"><strong>Dolphin</strong></span>.</p>
</li>
<li class="listitem">
<p>If you use an oldish version of Ubuntu Linux, you
may find the <a class="ulink" href=
"https://www.lesbonscomptes.com/recoll/faqsandhowtos/UnityLens"
target="_top">Ubuntu Unity Lens</a> module
useful.</p>
</li>
<li class="listitem">
<p>There is also an independently developed <a class=
"ulink" href=
"http://kde-apps.org/content/show.php/recollrunner?content=128203"
target="_top">Krunner plugin</a>.</p>
</li>
</ul>
</div>
<p>Here follow a few other things that may help.</p>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name="RCL.SEARCH.SHORTCUT" id=
"RCL.SEARCH.SHORTCUT"></a>3.10.1.&nbsp;Hotkeying
recoll</h3>
</div>
</div>
</div>
<p>It is surprisingly convenient to be able to show or
hide the <span class="application">Recoll</span> GUI with
a single keystroke. Recoll comes with a small Python
script, based on the <span class=
"application">libwnck</span> window manager interface
library, which will allow you to do just this. The
detailed instructions are on <a class="ulink" href=
"https://www.lesbonscomptes.com/recoll/faqsandhowtos/HotRecoll"
target="_top">this wiki page</a>.</p>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name="RCL.KICKER-APPLET" id=
"RCL.KICKER-APPLET"></a>3.10.2.&nbsp;The KDE Kicker
Recoll applet</h3>
</div>
</div>
</div>
<p>This is probably obsolete now. Anyway:</p>
<p>The <span class="application">Recoll</span> source
tree contains the source code to the <span class=
"application">recoll_applet</span>, a small application
derived from the <span class=
"application">find_applet</span>. This can be used to add
a small <span class="application">Recoll</span> launcher
to the KDE panel.</p>
<p>The applet is not automatically built with the main
<span class="application">Recoll</span> programs, nor is
it included with the main source distribution (because
the KDE build boilerplate makes it relatively big). You
can download its source from the recoll.org download
page. Use the omnipotent <strong class=
"userinput"><code>configure;make;make
install</code></strong> incantation to build and
install.</p>
<p>You can then add the applet to the panel by
right-clicking the panel and choosing the <span class=
"guilabel">Add applet</span> entry.</p>
<p>The <span class="application">recoll_applet</span> has
a small text window where you can type a <span class=
"application">Recoll</span> query (in query language
form), and an icon which can be used to restrict the
search to certain types of files. It is quite primitive,
and launches a new recoll GUI instance every time (even
if it is already running). You may find it useful
anyway.</p>
</div>
</div>
</div>
<div class="chapter">
<div class="titlepage">
<div>
<div>
<h1 class="title"><a name="RCL.PROGRAM" id=
"RCL.PROGRAM"></a>Chapter&nbsp;4.&nbsp;Programming
interface</h1>
</div>
</div>
</div>
<p><span class="application">Recoll</span> has an Application
Programming Interface, usable both for indexing and
searching, currently accessible from the <span class=
"application">Python</span> language.</p>
<p>Another less radical way to extend the application is to
write input handlers for new types of documents.</p>
<p>The processing of metadata attributes for documents
(<code class="literal">fields</code>) is highly
configurable.</p>
<div class="sect1">
<div class="titlepage">
<div>
<div>
<h2 class="title" style="clear: both"><a name=
"RCL.PROGRAM.FILTERS" id=
"RCL.PROGRAM.FILTERS"></a>4.1.&nbsp;Writing a
document input handler</h2>
</div>
</div>
</div>
<div class="note" style=
"margin-left: 0.5in; margin-right: 0.5in;">
<h3 class="title">Terminology</h3>
<p>The small programs or pieces of code which handle the
processing of the different document types for
<span class="application">Recoll</span> used to be called
<code class="literal">filters</code>, which is still
reflected in the name of the directory which holds them
and many configuration variables. They were named this
way because one of their primary functions is to filter
out the formatting directives and keep the text content.
However these modules may have other behaviours, and the
term <code class="literal">input handler</code> is now
progressively substituted in the documentation.
<code class="literal">filter</code> is still used in many
places though.</p>
</div>
<p><span class="application">Recoll</span> input handlers
cooperate to translate from the multitude of input document
formats, simple ones as <span class=
"application">opendocument</span>, <span class=
"application">acrobat</span>, or compound ones such as
<span class="application">Zip</span> or <span class=
"application">Email</span>, into the final <span class=
"application">Recoll</span> indexing input format, which is
plain text (in many cases the processing pipeline has an
intermediary HTML step, which may be used for better
previewing presentation). Most input handlers are
executable programs or scripts. A few handlers are coded in
C++ and live inside <span class=
"command"><strong>recollindex</strong></span>. This latter
kind will not be described here.</p>
<p>There are two kinds of external executable input
handlers:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style="list-style-type: disc;">
<li class="listitem">
<p>Simple <code class="literal">exec</code> handlers
run once and exit. They can be bare programs like
<span class=
"command"><strong>antiword</strong></span>, or
scripts using other programs. They are very simple to
write, because they just need to print the converted
document to the standard output. Their output can be
plain text or HTML. HTML is usually preferred because
it can store metadata fields and it allows preserving
some of the formatting for the GUI preview. However,
these handlers have limitations:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style=
"list-style-type: circle;">
<li class="listitem">
<p>They can only process one document per
file.</p>
</li>
<li class="listitem">
<p>The output MIME type must be known and
fixed.</p>
</li>
<li class="listitem">
<p>The character encoding, if relevant, must be
known and fixed (or possibly just depending on
location).</p>
</li>
</ul>
</div>
</li>
<li class="listitem">
<p>Multiple <code class="literal">execm</code>
handlers can process multiple files (sparing the
process startup time which can be very significant),
or multiple documents per file (e.g.: for archives or
multi-chapter publications). They communicate with
the indexer through a simple protocol, but are
nevertheless a bit more complicated than the older
kind. Most of the new handlers are written in
<span class="application">Python</span> (exception:
<span class="command"><strong>rclimg</strong></span>
which is written in Perl because <code class=
"literal">exiftool</code> has no real Python
equivalent). The Python handlers use common modules
to factor out the boilerplate, which can make them
very simple in favorable cases. The subdocuments
output by these handlers can be directly indexable
(text or HTML), or they can be other simple or
compound documents that will need to be processed by
another handler.</p>
</li>
</ul>
</div>
<p>In both cases, handlers deal with regular file system
files, and can process either a single document, or a
linear list of documents in each file. <span class=
"application">Recoll</span> is responsible for performing
up to date checks, deal with more complex embedding and
other upper level issues.</p>
<p>A simple handler returning a document in <code class=
"literal">text/plain</code> format, can transfer no
metadata to the indexer. Generic metadata, like document
size or modification date, will be gathered and stored by
the indexer.</p>
<p>Handlers that produce <code class=
"literal">text/html</code> format can return an arbitrary
amount of metadata inside HTML <code class=
"literal">meta</code> tags. These will be processed
according to the directives found in the <a class="link"
href="#RCL.PROGRAM.FIELDS" title=
"4.2.&nbsp;Field data processing"><code class=
"filename">fields</code> configuration file</a>.</p>
<p>The handlers that can handle multiple documents per file
return a single piece of data to identify each document
inside the file. This piece of data, called an <code class=
"literal">ipath</code> will be sent back by <span class=
"application">Recoll</span> to extract the document at
query time, for previewing, or for creating a temporary
file to be opened by a viewer. These handlers can also
return metadata either as HTML <code class=
"literal">meta</code> tags, or as named data through the
communication protocol.</p>
<p>The following section describes the simple handlers, and
the next one gives a few explanations about the
<code class="literal">execm</code> ones. You could
conceivably write a simple handler with only the elements
in the manual. This will not be the case for the other
ones, for which you will have to look at the code.</p>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.PROGRAM.FILTERS.SIMPLE" id=
"RCL.PROGRAM.FILTERS.SIMPLE"></a>4.1.1.&nbsp;Simple
input handlers</h3>
</div>
</div>
</div>
<p><span class="application">Recoll</span> simple
handlers are usually shell-scripts, but this is in no way
necessary. Extracting the text from the native format is
the difficult part. Outputting the format expected by
<span class="application">Recoll</span> is trivial.
Happily enough, most document formats have translators or
text extractors which can be called from the handler. In
some cases the output of the translating program is
completely appropriate, and no intermediate shell-script
is needed.</p>
<p>Input handlers are called with a single argument which
is the source file name. They should output the result to
stdout.</p>
<p>When writing a handler, you should decide if it will
output plain text or HTML. Plain text is simpler, but you
will not be able to add metadata or vary the output
character encoding (this will be defined in a
configuration file). Additionally, some formatting may be
easier to preserve when previewing HTML. Actually the
deciding factor is metadata: <span class=
"application">Recoll</span> has a way to <a class="link"
href="#RCL.PROGRAM.FILTERS.HTML" title=
"4.1.4.&nbsp;Input handler output">extract metadata from
the HTML header and use it for field searches.</a>.</p>
<p>The <code class=
"envar">RECOLL_FILTER_FORPREVIEW</code> environment
variable (values <code class="literal">yes</code>,
<code class="literal">no</code>) tells the handler if the
operation is for indexing or previewing. Some handlers
use this to output a slightly different format, for
example stripping uninteresting repeated keywords (ie:
<code class="literal">Subject:</code> for email) when
indexing. This is not essential.</p>
<p>You should look at one of the simple handlers, for
example <span class=
"command"><strong>rclps</strong></span> for a starting
point.</p>
<p>Don't forget to make your handler executable before
testing !</p>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.PROGRAM.FILTERS.MULTIPLE" id=
"RCL.PROGRAM.FILTERS.MULTIPLE"></a>4.1.2.&nbsp;"Multiple"
handlers</h3>
</div>
</div>
</div>
<p>If you can program and want to write an <code class=
"literal">execm</code> handler, it should not be too
difficult to make sense of one of the existing
handlers.</p>
<p>The existing handlers differ in the amount of helper
code which they are using:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style=
"list-style-type: disc;">
<li class="listitem">
<p><code class="literal">rclimg</code> is written
in Perl and handles the execm protocol all by
itself (showing how trivial it is).</p>
</li>
<li class="listitem">
<p>All the Python handlers share at least the
<code class="filename">rclexecm.py</code> module,
which handles the communication. Have a look at,
for example, <code class="filename">rclzip</code>
for a handler which uses <code class=
"filename">rclexecm.py</code> directly.</p>
</li>
<li class="listitem">
<p>Most Python handlers which process
single-document files by executing another command
are further abstracted by using the <code class=
"filename">rclexec1.py</code> module. See for
example <code class="filename">rclrtf.py</code> for
a simple one, or <code class=
"filename">rcldoc.py</code> for a slightly more
complicated one (possibly executing several
commands).</p>
</li>
<li class="listitem">
<p>Handlers which extract text from an XML document
by using an XSLT style sheet are now executed
inside <span class=
"command"><strong>recollindex</strong></span>, with
only the style sheet stored in the <code class=
"filename">filters/</code> directory. These can use
a single style sheet (e.g. <code class=
"filename">abiword.xsl</code>), or two sheets for
the data and metadata (e.g. <code class=
"filename">opendoc-body.xsl</code> and <code class=
"filename">opendoc-meta.xsl</code>). The
<code class="filename">mimeconf</code>
configuration file defines how the sheets are used,
have a look. Before the C++ import, the xsl-based
handlers used a common module <code class=
"filename">rclgenxslt.py</code>, it is still around
but unused at the moment. The handler for OpenXML
presentations is still the Python version because
the format did not fit with what the C++ code does.
It would be a good base for another similar
issue.</p>
</li>
</ul>
</div>
<p>There is a sample trivial handler based on
<code class="filename">rclexecm.py</code>, with many
comments, not actually used by <span class=
"application">Recoll</span>. It would index a text file
as one document per line. Look for <code class=
"filename">rcltxtlines.py</code> in the <code class=
"filename">src/filters</code> directory in the online
<span class="application">Recoll</span> <a class="ulink"
href="https://framagit.org/medoc92/recoll" target=
"_top">Git repository</a> (the sample not in the
distributed release at the moment).</p>
<p>You can also have a look at the slightly more complex
<span class="command"><strong>rclzip</strong></span>
which uses Zip file paths as identifiers (<code class=
"literal">ipath</code>).</p>
<p><code class="literal">execm</code> handlers sometimes
need to make a choice for the nature of the <code class=
"literal">ipath</code> elements that they use in
communication with the indexer. Here are a few
guidelines:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style=
"list-style-type: disc;">
<li class="listitem">
<p>Use ASCII or UTF-8 (if the identifier is an
integer print it, for example, like printf %d would
do).</p>
</li>
<li class="listitem">
<p>If at all possible, the data should make some
kind of sense when printed to a log file to help
with debugging.</p>
</li>
<li class="listitem">
<p><span class="application">Recoll</span> uses a
colon (<code class="literal">:</code>) as a
separator to store a complex path internally (for
deeper embedding). Colons inside the <code class=
"literal">ipath</code> elements output by a handler
will be escaped, but would be a bad choice as a
handler-specific separator (mostly, again, for
debugging issues).</p>
</li>
</ul>
</div>
<p>In any case, the main goal is that it should be easy
for the handler to extract the target document, given the
file name and the <code class="literal">ipath</code>
element.</p>
<p><code class="literal">execm</code> handlers will also
produce a document with a null <code class=
"literal">ipath</code> element. Depending on the type of
document, this may have some associated data (e.g. the
body of an email message), or none (typical for an
archive file). If it is empty, this document will be
useful anyway for some operations, as the parent of the
actual data documents.</p>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.PROGRAM.FILTERS.ASSOCIATION" id=
"RCL.PROGRAM.FILTERS.ASSOCIATION"></a>4.1.3.&nbsp;Telling
<span class="application">Recoll</span> about the
handler</h3>
</div>
</div>
</div>
<p>There are two elements that link a file to the handler
which should process it: the association of file to MIME
type and the association of a MIME type with a
handler.</p>
<p>The association of files to MIME types is mostly based
on name suffixes. The types are defined inside the
<a class="link" href="#RCL.INSTALL.CONFIG.MIMEMAP" title=
"5.4.4.&nbsp;The mimemap file"><code class=
"filename">mimemap</code> file</a>. Example:</p>
<pre class="programlisting">
.doc = application/msword
</pre>
<p>If no suffix association is found for the file name,
<span class="application">Recoll</span> will try to
execute a system command (typically <span class=
"command"><strong>file -i</strong></span> or <span class=
"command"><strong>xdg-mime</strong></span>) to determine
a MIME type.</p>
<p>The second element is the association of MIME types to
handlers in the <a class="link" href=
"#RCL.INSTALL.CONFIG.MIMECONF" title=
"5.4.5.&nbsp;The mimeconf file"><code class=
"filename">mimeconf</code> file</a>. A sample will
probably be better than a long explanation:</p>
<pre class="programlisting">
[index]
application/msword = exec antiword -t -i 1 -m UTF-8;\
mimetype = text/plain ; charset=utf-8
application/ogg = exec rclogg
text/rtf = exec unrtf --nopict --html; charset=iso-8859-1; mimetype=text/html
application/x-chm = execm rclchm
</pre>
<p>The fragment specifies that:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style=
"list-style-type: disc;">
<li class="listitem">
<p><code class="literal">application/msword</code>
files are processed by executing the <span class=
"command"><strong>antiword</strong></span> program,
which outputs <code class=
"literal">text/plain</code> encoded in <code class=
"literal">utf-8</code>.</p>
</li>
<li class="listitem">
<p><code class="literal">application/ogg</code>
files are processed by the <span class=
"command"><strong>rclogg</strong></span> script,
with default output type (<code class=
"literal">text/html</code>, with encoding specified
in the header, or <code class=
"literal">utf-8</code> by default).</p>
</li>
<li class="listitem">
<p><code class="literal">text/rtf</code> is
processed by <span class=
"command"><strong>unrtf</strong></span>, which
outputs <code class="literal">text/html</code>. The
<code class="literal">iso-8859-1</code> encoding is
specified because it is not the <code class=
"literal">utf-8</code> default, and not output by
<span class="command"><strong>unrtf</strong></span>
in the HTML header section.</p>
</li>
<li class="listitem">
<p><code class="literal">application/x-chm</code>
is processed by a persistent handler. This is
determined by the <code class=
"literal">execm</code> keyword.</p>
</li>
</ul>
</div>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.PROGRAM.FILTERS.HTML" id=
"RCL.PROGRAM.FILTERS.HTML"></a>4.1.4.&nbsp;Input
handler output</h3>
</div>
</div>
</div>
<p>Both the simple and persistent input handlers can
return any MIME type to Recoll, which will further
process the data according to the MIME configuration.</p>
<p>Most input filters filters produce either <code class=
"literal">text/plain</code> or <code class=
"literal">text/html</code> data. There are exceptions,
for example, filters which process archive file
(<code class="literal">zip</code>, <code class=
"literal">tar</code>, etc.) will usually return the
documents as they are found, without processing them
further.</p>
<p>There is nothing to say about <code class=
"literal">text/plain</code> output, except that its
character encoding should be consistent with what is
specified in the <code class="filename">mimeconf</code>
file.</p>
<p>For filters producing HTML, the output could be very
minimal like the following example:</p>
<pre class="programlisting">
&lt;html&gt;
&lt;head&gt;
&lt;meta http-equiv="Content-Type" content="text/html;charset=UTF-8"/&gt;
&lt;/head&gt;
&lt;body&gt;
Some text content
&lt;/body&gt;
&lt;/html&gt;
</pre>
<p>You should take care to escape some characters inside
the text by transforming them into appropriate entities.
At the very minimum, "<code class="literal">&amp;</code>"
should be transformed into "<code class=
"literal">&amp;amp;</code>", "<code class=
"literal">&lt;</code>" should be transformed into
"<code class="literal">&amp;lt;</code>". This is not
always properly done by external helper programs which
output HTML, and of course never by those which output
plain text.</p>
<p>When encapsulating plain text in an HTML body, the
display of a preview may be improved by enclosing the
text inside <code class="literal">&lt;pre&gt;</code>
tags.</p>
<p>The character set needs to be specified in the header.
It does not need to be UTF-8 (<span class=
"application">Recoll</span> will take care of translating
it), but it must be accurate for good results.</p>
<p><span class="application">Recoll</span> will process
<code class="literal">meta</code> tags inside the header
as possible document fields candidates. Documents fields
can be processed by the indexer in different ways, for
searching or displaying inside query results. This is
described in a <a class="link" href="#RCL.PROGRAM.FIELDS"
title="4.2.&nbsp;Field data processing">following
section.</a></p>
<p>By default, the indexer will process the standard
header fields if they are present: <code class=
"literal">title</code>, <code class=
"literal">meta/description</code>, and <code class=
"literal">meta/keywords</code> are both indexed and
stored for query-time display.</p>
<p>A predefined non-standard <code class=
"literal">meta</code> tag will also be processed by
<span class="application">Recoll</span> without further
configuration: if a <code class="literal">date</code> tag
is present and has the right format, it will be used as
the document date (for display and sorting), in
preference to the file modification date. The date format
should be as follows:</p>
<pre class="programlisting">
&lt;meta name="date" content="YYYY-mm-dd HH:MM:SS"&gt;
or
&lt;meta name="date" content="YYYY-mm-ddTHH:MM:SS"&gt;
</pre>
<p>Example:</p>
<pre class="programlisting">
&lt;meta name="date" content="2013-02-24 17:50:00"&gt;
</pre>
<p>Input handlers also have the possibility to "invent"
field names. This should also be output as meta tags:</p>
<pre class="programlisting">
&lt;meta name="somefield" content="Some textual data" /&gt;
</pre>
<p>You can embed HTML markup inside the content of custom
fields, for improving the display inside result lists. In
this case, add a (wildly non-standard) <code class=
"literal">markup</code> attribute to tell <span class=
"application">Recoll</span> that the value is HTML and
should not be escaped for display.</p>
<pre class="programlisting">
&lt;meta name="somefield" markup="html" content="Some &lt;i&gt;textual&lt;/i&gt; data" /&gt;
</pre>
<p>As written above, the processing of fields is
described in a <a class="link" href="#RCL.PROGRAM.FIELDS"
title="4.2.&nbsp;Field data processing">further
section</a>.</p>
<p>Persistent filters can use another, probably simpler,
method to produce metadata, by calling the <code class=
"literal">setfield()</code> helper method. This avoids
the necessity to produce HTML, and any issue with HTML
quoting. See, for example, <code class=
"filename">rclaudio</code> in <span class=
"application">Recoll</span> 1.23 and later for an example
of handler which outputs <code class=
"literal">text/plain</code> and uses <code class=
"literal">setfield()</code> to produce metadata.</p>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.PROGRAM.FILTERS.PAGES" id=
"RCL.PROGRAM.FILTERS.PAGES"></a>4.1.5.&nbsp;Page
numbers</h3>
</div>
</div>
</div>
<p>The indexer will interpret <code class=
"literal">^L</code> characters in the handler output as
indicating page breaks, and will record them. At query
time, this allows starting a viewer on the right page for
a hit or a snippet. Currently, only the PDF, Postscript
and DVI handlers generate page breaks.</p>
</div>
</div>
<div class="sect1">
<div class="titlepage">
<div>
<div>
<h2 class="title" style="clear: both"><a name=
"RCL.PROGRAM.FIELDS" id=
"RCL.PROGRAM.FIELDS"></a>4.2.&nbsp;Field data
processing</h2>
</div>
</div>
</div>
<p><code class="literal">Fields</code> are named pieces of
information in or about documents, like <code class=
"literal">title</code>, <code class=
"literal">author</code>, <code class=
"literal">abstract</code>.</p>
<p>The field values for documents can appear in several
ways during indexing: either output by input handlers as
<code class="literal">meta</code> fields in the HTML header
section, or extracted from file extended attributes, or
added as attributes of the <code class="literal">Doc</code>
object when using the API, or again synthetized internally
by <span class="application">Recoll</span>.</p>
<p>The <span class="application">Recoll</span> query
language allows searching for text in a specific field.</p>
<p><span class="application">Recoll</span> defines a number
of default fields. Additional ones can be output by
handlers, and described in the <code class=
"filename">fields</code> configuration file.</p>
<p>Fields can be:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style="list-style-type: disc;">
<li class="listitem">
<p><code class="literal">indexed</code>, meaning that
their terms are separately stored in inverted lists
(with a specific prefix), and that a field-specific
search is possible.</p>
</li>
<li class="listitem">
<p><code class="literal">stored</code>, meaning that
their value is recorded in the index data record for
the document, and can be returned and displayed with
search results.</p>
</li>
</ul>
</div>
<p>A field can be either or both indexed and stored. This
and other aspects of fields handling is defined inside the
<code class="filename">fields</code> configuration
file.</p>
<p>Some fields may also designated as supporting range
queries, meaning that the results may be selected for an
interval of its values. See the <a class="link" href=
"#RCL.INSTALL.CONFIG.FIELDS" title=
"5.4.3.&nbsp;The fields file">configuration section</a> for
more details.</p>
<p>The sequence of events for field processing is as
follows:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style="list-style-type: disc;">
<li class="listitem">
<p>During indexing, <span class=
"command"><strong>recollindex</strong></span> scans
all <code class="literal">meta</code> fields in HTML
documents (most document types are transformed into
HTML at some point). It compares the name for each
element to the configuration defining what should be
done with fields (the <code class=
"filename">fields</code> file)</p>
</li>
<li class="listitem">
<p>If the name for the <code class=
"literal">meta</code> element matches one for a field
that should be indexed, the contents are processed
and the terms are entered into the index with the
prefix defined in the <code class=
"filename">fields</code> file.</p>
</li>
<li class="listitem">
<p>If the name for the <code class=
"literal">meta</code> element matches one for a field
that should be stored, the content of the element is
stored with the document data record, from which it
can be extracted and displayed at query time.</p>
</li>
<li class="listitem">
<p>At query time, if a field search is performed, the
index prefix is computed and the match is only
performed against appropriately prefixed terms in the
index.</p>
</li>
<li class="listitem">
<p>At query time, the field can be displayed inside
the result list by using the appropriate directive in
the definition of the <a class="link" href=
"#RCL.SEARCH.GUI.CUSTOM.RESLIST" title=
"The result list format">result list paragraph
format</a>. All fields are displayed on the fields
screen of the preview window (which you can reach
through the right-click menu). This is independent of
the fact that the search which produced the results
used the field or not.</p>
</li>
</ul>
</div>
<p>You can find more information in the <a class="link"
href="#RCL.INSTALL.CONFIG.FIELDS" title=
"5.4.3.&nbsp;The fields file">section about the
<code class="filename">fields</code> file</a>, or in
comments inside the file.</p>
<p>You can also have a look at the <a class="ulink" href=
"https://www.lesbonscomptes.com/recoll/faqsandhowtos/HandleCustomField"
target="_top">example in the FAQs area</a>, detailing how
one could add a <span class="emphasis"><em>page
count</em></span> field to pdf documents for displaying
inside result lists.</p>
</div>
<div class="sect1">
<div class="titlepage">
<div>
<div>
<h2 class="title" style="clear: both"><a name=
"RCL.PROGRAM.PYTHONAPI" id=
"RCL.PROGRAM.PYTHONAPI"></a>4.3.&nbsp;Python API</h2>
</div>
</div>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.PROGRAM.PYTHONAPI.INTRO" id=
"RCL.PROGRAM.PYTHONAPI.INTRO"></a>4.3.1.&nbsp;Introduction</h3>
</div>
</div>
</div>
<p>The <span class="application">Recoll</span> Python
programming interface can be used both for searching and
for creating/updating an index. Bindings exist for
Python2 and Python3 (Jan 2021: python2 support will be
dropped soon).</p>
<p>The search interface is used in a number of active
projects: the <a class="ulink" href=
"https://www.lesbonscomptes.com/recoll/pages/download.html#gssp"
target="_top"><span class="application">Recoll</span>
<span class="application">Gnome Shell Search
Provider</span></a> , the <a class="ulink" href=
"https://framagit.org/medoc92/recollwebui" target=
"_top"><span class="application">Recoll</span> Web
UI</a>, and the <a class="ulink" href=
"https://www.lesbonscomptes.com/upmpdcli/upmpdcli-manual.html#UPRCL"
target="_top">upmpdcli UPnP Media Server</a>, in addition
to many small scripts.</p>
<p>The index update section of the API may be used to
create and update <span class="application">Recoll</span>
indexes on specific configurations (separate from the
ones created by <span class=
"command"><strong>recollindex</strong></span>). The
resulting databases can be queried alone, or in
conjunction with regular ones, through the GUI or any of
the query interfaces.</p>
<p>The search API is modeled along the Python database
API version 2.0 specification (early versions used the
version 1.0 spec).</p>
<p>The <code class="literal">recoll</code> package
contains two modules:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style=
"list-style-type: disc;">
<li class="listitem">
<p>The <code class="literal">recoll</code> module
contains functions and classes used to query (or
update) the index.</p>
</li>
<li class="listitem">
<p>The <code class="literal">rclextract</code>
module contains functions and classes used at query
time to access document data. The <code class=
"literal">recoll</code> module must be imported
before <code class="literal">rclextract</code></p>
</li>
</ul>
</div>
<p>There is a good chance that your system repository has
packages for the Recoll Python API, sometimes in a
package separate from the main one (maybe named something
like python-recoll). Else refer to the <a class="link"
href="#RCL.INSTALL.BUILDING" title=
"5.3.&nbsp;Building from source">Building from source
chapter</a>.</p>
<p>As an introduction, the following small sample will
run a query and list the title and url for each of the
results. The <code class="filename">python/samples</code>
source directory contains several examples of Python
programming with <span class="application">Recoll</span>,
exercising the extension more completely, and especially
its data extraction features.</p>
<pre class="programlisting">
#!/usr/bin/python3
from recoll import recoll
db = recoll.connect()
query = db.query()
nres = query.execute("some query")
results = query.fetchmany(20)
for doc in results:
print("%s %s" % (doc.url, doc.title))
</pre>
<p>You can also take a look at the source for the
<a class="ulink" href=
"https://framagit.org/medoc92/recollwebui/-/blob/master/webui.py"
target="_top">Recoll WebUI</a>, the <a class="ulink"
href="https://framagit.org/medoc92/upmpdcli/-/blob/master/src/mediaserver/cdplugins/uprcl/uprclfolders.py"
target="_top">upmpdcli local media server</a>, or the
<a class="ulink" href=
"https://framagit.org/medoc92/recoll-gssp/-/blob/master/gssp-recoll.py"
target="_top">Gnome Shell Search Provider</a>.</p>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.PROGRAM.PYTHONAPI.ELEMENTS" id=
"RCL.PROGRAM.PYTHONAPI.ELEMENTS"></a>4.3.2.&nbsp;Interface
elements</h3>
</div>
</div>
</div>
<p>A few elements in the interface are specific and and
need an explanation.</p>
<div class="variablelist">
<dl class="variablelist">
<dt><a name="RCL.PROGRAM.PYTHONAPI.ELEMENTS.IPATH"
id="RCL.PROGRAM.PYTHONAPI.ELEMENTS.IPATH"></a><span class="term">ipath</span></dt>
<dd>
<p>This data value (set as a field in the Doc
object) is stored, along with the URL, but not
indexed by <span class="application">Recoll</span>.
Its contents are not interpreted by the index
layer, and its use is up to the application. For
example, the <span class=
"application">Recoll</span> file system indexer
uses the <code class="literal">ipath</code> to
store the part of the document access path internal
to (possibly imbricated) container documents.
<code class="literal">ipath</code> in this case is
a vector of access elements (e.g, the first part
could be a path inside a zip file to an archive
member which happens to be an mbox file, the second
element would be the message sequential number
inside the mbox etc.). <code class=
"literal">url</code> and <code class=
"literal">ipath</code> are returned in every search
result and define the access to the original
document. <code class="literal">ipath</code> is
empty for top-level document/files (e.g. a PDF
document which is a filesystem file). The
<span class="application">Recoll</span> GUI knows
about the structure of the <code class=
"literal">ipath</code> values used by the
filesystem indexer, and uses it for such functions
as opening the parent of a given document.</p>
</dd>
<dt><a name="RCL.PROGRAM.PYTHONAPI.ELEMENTS.UDI" id=
"RCL.PROGRAM.PYTHONAPI.ELEMENTS.UDI"></a><span class=
"term">udi</span></dt>
<dd>
<p>An <code class="literal">udi</code> (unique
document identifier) identifies a document. Because
of limitations inside the index engine, it is
restricted in length (to 200 bytes), which is why a
regular URI cannot be used. The structure and
contents of the <code class="literal">udi</code> is
defined by the application and opaque to the index
engine. For example, the internal file system
indexer uses the complete document path (file path
+ internal path), truncated to length, the
suppressed part being replaced by a hash value. The
<code class="literal">udi</code> is not explicit in
the query interface (it is used "under the hood" by
the <code class="filename">rclextract</code>
module), but it is an explicit element of the
update interface.</p>
</dd>
<dt><a name=
"RCL.PROGRAM.PYTHONAPI.ELEMENTS.PARENTUDI" id=
"RCL.PROGRAM.PYTHONAPI.ELEMENTS.PARENTUDI"></a><span class="term">parent_udi</span></dt>
<dd>
<p>If this attribute is set on a document when
entering it in the index, it designates its
physical container document. In a multilevel
hierarchy, this may not be the immediate parent.
<code class="literal">parent_udi</code> is
optional, but its use by an indexer may simplify
index maintenance, as <span class=
"application">Recoll</span> will automatically
delete all children defined by <code class=
"literal">parent_udi == udi</code> when the
document designated by <code class=
"literal">udi</code> is destroyed. e.g. if a
<code class="literal">Zip</code> archive contains
entries which are themselves containers, like
<code class="literal">mbox</code> files, all the
subdocuments inside the <code class=
"literal">Zip</code> file (mbox, messages, message
attachments, etc.) would have the same <code class=
"literal">parent_udi</code>, matching the
<code class="literal">udi</code> for the
<code class="literal">Zip</code> file, and all
would be destroyed when the <code class=
"literal">Zip</code> file (identified by its
<code class="literal">udi</code>) is removed from
the index. The standard filesystem indexer uses
<code class="literal">parent_udi</code>.</p>
</dd>
<dt><span class="term">Stored and indexed
fields</span></dt>
<dd>
<p>The <a class="link" href=
"#RCL.INSTALL.CONFIG.FIELDS" title=
"5.4.3.&nbsp;The fields file"><code class=
"filename">fields</code> file</a> inside the
<span class="application">Recoll</span>
configuration defines which document fields are
either <code class="literal">indexed</code>
(searchable), <code class="literal">stored</code>
(retrievable with search results), or both. Apart
from a few standard/internal fields, only the
<code class="literal">stored</code> fields are
retrievable through the Python search
interface.</p>
</dd>
</dl>
</div>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.PROGRAM.PYTHONAPI.LOG" id=
"RCL.PROGRAM.PYTHONAPI.LOG"></a>4.3.3.&nbsp;Log
messages for Python scripts</h3>
</div>
</div>
</div>
<p>Two specific configuration variables: <code class=
"literal">pyloglevel</code> and <code class=
"literal">pylogfilename</code> allow overriding the
generic values for Python programs. Set <code class=
"literal">pyloglevel</code> to 2 to suppress default
startup messages (printed at level 3).</p>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.PROGRAM.PYTHONAPI.SEARCH" id=
"RCL.PROGRAM.PYTHONAPI.SEARCH"></a>4.3.4.&nbsp;Python
search interface</h3>
</div>
</div>
</div>
<div class="sect3">
<div class="titlepage">
<div>
<div>
<h4 class="title"><a name=
"RCL.PROGRAM.PYTHONAPI.RECOLL" id=
"RCL.PROGRAM.PYTHONAPI.RECOLL"></a>The recoll
module</h4>
</div>
</div>
</div>
<div class="simplesect">
<div class="titlepage">
<div>
<div>
<h5 class="title"><a name=
"RCL.PROGRAM.PYTHONAPI.RECOLL.CONNECT" id=
"RCL.PROGRAM.PYTHONAPI.RECOLL.CONNECT"></a>connect(confdir=None,
extra_dbs=None, writable = False)</h5>
</div>
</div>
</div>
<p>The <code class="literal">connect()</code>
function connects to one or several <span class=
"application">Recoll</span> index(es) and returns a
<code class="literal">Db</code> object.</p>
<p>This call initializes the recoll module, and it
should always be performed before any other call or
object creation.</p>
<div class="itemizedlist">
<ul class="itemizedlist" style=
"list-style-type: disc;">
<li class="listitem">
<p><code class="literal">confdir</code> may
specify a configuration directory. The usual
defaults apply.</p>
</li>
<li class="listitem">
<p><code class="literal">extra_dbs</code> is a
list of additional indexes (Xapian
directories).</p>
</li>
<li class="listitem">
<p><code class="literal">writable</code>
decides if we can index new data through this
connection.</p>
</li>
</ul>
</div>
</div>
<div class="simplesect">
<div class="titlepage">
<div>
<div>
<h5 class="title"><a name=
"RCL.PROGRAM.PYTHONAPI.RECOLL.DB" id=
"RCL.PROGRAM.PYTHONAPI.RECOLL.DB"></a>The Db
class</h5>
</div>
</div>
</div>
<p>A Db object is created by a <code class=
"literal">connect()</code> call and holds a
connection to a Recoll index.</p>
<div class="variablelist">
<dl class="variablelist">
<dt><span class="term">Db.close()</span></dt>
<dd>
<p>Closes the connection. You can't do anything
with the <code class="literal">Db</code> object
after this.</p>
</dd>
<dt><span class="term">Db.query(),
Db.cursor()</span></dt>
<dd>
<p>These aliases return a blank <code class=
"literal">Query</code> object for this
index.</p>
</dd>
<dt><span class=
"term">Db.setAbstractParams(maxchars,
contextwords)</span></dt>
<dd>
<p>Set the parameters used to build snippets
(sets of keywords in context text fragments).
<code class="literal">maxchars</code> defines
the maximum total size of the abstract.
<code class="literal">contextwords</code>
defines how many terms are shown around the
keyword.</p>
</dd>
<dt><span class="term">Db.termMatch(match_type,
expr, field='', maxlen=-1, casesens=False,
diacsens=False, lang='english')</span></dt>
<dd>
<p>Expand an expression against the index term
list. Performs the basic function from the GUI
term explorer tool. <code class=
"literal">match_type</code> can be either of
<code class="literal">wildcard</code>,
<code class="literal">regexp</code> or
<code class="literal">stem</code>. Returns a
list of terms expanded from the input
expression.</p>
</dd>
</dl>
</div>
</div>
<div class="simplesect">
<div class="titlepage">
<div>
<div>
<h5 class="title"><a name=
"RCL.PROGRAM.PYTHONAPI.RECOLL.CLASSES.QUERY"
id="RCL.PROGRAM.PYTHONAPI.RECOLL.CLASSES.QUERY">
</a>The Query class</h5>
</div>
</div>
</div>
<p>A <code class="literal">Query</code> object
(equivalent to a cursor in the Python DB API) is
created by a <code class="literal">Db.query()</code>
call. It is used to execute index searches.</p>
<div class="variablelist">
<dl class="variablelist">
<dt><span class="term">Query.sortby(fieldname,
ascending=True)</span></dt>
<dd>
<p>Sort results by <em class=
"replaceable"><code>fieldname</code></em>, in
ascending or descending order. Must be called
before executing the search.</p>
</dd>
<dt><span class=
"term">Query.execute(query_string, stemming=1,
stemlang="english", fetchtext=False,
collapseduplicates=False)</span></dt>
<dd>
<p>Starts a search for <em class=
"replaceable"><code>query_string</code></em>, a
<span class="application">Recoll</span> search
language string. If the index stores the
document texts and <code class=
"literal">fetchtext</code> is True, store the
document extracted text in <code class=
"literal">doc.text</code>.</p>
</dd>
<dt><span class=
"term">Query.executesd(SearchData,
fetchtext=False,
collapseduplicates=False)</span></dt>
<dd>
<p>Starts a search for the query defined by the
SearchData object. If the index stores the
document texts and <code class=
"literal">fetchtext</code> is True, store the
document extracted text in <code class=
"literal">doc.text</code>.</p>
</dd>
<dt><span class=
"term">Query.fetchmany(size=query.arraysize)</span></dt>
<dd>
<p>Fetches the next <code class=
"literal">Doc</code> objects in the current
search results, and returns them as an array of
the required size, which is by default the
value of the <code class=
"literal">arraysize</code> data member.</p>
</dd>
<dt><span class=
"term">Query.fetchone()</span></dt>
<dd>
<p>Fetches the next <code class=
"literal">Doc</code> object from the current
search results. Generates a StopIteration
exception if there are no results left.</p>
</dd>
<dt><span class="term">Query.close()</span></dt>
<dd>
<p>Closes the query. The object is unusable
after the call.</p>
</dd>
<dt><span class="term">Query.scroll(value,
mode='relative')</span></dt>
<dd>
<p>Adjusts the position in the current result
set. <code class="literal">mode</code> can be
<code class="literal">relative</code> or
<code class="literal">absolute</code>.</p>
</dd>
<dt><span class=
"term">Query.getgroups()</span></dt>
<dd>
<p>Retrieves the expanded query terms as a list
of pairs. Meaningful only after executexx In
each pair, the first entry is a list of user
terms (of size one for simple terms, or more
for group and phrase clauses), the second a
list of query terms as derived from the user
terms and used in the Xapian Query.</p>
</dd>
<dt><span class=
"term">Query.getxquery()</span></dt>
<dd>
<p>Return the Xapian query description as a
Unicode string. Meaningful only after
executexx.</p>
</dd>
<dt><span class="term">Query.highlight(text,
ishtml = 0, methods = object)</span></dt>
<dd>
<p>Will insert &lt;span "class=rclmatch"&gt;,
&lt;/span&gt; tags around the match areas in
the input text and return the modified text.
<code class="literal">ishtml</code> can be set
to indicate that the input text is HTML and
that HTML special characters should not be
escaped. <code class="literal">methods</code>
if set should be an object with methods
startMatch(i) and endMatch() which will be
called for each match and should return a begin
and end tag</p>
</dd>
<dt><span class="term">Query.makedocabstract(doc,
methods = object))</span></dt>
<dd>
<p>Create a snippets abstract for <code class=
"literal">doc</code> (a <code class=
"literal">Doc</code> object) by selecting text
around the match terms. If methods is set, will
also perform highlighting. See the highlight
method.</p>
</dd>
<dt><span class="term">Query.getsnippets(doc,
maxoccs = -1, ctxwords = -1, sortbypage=False,
methods = object)</span></dt>
<dd>
<p>Will return a list of extracts from the
result document by selecting text around the
match terms. Each entry in the result list is a
triple: page number, term, text. By default,
the most relevants snippets appear first in the
list. Set <code class=
"literal">sortbypage</code> to sort by page
number instead. If <code class=
"literal">methods</code> is set, the fragments
will be highlighted (see the highlight method).
If <code class="literal">maxoccs</code> is set,
it defines the maximum result list length.
<code class="literal">ctxwords</code> allows
adjusting the individual snippet context
size.</p>
</dd>
<dt><span class="term">Query.__iter__() and
Query.next()</span></dt>
<dd>
<p>So that things like <code class=
"literal">for doc in query:</code> will
work.</p>
</dd>
</dl>
</div>
<div class="variablelist">
<dl class="variablelist">
<dt><span class=
"term">Query.arraysize</span></dt>
<dd>
<p>Default number of records processed by
fetchmany (r/w).</p>
</dd>
<dt><span class="term">Query.rowcount</span></dt>
<dd>
<p>Number of records returned by the last
execute.</p>
</dd>
<dt><span class=
"term">Query.rownumber</span></dt>
<dd>
<p>Next index to be fetched from results.
Normally increments after each fetchone() call,
but can be set/reset before the call to effect
seeking (equivalent to using <code class=
"literal">scroll()</code>). Starts at 0.</p>
</dd>
</dl>
</div>
</div>
<div class="simplesect">
<div class="titlepage">
<div>
<div>
<h5 class="title"><a name=
"RCL.PROGRAM.PYTHONAPI.RECOLL.CLASSES.DOC" id=
"RCL.PROGRAM.PYTHONAPI.RECOLL.CLASSES.DOC"></a>The
Doc class</h5>
</div>
</div>
</div>
<p>A <code class="literal">Doc</code> object contains
index data for a given document. The data is
extracted from the index when searching, or set by
the indexer program when updating. The Doc object has
many attributes to be read or set by its user. It
mostly matches the Rcl::Doc C++ object. Some of the
attributes are predefined, but, especially when
indexing, others can be set, the name of which will
be processed as field names by the indexing
configuration. Inputs can be specified as Unicode or
strings. Outputs are Unicode objects. All dates are
specified as Unix timestamps, printed as strings.
Please refer to the <code class=
"filename">rcldb/rcldoc.cpp</code> C++ file for a
full description of the predefined attributes. Here
follows a short list.</p>
<div class="itemizedlist">
<ul class="itemizedlist" style=
"list-style-type: disc;">
<li class="listitem">
<p><code class="literal">url</code> the
document URL but see also <code class=
"literal">getbinurl()</code></p>
</li>
<li class="listitem">
<p><code class="literal">ipath</code> the
document <code class="literal">ipath</code> for
embedded documents.</p>
</li>
<li class="listitem">
<p><code class="literal">fbytes, dbytes</code>
the document file and text sizes.</p>
</li>
<li class="listitem">
<p><code class="literal">fmtime, dmtime</code>
the document file and document times.</p>
</li>
<li class="listitem">
<p><code class="literal">xdocid</code> the
document Xapian document ID. This is useful if
you want to access the document through a
direct Xapian operation.</p>
</li>
<li class="listitem">
<p><code class="literal">mtype</code> the
document MIME type.</p>
</li>
<li class="listitem">
<p>Fields stored by default: <code class=
"literal">author</code>, <code class=
"literal">filename</code>, <code class=
"literal">keywords</code>, <code class=
"literal">recipient</code></p>
</li>
</ul>
</div>
<p>At query time, only the fields that are defined as
<code class="literal">stored</code> either by default
or in the <code class="filename">fields</code>
configuration file will be meaningful in the
<code class="literal">Doc</code> object. The document
processed text may be present or not, depending if
the index stores the text at all, and if it does, on
the <code class="literal">fetchtext</code> query
execute option. See also the <code class=
"literal">rclextract</code> module for accessing
document contents.</p>
<div class="variablelist">
<dl class="variablelist">
<dt><span class="term">get(key), []
operator</span></dt>
<dd>
<p>Retrieve the named document attribute. You
can also use <code class="literal">getattr(doc,
key)</code> or <code class=
"literal">doc.key</code>.</p>
</dd>
<dt><span class="term">doc.key =
value</span></dt>
<dd>
<p>Set the the named document attribute. You
can also use <code class="literal">setattr(doc,
key, value)</code>.</p>
</dd>
<dt><span class="term">getbinurl()</span></dt>
<dd>
<p>Retrieve the URL in byte array format (no
transcoding), for use as parameter to a system
call.</p>
</dd>
<dt><span class="term">setbinurl(url)</span></dt>
<dd>
<p>Set the URL in byte array format (no
transcoding).</p>
</dd>
<dt><span class="term">items()</span></dt>
<dd>
<p>Return a dictionary of doc object
keys/values</p>
</dd>
<dt><span class="term">keys()</span></dt>
<dd>
<p>list of doc object keys (attribute
names).</p>
</dd>
</dl>
</div>
</div>
<div class="simplesect">
<div class="titlepage">
<div>
<div>
<h5 class="title"><a name=
"RCL.PROGRAM.PYTHONAPI.RECOLL.CLASSES.SEARCHDATA"
id=
"RCL.PROGRAM.PYTHONAPI.RECOLL.CLASSES.SEARCHDATA">
</a>The SearchData class</h5>
</div>
</div>
</div>
<p>A <code class="literal">SearchData</code> object
allows building a query by combining clauses, for
execution by <code class=
"literal">Query.executesd()</code>. It can be used in
replacement of the query language approach. The
interface is going to change a little, so no detailed
doc for now...</p>
<div class="variablelist">
<dl class="variablelist">
<dt><span class=
"term">addclause(type='and'|'or'|'excl'|'phrase'|'near'|'sub',
qstring=string, slack=0, field='', stemming=1,
subSearch=SearchData)</span></dt>
<dd></dd>
</dl>
</div>
</div>
</div>
<div class="sect3">
<div class="titlepage">
<div>
<div>
<h4 class="title"><a name=
"RCL.PROGRAM.PYTHONAPI.RCLEXTRACT" id=
"RCL.PROGRAM.PYTHONAPI.RCLEXTRACT"></a>The
rclextract module</h4>
</div>
</div>
</div>
<p>Prior to <span class="application">Recoll</span>
1.25, index queries could not provide document content
because it was never stored. <span class=
"application">Recoll</span> 1.25 and later usually
store the document text, which can be optionally
retrieved when running a query (see <code class=
"literal">query.execute()</code> above - the result is
always plain text).</p>
<p>The <code class="literal">rclextract</code> module
can give access to the original document and to the
document text content (if not stored by the index, or
to access an HTML version of the text). Accessing the
original document is particularly useful if it is
embedded (e.g. an email attachment).</p>
<p>You need to import the <code class=
"literal">recoll</code> module before the <code class=
"literal">rclextract</code> module.</p>
<div class="simplesect">
<div class="titlepage">
<div>
<div>
<h5 class="title"><a name=
"RCL.PROGRAM.PYTHONAPI.RCLEXTRACT.CLASSES.EXTRACTOR"
id=
"RCL.PROGRAM.PYTHONAPI.RCLEXTRACT.CLASSES.EXTRACTOR">
</a>The Extractor class</h5>
</div>
</div>
</div>
<div class="variablelist">
<dl class="variablelist">
<dt><span class="term">Extractor(doc)</span></dt>
<dd>
<p>An <code class="literal">Extractor</code>
object is built from a <code class=
"literal">Doc</code> object, output from a
query.</p>
</dd>
<dt><span class=
"term">Extractor.textextract(ipath)</span></dt>
<dd>
<p>Extract document defined by <em class=
"replaceable"><code>ipath</code></em> and
return a <code class="literal">Doc</code>
object. The <code class=
"literal">doc.text</code> field has the
document text converted to either text/plain or
text/html according to <code class=
"literal">doc.mimetype</code>. The typical use
would be as follows:</p>
<pre class="programlisting">
from recoll import recoll, rclextract
qdoc = query.fetchone()
extractor = recoll.Extractor(qdoc)
doc = extractor.textextract(qdoc.ipath)
# use doc.text, e.g. for previewing</pre>
<p>Passing <code class=
"literal">qdoc.ipath</code> to <code class=
"literal">textextract()</code> is redundant,
but reflects the fact that the <code class=
"literal">Extractor</code> object actually has
the capability to access the other entries in a
compound document.</p>
</dd>
<dt><span class=
"term">Extractor.idoctofile(ipath, targetmtype,
outfile='')</span></dt>
<dd>
<p>Extracts document into an output file, which
can be given explicitly or will be created as a
temporary file to be deleted by the caller.
Typical use:</p>
<pre class="programlisting">
from recoll import recoll, rclextract
qdoc = query.fetchone()
extractor = recoll.Extractor(qdoc)
filename = extractor.idoctofile(qdoc.ipath, qdoc.mimetype)</pre>
<p>In all cases the output is a copy, even if
the requested document is a regular system
file, which may be wasteful in some cases. If
you want to avoid this, you can test for a
simple file document as follows:</p>
<pre class="programlisting">
not doc.ipath and (not "rclbes" in doc.keys() or doc["rclbes"] == "FS")
</pre>
</dd>
</dl>
</div>
</div>
</div>
<div class="sect3">
<div class="titlepage">
<div>
<div>
<h4 class="title"><a name=
"RCL.PROGRAM.PYTHONAPI.SEARCH.EXAMPLE" id=
"RCL.PROGRAM.PYTHONAPI.SEARCH.EXAMPLE"></a>Search
API usage example</h4>
</div>
</div>
</div>
<p>The following sample would query the index with a
user language string. See the <code class=
"filename">python/samples</code> directory inside the
<span class="application">Recoll</span> source for
other examples. The <code class=
"filename">recollgui</code> subdirectory has a very
embryonic GUI which demonstrates the highlighting and
data extraction functions.</p>
<pre class="programlisting">
#!/usr/bin/python3
from recoll import recoll
db = recoll.connect()
db.setAbstractParams(maxchars=80, contextwords=4)
query = db.query()
nres = query.execute("some user question")
print("Result count: %d" % nres)
if nres &gt; 5:
nres = 5
for i in range(nres):
doc = query.fetchone()
print("Result #%d" % (query.rownumber))
for k in ("title", "size"):
print("%s : %s" % (k, getattr(doc, k)))
print("%s\n" % db.makeDocAbstract(doc, query))
</pre>
</div>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.PROGRAM.PYTHONAPI.UPDATE" id=
"RCL.PROGRAM.PYTHONAPI.UPDATE"></a>4.3.5.&nbsp;Creating
Python external indexers</h3>
</div>
</div>
</div>
<p>The update API can be used to create an index from
data which is not accessible to the regular <span class=
"application">Recoll</span> indexer, or structured to
present difficulties to the <span class=
"application">Recoll</span> input handlers.</p>
<p>An indexer created using this API will be have
equivalent work to do as the the Recoll file system
indexer: look for modified documents, extract their text,
call the API for indexing it, take care of purging the
index out of data from documents which do not exist in
the document store any more.</p>
<p>The data for such an external indexer should be stored
in an index separate from any used by the <span class=
"application">Recoll</span> internal file system indexer.
The reason is that the main document indexer purge pass
(removal of deleted documents) would also remove all the
documents belonging to the external indexer, as they were
not seen during the filesystem walk. The main indexer
documents would also probably be a problem for the
external indexer own purge operation.</p>
<p>While there would be ways to enable multiple foreign
indexers to cooperate on a single index, it is just
simpler to use separate ones, and use the multiple index
access capabilities of the query interface, if
needed.</p>
<p>There are two parts in the update interface:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style=
"list-style-type: disc;">
<li class="listitem">
<p>Methods inside the <code class=
"filename">recoll</code> module allow inserting
data into the index, to make it accessible by the
normal query interface.</p>
</li>
<li class="listitem">
<p>An interface based on scripts execution is
defined to allow either the GUI or the <code class=
"filename">rclextract</code> module to access
original document data for previewing or
editing.</p>
</li>
</ul>
</div>
<div class="sect3">
<div class="titlepage">
<div>
<div>
<h4 class="title"><a name=
"RCL.PROGRAM.PYTHONAPI.UPDATE.UPDATE" id=
"RCL.PROGRAM.PYTHONAPI.UPDATE.UPDATE"></a>Python
update interface</h4>
</div>
</div>
</div>
<p>The update methods are part of the <code class=
"filename">recoll</code> module described above. The
connect() method is used with a <code class=
"literal">writable=true</code> parameter to obtain a
writable <code class="literal">Db</code> object. The
following <code class="literal">Db</code> object
methods are then available.</p>
<div class="variablelist">
<dl class="variablelist">
<dt><span class="term">addOrUpdate(udi, doc,
parent_udi=None)</span></dt>
<dd>
<p>Add or update index data for a given document
The <code class="literal"><a class="link" href=
"#RCL.PROGRAM.PYTHONAPI.ELEMENTS.UDI">udi</a></code>
string must define a unique id for the document.
It is an opaque interface element and not
interpreted inside Recoll. <code class=
"literal">doc</code> is a <code class=
"literal"><a class="link" href=
"#RCL.PROGRAM.PYTHONAPI.RECOLL.CLASSES.DOC"
title="The Doc class">Doc</a></code> object,
created from the data to be indexed (the main
text should be in <code class=
"literal">doc.text</code>). If <code class=
"literal"><a class="link" href=
"#RCL.PROGRAM.PYTHONAPI.ELEMENTS.PARENTUDI">parent_udi</a></code>
is set, this is a unique identifier for the
top-level container (e.g. for the filesystem
indexer, this would be the one which is an actual
file).</p>
</dd>
<dt><span class="term">delete(udi)</span></dt>
<dd>
<p>Purge index from all data for <code class=
"literal">udi</code>, and all documents (if any)
which have a matrching <code class=
"literal">parent_udi</code>.</p>
</dd>
<dt><span class="term">needUpdate(udi,
sig)</span></dt>
<dd>
<p>Test if the index needs to be updated for the
document identified by <code class=
"literal">udi</code>. If this call is to be used,
the <code class="literal">doc.sig</code> field
should contain a signature value when calling
<code class="literal">addOrUpdate()</code>. The
<code class="literal">needUpdate()</code> call
then compares its parameter value with the stored
<code class="literal">sig</code> for <code class=
"literal">udi</code>. <code class=
"literal">sig</code> is an opaque value, compared
as a string.</p>
<p>The filesystem indexer uses a concatenation of
the decimal string values for file size and
update time, but a hash of the contents could
also be used.</p>
<p>As a side effect, if the return value is false
(the index is up to date), the call will set the
existence flag for the document (and any
subdocument defined by its <code class=
"literal">parent_udi</code>), so that a later
<code class="literal">purge()</code> call will
preserve them).</p>
<p>The use of <code class=
"literal">needUpdate()</code> and <code class=
"literal">purge()</code> is optional, and the
indexer may use another method for checking the
need to reindex or to delete stale entries.</p>
</dd>
<dt><span class="term">purge()</span></dt>
<dd>
<p>Delete all documents that were not touched
during the just finished indexing pass (since
open-for-write). These are the documents for the
needUpdate() call was not performed, indicating
that they no longer exist in the primary storage
system.</p>
</dd>
</dl>
</div>
</div>
<div class="sect3">
<div class="titlepage">
<div>
<div>
<h4 class="title"><a name=
"RCL.PROGRAM.PYTHONAPI.UPDATE.ACCESS" id=
"RCL.PROGRAM.PYTHONAPI.UPDATE.ACCESS"></a>Query
data access for external indexers (1.23)</h4>
</div>
</div>
</div>
<p><span class="application">Recoll</span> has internal
methods to access document data for its internal
(filesystem) indexer. An external indexer needs to
provide data access methods if it needs integration
with the GUI (e.g. preview function), or support for
the <code class="filename">rclextract</code>
module.</p>
<p>The index data and the access method are linked by
the <code class="literal">rclbes</code> (recoll backend
storage) <code class="literal">Doc</code> field. You
should set this to a short string value identifying
your indexer (e.g. the filesystem indexer uses either
"FS" or an empty value, the Web history indexer uses
"BGL").</p>
<p>The link is actually performed inside a <code class=
"filename">backends</code> configuration file (stored
in the configuration directory). This defines commands
to execute to access data from the specified indexer.
Example, for the mbox indexing sample found in the
Recoll source (which sets <code class=
"literal">rclbes="MBOX"</code>):</p>
<pre class="programlisting">[MBOX]
fetch = /path/to/recoll/src/python/samples/rclmbox.py fetch
makesig = path/to/recoll/src/python/samples/rclmbox.py makesig
</pre>
<p><code class="literal">fetch</code> and <code class=
"literal">makesig</code> define two commands to execute
to respectively retrieve the document text and compute
the document signature (the example implementation uses
the same script with different first parameters to
perform both operations).</p>
<p>The scripts are called with three additional
arguments: <code class="literal">udi</code>,
<code class="literal">url</code>, <code class=
"literal">ipath</code>, stored with the document when
it was indexed, and may use any or all to perform the
requested operation. The caller expects the result data
on <code class="literal">stdout</code>.</p>
</div>
<div class="sect3">
<div class="titlepage">
<div>
<div>
<h4 class="title"><a name=
"RCL.PROGRAM.PYTHONAPI.UPDATE.SAMPLES" id=
"RCL.PROGRAM.PYTHONAPI.UPDATE.SAMPLES"></a>External
indexer samples</h4>
</div>
</div>
</div>
<p>The Recoll source tree has two samples of external
indexers in the <code class=
"filename">src/python/samples</code> directory. The
more interesting one is <code class=
"filename">rclmbox.py</code> which indexes a directory
containing <code class="literal">mbox</code> folder
files. It exercises most features in the update
interface, and has a data access interface.</p>
<p>See the comments inside the file for more
information.</p>
</div>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.PROGRAM.PYTHONAPI.COMPAT" id=
"RCL.PROGRAM.PYTHONAPI.COMPAT"></a>4.3.6.&nbsp;Package
compatibility with the previous version</h3>
</div>
</div>
</div>
<p>The following code fragments can be used to ensure
that code can run with both the old and the new API (as
long as it does not use the new abilities of the new API
of course).</p>
<p>Adapting to the new package structure:</p>
<pre class="programlisting">
try:
from recoll import recoll
from recoll import rclextract
hasextract = True
except:
import recoll
hasextract = False
</pre>
<p>Adapting to the change of nature of the <code class=
"literal">next</code> <code class="literal">Query</code>
member. The same test can be used to choose to use the
<code class="literal">scroll()</code> method (new) or set
the <code class="literal">next</code> value (old).</p>
<pre class=
"programlisting">rownum = query.next if type(query.next) == int else query.rownumber</pre>
</div>
</div>
</div>
<div class="chapter">
<div class="titlepage">
<div>
<div>
<h1 class="title"><a name="RCL.INSTALL" id=
"RCL.INSTALL"></a>Chapter&nbsp;5.&nbsp;Installation and
configuration</h1>
</div>
</div>
</div>
<div class="sect1">
<div class="titlepage">
<div>
<div>
<h2 class="title" style="clear: both"><a name=
"RCL.INSTALL.BINARY" id=
"RCL.INSTALL.BINARY"></a>5.1.&nbsp;Installing a
binary copy</h2>
</div>
</div>
</div>
<p><span class="application">Recoll</span> binary copies
are always distributed as regular packages for your system.
They can be obtained either through the system's normal
software distribution framework (e.g. <span class=
"application">Debian/Ubuntu apt</span>, <span class=
"application">FreeBSD</span> ports, etc.), or from some
type of "backports" repository providing versions newer
than the standard ones, or found on the <span class=
"application">Recoll</span> Web site in some cases. The
most up-to-date information about Recoll packages can
usually be found on the <a class="ulink" href=
"http://www.recoll.org/pages/download.html" target=
"_top"><span class="application">Recoll</span> Web site
downloads page</a></p>
<p>The <span class="application">Windows</span> version of
Recoll comes in a self-contained setup file, there is
nothing else to install.</p>
<p>On <span class="application">Unix</span>-like systems,
the package management tools will automatically install
hard dependencies for packages obtained from a proper
package repository. You will have to deal with them by hand
for downloaded packages (for example, when <span class=
"command"><strong>dpkg</strong></span> complains about
missing dependencies).</p>
<p>In all cases, you will have to check or install
<a class="link" href="#RCL.INSTALL.EXTERNAL" title=
"5.2.&nbsp;Supporting packages">supporting applications</a>
for the file types that you want to index beyond those that
are natively processed by <span class=
"application">Recoll</span> (text, HTML, email files, and a
few others).</p>
<p>You should also maybe have a look at the <a class="link"
href="#RCL.INSTALL.CONFIG" title=
"5.4.&nbsp;Configuration overview">configuration
section</a> (but this may not be necessary for a quick test
with default parameters). Most parameters can be more
conveniently set from the GUI interface.</p>
</div>
<div class="sect1">
<div class="titlepage">
<div>
<div>
<h2 class="title" style="clear: both"><a name=
"RCL.INSTALL.EXTERNAL" id=
"RCL.INSTALL.EXTERNAL"></a>5.2.&nbsp;Supporting
packages</h2>
</div>
</div>
</div>
<div class="note" style=
"margin-left: 0.5in; margin-right: 0.5in;">
<h3 class="title">Note</h3>
<p>The <span class="application">Windows</span>
installation of <span class="application">Recoll</span>
is self-contained. <span class=
"application">Windows</span> users can skip this
section.</p>
</div>
<p><span class="application">Recoll</span> uses external
applications to index some file types. You need to install
them for the file types that you wish to have indexed
(these are run-time optional dependencies. None is needed
for building or running <span class=
"application">Recoll</span> except for indexing their
specific file type).</p>
<p>After an indexing pass, the commands that were found
missing can be displayed from the <span class=
"command"><strong>recoll</strong></span> <span class=
"guilabel">File</span> menu. The list is stored in the
<code class="filename">missing</code> text file inside the
configuration directory.</p>
<p>The past has proven that I was unable to maintain an up
to date application list in this manual. Please check
<a class="ulink" href=
"http://www.recoll.org/pages/features.html#doctypes"
target="_top">http://www.recoll.org/pages/features.html</a>
for a complete list along with links to the home pages or
best source/patches pages, and misc tips. What follows is
only a very short extract of the stable essentials.</p>
<div class="itemizedlist">
<ul class="itemizedlist" style="list-style-type: disc;">
<li class="listitem">
<p>PDF files need <span class=
"command"><strong>pdftotext</strong></span> which is
part of <span class="application">Poppler</span>
(usually comes with the <code class=
"literal">poppler-utils</code> package). Avoid the
original one from <span class=
"application">Xpdf</span>.</p>
</li>
<li class="listitem">
<p>MS Word documents need <span class=
"command"><strong>antiword</strong></span>. It is
also useful to have <span class=
"command"><strong>wvWare</strong></span> installed as
it may be be used as a fallback for some files which
<span class=
"command"><strong>antiword</strong></span> does not
handle.</p>
</li>
<li class="listitem">
<p>RTF files need <span class=
"command"><strong>unrtf</strong></span>, which, in
its older versions, has much trouble with non-western
character sets. Many Linux distributions carry
outdated <span class=
"command"><strong>unrtf</strong></span> versions.
Check <a class="ulink" href=
"http://www.recoll.org/pages/features.html#doctypes"
target=
"_top">http://www.recoll.org/pages/features.html</a>
for details.</p>
</li>
<li class="listitem">
<p>Pictures: <span class="application">Recoll</span>
uses the <span class="application">Exiftool</span>
<span class="application">Perl</span> package to
extract tag information. Most image file formats are
supported.</p>
</li>
<li class="listitem">
<p>Up to <span class="application">Recoll</span>
1.24, many XML-based formats need the <span class=
"command"><strong>xsltproc</strong></span> command,
which usually comes with <span class=
"application">libxslt</span>. These are: abiword, fb2
ebooks, kword, openoffice, opendocument svg.
<span class="application">Recoll</span> 1.25 and
later process them internally (using libxslt).</p>
</li>
</ul>
</div>
</div>
<div class="sect1">
<div class="titlepage">
<div>
<div>
<h2 class="title" style="clear: both"><a name=
"RCL.INSTALL.BUILDING" id=
"RCL.INSTALL.BUILDING"></a>5.3.&nbsp;Building from
source</h2>
</div>
</div>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.INSTALL.BUILDING.PREREQS" id=
"RCL.INSTALL.BUILDING.PREREQS"></a>5.3.1.&nbsp;Prerequisites</h3>
</div>
</div>
</div>
<p>The following prerequisites are described in broad
terms and not as specific package names (which will
depend on the exact platform). The dependencies should be
available as packages on most common Unix derivatives,
and it should be quite uncommon that you would have to
build one of them.</p>
<p>If you do not need the GUI, you can avoid all GUI
dependencies by disabling its build. (See the configure
section further).</p>
<p>The shopping list:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style=
"list-style-type: disc;">
<li class="listitem">
<p>If you start from git code, you will need the
<span class=
"command"><strong>autoconf</strong></span>,
<span class=
"command"><strong>automake</strong></span> and
<span class=
"command"><strong>libtool</strong></span> triad.
They are not needed for building from tar
distributions.</p>
</li>
<li class="listitem">
<p>C++ compiler. Recent versions require C++11
compatibility (1.23 and later).</p>
</li>
<li class="listitem">
<p><span class=
"command"><strong>bison</strong></span> command
(for <span class="application">Recoll</span> 1.21
and later).</p>
</li>
<li class="listitem">
<p>For building the documentation: the <span class=
"command"><strong>xsltproc</strong></span> command,
and the Docbook XML and style sheet files. You can
avoid this dependency by disabling documentation
building with the <code class=
"literal">--disable-userdoc</code> <span class=
"command"><strong>configure</strong></span>
option.</p>
</li>
<li class="listitem">
<p>Development files for <a class="ulink" href=
"http://www.xapian.org" target="_top"><span class=
"application">Xapian core</span></a>.</p>
<div class="important" style=
"margin-left: 0.5in; margin-right: 0.5in;">
<h3 class="title">Important</h3>
<p>If you are building Xapian for an older CPU
(before Pentium 4 or Athlon 64), you need to add
the <code class="option">--disable-sse</code>
flag to the configure command. Else all Xapian
application will crash with an <code class=
"literal">illegal instruction</code> error.</p>
</div>
</li>
<li class="listitem">
<p>Development files for <a class="ulink" href=
"http://qt-project.org/downloads" target=
"_top"><span class="application">Qt 5</span></a> .
and its own dependencies (X11 etc.)</p>
</li>
<li class="listitem">
<p>Development files for libxslt</p>
</li>
<li class="listitem">
<p>Development files for <span class=
"application">zlib</span>.</p>
</li>
<li class="listitem">
<p>Development files for <span class=
"application">Python</span> (or use <code class=
"literal">--disable-python-module</code>).</p>
</li>
<li class="listitem">
<p>Development files for libchm</p>
</li>
<li class="listitem">
<p>You may also need <a class="ulink" href=
"http://www.gnu.org/software/libiconv/" target=
"_top">libiconv</a>. On <span class=
"application">Linux</span> systems, the iconv
interface is part of libc and you should not need
to do anything special.</p>
</li>
</ul>
</div>
<p>Check the <a class="ulink" href=
"http://www.recoll.org/pages/download.html" target=
"_top"><span class="application">Recoll</span> download
page</a> for up to date version information.</p>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.INSTALL.BUILDING.BUILDING" id=
"RCL.INSTALL.BUILDING.BUILDING"></a>5.3.2.&nbsp;Building</h3>
</div>
</div>
</div>
<p><span class="application">Recoll</span> has been built
on Linux, FreeBSD, Mac OS X, and Solaris, most versions
after 2005 should be ok, maybe some older ones too
(Solaris 8 used to be ok). If you build on another
system, and need to modify things, <a class="ulink" href=
"mailto:jfd@recoll.org" target="_top">I would very much
welcome patches</a>.</p>
<p><b>Configure options:&nbsp;</b></p>
<div class="itemizedlist">
<ul class="itemizedlist" style=
"list-style-type: disc;">
<li class="listitem">
<p><code class="option">--without-aspell</code>
will disable the code for phonetic matching of
search terms.</p>
</li>
<li class="listitem">
<p><code class="option">--with-fam</code> or
<code class="option">--with-inotify</code> will
enable the code for real time indexing. Inotify
support is enabled by default on Linux systems.</p>
</li>
<li class="listitem">
<p><code class="option">--with-qzeitgeist</code>
will enable sending <span class=
"application">Zeitgeist</span> events about the
visited search results, and needs the <span class=
"application">qzeitgeist</span> package.</p>
</li>
<li class="listitem">
<p><code class="option">--disable-webkit</code> is
available from version 1.17 to implement the result
list with a <span class="application">Qt</span>
QTextBrowser instead of a WebKit widget if you do
not or can't depend on the latter.</p>
</li>
<li class="listitem">
<p><code class="option">--disable-qtgui</code>
Disable the Qt interface. Will allow building the
indexer and the command line search program in
absence of a Qt environment.</p>
</li>
<li class="listitem">
<p><code class="option">--enable-webengine</code>
Enable the use of Qt Webengine (only meaningful if
the Qt GUI is enabled), in place or Qt Webkit.</p>
</li>
<li class="listitem">
<p><code class="option">--disable-idxthreads</code>
is available from version 1.19 to suppress
multithreading inside the indexing process. You can
also use the run-time configuration to restrict
<span class=
"command"><strong>recollindex</strong></span> to
using a single thread, but the compile-time option
may disable a few more unused locks. This only
applies to the use of multithreading for the core
index processing (data input). The <span class=
"application">Recoll</span> monitor mode always
uses at least two threads of execution.</p>
</li>
<li class="listitem">
<p><code class=
"option">--disable-python-module</code> will avoid
building the <span class=
"application">Python</span> module.</p>
</li>
<li class="listitem">
<p><code class="option">--disable-python-chm</code>
will avoid building the Python libchm interface
used to index CHM files.</p>
</li>
<li class="listitem">
<p><code class="option">--enable-camelcase</code>
will enable splitting <em class=
"replaceable"><code>camelCase</code></em> words.
This is not enabled by default as it has the
unfortunate side-effect of making some phrase
searches quite confusing: ie, <code class=
"literal">"MySQL manual"</code> would be matched by
<code class="literal">"MySQL manual"</code> and
<code class="literal">"my sql manual"</code> but
not <code class="literal">"mysql manual"</code>
(only inside phrase searches).</p>
</li>
<li class="listitem">
<p><code class="option">--with-file-command</code>
Specify the version of the 'file' command to use
(ie: --with-file-command=/usr/local/bin/file). Can
be useful to enable the gnu version on systems
where the native one is bad.</p>
</li>
<li class="listitem">
<p><code class="option">--disable-x11mon</code>
Disable <span class="application">X11</span>
connection monitoring inside recollindex. Together
with --disable-qtgui, this allows building recoll
without <span class="application">Qt</span> and
<span class="application">X11</span>.</p>
</li>
<li class="listitem">
<p><code class="option">--disable-userdoc</code>
will avoid building the user manual. This avoids
having to install the Docbook XML/XSL files and the
TeX toolchain used for translating the manual to
PDF.</p>
</li>
<li class="listitem">
<p><code class="option">--enable-recollq</code>
Enable building the <span class=
"command"><strong>recollq</strong></span> command
line query tool (recoll -t without need for Qt).
This is done by default if --disable-qtgui is set
but this option enables forcing it.</p>
</li>
<li class="listitem">
<p><code class="option">--disable-pic</code>
(<span class="application">Recoll</span> versions
up to 1.21 only) will compile <span class=
"application">Recoll</span> with position-dependant
code. This is incompatible with building the KIO or
the <span class="application">Python</span> or
<span class="application">PHP</span> extensions,
but might yield very marginally faster code.</p>
</li>
<li class="listitem">
<p>Of course the usual <span class=
"application">autoconf</span> <span class=
"command"><strong>configure</strong></span>
options, like <code class="option">--prefix</code>
apply.</p>
</li>
</ul>
</div>
<p>Normal procedure (for source extracted from a tar
distribution):</p>
<pre class="screen">
<strong class=
"userinput"><code>cd recoll-xxx</code></strong>
<strong class=
"userinput"><code>./configure</code></strong>
<strong class="userinput"><code>make</code></strong>
<strong class=
"userinput"><code>(practices usual hardship-repelling invocations)</code></strong>
</pre>
<p>When building from source cloned from the git
repository, you also need to install <span class=
"application">autoconf</span>, <span class=
"application">automake</span>, and <span class=
"application">libtool</span> and you must execute
<code class="literal">sh autogen.sh</code> in the top
source directory before running <code class=
"literal">configure</code>.</p>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.INSTALL.BUILDING.INSTALL" id=
"RCL.INSTALL.BUILDING.INSTALL"></a>5.3.3.&nbsp;Installing</h3>
</div>
</div>
</div>
<p>Use <strong class="userinput"><code>make
install</code></strong> in the root of the source tree.
This will copy the commands to <code class=
"filename"><em class=
"replaceable"><code>prefix</code></em>/bin</code> and the
sample configuration files, scripts and other shared data
to <code class="filename"><em class=
"replaceable"><code>prefix</code></em>/share/recoll</code>.</p>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.INSTALL.BUILDING.PYTHON" id=
"RCL.INSTALL.BUILDING.PYTHON"></a>5.3.4.&nbsp;Python
API package</h3>
</div>
</div>
</div>
<p>The Python interface can be found in the source tree,
under the <code class="filename">python/recoll</code>
directory.</p>
<p>As of <span class="application">Recoll</span> 1.19,
the module can be compiled for Python3.</p>
<p>The normal <span class="application">Recoll</span>
build procedure (see above) installs the API package for
the default system version (python) along with the main
code. The package for other Python versions (e.g. python3
if the system default is python2) must be explicitly
built and installed.</p>
<p>The <code class="filename">python/recoll/</code>
directory contains the usual <code class=
"filename">setup.py</code>. After configuring and
building the main <span class="application">Recoll</span>
code, you can use the script to build and install the
Python module:</p>
<pre class="screen">
<strong class=
"userinput"><code>cd recoll-xxx/python/recoll</code></strong>
<strong class=
"userinput"><code>pythonX setup.py build</code></strong>
<strong class=
"userinput"><code>sudo pythonX setup.py install</code></strong>
</pre>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.INSTALL.BUILDING.SOLARIS" id=
"RCL.INSTALL.BUILDING.SOLARIS"></a>5.3.5.&nbsp;Building
on Solaris</h3>
</div>
</div>
</div>
<p>We did not test building the GUI on Solaris for recent
versions. You will need at least Qt 4.4. There are some
hints on <a class="ulink" href=
"http://www.recoll.org/download-1.14.html" target=
"_top">an old web site page</a>, they may still be
valid.</p>
<p>Someone did test the 1.19 indexer and Python module
build, they do work, with a few minor glitches. Be sure
to use GNU <span class=
"command"><strong>make</strong></span> and <span class=
"command"><strong>install</strong></span>.</p>
</div>
</div>
<div class="sect1">
<div class="titlepage">
<div>
<div>
<h2 class="title" style="clear: both"><a name=
"RCL.INSTALL.CONFIG" id=
"RCL.INSTALL.CONFIG"></a>5.4.&nbsp;Configuration
overview</h2>
</div>
</div>
</div>
<p>Most of the parameters specific to the <span class=
"command"><strong>recoll</strong></span> GUI are set
through the <span class="guilabel">Preferences</span> menu
and stored in the standard Qt place (<code class=
"filename">$HOME/.config/Recoll.org/recoll.conf</code>).
You probably do not want to edit this by hand.</p>
<p><span class="application">Recoll</span> indexing options
are set inside text configuration files located in a
configuration directory. There can be several such
directories, each of which defines the parameters for one
index.</p>
<p>The configuration files can be edited by hand or through
the <span class="guilabel">Index configuration</span>
dialog (<span class="guilabel">Preferences</span> menu).
The GUI tool will try to respect your formatting and
comments as much as possible, so it is quite possible to
use both approaches on the same configuration.</p>
<p>The most accurate documentation for the configuration
parameters is given by comments inside the default files,
and we will just give a general overview here.</p>
<p>For each index, there are at least two sets of
configuration files. System-wide configuration files are
kept in a directory named like <code class=
"filename">/usr/share/recoll/examples</code>, and define
default values, shared by all indexes. For each index, a
parallel set of files defines the customized
parameters.</p>
<p>The default location of the customized configuration is
the <code class="filename">.recoll</code> directory in your
home. Most people will only use this directory.</p>
<p>This location can be changed, or others can be added
with the <code class="envar">RECOLL_CONFDIR</code>
environment variable or the <code class="option">-c</code>
option parameter to <span class=
"command"><strong>recoll</strong></span> and <span class=
"command"><strong>recollindex</strong></span>.</p>
<p>In addition (as of <span class=
"application">Recoll</span> version 1.19.7), it is possible
to specify two additional configuration directories which
will be stacked before and after the user configuration
directory. These are defined by the <code class=
"envar">RECOLL_CONFTOP</code> and <code class=
"envar">RECOLL_CONFMID</code> environment variables. Values
from configuration files inside the top directory will
override user ones, values from configuration files inside
the middle directory will override system ones and be
overridden by user ones. These two variables may be of use
to applications which augment <span class=
"application">Recoll</span> functionality, and need to add
configuration data without disturbing the user's files.
Please note that the two, currently single, values will
probably be interpreted as colon-separated lists in the
future: do not use colon characters inside the directory
paths.</p>
<p>If the <code class="filename">.recoll</code> directory
does not exist when <span class=
"command"><strong>recoll</strong></span> or <span class=
"command"><strong>recollindex</strong></span> are started,
it will be created with a set of empty configuration files.
<span class="command"><strong>recoll</strong></span> will
give you a chance to edit the configuration file before
starting indexing. <span class=
"command"><strong>recollindex</strong></span> will proceed
immediately. To avoid mistakes, the automatic directory
creation will only occur for the default location, not if
<code class="option">-c</code> or <code class=
"envar">RECOLL_CONFDIR</code> were used (in the latter
cases, you will have to create the directory).</p>
<p>All configuration files share the same format. For
example, a short extract of the main configuration file
might look as follows:</p>
<pre class="programlisting">
# Space-separated list of files and directories to index.
topdirs = ~/docs /usr/share/doc
[~/somedirectory-with-utf8-txt-files]
defaultcharset = utf-8
</pre>
<p>There are three kinds of lines:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style="list-style-type: disc;">
<li class="listitem">
<p>Comment (starts with <span class=
"emphasis"><em>#</em></span>) or empty.</p>
</li>
<li class="listitem">
<p>Parameter affectation (<span class=
"emphasis"><em>name = value</em></span>).</p>
</li>
<li class="listitem">
<p>Section definition ([<span class=
"emphasis"><em>somedirname</em></span>]).</p>
</li>
</ul>
</div>
<p>Long lines can be broken by ending each incomplete part
with a backslash (<code class="literal">\</code>).</p>
<p>Depending on the type of configuration file, section
definitions either separate groups of parameters or allow
redefining some parameters for a directory sub-tree. They
stay in effect until another section definition, or the end
of file, is encountered. Some of the parameters used for
indexing are looked up hierarchically from the current
directory location upwards. Not all parameters can be
meaningfully redefined, this is specified for each in the
next section.</p>
<div class="important" style=
"margin-left: 0.5in; margin-right: 0.5in;">
<h3 class="title">Important</h3>
<p>Global parameters <span class="emphasis"><em>must
not</em></span> be defined in a directory subsection,
else they will not be found at all by the <span class=
"application">Recoll</span> code, which looks for them at
the top level (e.g. <code class=
"literal">skippedPaths</code>).</p>
</div>
<p>When found at the beginning of a file path, the tilde
character (~) is expanded to the name of the user's home
directory, as a shell would do.</p>
<p>Some parameters are lists of strings. White space is
used for separation. List elements with embedded spaces can
be quoted using double-quotes. Double quotes inside these
elements can be escaped with a backslash.</p>
<p>No value inside a configuration file can contain a
newline character. Long lines can be continued by escaping
the physical newline with backslash, even inside quoted
strings.</p>
<pre class="programlisting">
astringlist = "some string \
with spaces"
thesame = "some string with spaces"
</pre>
<p>Parameters which are not part of string lists can't be
quoted, and leading and trailing space characters are
stripped before the value is used.</p>
<p><b>Encoding issues.&nbsp;</b>Most of the configuration
parameters are plain ASCII. Two particular sets of values
may cause encoding issues:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style="list-style-type: disc;">
<li class="listitem">
<p>File path parameters may contain non-ascii
characters and should use the exact same byte values
as found in the file system directory. Usually, this
means that the configuration file should use the
system default locale encoding.</p>
</li>
<li class="listitem">
<p>The <code class="envar">unac_except_trans</code>
parameter should be encoded in UTF-8. If your system
locale is not UTF-8, and you need to also specify
non-ascii file paths, this poses a difficulty because
common text editors cannot handle multiple encodings
in a single file. In this relatively unlikely case,
you can edit the configuration file as two separate
text files with appropriate encodings, and
concatenate them to create the complete
configuration.</p>
</li>
</ul>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.INSTALL.CONFIG.ENVIR" id=
"RCL.INSTALL.CONFIG.ENVIR"></a>5.4.1.&nbsp;Environment
variables</h3>
</div>
</div>
</div>
<div class="variablelist">
<dl class="variablelist">
<dt><span class="term"><code class=
"varname">RECOLL_CONFDIR</code></span></dt>
<dd>
<p>Defines the main configuration directory.</p>
</dd>
<dt><span class="term"><code class=
"varname">RECOLL_TMPDIR, TMPDIR</code></span></dt>
<dd>
<p>Locations for temporary files, in this order of
priority. The default if none of these is set is to
use <code class="filename">/tmp</code>. Big
temporary files may be created during indexing,
mostly for decompressing, and also for processing,
e.g. email attachments.</p>
</dd>
<dt><span class="term"><code class=
"varname">RECOLL_CONFTOP,
RECOLL_CONFMID</code></span></dt>
<dd>
<p>Allow adding configuration directories with
priorities below and above the user directory (see
above the Configuration overview section for
details).</p>
</dd>
<dt><span class="term"><code class=
"varname">RECOLL_EXTRA_DBS,
RECOLL_ACTIVE_EXTRA_DBS</code></span></dt>
<dd>
<p>Help for setting up external indexes. See
<a class="link" href="#RCL.SEARCH.GUI.MULTIDB"
title="3.2.10.&nbsp;Multiple indexes">this
paragraph</a> for explanations.</p>
</dd>
<dt><span class="term"><code class=
"varname">RECOLL_DATADIR</code></span></dt>
<dd>
<p>Defines replacement for the default location of
Recoll data files, normally found in, e.g.,
<code class=
"filename">/usr/share/recoll</code>).</p>
</dd>
<dt><span class="term"><code class=
"varname">RECOLL_FILTERSDIR</code></span></dt>
<dd>
<p>Defines replacement for the default location of
Recoll filters, normally found in, e.g.,
<code class=
"filename">/usr/share/recoll/filters</code>).</p>
</dd>
<dt><span class="term"><code class=
"varname">ASPELL_PROG</code></span></dt>
<dd>
<p><span class=
"command"><strong>aspell</strong></span> program to
use for creating the spelling dictionary. The
result has to be compatible with the <code class=
"filename">libaspell</code> which <span class=
"application">Recoll</span> is using.</p>
</dd>
</dl>
</div>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF" id=
"RCL.INSTALL.CONFIG.RECOLLCONF"></a>5.4.2.&nbsp;Recoll
main configuration file, recoll.conf</h3>
</div>
</div>
</div>
<div class="sect3">
<div class="titlepage">
<div>
<div>
<h4 class="title"><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.WHATDOCS" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.WHATDOCS"></a>Parameters
affecting what documents we index</h4>
</div>
</div>
</div>
<div class="variablelist">
<dl class="variablelist">
<dt><a name="RCL.INSTALL.CONFIG.RECOLLCONF.TOPDIRS"
id=
"RCL.INSTALL.CONFIG.RECOLLCONF.TOPDIRS"></a><span class="term"><code class="varname">topdirs</code></span></dt>
<dd>
<p>Space-separated list of files or directories
to recursively index. Default to ~ (indexes
$HOME). You can use symbolic links in the list,
they will be followed, independently of the value
of the followLinks variable.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.MONITORDIRS" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.MONITORDIRS"></a><span class="term"><code class="varname">monitordirs</code></span></dt>
<dd>
<p>Space-separated list of files or directories
to monitor for updates. When running the
real-time indexer, this allows monitoring only a
subset of the whole indexed area. The elements
must be included in the tree defined by the
'topdirs' members.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.SKIPPEDNAMES" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.SKIPPEDNAMES"></a><span class="term"><code class="varname">skippedNames</code></span></dt>
<dd>
<p>Files and directories which should be ignored.
White space separated list of wildcard patterns
(simple ones, not paths, must contain no / ),
which will be tested against file and directory
names. The list in the default configuration does
not exclude hidden directories (names beginning
with a dot), which means that it may index quite
a few things that you do not want. On the other
hand, email user agents like Thunderbird usually
store messages in hidden directories, and you
probably want this indexed. One possible solution
is to have ".*" in "skippedNames", and add things
like "~/.thunderbird" "~/.evolution" to
"topdirs". Not even the file names are indexed
for patterns in this list, see the
"noContentSuffixes" variable for an alternative
approach which indexes the file names. Can be
redefined for any subtree.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.SKIPPEDNAMES-" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.SKIPPEDNAMES-"></a><span class="term"><code class="varname">skippedNames-</code></span></dt>
<dd>
<p>List of name endings to remove from the
default skippedNames list.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.SKIPPEDNAMES+" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.SKIPPEDNAMES+"></a><span class="term"><code class="varname">skippedNames+</code></span></dt>
<dd>
<p>List of name endings to add to the default
skippedNames list.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.ONLYNAMES" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.ONLYNAMES"></a><span class="term"><code class="varname">onlyNames</code></span></dt>
<dd>
<p>Regular file name filter patterns If this is
set, only the file names not in skippedNames and
matching one of the patterns will be considered
for indexing. Can be redefined per subtree. Does
not apply to directories.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.NOCONTENTSUFFIXES"
id=
"RCL.INSTALL.CONFIG.RECOLLCONF.NOCONTENTSUFFIXES"></a><span class="term"><code class="varname">noContentSuffixes</code></span></dt>
<dd>
<p>List of name endings (not necessarily
dot-separated suffixes) for which we don't try
MIME type identification, and don't uncompress or
index content. Only the names will be indexed.
This complements the now obsoleted recoll_noindex
list from the mimemap file, which will go away in
a future release (the move from mimemap to
recoll.conf allows editing the list through the
GUI). This is different from skippedNames because
these are name ending matches only (not wildcard
patterns), and the file name itself gets indexed
normally. This can be redefined for
subdirectories.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.NOCONTENTSUFFIXES-"
id=
"RCL.INSTALL.CONFIG.RECOLLCONF.NOCONTENTSUFFIXES-"></a><span class="term"><code class="varname">noContentSuffixes-</code></span></dt>
<dd>
<p>List of name endings to remove from the
default noContentSuffixes list.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.NOCONTENTSUFFIXES+"
id=
"RCL.INSTALL.CONFIG.RECOLLCONF.NOCONTENTSUFFIXES+"></a><span class="term"><code class="varname">noContentSuffixes+</code></span></dt>
<dd>
<p>List of name endings to add to the default
noContentSuffixes list.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.SKIPPEDPATHS" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.SKIPPEDPATHS"></a><span class="term"><code class="varname">skippedPaths</code></span></dt>
<dd>
<p>Absolute paths we should not go into.
Space-separated list of wildcard expressions for
absolute filesystem paths. Must be defined at the
top level of the configuration file, not in a
subsection. Can contain files and directories.
The database and configuration directories will
automatically be added. The expressions are
matched using 'fnmatch(3)' with the FNM_PATHNAME
flag set by default. This means that '/'
characters must be matched explicitly. You can
set 'skippedPathsFnmPathname' to 0 to disable the
use of FNM_PATHNAME (meaning that '/*/dir3' will
match '/dir1/dir2/dir3'). The default value
contains the usual mount point for removable
media to remind you that it is a bad idea to have
Recoll work on these (esp. with the monitor:
media gets indexed on mount, all data gets erased
on unmount). Explicitly adding '/media/xxx' to
the 'topdirs' variable will override this.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.SKIPPEDPATHSFNMPATHNAME"
id=
"RCL.INSTALL.CONFIG.RECOLLCONF.SKIPPEDPATHSFNMPATHNAME">
</a><span class="term"><code class=
"varname">skippedPathsFnmPathname</code></span></dt>
<dd>
<p>Set to 0 to override use of FNM_PATHNAME for
matching skipped paths.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.NOWALKFN" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.NOWALKFN"></a><span class="term"><code class="varname">nowalkfn</code></span></dt>
<dd>
<p>File name which will cause its parent
directory to be skipped. Any directory containing
a file with this name will be skipped as if it
was part of the skippedPaths list. Ex:
.recoll-noindex</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.DAEMSKIPPEDPATHS"
id="RCL.INSTALL.CONFIG.RECOLLCONF.DAEMSKIPPEDPATHS">
</a><span class="term"><code class=
"varname">daemSkippedPaths</code></span></dt>
<dd>
<p>skippedPaths equivalent specific to real time
indexing. This enables having parts of the tree
which are initially indexed but not monitored. If
daemSkippedPaths is not set, the daemon uses
skippedPaths.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.ZIPUSESKIPPEDNAMES"
id=
"RCL.INSTALL.CONFIG.RECOLLCONF.ZIPUSESKIPPEDNAMES"></a><span class="term"><code class="varname">zipUseSkippedNames</code></span></dt>
<dd>
<p>Use skippedNames inside Zip archives. Fetched
directly by the rclzip handler. Skip the patterns
defined by skippedNames inside Zip archives. Can
be redefined for subdirectories. See
https://www.lesbonscomptes.com/recoll/faqsandhowtos/FilteringOutZipArchiveMembers.html</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.ZIPSKIPPEDNAMES" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.ZIPSKIPPEDNAMES"></a><span class="term"><code class="varname">zipSkippedNames</code></span></dt>
<dd>
<p>Space-separated list of wildcard expressions
for names that should be ignored inside zip
archives. This is used directly by the zip
handler. If zipUseSkippedNames is not set,
zipSkippedNames defines the patterns to be
skipped inside archives. If zipUseSkippedNames is
set, the two lists are concatenated and used. Can
be redefined for subdirectories. See
https://www.lesbonscomptes.com/recoll/faqsandhowtos/FilteringOutZipArchiveMembers.html</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.FOLLOWLINKS" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.FOLLOWLINKS"></a><span class="term"><code class="varname">followLinks</code></span></dt>
<dd>
<p>Follow symbolic links during indexing. The
default is to ignore symbolic links to avoid
multiple indexing of linked files. No effort is
made to avoid duplication when this option is set
to true. This option can be set individually for
each of the 'topdirs' members by using sections.
It can not be changed below the 'topdirs' level.
Links in the 'topdirs' list itself are always
followed.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.INDEXEDMIMETYPES"
id="RCL.INSTALL.CONFIG.RECOLLCONF.INDEXEDMIMETYPES">
</a><span class="term"><code class=
"varname">indexedmimetypes</code></span></dt>
<dd>
<p>Restrictive list of indexed mime types.
Normally not set (in which case all supported
types are indexed). If it is set, only the types
from the list will have their contents indexed.
The names will be indexed anyway if
indexallfilenames is set (default). MIME type
names should be taken from the mimemap file (the
values may be different from xdg-mime or file -i
output in some cases). Can be redefined for
subtrees.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.EXCLUDEDMIMETYPES"
id=
"RCL.INSTALL.CONFIG.RECOLLCONF.EXCLUDEDMIMETYPES"></a><span class="term"><code class="varname">excludedmimetypes</code></span></dt>
<dd>
<p>List of excluded MIME types. Lets you exclude
some types from indexing. MIME type names should
be taken from the mimemap file (the values may be
different from xdg-mime or file -i output in some
cases) Can be redefined for subtrees.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.NOMD5TYPES" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.NOMD5TYPES"></a><span class="term"><code class="varname">nomd5types</code></span></dt>
<dd>
<p>Don't compute md5 for these types. md5
checksums are used only for deduplicating
results, and can be very expensive to compute on
multimedia or other big files. This list lets you
turn off md5 computation for selected types. It
is global (no redefinition for subtrees). At the
moment, it only has an effect for external
handlers (exec and execm). The file types can be
specified by listing either MIME types (e.g.
audio/mpeg) or handler names (e.g. rclaudio).</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.COMPRESSEDFILEMAXKBS"
id=
"RCL.INSTALL.CONFIG.RECOLLCONF.COMPRESSEDFILEMAXKBS">
</a><span class="term"><code class=
"varname">compressedfilemaxkbs</code></span></dt>
<dd>
<p>Size limit for compressed files. We need to
decompress these in a temporary directory for
identification, which can be wasteful in some
cases. Limit the waste. Negative means no limit.
0 results in no processing of any compressed
file. Default 50 MB.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.TEXTFILEMAXMBS" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.TEXTFILEMAXMBS"></a><span class="term"><code class="varname">textfilemaxmbs</code></span></dt>
<dd>
<p>Size limit for text files. Mostly for skipping
monster logs. Default 20 MB.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.INDEXALLFILENAMES"
id=
"RCL.INSTALL.CONFIG.RECOLLCONF.INDEXALLFILENAMES"></a><span class="term"><code class="varname">indexallfilenames</code></span></dt>
<dd>
<p>Index the file names of unprocessed files
Index the names of files the contents of which we
don't index because of an excluded or unsupported
MIME type.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.USESYSTEMFILECOMMAND"
id=
"RCL.INSTALL.CONFIG.RECOLLCONF.USESYSTEMFILECOMMAND">
</a><span class="term"><code class=
"varname">usesystemfilecommand</code></span></dt>
<dd>
<p>Use a system command for file MIME type
guessing as a final step in file type
identification This is generally useful, but will
usually cause the indexing of many bogus 'text'
files. See 'systemfilecommand' for the command
used.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.SYSTEMFILECOMMAND"
id=
"RCL.INSTALL.CONFIG.RECOLLCONF.SYSTEMFILECOMMAND"></a><span class="term"><code class="varname">systemfilecommand</code></span></dt>
<dd>
<p>Command used to guess MIME types if the
internal methods fails This should be a "file -i"
workalike. The file path will be added as a last
parameter to the command line. "xdg-mime" works
better than the traditional "file" command, and
is now the configured default (with a hard-coded
fallback to "file")</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.PROCESSWEBQUEUE" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.PROCESSWEBQUEUE"></a><span class="term"><code class="varname">processwebqueue</code></span></dt>
<dd>
<p>Decide if we process the Web queue. The queue
is a directory where the Recoll Web browser
plugins create the copies of visited pages.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.TEXTFILEPAGEKBS" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.TEXTFILEPAGEKBS"></a><span class="term"><code class="varname">textfilepagekbs</code></span></dt>
<dd>
<p>Page size for text files. If this is set,
text/plain files will be divided into documents
of approximately this size. Will reduce memory
usage at index time and help with loading data in
the preview window at query time. Particularly
useful with very big files, such as application
or system logs. Also see textfilemaxmbs and
compressedfilemaxkbs.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.MEMBERMAXKBS" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.MEMBERMAXKBS"></a><span class="term"><code class="varname">membermaxkbs</code></span></dt>
<dd>
<p>Size limit for archive members. This is passed
to the filters in the environment as
RECOLL_FILTER_MAXMEMBERKB.</p>
</dd>
</dl>
</div>
</div>
<div class="sect3">
<div class="titlepage">
<div>
<div>
<h4 class="title"><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.TERMS" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.TERMS"></a>Parameters
affecting how we generate terms and organize the
index</h4>
</div>
</div>
</div>
<div class="variablelist">
<dl class="variablelist">
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.INDEXSTRIPCHARS" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.INDEXSTRIPCHARS"></a><span class="term"><code class="varname">indexStripChars</code></span></dt>
<dd>
<p>Decide if we store character case and
diacritics in the index. If we do, searches
sensitive to case and diacritics can be
performed, but the index will be bigger, and some
marginal weirdness may sometimes occur. The
default is a stripped index. When using multiple
indexes for a search, this parameter must be
defined identically for all. Changing the value
implies an index reset.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.INDEXSTOREDOCTEXT"
id=
"RCL.INSTALL.CONFIG.RECOLLCONF.INDEXSTOREDOCTEXT"></a><span class="term"><code class="varname">indexStoreDocText</code></span></dt>
<dd>
<p>Decide if we store the documents' text content
in the index. Storing the text allows extracting
snippets from it at query time, instead of
building them from index position data. Newer
Xapian index formats have rendered our use of
positions list unacceptably slow in some cases.
The last Xapian index format with good
performance for the old method is Chert, which is
default for 1.2, still supported but not default
in 1.4 and will be dropped in 1.6. The stored
document text is translated from its original
format to UTF-8 plain text, but not stripped of
upper-case, diacritics, or punctuation signs.
Storing it increases the index size by 10-20%
typically, but also allows for nicer snippets, so
it may be worth enabling it even if not strictly
needed for performance if you can afford the
space. The variable only has an effect when
creating an index, meaning that the xapiandb
directory must not exist yet. Its exact effect
depends on the Xapian version. For Xapian 1.4, if
the variable is set to 0, the Chert format will
be used, and the text will not be stored. If the
variable is 1, Glass will be used, and the text
stored. For Xapian 1.2, and for versions after
1.5 and newer, the index format is always the
default, but the variable controls if the text is
stored or not, and the abstract generation
method. With Xapian 1.5 and later, and the
variable set to 0, abstract generation may be
very slow, but this setting may still be useful
to save space if you do not use abstract
generation at all.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.NONUMBERS" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.NONUMBERS"></a><span class="term"><code class="varname">nonumbers</code></span></dt>
<dd>
<p>Decides if terms will be generated for
numbers. For example "123", "1.5e6", 192.168.1.4,
would not be indexed if nonumbers is set
("value123" would still be). Numbers are often
quite interesting to search for, and this should
probably not be set except for special
situations, ie, scientific documents with huge
amounts of numbers in them, where setting
nonumbers will reduce the index size. This can
only be set for a whole index, not for a
subtree.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.DEHYPHENATE" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.DEHYPHENATE"></a><span class="term"><code class="varname">dehyphenate</code></span></dt>
<dd>
<p>Determines if we index 'coworker' also when
the input is 'co-worker'. This is new in version
1.22, and on by default. Setting the variable to
off allows restoring the previous behaviour.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.BACKSLASHASLETTER"
id=
"RCL.INSTALL.CONFIG.RECOLLCONF.BACKSLASHASLETTER"></a><span class="term"><code class="varname">backslashasletter</code></span></dt>
<dd>
<p>Process backslash as normal letter. This may
make sense for people wanting to index TeX
commands as such but is not of much general
use.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.UNDERSCOREASLETTER"
id=
"RCL.INSTALL.CONFIG.RECOLLCONF.UNDERSCOREASLETTER"></a><span class="term"><code class="varname">underscoreasletter</code></span></dt>
<dd>
<p>Process underscore as normal letter. This
makes sense in so many cases that one wonders if
it should not be the default.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.MAXTERMLENGTH" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.MAXTERMLENGTH"></a><span class="term"><code class="varname">maxtermlength</code></span></dt>
<dd>
<p>Maximum term length. Words longer than this
will be discarded. The default is 40 and used to
be hard-coded, but it can now be adjusted. You
need an index reset if you change the value.</p>
</dd>
<dt><a name="RCL.INSTALL.CONFIG.RECOLLCONF.NOCJK"
id=
"RCL.INSTALL.CONFIG.RECOLLCONF.NOCJK"></a><span class="term"><code class="varname">nocjk</code></span></dt>
<dd>
<p>Decides if specific East Asian (Chinese Korean
Japanese) characters/word splitting is turned
off. This will save a small amount of CPU if you
have no CJK documents. If your document base does
include such text but you are not interested in
searching it, setting nocjk may be a significant
time and space saver.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.CJKNGRAMLEN" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.CJKNGRAMLEN"></a><span class="term"><code class="varname">cjkngramlen</code></span></dt>
<dd>
<p>This lets you adjust the size of n-grams used
for indexing CJK text. The default value of 2 is
probably appropriate in most cases. A value of 3
would allow more precision and efficiency on
longer words, but the index will be approximately
twice as large.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.INDEXSTEMMINGLANGUAGES"
id=
"RCL.INSTALL.CONFIG.RECOLLCONF.INDEXSTEMMINGLANGUAGES">
</a><span class="term"><code class=
"varname">indexstemminglanguages</code></span></dt>
<dd>
<p>Languages for which to create stemming
expansion data. Stemmer names can be found by
executing 'recollindex -l', or this can also be
set from a list in the GUI. The values are full
language names, e.g. english, french...</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.DEFAULTCHARSET" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.DEFAULTCHARSET"></a><span class="term"><code class="varname">defaultcharset</code></span></dt>
<dd>
<p>Default character set. This is used for files
which do not contain a character set definition
(e.g.: text/plain). Values found inside files,
e.g. a 'charset' tag in HTML documents, will
override it. If this is not set, the default
character set is the one defined by the NLS
environment ($LC_ALL, $LC_CTYPE, $LANG), or
ultimately iso-8859-1 (cp-1252 in fact). If for
some reason you want a general default which does
not match your LANG and is not 8859-1, use this
variable. This can be redefined for any
sub-directory.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.UNAC_EXCEPT_TRANS"
id=
"RCL.INSTALL.CONFIG.RECOLLCONF.UNAC_EXCEPT_TRANS"></a><span class="term"><code class="varname">unac_except_trans</code></span></dt>
<dd>
<p>A list of characters, encoded in UTF-8, which
should be handled specially when converting text
to unaccented lowercase. For example, in Swedish,
the letter a with diaeresis has full alphabet
citizenship and should not be turned into an a.
Each element in the space-separated list has the
special character as first element and the
translation following. The handling of both the
lowercase and upper-case versions of a character
should be specified, as appartenance to the list
will turn-off both standard accent and case
processing. The value is global and affects both
indexing and querying. Examples: Swedish:
unac_except_trans = ää Ää öö Öö üü Üü ßss œoe Œoe
æae Æae ffff fifi flfl åå Åå . German:
unac_except_trans = ää Ää öö Öö üü Üü ßss œoe Œoe
æae Æae ffff fifi flfl In French, you probably want
to decompose oe and ae and nobody would type a
German ß unac_except_trans = ßss œoe Œoe æae Æae
ffff fifi flfl . The default for all until someone
protests follows. These decompositions are not
performed by unac, but it is unlikely that
someone would type the composed forms in a
search. unac_except_trans = ßss œoe Œoe æae Æae
ffff fifi flfl</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.MAILDEFCHARSET" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.MAILDEFCHARSET"></a><span class="term"><code class="varname">maildefcharset</code></span></dt>
<dd>
<p>Overrides the default character set for email
messages which don't specify one. This is mainly
useful for readpst (libpst) dumps, which are
utf-8 but do not say so.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.LOCALFIELDS" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.LOCALFIELDS"></a><span class="term"><code class="varname">localfields</code></span></dt>
<dd>
<p>Set fields on all files (usually of a specific
fs area). Syntax is the usual: name = value ;
attr1 = val1 ; [...] value is empty so this needs
an initial semi-colon. This is useful, e.g., for
setting the rclaptg field for application
selection inside mimeview.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.TESTMODIFUSEMTIME"
id=
"RCL.INSTALL.CONFIG.RECOLLCONF.TESTMODIFUSEMTIME"></a><span class="term"><code class="varname">testmodifusemtime</code></span></dt>
<dd>
<p>Use mtime instead of ctime to test if a file
has been modified. The time is used in addition
to the size, which is always used. Setting this
can reduce re-indexing on systems where extended
attributes are used (by some other application),
but not indexed, because changing extended
attributes only affects ctime. Notes: - This may
prevent detection of change in some marginal file
rename cases (the target would need to have the
same size and mtime). - You should probably also
set noxattrfields to 1 in this case, except if
you still prefer to perform xattr indexing, for
example if the local file update pattern makes it
of value (as in general, there is a risk for pure
extended attributes updates without file
modification to go undetected). Perform a full
index reset after changing this.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.NOXATTRFIELDS" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.NOXATTRFIELDS"></a><span class="term"><code class="varname">noxattrfields</code></span></dt>
<dd>
<p>Disable extended attributes conversion to
metadata fields. This probably needs to be set if
testmodifusemtime is set.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.METADATACMDS" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.METADATACMDS"></a><span class="term"><code class="varname">metadatacmds</code></span></dt>
<dd>
<p>Define commands to gather external metadata,
e.g. tmsu tags. There can be several entries,
separated by semi-colons, each defining which
field name the data goes into and the command to
use. Don't forget the initial semi-colon. All the
field names must be different. You can use
aliases in the "field" file if necessary. As a
not too pretty hack conceded to convenience, any
field name beginning with "rclmulti" will be
taken as an indication that the command returns
multiple field values inside a text blob
formatted as a recoll configuration file
("fieldname = fieldvalue" lines). The rclmultixx
name will be ignored, and field names and values
will be parsed from the data. Example:
metadatacmds = ; tags = tmsu tags %f; rclmulti1 =
cmdOutputsConf %f</p>
</dd>
</dl>
</div>
</div>
<div class="sect3">
<div class="titlepage">
<div>
<div>
<h4 class="title"><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.STORE" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.STORE"></a>Parameters
affecting where and how we store things</h4>
</div>
</div>
</div>
<div class="variablelist">
<dl class="variablelist">
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.CACHEDIR" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.CACHEDIR"></a><span class="term"><code class="varname">cachedir</code></span></dt>
<dd>
<p>Top directory for Recoll data. Recoll data
directories are normally located relative to the
configuration directory (e.g. ~/.recoll/xapiandb,
~/.recoll/mboxcache). If 'cachedir' is set, the
directories are stored under the specified value
instead (e.g. if cachedir is ~/.cache/recoll, the
default dbdir would be ~/.cache/recoll/xapiandb).
This affects dbdir, webcachedir, mboxcachedir,
aspellDicDir, which can still be individually
specified to override cachedir. Note that if you
have multiple configurations, each must have a
different cachedir, there is no automatic
computation of a subpath under cachedir.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.MAXFSOCCUPPC" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.MAXFSOCCUPPC"></a><span class="term"><code class="varname">maxfsoccuppc</code></span></dt>
<dd>
<p>Maximum file system occupation over which we
stop indexing. The value is a percentage,
corresponding to what the "Capacity" df output
column shows. The default value is 0, meaning no
checking.</p>
</dd>
<dt><a name="RCL.INSTALL.CONFIG.RECOLLCONF.DBDIR"
id=
"RCL.INSTALL.CONFIG.RECOLLCONF.DBDIR"></a><span class="term"><code class="varname">dbdir</code></span></dt>
<dd>
<p>Xapian database directory location. This will
be created on first indexing. If the value is not
an absolute path, it will be interpreted as
relative to cachedir if set, or the configuration
directory (-c argument or $RECOLL_CONFDIR). If
nothing is specified, the default is then
~/.recoll/xapiandb/</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.IDXSTATUSFILE" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.IDXSTATUSFILE"></a><span class="term"><code class="varname">idxstatusfile</code></span></dt>
<dd>
<p>Name of the scratch file where the indexer
process updates its status. Default:
idxstatus.txt inside the configuration
directory.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.MBOXCACHEDIR" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.MBOXCACHEDIR"></a><span class="term"><code class="varname">mboxcachedir</code></span></dt>
<dd>
<p>Directory location for storing mbox message
offsets cache files. This is normally 'mboxcache'
under cachedir if set, or else under the
configuration directory, but it may be useful to
share a directory between different
configurations.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.MBOXCACHEMINMBS" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.MBOXCACHEMINMBS"></a><span class="term"><code class="varname">mboxcacheminmbs</code></span></dt>
<dd>
<p>Minimum mbox file size over which we cache the
offsets. There is really no sense in caching
offsets for small files. The default is 5 MB.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.MBOXMAXMSGMBS" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.MBOXMAXMSGMBS"></a><span class="term"><code class="varname">mboxmaxmsgmbs</code></span></dt>
<dd>
<p>Maximum mbox member message size in megabytes.
Size over which we assume that the mbox format is
bad or we misinterpreted it, at which point we
just stop processing the file.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.WEBCACHEDIR" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.WEBCACHEDIR"></a><span class="term"><code class="varname">webcachedir</code></span></dt>
<dd>
<p>Directory where we store the archived web
pages. This is only used by the web history
indexing code Default: cachedir/webcache if
cachedir is set, else
$RECOLL_CONFDIR/webcache</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.WEBCACHEMAXMBS" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.WEBCACHEMAXMBS"></a><span class="term"><code class="varname">webcachemaxmbs</code></span></dt>
<dd>
<p>Maximum size in MB of the Web archive. This is
only used by the web history indexing code.
Default: 40 MB. Reducing the size will not
physically truncate the file.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.WEBQUEUEDIR" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.WEBQUEUEDIR"></a><span class="term"><code class="varname">webqueuedir</code></span></dt>
<dd>
<p>The path to the Web indexing queue. This used
to be hard-coded in the old plugin as
~/.recollweb/ToIndex so there would be no need or
possibility to change it, but the WebExtensions
plugin now downloads the files to the user
Downloads directory, and a script moves them to
webqueuedir. The script reads this value from the
config so it has become possible to change
it.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.WEBDOWNLOADSDIR" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.WEBDOWNLOADSDIR"></a><span class="term"><code class="varname">webdownloadsdir</code></span></dt>
<dd>
<p>The path to browser downloads directory. This
is where the new browser add-on extension has to
create the files. They are then moved by a script
to webqueuedir.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.ASPELLDICDIR" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.ASPELLDICDIR"></a><span class="term"><code class="varname">aspellDicDir</code></span></dt>
<dd>
<p>Aspell dictionary storage directory location.
The aspell dictionary (aspdict.(lang).rws) is
normally stored in the directory specified by
cachedir if set, or under the configuration
directory.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.FILTERSDIR" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.FILTERSDIR"></a><span class="term"><code class="varname">filtersdir</code></span></dt>
<dd>
<p>Directory location for executable input
handlers. If RECOLL_FILTERSDIR is set in the
environment, we use it instead. Defaults to
$prefix/share/recoll/filters. Can be redefined
for subdirectories.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.ICONSDIR" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.ICONSDIR"></a><span class="term"><code class="varname">iconsdir</code></span></dt>
<dd>
<p>Directory location for icons. The only reason
to change this would be if you want to change the
icons displayed in the result list. Defaults to
$prefix/share/recoll/images</p>
</dd>
</dl>
</div>
</div>
<div class="sect3">
<div class="titlepage">
<div>
<div>
<h4 class="title"><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.PERFS" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.PERFS"></a>Parameters
affecting indexing performance and resource
usage</h4>
</div>
</div>
</div>
<div class="variablelist">
<dl class="variablelist">
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.IDXFLUSHMB" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.IDXFLUSHMB"></a><span class="term"><code class="varname">idxflushmb</code></span></dt>
<dd>
<p>Threshold (megabytes of new data) where we
flush from memory to disk index. Setting this
allows some control over memory usage by the
indexer process. A value of 0 means no explicit
flushing, which lets Xapian perform its own
thing, meaning flushing every
$XAPIAN_FLUSH_THRESHOLD documents created,
modified or deleted: as memory usage depends on
average document size, not only document count,
the Xapian approach is is not very useful, and
you should let Recoll manage the flushes. The
program compiled value is 0. The configured
default value (from this file) is now 50 MB, and
should be ok in many cases. You can set it as low
as 10 to conserve memory, but if you are looking
for maximum speed, you may want to experiment
with values between 20 and 200. In my experience,
values beyond this are always counterproductive.
If you find otherwise, please drop me a note.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.FILTERMAXSECONDS"
id="RCL.INSTALL.CONFIG.RECOLLCONF.FILTERMAXSECONDS">
</a><span class="term"><code class=
"varname">filtermaxseconds</code></span></dt>
<dd>
<p>Maximum external filter execution time in
seconds. Default 1200 (20mn). Set to 0 for no
limit. This is mainly to avoid infinite loops in
postscript files (loop.ps)</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.FILTERMAXMBYTES" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.FILTERMAXMBYTES"></a><span class="term"><code class="varname">filtermaxmbytes</code></span></dt>
<dd>
<p>Maximum virtual memory space for filter
processes (setrlimit(RLIMIT_AS)), in megabytes.
Note that this includes any mapped libs (there is
no reliable Linux way to limit the data space
only), so we need to be a bit generous here.
Anything over 2000 will be ignored on 32 bits
machines.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.THRQSIZES" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.THRQSIZES"></a><span class="term"><code class="varname">thrQSizes</code></span></dt>
<dd>
<p>Stage input queues configuration. There are
three internal queues in the indexing pipeline
stages (file data extraction, terms generation,
index update). This parameter defines the queue
depths for each stage (three integer values). If
a value of -1 is given for a given stage, no
queue is used, and the thread will go on
performing the next stage. In practise, deep
queues have not been shown to increase
performance. Default: a value of 0 for the first
queue tells Recoll to perform autoconfiguration
based on the detected number of CPUs (no need for
the two other values in this case). Use thrQSizes
= -1 -1 -1 to disable multithreading
entirely.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.THRTCOUNTS" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.THRTCOUNTS"></a><span class="term"><code class="varname">thrTCounts</code></span></dt>
<dd>
<p>Number of threads used for each indexing
stage. The three stages are: file data
extraction, terms generation, index update). The
use of the counts is also controlled by some
special values in thrQSizes: if the first queue
depth is 0, all counts are ignored
(autoconfigured); if a value of -1 is used for a
queue depth, the corresponding thread count is
ignored. It makes no sense to use a value other
than 1 for the last stage because updating the
Xapian index is necessarily single-threaded (and
protected by a mutex).</p>
</dd>
</dl>
</div>
</div>
<div class="sect3">
<div class="titlepage">
<div>
<div>
<h4 class="title"><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.MISC" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.MISC"></a>Miscellaneous
parameters</h4>
</div>
</div>
</div>
<div class="variablelist">
<dl class="variablelist">
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.LOGLEVEL" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.LOGLEVEL"></a><span class="term"><code class="varname">loglevel</code></span></dt>
<dd>
<p>Log file verbosity 1-6. A value of 2 will
print only errors and warnings. 3 will print
information like document updates, 4 is quite
verbose and 6 very verbose.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.LOGFILENAME" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.LOGFILENAME"></a><span class="term"><code class="varname">logfilename</code></span></dt>
<dd>
<p>Log file destination. Use 'stderr' (default)
to write to the console.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.IDXLOGLEVEL" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.IDXLOGLEVEL"></a><span class="term"><code class="varname">idxloglevel</code></span></dt>
<dd>
<p>Override loglevel for the indexer.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.IDXLOGFILENAME" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.IDXLOGFILENAME"></a><span class="term"><code class="varname">idxlogfilename</code></span></dt>
<dd>
<p>Override logfilename for the indexer.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.DAEMLOGLEVEL" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.DAEMLOGLEVEL"></a><span class="term"><code class="varname">daemloglevel</code></span></dt>
<dd>
<p>Override loglevel for the indexer in real time
mode. The default is to use the idx... values if
set, else the log... values.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.DAEMLOGFILENAME" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.DAEMLOGFILENAME"></a><span class="term"><code class="varname">daemlogfilename</code></span></dt>
<dd>
<p>Override logfilename for the indexer in real
time mode. The default is to use the idx...
values if set, else the log... values.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.PYLOGLEVEL" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.PYLOGLEVEL"></a><span class="term"><code class="varname">pyloglevel</code></span></dt>
<dd>
<p>Override loglevel for the python module.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.PYLOGFILENAME" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.PYLOGFILENAME"></a><span class="term"><code class="varname">pylogfilename</code></span></dt>
<dd>
<p>Override logfilename for the python
module.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.ORGIDXCONFDIR" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.ORGIDXCONFDIR"></a><span class="term"><code class="varname">orgidxconfdir</code></span></dt>
<dd>
<p>Original location of the configuration
directory. This is used exclusively for movable
datasets. Locating the configuration directory
inside the directory tree makes it possible to
provide automatic query time path translations
once the data set has moved (for example, because
it has been mounted on another location).</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.CURIDXCONFDIR" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.CURIDXCONFDIR"></a><span class="term"><code class="varname">curidxconfdir</code></span></dt>
<dd>
<p>Current location of the configuration
directory. Complement orgidxconfdir for movable
datasets. This should be used if the
configuration directory has been copied from the
dataset to another location, either because the
dataset is readonly and an r/w copy is desired,
or for performance reasons. This records the
original moved location before copy, to allow
path translation computations. For example if a
dataset originally indexed as
'/home/me/mydata/config' has been mounted to
'/media/me/mydata', and the GUI is running from a
copied configuration, orgidxconfdir would be
'/home/me/mydata/config', and curidxconfdir (as
set in the copied configuration) would be
'/media/me/mydata/config'.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.IDXRUNDIR" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.IDXRUNDIR"></a><span class="term"><code class="varname">idxrundir</code></span></dt>
<dd>
<p>Indexing process current directory. The input
handlers sometimes leave temporary files in the
current directory, so it makes sense to have
recollindex chdir to some temporary directory. If
the value is empty, the current directory is not
changed. If the value is (literal) tmp, we use
the temporary directory as set by the environment
(RECOLL_TMPDIR else TMPDIR else /tmp). If the
value is an absolute path to a directory, we go
there.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.CHECKNEEDRETRYINDEXSCRIPT"
id=
"RCL.INSTALL.CONFIG.RECOLLCONF.CHECKNEEDRETRYINDEXSCRIPT">
</a><span class="term"><code class=
"varname">checkneedretryindexscript</code></span></dt>
<dd>
<p>Script used to heuristically check if we need
to retry indexing files which previously failed.
The default script checks the modified dates on
/usr/bin and /usr/local/bin. A relative path will
be looked up in the filters dirs, then in the
path. Use an absolute path to do otherwise.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.RECOLLHELPERPATH"
id="RCL.INSTALL.CONFIG.RECOLLCONF.RECOLLHELPERPATH">
</a><span class="term"><code class=
"varname">recollhelperpath</code></span></dt>
<dd>
<p>Additional places to search for helper
executables. This is only used on Windows for
now.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.IDXABSMLEN" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.IDXABSMLEN"></a><span class="term"><code class="varname">idxabsmlen</code></span></dt>
<dd>
<p>Length of abstracts we store while indexing.
Recoll stores an abstract for each indexed file.
The text can come from an actual 'abstract'
section in the document or will just be the
beginning of the document. It is stored in the
index so that it can be displayed inside the
result lists without decoding the original file.
The idxabsmlen parameter defines the size of the
stored abstract. The default value is 250 bytes.
The search interface gives you the choice to
display this stored text or a synthetic abstract
built by extracting text around the search terms.
If you always prefer the synthetic abstract, you
can reduce this value and save a little
space.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.IDXMETASTOREDLEN"
id="RCL.INSTALL.CONFIG.RECOLLCONF.IDXMETASTOREDLEN">
</a><span class="term"><code class=
"varname">idxmetastoredlen</code></span></dt>
<dd>
<p>Truncation length of stored metadata fields.
This does not affect indexing (the whole field is
processed anyway), just the amount of data stored
in the index for the purpose of displaying fields
inside result lists or previews. The default
value is 150 bytes which may be too low if you
have custom fields.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.IDXTEXTTRUNCATELEN"
id=
"RCL.INSTALL.CONFIG.RECOLLCONF.IDXTEXTTRUNCATELEN"></a><span class="term"><code class="varname">idxtexttruncatelen</code></span></dt>
<dd>
<p>Truncation length for all document texts. Only
index the beginning of documents. This is not
recommended except if you are sure that the
interesting keywords are at the top and have
severe disk space issues.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.ASPELLLANGUAGE" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.ASPELLLANGUAGE"></a><span class="term"><code class="varname">aspellLanguage</code></span></dt>
<dd>
<p>Language definitions to use when creating the
aspell dictionary. The value must match a set of
aspell language definition files. You can type
"aspell dicts" to see a list The default if this
is not set is to use the NLS environment to guess
the value. The values are the 2-letter language
codes (e.g. 'en', 'fr'...)</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.ASPELLADDCREATEPARAM"
id=
"RCL.INSTALL.CONFIG.RECOLLCONF.ASPELLADDCREATEPARAM">
</a><span class="term"><code class=
"varname">aspellAddCreateParam</code></span></dt>
<dd>
<p>Additional option and parameter to aspell
dictionary creation command. Some aspell packages
may need an additional option (e.g. on Debian
Jessie: --local-data-dir=/usr/lib/aspell). See
Debian bug 772415.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.ASPELLKEEPSTDERR"
id="RCL.INSTALL.CONFIG.RECOLLCONF.ASPELLKEEPSTDERR">
</a><span class="term"><code class=
"varname">aspellKeepStderr</code></span></dt>
<dd>
<p>Set this to have a look at aspell dictionary
creation errors. There are always many, so this
is mostly for debugging.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.NOASPELL" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.NOASPELL"></a><span class="term"><code class="varname">noaspell</code></span></dt>
<dd>
<p>Disable aspell use. The aspell dictionary
generation takes time, and some combinations of
aspell version, language, and local terms, result
in aspell crashing, so it sometimes makes sense
to just disable the thing.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.MONAUXINTERVAL" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.MONAUXINTERVAL"></a><span class="term"><code class="varname">monauxinterval</code></span></dt>
<dd>
<p>Auxiliary database update interval. The real
time indexer only updates the auxiliary databases
(stemdb, aspell) periodically, because it would
be too costly to do it for every document change.
The default period is one hour.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.MONIXINTERVAL" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.MONIXINTERVAL"></a><span class="term"><code class="varname">monixinterval</code></span></dt>
<dd>
<p>Minimum interval (seconds) between processings
of the indexing queue. The real time indexer does
not process each event when it comes in, but lets
the queue accumulate, to diminish overhead and to
aggregate multiple events affecting the same
file. Default 30 S.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.MONDELAYPATTERNS"
id="RCL.INSTALL.CONFIG.RECOLLCONF.MONDELAYPATTERNS">
</a><span class="term"><code class=
"varname">mondelaypatterns</code></span></dt>
<dd>
<p>Timing parameters for the real time indexing.
Definitions for files which get a longer delay
before reindexing is allowed. This is for
fast-changing files, that should only be
reindexed once in a while. A list of
wildcardPattern:seconds pairs. The patterns are
matched with fnmatch(pattern, path, 0) You can
quote entries containing white space with double
quotes (quote the whole entry, not the pattern).
The default is empty. Example: mondelaypatterns =
*.log:20 "*with spaces.*:30"</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.IDXNICEPRIO" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.IDXNICEPRIO"></a><span class="term"><code class="varname">idxniceprio</code></span></dt>
<dd>
<p>"nice" process priority for the indexing
processes. Default: 19 (lowest) Appeared with
1.26.5. Prior versions were fixed at 19.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.MONIONICECLASS" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.MONIONICECLASS"></a><span class="term"><code class="varname">monioniceclass</code></span></dt>
<dd>
<p>ionice class for the indexing process. Despite
the misleading name, and on platforms where this
is supported, this affects all indexing
processes, not only the real time/monitoring
ones. The default value is 3 (use lowest "Idle"
priority).</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.MONIONICECLASSDATA"
id=
"RCL.INSTALL.CONFIG.RECOLLCONF.MONIONICECLASSDATA"></a><span class="term"><code class="varname">monioniceclassdata</code></span></dt>
<dd>
<p>ionice class level parameter if the class
supports it. The default is empty, as the default
"Idle" class has no levels.</p>
</dd>
</dl>
</div>
</div>
<div class="sect3">
<div class="titlepage">
<div>
<div>
<h4 class="title"><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.QUERY" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.QUERY"></a>Query-time
parameters (no impact on the index)</h4>
</div>
</div>
</div>
<div class="variablelist">
<dl class="variablelist">
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.AUTODIACSENS" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.AUTODIACSENS"></a><span class="term"><code class="varname">autodiacsens</code></span></dt>
<dd>
<p>auto-trigger diacritics sensitivity (raw index
only). IF the index is not stripped, decide if we
automatically trigger diacritics sensitivity if
the search term has accented characters (not in
unac_except_trans). Else you need to use the
query language and the "D" modifier to specify
diacritics sensitivity. Default is no.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.AUTOCASESENS" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.AUTOCASESENS"></a><span class="term"><code class="varname">autocasesens</code></span></dt>
<dd>
<p>auto-trigger case sensitivity (raw index
only). IF the index is not stripped (see
indexStripChars), decide if we automatically
trigger character case sensitivity if the search
term has upper-case characters in any but the
first position. Else you need to use the query
language and the "C" modifier to specify
character-case sensitivity. Default is yes.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.MAXTERMEXPAND" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.MAXTERMEXPAND"></a><span class="term"><code class="varname">maxTermExpand</code></span></dt>
<dd>
<p>Maximum query expansion count for a single
term (e.g.: when using wildcards). This only
affects queries, not indexing. We used to not
limit this at all (except for filenames where the
limit was too low at 1000), but it is
unreasonable with a big index. Default 10000.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.MAXXAPIANCLAUSES"
id="RCL.INSTALL.CONFIG.RECOLLCONF.MAXXAPIANCLAUSES">
</a><span class="term"><code class=
"varname">maxXapianClauses</code></span></dt>
<dd>
<p>Maximum number of clauses we add to a single
Xapian query. This only affects queries, not
indexing. In some cases, the result of term
expansion can be multiplicative, and we want to
avoid eating all the memory. Default 50000.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.SNIPPETMAXPOSWALK"
id=
"RCL.INSTALL.CONFIG.RECOLLCONF.SNIPPETMAXPOSWALK"></a><span class="term"><code class="varname">snippetMaxPosWalk</code></span></dt>
<dd>
<p>Maximum number of positions we walk while
populating a snippet for the result list. The
default of 1,000,000 may be insufficient for very
big documents, the consequence would be snippets
with possibly meaning-altering missing words.</p>
</dd>
</dl>
</div>
</div>
<div class="sect3">
<div class="titlepage">
<div>
<div>
<h4 class="title"><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.PDF" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.PDF"></a>Parameters
for the PDF input script</h4>
</div>
</div>
</div>
<div class="variablelist">
<dl class="variablelist">
<dt><a name="RCL.INSTALL.CONFIG.RECOLLCONF.PDFOCR"
id=
"RCL.INSTALL.CONFIG.RECOLLCONF.PDFOCR"></a><span class="term"><code class="varname">pdfocr</code></span></dt>
<dd>
<p>Attempt OCR of PDF files with no text content.
This can be defined in subdirectories. The
default is off because OCR is so very slow.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.PDFATTACH" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.PDFATTACH"></a><span class="term"><code class="varname">pdfattach</code></span></dt>
<dd>
<p>Enable PDF attachment extraction by executing
pdftk (if available). This is normally disabled,
because it does slow down PDF indexing a bit even
if not one attachment is ever found.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.PDFEXTRAMETA" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.PDFEXTRAMETA"></a><span class="term"><code class="varname">pdfextrameta</code></span></dt>
<dd>
<p>Extract text from selected XMP metadata tags.
This is a space-separated list of qualified XMP
tag names. Each element can also include a
translation to a Recoll field name, separated by
a '|' character. If the second element is absent,
the tag name is used as the Recoll field names.
You will also need to add specifications to the
"fields" file to direct processing of the
extracted data.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.PDFEXTRAMETAFIX" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.PDFEXTRAMETAFIX"></a><span class="term"><code class="varname">pdfextrametafix</code></span></dt>
<dd>
<p>Define name of XMP field editing script. This
defines the name of a script to be loaded for
editing XMP field values. The script should
define a 'MetaFixer' class with a metafix()
method which will be called with the qualified
tag name and value of each selected field, for
editing or erasing. A new instance is created for
each document, so that the object can keep state
for, e.g. eliminating duplicate values.</p>
</dd>
</dl>
</div>
</div>
<div class="sect3">
<div class="titlepage">
<div>
<div>
<h4 class="title"><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.OCR" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.OCR"></a>Parameters
for OCR processing</h4>
</div>
</div>
</div>
<div class="variablelist">
<dl class="variablelist">
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.OCRPROGS" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.OCRPROGS"></a><span class="term"><code class="varname">ocrprogs</code></span></dt>
<dd>
<p>OCR modules to try. The top OCR script will
try to load the corresponding modules in order
and use the first which reports being capable of
performing OCR on the input file. Modules for
tesseract (tesseract) and ABBYY FineReader
(abbyy) are present in the standard distribution.
For compatibility with the previous version, if
this is not defined at all, the default value is
"tesseract". Use an explicit empty value if
needed. A value of "abbyy tesseract" will try
everything.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.OCRCACHEDIR" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.OCRCACHEDIR"></a><span class="term"><code class="varname">ocrcachedir</code></span></dt>
<dd>
<p>Location for caching OCR data. The default if
this is empty or undefined is to store the cached
OCR data under $RECOLL_CONFDIR/ocrcache.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.TESSERACTLANG" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.TESSERACTLANG"></a><span class="term"><code class="varname">tesseractlang</code></span></dt>
<dd>
<p>Language to assume for tesseract OCR.
Important for improving the OCR accuracy. This
can also be set through the contents of a file in
the currently processed directory. See the
rclocrtesseract.py script. Example values: eng,
fra... See the tesseract documentation.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.TESSERACTCMD" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.TESSERACTCMD"></a><span class="term"><code class="varname">tesseractcmd</code></span></dt>
<dd>
<p>Path for the tesseract command. Do not quote.
This is mostly useful on Windows, or for
specifying a non-default tesseract command. E.g.
on Windows. tesseractcmd =
C:/ProgramFiles(x86)/Tesseract-OCR/tesseract.exe</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.ABBYYLANG" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.ABBYYLANG"></a><span class="term"><code class="varname">abbyylang</code></span></dt>
<dd>
<p>Language to assume for abbyy OCR. Important
for improving the OCR accuracy. This can also be
set through the contents of a file in the
currently processed directory. See the
rclocrabbyy.py script. Typical values: English,
French... See the ABBYY documentation.</p>
</dd>
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.ABBYYCMD" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.ABBYYCMD"></a><span class="term"><code class="varname">abbyycmd</code></span></dt>
<dd>
<p>Path for the abbyy command The ABBY directory
is usually not in the path, so you should set
this.</p>
</dd>
</dl>
</div>
</div>
<div class="sect3">
<div class="titlepage">
<div>
<div>
<h4 class="title"><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.SPECLOCATIONS" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.SPECLOCATIONS"></a>Parameters
set for specific locations</h4>
</div>
</div>
</div>
<div class="variablelist">
<dl class="variablelist">
<dt><a name=
"RCL.INSTALL.CONFIG.RECOLLCONF.MHMBOXQUIRKS" id=
"RCL.INSTALL.CONFIG.RECOLLCONF.MHMBOXQUIRKS"></a><span class="term"><code class="varname">mhmboxquirks</code></span></dt>
<dd>
<p>Enable thunderbird/mozilla-seamonkey mbox
format quirks Set this for the directory where
the email mbox files are stored.</p>
</dd>
</dl>
</div>
</div>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.INSTALL.CONFIG.FIELDS" id=
"RCL.INSTALL.CONFIG.FIELDS"></a>5.4.3.&nbsp;The
fields file</h3>
</div>
</div>
</div>
<p>This file contains information about dynamic fields
handling in <span class="application">Recoll</span>. Some
very basic fields have hard-wired behaviour, and, mostly,
you should not change the original data inside the
<code class="filename">fields</code> file. But you can
create custom fields fitting your data and handle them
just like they were native ones.</p>
<p>The <code class="filename">fields</code> file has
several sections, which each define an aspect of fields
processing. Quite often, you'll have to modify several
sections to obtain the desired behaviour.</p>
<p>We will only give a short description here, you should
refer to the comments inside the default file for more
detailed information.</p>
<p>Field names should be lowercase alphabetic ASCII.</p>
<div class="variablelist">
<dl class="variablelist">
<dt><span class="term">[prefixes]</span></dt>
<dd>
<p>A field becomes indexed (searchable) by having a
prefix defined in this section. There is a more
complete explanation of what prefixes are in used
by a standard recoll installation. In a nutshell:
extension prefixes should be all caps, begin with
XY, and short. E.g. XYMFLD.</p>
</dd>
<dt><span class="term">[values]</span></dt>
<dd>
<p>Fields listed in this section will be stored as
<span class="application">Xapian</span>
<code class="literal">values</code> inside the
index. This makes them available for range queries,
allowing to filter results according to the field
value. This feature currently supports string and
integer data. See the comments in the file for more
detail</p>
</dd>
<dt><span class="term">[stored]</span></dt>
<dd>
<p>A field becomes stored (displayable inside
results) by having its name listed in this section
(typically with an empty value).</p>
</dd>
<dt><span class="term">[aliases]</span></dt>
<dd>
<p>This section defines lists of synonyms for the
canonical names used inside the <code class=
"literal">[prefixes]</code> and <code class=
"literal">[stored]</code> sections</p>
</dd>
<dt><span class="term">[queryaliases]</span></dt>
<dd>
<p>This section also defines aliases for the
canonic field names, with the difference that the
substitution will only be used at query time,
avoiding any possibility that the value would
pick-up random metadata from documents.</p>
</dd>
<dt><span class="term">handler-specific
sections</span></dt>
<dd>
<p>Some input handlers may need specific
configuration for handling fields. Only the email
message handler currently has such a section (named
<code class="literal">[mail]</code>). It allows
indexing arbitrary email headers in addition to the
ones indexed by default. Other such sections may
appear in the future.</p>
</dd>
</dl>
</div>
<p>Here follows a small example of a personal
<code class="filename">fields</code> file. This would
extract a specific email header and use it as a
searchable field, with data displayable inside result
lists. (Side note: as the email handler does no decoding
on the values, only plain ascii headers can be indexed,
and only the first occurrence will be used for headers
that occur several times).</p>
<pre class="programlisting">[prefixes]
# Index mailmytag contents (with the given prefix)
mailmytag = XMTAG
[stored]
# Store mailmytag inside the document data record (so that it can be
# displayed - as %(mailmytag) - in result lists).
mailmytag =
[queryaliases]
filename = fn
containerfilename = cfn
[mail]
# Extract the X-My-Tag mail header, and use it internally with the
# mailmytag field name
x-my-tag = mailmytag
</pre>
<div class="sect3">
<div class="titlepage">
<div>
<div>
<h4 class="title"><a name=
"RCL.INSTALL.CONFIG.FIELDS.XATTR" id=
"RCL.INSTALL.CONFIG.FIELDS.XATTR"></a>Extended
attributes in the fields file</h4>
</div>
</div>
</div>
<p><span class="application">Recoll</span> versions
1.19 and later process user extended file attributes as
documents fields by default.</p>
<p>Attributes are processed as fields of the same name,
after removing the <code class="literal">user</code>
prefix on Linux.</p>
<p>The <code class="literal">[xattrtofields]</code>
section of the <code class="filename">fields</code>
file allows specifying translations from extended
attributes names to <span class=
"application">Recoll</span> field names. An empty
translation disables use of the corresponding attribute
data.</p>
</div>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.INSTALL.CONFIG.MIMEMAP" id=
"RCL.INSTALL.CONFIG.MIMEMAP"></a>5.4.4.&nbsp;The
mimemap file</h3>
</div>
</div>
</div>
<p><code class="filename">mimemap</code> specifies the
file name extension to MIME type mappings.</p>
<p>For file names without an extension, or with an
unknown one, a system command (<span class=
"command"><strong>file</strong></span> <code class=
"option">-i</code>, or <span class=
"command"><strong>xdg-mime</strong></span>) will be
executed to determine the MIME type (this can be switched
off, or the command changed inside the main configuration
file).</p>
<p>All extension values in <code class=
"filename">mimemap</code> must be entered in lower case.
File names extensions are lower-cased for comparison
during indexing, meaning that an upper case <code class=
"filename">mimemap</code> entry will never be
matched.</p>
<p>The mappings can be specified on a per-subtree basis,
which may be useful in some cases. Example: <span class=
"application">okular</span> notes have a <code class=
"filename">.xml</code> extension but should be handled
specially, which is possible because they are usually all
located in one place. Example:</p>
<pre class=
"programlisting">[~/.kde/share/apps/okular/docdata]
.xml = application/x-okular-notes</pre>
<p>The <code class="varname">recoll_noindex</code>
<code class="filename">mimemap</code> variable has been
moved to <code class="filename">recoll.conf</code> and
renamed to <code class=
"varname">noContentSuffixes</code>, while keeping the
same function, as of <span class=
"application">Recoll</span> version 1.21. For older
<span class="application">Recoll</span> versions, see the
documentation for <code class=
"varname">noContentSuffixes</code> but use <code class=
"varname">recoll_noindex</code> in <code class=
"filename">mimemap</code>.</p>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.INSTALL.CONFIG.MIMECONF" id=
"RCL.INSTALL.CONFIG.MIMECONF"></a>5.4.5.&nbsp;The
mimeconf file</h3>
</div>
</div>
</div>
<p>The main purpose of the <code class=
"filename">mimeconf</code> file is to specify how the
different MIME types are handled for indexing. This is
done in the <code class="literal">[index]</code> section,
which should not be modified casually. See the comments
in the file.</p>
<p>The file also contains other definitions which affect
the query language and the GUI, and which, in retrospect,
should have been stored elsewhere.</p>
<p>The <code class="literal">[icons]</code> section
allows you to change the icons which are displayed by the
<span class="command"><strong>recoll</strong></span> GUI
in the result lists (the values are the basenames of the
<code class="literal">png</code> images inside the
<code class="filename">iconsdir</code> directory (which
is itself defined in <code class=
"filename">recoll.conf</code>).</p>
<p>The <code class="literal">[categories]</code> section
defines the groupings of MIME types into <code class=
"literal">categories</code> as used when adding an
<code class="literal">rclcat</code> clause to a <a class=
"link" href="#RCL.SEARCH.LANG" title=
"3.5.&nbsp;The query language">query language</a> query.
<code class="literal">rclcat</code> clauses are also used
by the default <code class="literal">guifilters</code>
buttons in the GUI (see next).</p>
<p>The filter controls appear at the top of the
<span class="command"><strong>recoll</strong></span> GUI,
either as checkboxes just above the result list, or as a
dropbox in the tool area.</p>
<p>By default, they are labeled: <code class=
"literal">media</code>, <code class=
"literal">message</code>, <code class=
"literal">other</code>, <code class=
"literal">presentation</code>, <code class=
"literal">spreadsheet</code> and <code class=
"literal">text</code>, and each maps to a document
category. This is determined in the <code class=
"literal">[guifilters]</code> section, where each control
is defined by a variable naming a query language
fragment.</p>
<p>A simple example will hopefully make things
clearer.</p>
<pre class="programlisting">[guifilters]
Big Books = dir:"~/My Books" size&gt;10K
My Docs = dir:"~/My Documents"
Small Books = dir:"~/My Books" size&lt;10K
System Docs = dir:/usr/share/doc
</pre>
<p>The above definition would create four filter
checkboxes, labelled <code class="literal">Big
Books</code>, <code class="literal">My Docs</code>,
etc.</p>
<p>The text after the equal sign must be a valid query
language fragment, and, when the button is checked, it
will be combined with the rest of the query with an AND
conjunction.</p>
<p>Any name text before a colon character will be erased
in the display, but used for sorting. You can use this to
display the checkboxes in any order you like. For
example, the following would do exactly the same as
above, but ordering the checkboxes in the reverse
order.</p>
<pre class="programlisting">[guifilters]
d:Big Books = dir:"~/My Books" size&gt;10K
c:My Docs = dir:"~/My Documents"
b:Small Books = dir:"~/My Books" size&lt;10K
a:System Docs = dir:/usr/share/doc
</pre>
<p>As you may have guessed, The default <code class=
"literal">[guifilters]</code> section looks like:</p>
<pre class="programlisting">[guifilters]
text = rclcat:text
spreadsheet = rclcat:spreadsheet
presentation = rclcat:presentation
media = rclcat:media
message = rclcat:message
other = rclcat:other
</pre>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.INSTALL.CONFIG.MIMEVIEW" id=
"RCL.INSTALL.CONFIG.MIMEVIEW"></a>5.4.6.&nbsp;The
mimeview file</h3>
</div>
</div>
</div>
<p><code class="filename">mimeview</code> specifies which
programs are started when you click on an <span class=
"guilabel">Open</span> link in a result list. Ie: HTML is
normally displayed using <span class=
"application">firefox</span>, but you may prefer
<span class="application">Konqueror</span>, your
<span class="application">openoffice.org</span> program
might be named <span class=
"command"><strong>oofice</strong></span> instead of
<span class="command"><strong>openoffice</strong></span>
etc.</p>
<p>Changes to this file can be done by direct editing, or
through the <span class=
"command"><strong>recoll</strong></span> GUI preferences
dialog.</p>
<p>If <span class="guilabel">Use desktop preferences to
choose document editor</span> is checked in the
<span class="application">Recoll</span> GUI preferences,
all <code class="filename">mimeview</code> entries will
be ignored except the one labelled <code class=
"literal">application/x-all</code> (which is set to use
<span class="command"><strong>xdg-open</strong></span> by
default).</p>
<p>In this case, the <code class=
"literal">xallexcepts</code> top level variable defines a
list of MIME type exceptions which will be processed
according to the local entries instead of being passed to
the desktop. This is so that specific <span class=
"application">Recoll</span> options such as a page number
or a search string can be passed to applications that
support them, such as the <span class=
"application">evince</span> viewer.</p>
<p>As for the other configuration files, the normal usage
is to have a <code class="filename">mimeview</code>
inside your own configuration directory, with just the
non-default entries, which will override those from the
central configuration file.</p>
<p>All viewer definition entries must be placed under a
<code class="literal">[view]</code> section.</p>
<p>The keys in the file are normally MIME types. You can
add an application tag to specialize the choice for an
area of the filesystem (using a <code class=
"varname">localfields</code> specification in
<code class="filename">mimeconf</code>). The syntax for
the key is <em class=
"replaceable"><code>mimetype</code></em><code class=
"literal">|</code><em class=
"replaceable"><code>tag</code></em></p>
<p>The <code class="varname">nouncompforviewmts</code>
entry, (placed at the top level, outside of the
<code class="literal">[view]</code> section), holds a
list of MIME types that should not be uncompressed before
starting the viewer (if they are found compressed, ie:
<em class=
"replaceable"><code>mydoc.doc.gz</code></em>).</p>
<p>The right side of each assignment holds a command to
be executed for opening the file. The following
substitutions are performed:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style=
"list-style-type: disc;">
<li class="listitem">
<p><b>%D.&nbsp;</b>Document date</p>
</li>
<li class="listitem">
<p><b>%f.&nbsp;</b>File name. This may be the name
of a temporary file if it was necessary to create
one (ie: to extract a subdocument from a
container).</p>
</li>
<li class="listitem">
<p><b>%i.&nbsp;</b>Internal path, for subdocuments
of containers. The format depends on the container
type. If this appears in the command line,
<span class="application">Recoll</span> will not
create a temporary file to extract the subdocument,
expecting the called application (possibly a
script) to be able to handle it.</p>
</li>
<li class="listitem">
<p><b>%M.&nbsp;</b>MIME type</p>
</li>
<li class="listitem">
<p><b>%p.&nbsp;</b>Page index. Only significant for
a subset of document types, currently only PDF,
Postscript and DVI files. Can be used to start the
editor at the right page for a match or
snippet.</p>
</li>
<li class="listitem">
<p><b>%s.&nbsp;</b>Search term. The value will only
be set for documents with indexed page numbers (ie:
PDF). The value will be one of the matched search
terms. It would allow pre-setting the value in the
"Find" entry inside Evince for example, for easy
highlighting of the term.</p>
</li>
<li class="listitem">
<p><b>%u.&nbsp;</b>Url.</p>
</li>
</ul>
</div>
<p>In addition to the predefined values above, all
strings like <code class="literal">%(fieldname)</code>
will be replaced by the value of the field named
<code class="literal">fieldname</code> for the document.
This could be used in combination with field
customisation to help with opening the document.</p>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.INSTALL.CONFIG.PTRANS" id=
"RCL.INSTALL.CONFIG.PTRANS"></a>5.4.7.&nbsp;The
<code class="filename">ptrans</code> file</h3>
</div>
</div>
</div>
<p><code class="filename">ptrans</code> specifies
query-time path translations. These can be useful in
<a class="link" href="#RCL.SEARCH.PTRANS" title=
"3.8.&nbsp;Path translations">multiple cases</a>.</p>
<p>The file has a section for any index which needs
translations, either the main one or additional query
indexes. The sections are named with the <span class=
"application">Xapian</span> index directory names. No
slash character should exist at the end of the paths (all
comparisons are textual). An example should make things
sufficiently clear</p>
<pre class="programlisting">
[/home/me/.recoll/xapiandb]
/this/directory/moved = /to/this/place
[/path/to/additional/xapiandb]
/server/volume1/docdir = /net/server/volume1/docdir
/server/volume2/docdir = /net/server/volume2/docdir
</pre>
</div>
<div class="sect2">
<div class="titlepage">
<div>
<div>
<h3 class="title"><a name=
"RCL.INSTALL.CONFIG.EXAMPLES" id=
"RCL.INSTALL.CONFIG.EXAMPLES"></a>5.4.8.&nbsp;Examples
of configuration adjustments</h3>
</div>
</div>
</div>
<div class="sect3">
<div class="titlepage">
<div>
<div>
<h4 class="title"><a name=
"RCL.INSTALL.CONFIG.EXAMPLES.ADDVIEW" id=
"RCL.INSTALL.CONFIG.EXAMPLES.ADDVIEW"></a>Adding
an external viewer for an non-indexed type</h4>
</div>
</div>
</div>
<p>Imagine that you have some kind of file which does
not have indexable content, but for which you would
like to have a functional <span class=
"guilabel">Open</span> link in the result list (when
found by file name). The file names end in <em class=
"replaceable"><code>.blob</code></em> and can be
displayed by application <em class=
"replaceable"><code>blobviewer</code></em>.</p>
<p>You need two entries in the configuration files for
this to work:</p>
<div class="itemizedlist">
<ul class="itemizedlist" style=
"list-style-type: disc;">
<li class="listitem">
<p>In <code class=
"filename">$RECOLL_CONFDIR/mimemap</code>
(typically <code class=
"filename">~/.recoll/mimemap</code>), add the
following line:</p>
<pre class="programlisting">
.blob = application/x-blobapp
</pre>
<p>Note that the MIME type is made up here, and
you could call it <em class=
"replaceable"><code>diesel/oil</code></em> just
the same.</p>
</li>
<li class="listitem">
<p>In <code class=
"filename">$RECOLL_CONFDIR/mimeview</code> under
the <code class="literal">[view]</code> section,
add:</p>
<pre class="programlisting">
application/x-blobapp = blobviewer %f
</pre>
<p>We are supposing that <em class=
"replaceable"><code>blobviewer</code></em> wants
a file name parameter here, you would use
<code class="literal">%u</code> if it liked URLs
better.</p>
</li>
</ul>
</div>
<p>If you just wanted to change the application used by
<span class="application">Recoll</span> to display a
MIME type which it already knows, you would just need
to edit <code class="filename">mimeview</code>. The
entries you add in your personal file override those in
the central configuration, which you do not need to
alter. <code class="filename">mimeview</code> can also
be modified from the Gui.</p>
</div>
<div class="sect3">
<div class="titlepage">
<div>
<div>
<h4 class="title"><a name=
"RCL.INSTALL.CONFIG.EXAMPLES.ADDINDEX" id=
"RCL.INSTALL.CONFIG.EXAMPLES.ADDINDEX"></a>Adding
indexing support for a new file type</h4>
</div>
</div>
</div>
<p>Let us now imagine that the above <em class=
"replaceable"><code>.blob</code></em> files actually
contain indexable text and that you know how to extract
it with a command line program. Getting <span class=
"application">Recoll</span> to index the files is easy.
You need to perform the above alteration, and also to
add data to the <code class="filename">mimeconf</code>
file (typically in <code class=
"filename">~/.recoll/mimeconf</code>):</p>
<div class="itemizedlist">
<ul class="itemizedlist" style=
"list-style-type: disc;">
<li class="listitem">
<p>Under the <code class="literal">[index]</code>
section, add the following line (more about the
<em class="replaceable"><code>rclblob</code></em>
indexing script later):</p>
<pre class="programlisting">
application/x-blobapp = exec rclblob</pre>
<p>Or if the files are mostly text and you don't
need to process them for indexing:</p>
<pre class="programlisting">
application/x-blobapp = internal text/plain</pre>
</li>
<li class="listitem">
<p>Under the <code class="literal">[icons]</code>
section, you should choose an icon to be
displayed for the files inside the result lists.
Icons are normally 64x64 pixels PNG files which
live in <code class=
"filename">/usr/share/recoll/images</code>.</p>
</li>
<li class="listitem">
<p>Under the <code class=
"literal">[categories]</code> section, you should
add the MIME type where it makes sense (you can
also create a category). Categories may be used
for filtering in advanced search.</p>
</li>
</ul>
</div>
<p>The <em class=
"replaceable"><code>rclblob</code></em> handler should
be an executable program or script which exists inside
<code class=
"filename">/usr/share/recoll/filters</code>. It will be
given a file name as argument and should output the
text or html contents on the standard output.</p>
<p>The <a class="link" href="#RCL.PROGRAM.FILTERS"
title=
"4.1.&nbsp;Writing a document input handler">filter
programming</a> section describes in more detail how to
write an input handler.</p>
</div>
</div>
</div>
</div>
</div>
</body>
</html>