Merge branch 'master' of https://framagit.org/medoc92/recoll

packaging/debian/buildppa.sh

@@ -13,7 +13,7 @@
 
 PPA_KEYID=7808CE96D38B9201
 
-RCLVERS=1.31.5
+RCLVERS=1.31.6
 SCOPEVERS=1.20.2.4
 GSSPVERS=1.1.1
 PPAVERS=1
@@ -85,7 +85,7 @@ done
 
 
 ### KIO.
-series="bionic focal groovy hirsute impish"
+#series="bionic focal hirsute impish"
 series=
 
 debdir=debiankio

packaging/debian/debian/changelog

@@ -1,3 +1,9 @@
+recoll (1.31.6-1~ppaPPAVERS~SERIES1) SERIES; urgency=low
+
+  * Almost no change: translation files update.
+    
+ -- Jean-Francois Dockes <jf@dockes.org>  Sat, 20 Dec 2021 09:25:00 +0100
+
 recoll (1.31.5-1~ppaPPAVERS~SERIES1) SERIES; urgency=low
 
   * Fix ennoying bug in tesseract OCR temporary files cleanup.

packaging/debian/debiankio/changelog

@@ -1,3 +1,9 @@
+kio-recoll (1.31.6-1~ppaPPAVERS~SERIES1) SERIES; urgency=low
+
+  * Follow
+    
+ -- Jean-Francois Dockes <jf@dockes.org>  Sat, 20 Dec 2021 09:25:00 +0100
+
 kio-recoll (1.31.5-1~ppaPPAVERS~SERIES1) SERIES; urgency=low
 
   * Follow recoll version

src/RECOLL-VERSION.txt

@@ -1 +1 @@
-1.31.5
+1.31.6

src/common/autoconfig-mac.h

@@ -125,7 +125,7 @@
 #define PACKAGE_NAME "Recoll"
 
 /* Define to the full name and version of this package. */
-#define PACKAGE_STRING "Recoll 1.31.5"
+#define PACKAGE_STRING "Recoll 1.31.6"
 
 /* Define to the one symbol short name of this package. */
 #define PACKAGE_TARNAME "recoll"
@@ -134,7 +134,7 @@
 #define PACKAGE_URL ""
 
 /* Define to the version of this package. */
-#define PACKAGE_VERSION "1.31.5"
+#define PACKAGE_VERSION "1.31.6"
 
 /* putenv parameter is const */
 /* #undef PUTENV_ARG_CONST */

src/common/autoconfig-win.h

@@ -118,7 +118,7 @@
 #define PACKAGE_NAME "Recoll"
 
 /* Define to the full name and version of this package. */
-#define PACKAGE_STRING "Recoll 1.31.5"
+#define PACKAGE_STRING "Recoll 1.31.6"
 
 /* Define to the one symbol short name of this package. */
 #define PACKAGE_TARNAME "recoll"
@@ -127,7 +127,7 @@
 #define PACKAGE_URL ""
 
 /* Define to the version of this package. */
-#define PACKAGE_VERSION "1.31.5"
+#define PACKAGE_VERSION "1.31.6"
 
 /* putenv parameter is const */
 /* #undef PUTENV_ARG_CONST */

src/common/textsplit.cpp

@@ -596,6 +596,7 @@ bool TextSplit::text_to_words(const string &in)
     clearsplitstate();
     
     bool pagepending = false;
+    bool nlpending = false;
     bool softhyphenpending = false;
 
     // Running count of non-alphanum chars. Reset when we see one;
@@ -705,6 +706,10 @@ bool TextSplit::text_to_words(const string &in)
                 pagepending = false;
                 newpage(m_wordpos);
             }
+            if (nlpending) {
+                nlpending = false;
+                newline(m_wordpos);
+            }
             break;
 
         case WILD:
@@ -745,6 +750,12 @@ bool TextSplit::text_to_words(const string &in)
                         break;
                     }
                 } else {
+                    // Note about dangling hyphens: we always strip '-' found before whitespace,
+                    // even before a newline, then generate two terms, before and after the line
+                    // break. We have no way to know if '-' is there because a word was broken by
+                    // justification or if it was part of an actual compound word (would need a
+                    // dictionary to check). As soft-hyphen *should* be used if the '-' is not part
+                    // of the text.
                     if (nextc == -1 || isvisiblewhite(nextc)) {
                         goto SPACE;
                     }
@@ -844,19 +855,10 @@ bool TextSplit::text_to_words(const string &in)
             break;
 
         case '\n':
+            nlpending = true;
+            /* FALLTHROUGH */
         case '\r':
-            if (m_span.length() && *m_span.rbegin() == '-') {
+            if (softhyphenpending) {
-                // if '-' is the last char before end of line, we
-                // strip it.  We have no way to know if this is added
-                // because of the line split or if it was part of an
-                // actual compound word (would need a dictionary to
-                // check).  As soft-hyphen *should* be used if the '-'
-                // is not part of the text, it is better to properly
-                // process a real compound word, and produce wrong
-                // output from wrong text. The word-emitting routine
-                // will strip the trailing '-'.
-                goto SPACE;
-            } else if (softhyphenpending) {
                 // Don't reset soft-hyphen
                 continue;
             } else {

src/common/textsplit.h

@@ -73,6 +73,9 @@ public:
      * just don't know about pages. */
     virtual void newpage(int /*pos*/) {}
 
+    /** Called when we encounter newline \n 0x0a. Override to use the event. */
+    virtual void newline(int /*pos*/) {}
+
     // Static utility functions:
 
     /** Count words in string, as the splitter would generate them */

src/doc/man/recoll.conf.5

@@ -613,8 +613,7 @@ 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
+curidxconfdir (as set in the copied configuration) would be '/media/me/mydata/config'.
-'/media/me/mydata/config'.
 .TP
 .BI "idxrundir = "dfn
 Indexing process current directory. The input

src/doc/user/recoll.conf.xml

@@ -170,13 +170,12 @@ listing either MIME types (e.g. audio/mpeg) or handler names
 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.
+processing of any compressed file. Default 100 MB.
 </para></listitem></varlistentry>
 <varlistentry id="RCL.INSTALL.CONFIG.RECOLLCONF.TEXTFILEMAXMBS">
 <term><varname>textfilemaxmbs</varname></term>
-<listitem><para>Size limit for text
+<listitem><para>Size limit for text files. Mostly for skipping monster logs. Default 20 MB. Use a value of -1 to
-files. Mostly for skipping monster
+disable.
-logs. Default 20 MB.
 </para></listitem></varlistentry>
 <varlistentry id="RCL.INSTALL.CONFIG.RECOLLCONF.INDEXALLFILENAMES">
 <term><varname>indexallfilenames</varname></term>

src/doc/user/usermanual.html

@@ -10,7 +10,7 @@
   <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.">
+  "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 1.31.">
 </head>
 <body bgcolor="white" text="black" link="#0000FF" vlink="#840084"
 alink="#0000FF">
@@ -53,7 +53,7 @@ alink="#0000FF">
             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>
+            "application">Recoll</span> 1.31.</p>
           </div>
         </div>
       </div>
@@ -443,7 +443,7 @@ alink="#0000FF">
       <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>
+      <span class="application">Recoll</span> 1.31.</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>.
@@ -9169,14 +9169,15 @@ hasextract = False
                   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>
+                  file. Default 100 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>
+                  monster logs. Default 20 MB. Use a value of -1 to
+                  disable.</p>
                 </dd>
                 <dt><a name=
                 "RCL.INSTALL.CONFIG.RECOLLCONF.INDEXALLFILENAMES"
@@ -10757,17 +10758,28 @@ other = rclcat:other
               <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
+                Postscript and DVI files. If it is set, a
-                editor at the right page for a match or
+                significant term will be chosen in the query, and
-                snippet.</p>
+                %p will be substituted with the first page where
+                the term appears. Can be used to start the editor
+                at the right page for a match or snippet.</p>
+              </li>
+              <li class="listitem">
+                <p><b>%l.&nbsp;</b>Line number. Only significant
+                for document types with relevant line breaks,
+                mostly text/plain and analogs. If it is set, a
+                significant term will be chosen in the query, and
+                %p will be substituted with the first line where
+                the term appears.</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:
+                be set for documents with indexed page or line
-                PDF). The value will be one of the matched search
+                numbers and if %p or %l is also used. The value
-                terms. It would allow pre-setting the value in the
+                will be one of the matched search terms. It would
-                "Find" entry inside Evince for example, for easy
+                allow pre-setting the value in the "Find" entry
-                highlighting of the term.</p>
+                inside Evince for example, for easy highlighting of
+                the term.</p>
               </li>
               <li class="listitem">
                 <p><b>%u.&nbsp;</b>Url.</p>

src/doc/user/usermanual.xml

@@ -5,7 +5,7 @@
 
 <!ENTITY RCL "<application>Recoll</application>">
 <!ENTITY RCLAPPS "<ulink url='http://www.recoll.org/pages/features.html#doctypes'>http://www.recoll.org/pages/features.html</ulink>">
-<!ENTITY RCLVERSION "1.29">
+<!ENTITY RCLVERSION "1.32">
 <!ENTITY XAP "<application>Xapian</application>">
 <!ENTITY WIN "<application>Windows</application>">
 <!ENTITY LIN "<application>Unix</application>-like systems">
@@ -7114,28 +7114,37 @@ other = rclcat:other
           (possibly a script) to be able to handle it.</para></formalpara>
           </listitem>
 
-          <listitem><formalpara><title>%M</title>
+          <listitem>
-          <para>MIME type</para></formalpara> 
+            <formalpara><title>%M</title><para>MIME type</para></formalpara> 
           </listitem>
 
-          <listitem><formalpara><title>%p</title>
+          <listitem>
-          <para>Page index. Only significant for a subset of document
+            <formalpara><title>%p</title><para>Page index. Only significant for a subset of
-          types, currently only PDF, Postscript and DVI files. Can be
+                document types, currently only PDF, Postscript and DVI files. If it is set, a
-          used to start the editor at the right page for a match or
+                significant term will be chosen in the query, and %p will be substituted with the
-          snippet.</para></formalpara>
+                first page where the term appears. Can be used to start the editor at the right page
+                for a match or snippet.</para></formalpara>
           </listitem>
 
-          <listitem><formalpara><title>%s</title>
+          <listitem>
-          <para>Search term. The value will only be set for documents
+            <formalpara><title>%l</title><para>Line number. Only significant for document
-          with indexed page numbers (ie: PDF). The value will be one of
+                types with relevant line breaks, mostly text/plain and analogs. If it is set, a
-          the matched search terms. It would allow pre-setting the
+                significant term will be chosen in the query, and %p will be substituted with the
-          value in the "Find" entry inside Evince for example, for easy
+                first line where the term appears.</para></formalpara>
-          highlighting of the term.</para></formalpara>  
           </listitem>
 
-          <listitem><formalpara><title>%u</title>
+          <listitem>
-          <para>Url.</para></formalpara> 
+            <formalpara><title>%s</title><para>Search term. The value will only be set for
+                documents with indexed page or line numbers and if %p or %l is also used. 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.</para></formalpara>
           </listitem>
+
+          <listitem>
+            <formalpara><title>%u</title><para>Url.</para></formalpara> 
+          </listitem>
+
         </itemizedlist>
 
         <para>In addition to the predefined values above, all strings like

src/filters/rclorgmode.py

@@ -23,7 +23,7 @@ class OrgModeExtractor:
         iseof = rclexecm.RclExecM.noteof
         if self.currentindex >= len(self.docs) -1:
             iseof = rclexecm.RclExecM.eofnext
-        self.em.setmimetype("text/plain")
+        self.em.setmimetype("text/x-orgmode-sub")
         try:
             self.em.setfield("title", docdata.splitlines()[0])
         except:

src/index/indexer.cpp

@@ -59,8 +59,9 @@ bool runWebFilesMoverScript(RclConfig *config)
     static string downloadsdir;
     if (downloadsdir.empty()) {
         if (!config->getConfParam("webdownloadsdir", downloadsdir)) {
-            downloadsdir = path_tildexpand("~/Downloads");
+            downloadsdir = "~/Downloads";
         }
+        downloadsdir = path_tildexpand(downloadsdir);
     }
     vector<string> cmdvec;
     config->pythonCmd("recoll-we-move-files.py", cmdvec);

src/python/recoll/recoll/conftree.py

@@ -304,3 +304,14 @@ def stringsToString(vs):
             out.append(s)
     return " ".join(out)
 
+def valToBool(s):
+    if not s:
+        return False
+    try:
+        val = int(s)
+        return val != 0
+    except:
+        pass
+    if type(s) == type(b''):
+        s = s.decode("UTF-8")
+    return s[0] in "tTyY"

src/qtgui/fragbuts.cpp

@@ -50,7 +50,7 @@ public:
 
     void startElement(const std::string &nm,
                       const std::map<std::string, std::string>&) override {
-        std::cerr << "startElement [" << nm << "]\n";
+        //std::cerr << "startElement [" << nm << "]\n";
         currentText.clear();
         if (nm == "buttons") {
             radio = false;
@@ -67,7 +67,7 @@ public:
         }
     }        
     void endElement(const std::string& nm) override {
-        std::cerr << "endElement [" << nm << "]\n";
+        //std::cerr << "endElement [" << nm << "]\n";
 
         if (nm == "label") {
             label = u8s2qs(currentText);
@@ -102,7 +102,7 @@ public:
         }
     }
     void characterData(const std::string &str) override {
-        std::cerr << "characterData [" << str << "]\n";
+        //std::cerr << "characterData [" << str << "]\n";
         currentText += str;
     }
 

src/qtgui/rclm_view.cpp

@@ -34,6 +34,7 @@
 #include "rclmain_w.h"
 #include "rclzg.h"
 #include "pathut.h"
+#include "unacpp.h"
 
 using namespace std;
 
@@ -42,7 +43,6 @@ static const vector<string> browser_list{
     "opera", "google-chrome", "chromium-browser",
     "palemoon", "iceweasel", "firefox", "konqueror", "epiphany"};
 
-
 // Start native viewer or preview for input Doc. This is used to allow
 // using recoll from another app (e.g. Unity Scope) to view embedded
 // result docs (docs with an ipath). . We act as a proxy to extract
@@ -155,13 +155,27 @@ void RclMain::openWith(Rcl::Doc doc, string cmdspec)
     execViewer(subs, false, execname, lcmd, cmdspec, doc);
 }
 
-void RclMain::startNativeViewer(Rcl::Doc doc, int pagenum, QString term)
+static bool pagenumNeeded(const std::string& cmd)
 {
+    return cmd.find("%p") != std::string::npos;
+}
+static bool linenumNeeded(const std::string& cmd)
+{
+    return cmd.find("%l") != std::string::npos;
+}
+static bool termNeeded(const std::string& cmd)
+{
+    return cmd.find("%s") != std::string::npos;
+}
+
+void RclMain::startNativeViewer(Rcl::Doc doc, int pagenum, QString qterm)
+{
+    std::string term = qs2utf8s(qterm);
     string apptag;
     doc.getmeta(Rcl::Doc::keyapptg, &apptag);
     LOGDEB("RclMain::startNativeViewer: mtype [" << doc.mimetype <<
            "] apptag ["  << apptag << "] page "  << pagenum << " term ["  <<
-           qs2utf8s(term) << "] url ["  << doc.url << "] ipath [" <<
+           term << "] url ["  << doc.url << "] ipath [" <<
            doc.ipath << "]\n");
 
     // Look for appropriate viewer
@@ -377,19 +391,19 @@ void RclMain::startNativeViewer(Rcl::Doc doc, int pagenum, QString term)
 
     // If we are not called with a page number (which would happen for a call
     // from the snippets window), see if we can compute a page number anyway.
-    if (pagenum == -1) {
+    if (m_source && pagenum == -1 && (pagenumNeeded(cmd) || termNeeded(cmd)|| linenumNeeded(cmd))) {
-        pagenum = 1;
+        pagenum = m_source->getFirstMatchPage(doc, term);
-        string lterm;
-        if (m_source)
-            pagenum = m_source->getFirstMatchPage(doc, lterm);
         if (pagenum == -1)
             pagenum = 1;
-        else // We get the match term used to compute the page
-            term = QString::fromUtf8(lterm.c_str());
     }
-    char cpagenum[20];
-    sprintf(cpagenum, "%d", pagenum);
 
+    int line = 1;
+    if (m_source && !term.empty() && linenumNeeded(cmd)) {
+        if (doc.text.empty()) {
+            rcldb->getDocRawText(doc);
+        }
+        line = m_source->getFirstMatchLine(doc, term);
+    }
 
     // Substitute %xx inside arguments
     string efftime;
@@ -408,9 +422,10 @@ void RclMain::startNativeViewer(Rcl::Doc doc, int pagenum, QString term)
     subs["f"] = fn;
     subs["F"] = fn;
     subs["i"] = FileInterner::getLastIpathElt(doc.ipath);
+    subs["l"] = ulltodecstr(line);
     subs["M"] = doc.mimetype;
-    subs["p"] = cpagenum;
+    subs["p"] = ulltodecstr(pagenum);
-    subs["s"] = (const char*)term.toLocal8Bit();
+    subs["s"] = term;
     subs["U"] = url_encode(url);
     subs["u"] = url;
     // Let %(xx) access all metadata.

src/qtgui/rclmain_w.h

@@ -140,8 +140,7 @@ public slots:
     virtual void showActionsSearch();
     virtual void startPreview(int docnum, Rcl::Doc doc, int keymods);
     virtual void startPreview(Rcl::Doc);
-    virtual void startNativeViewer(Rcl::Doc, int pagenum = -1,
+    virtual void startNativeViewer(Rcl::Doc, int pagenum = -1, QString term = QString());
-                                   QString term = QString());
     virtual void openWith(Rcl::Doc, string);
     virtual void saveDocToFile(Rcl::Doc);
     virtual void previewNextInTab(Preview *, int sid, int docnum);

src/qtgui/snippets_w.cpp

@@ -1,4 +1,4 @@
-/* Copyright (C) 2012 J.F.Dockes
+/* Copyright (C) 2012-2021 J.F.Dockes
  *   This program is free software; you can redistribute it and/or modify
  *   it under the terms of the GNU General Public License as published by
  *   the Free Software Foundation; either version 2 of the License, or
@@ -67,8 +67,7 @@ using namespace std;
 class PlainToRichQtSnippets : public PlainToRich {
 public:
     virtual string startMatch(unsigned int) {
-        return string("<span class='rclmatch' style='")
+        return string("<span class='rclmatch' style='") + qs2utf8s(prefs.qtermstyle) + string("'>");
-            + qs2utf8s(prefs.qtermstyle) + string("'>");
     }
     virtual string endMatch() {
         return string("</span>");
@@ -82,12 +81,10 @@ void SnippetsW::init()
     QPushButton *searchButton = new QPushButton(tr("Search"));
     searchButton->setAutoDefault(false);
     buttonBox->addButton(searchButton, QDialogButtonBox::ActionRole);
-//    setWindowFlags(Qt::WindowStaysOnTopHint);
     searchFM->hide();
 
     onNewShortcuts();
-    connect(&SCBase::scBase(), SIGNAL(shortcutsChanged()),
+    connect(&SCBase::scBase(), SIGNAL(shortcutsChanged()), this, SLOT(onNewShortcuts()));
-            this, SLOT(onNewShortcuts()));
 
     QPushButton *closeButton = buttonBox->button(QDialogButtonBox::Close);
     if (closeButton)
@@ -105,11 +102,9 @@ void SnippetsW::init()
     browserw = new QWebView(this);
     verticalLayout->insertWidget(0, browserw);
     browser->setUrl(QUrl(QString::fromUtf8("about:blank")));
-    connect(browser, SIGNAL(linkClicked(const QUrl &)), 
+    connect(browser, SIGNAL(linkClicked(const QUrl &)), this, SLOT(onLinkClicked(const QUrl &)));
-            this, SLOT(onLinkClicked(const QUrl &)));
     browser->page()->setLinkDelegationPolicy(QWebPage::DelegateAllLinks);
-    browser->page()->currentFrame()->setScrollBarPolicy(Qt::Horizontal,
+    browser->page()->currentFrame()->setScrollBarPolicy(Qt::Horizontal, Qt::ScrollBarAlwaysOff);
-                                                        Qt::ScrollBarAlwaysOff);
     QWEBSETTINGS *ws = browser->page()->settings();
     if (prefs.reslistfontfamily != "") {
         ws->setFontFamily(QWEBSETTINGS::StandardFont, prefs.reslistfontfamily);
@@ -136,8 +131,7 @@ void SnippetsW::init()
 #else
     browserw = new QTextBrowser(this);
     verticalLayout->insertWidget(0, browserw);
-    connect(browser, SIGNAL(anchorClicked(const QUrl &)), 
+    connect(browser, SIGNAL(anchorClicked(const QUrl &)), this, SLOT(onLinkClicked(const QUrl &)));
-            this, SLOT(onLinkClicked(const QUrl &)));
     browser->setReadOnly(true);
     browser->setUndoRedoEnabled(false);
     browser->setOpenLinks(false);
@@ -183,8 +177,7 @@ void SnippetsW::createPopupMenu(const QPoint& pos)
 {
     QMenu *popup = new QMenu(this);
     if (m_sortingByPage) {
-        popup->addAction(tr("Sort By Relevance"), this,
+        popup->addAction(tr("Sort By Relevance"), this, SLOT(reloadByRelevance()));
-                         SLOT(reloadByRelevance()));
     } else {
         popup->addAction(tr("Sort By Page"), this, SLOT(reloadByPage()));
     }
@@ -230,29 +223,22 @@ void SnippetsW::onSetDoc(Rcl::Doc doc, std::shared_ptr<DocSequence> source)
     source->getTerms(hdata);
 
     ostringstream oss;
-    oss << 
+    oss << "<html><head>"
-        "<html><head>"
+        "<meta http-equiv=\"content-type\" content=\"text/html; charset=utf-8\">";
-        "<meta http-equiv=\"content-type\" "
-        "content=\"text/html; charset=utf-8\">";
 
     oss << "<style type=\"text/css\">\nbody,table,select,input {\n";
     oss << "color: " + qs2utf8s(prefs.fontcolor) + ";\n";
     oss << "}\n</style>\n";
     oss << qs2utf8s(prefs.darkreslistheadertext) << qs2utf8s(prefs.reslistheadertext);
 
-    oss << 
+    oss << "</head><body><table class=\"snippets\">";
-        "</head>"
-        "<body>"
-        "<table class=\"snippets\">"
-        ;
 
     g_hiliter.set_inputhtml(false);
     bool nomatch = true;
 
     for (const auto& snippet : vpabs) {
         if (snippet.page == -1) {
-            oss << "<tr><td colspan=\"2\">" << 
+            oss << "<tr><td colspan=\"2\">" << snippet.snippet << "</td></tr>" << "\n";
-                snippet.snippet << "</td></tr>" << endl;
             continue;
         }
         list<string> lr;
@@ -263,13 +249,12 @@ void SnippetsW::onSetDoc(Rcl::Doc doc, std::shared_ptr<DocSequence> source)
         nomatch = false;
         oss << "<tr><td>";
         if (snippet.page > 0) {
-            oss << "<a href=\"http://h/P" << snippet.page << "T" <<
+            oss << "<a href=\"http://h/P" << snippet.page << "T" << snippet.term << "\">" <<
-                snippet.term << "\">" 
+                "P.&nbsp;" << snippet.page << "</a>";
-                << "P.&nbsp;" << snippet.page << "</a>";
         }
-        oss << "</td><td>" << lr.front().c_str() << "</td></tr>" << endl;
+        oss << "</td><td>" << lr.front().c_str() << "</td></tr>" << "\n";
     }
-    oss << "</table>" << endl;
+    oss << "</table>" << "\n";
     if (nomatch) {
         oss.str("<html><head></head><body>\n");
         oss << qs2utf8s(tr("<p>Sorry, no exact match was found within limits. "
@@ -278,12 +263,12 @@ void SnippetsW::onSetDoc(Rcl::Doc doc, std::shared_ptr<DocSequence> source)
     }
     oss << "\n</body></html>";
 #if defined(USING_WEBKIT) || defined(USING_WEBENGINE)
-    browser->setHtml(QString::fromUtf8(oss.str().c_str()));
+    browser->setHtml(u8s2qs(oss.str()));
 #else
     browser->clear();
     browser->append(".");
     browser->clear();
-    browser->insertHtml(QString::fromUtf8(oss.str().c_str()));
+    browser->insertHtml(u8s2qs(oss.str()));
     browser->moveCursor (QTextCursor::Start);
     browser->ensureCursorVisible();
 #endif
@@ -354,8 +339,7 @@ void SnippetsW::onLinkClicked(const QUrl &url)
             string term;
             if (termpos != string::npos)
                 term = ascurl.substr(termpos+1);
-            emit startNativeViewer(m_doc, page, 
+            emit startNativeViewer(m_doc, page, u8s2qs(term));
-                                   QString::fromUtf8(term.c_str()));
             return;
         }
         }

src/query/docseq.h

@@ -111,6 +111,9 @@ public:
     virtual int getFirstMatchPage(Rcl::Doc&, std::string&) {
         return -1;
     }
+    virtual int getFirstMatchLine(const Rcl::Doc&, const std::string&) {
+        return 1;
+    }
     /** Get duplicates. */
     virtual bool docDups(const Rcl::Doc&, std::vector<Rcl::Doc>&) {
         return false;

src/query/docseqdb.cpp

@@ -126,6 +126,17 @@ int DocSequenceDb::getFirstMatchPage(Rcl::Doc &doc, string& term)
     return -1;
 }
 
+int DocSequenceDb::getFirstMatchLine(const Rcl::Doc &doc, const string& term)
+{
+    std::unique_lock<std::mutex> locker(o_dblock);
+    if (!setQuery())
+        return false;
+    if (m_q->whatDb()) {
+        return m_q->getFirstMatchLine(doc, term);
+    }
+    return 1;
+}
+
 list<string> DocSequenceDb::expand(Rcl::Doc &doc)
 {
     std::unique_lock<std::mutex> locker(o_dblock);

src/query/docseqdb.h

@@ -43,6 +43,7 @@ public:
 
     virtual bool getAbstract(Rcl::Doc &doc, std::vector<std::string>&) override;
     virtual int getFirstMatchPage(Rcl::Doc&, std::string& term) override;
+    virtual int getFirstMatchLine(const Rcl::Doc&, const std::string& term) override;
     virtual bool docDups(const Rcl::Doc& doc, std::vector<Rcl::Doc>& dups)
         override;
     virtual std::string getDescription() override;

src/query/plaintorich.h

@@ -1,4 +1,4 @@
-/* Copyright (C) 2004 J.F.Dockes
+/* Copyright (C) 2004-2021 J.F.Dockes
  *   This program is free software; you can redistribute it and/or modify
  *   it under the terms of the GNU General Public License as published by
  *   the Free Software Foundation; either version 2 of the License, or
@@ -60,8 +60,7 @@ public:
      * @param in    raw text out of internfile.
      * @param out   rich text output, divided in chunks (to help our caller
      *   avoid inserting half tags into textedit which doesnt like it)
-     * @param in hdata terms and groups to be highlighted. These are
+     * @param in hdata terms and groups to be highlighted. See utils/hldata.h
-     *   lowercase and unaccented.
      * @param chunksize max size of chunks in output list
      */
     virtual bool plaintorich(const std::string &in, std::list<std::string> &out,

src/rcldb/rclabsfromtext.cpp

@@ -141,11 +141,9 @@ public:
     // add/update fragment definition.
     virtual bool takeword(const std::string& term, int pos, int bts, int bte) {
         LOGDEB1("takeword: [" << term << "] bytepos: "<<bts<<":"<<bte<<endl);
-        // Limit time taken with monster documents. The resulting
+        // Limit time taken with monster documents. The resulting abstract will be incorrect or
-        // abstract will be incorrect or inexistent, but this is
+        // inexistent, but this is better than taking forever (the default cutoff value comes from
-        // better than taking forever (the default cutoff value comes
+        // the snippetMaxPosWalk configuration parameter, and is 10E6)
-        // from the snippetMaxPosWalk configuration parameter, and is
-        // 10E6)
         if (maxtermcount && termcount++ > maxtermcount) {
             LOGINF("Rclabsfromtext: stopping because maxtermcount reached: "<<
                    maxtermcount << endl);
@@ -154,8 +152,7 @@ public:
         }
         // Also limit the number of fragments (just in case safety)
         if (m_fragments.size() > maxtermcount / 100) {
-            LOGINF("Rclabsfromtext: stopping because maxfragments reached: "<<
+            LOGINF("Rclabsfromtext: stopping: max fragments count: " << maxtermcount/100 << "\n");
-                   maxtermcount/100 << endl);
             retflags |= ABSRES_TRUNC;
             return false;
         }
@@ -193,8 +190,7 @@ public:
                 m_curterm = term;
                 m_curtermcoef = coef;
             } else {
-                LOGDEB2("Extending current fragment: " << m_remainingWords <<
+                LOGDEB2("Extending current fragment: "<<m_remainingWords<<" -> "<<m_ctxwords<< "\n");
-                        " -> " << m_ctxwords << endl);
                 m_extcount++;
 #ifdef COMPUTE_HLZONES
                 if (m_prevwordhit) {
@@ -215,9 +211,8 @@ public:
             m_curfragcoef += coef;
             m_remainingWords = m_ctxwords + 1;
             if (m_extcount > 5) {
-                // Limit expansion of contiguous fragments (this is to
+                // Limit expansion of contiguous fragments (this is to avoid common terms in search
-                // avoid common terms in search causing long
+                // causing long heavyweight meaningless fragments. Also, limit length).
-                // heavyweight meaningless fragments. Also, limit length).
                 m_remainingWords = 1;
                 m_extcount = 0;
             }
@@ -247,18 +242,14 @@ public:
                 LOGDEB1("FRAGMENT: from byte " << m_curfrag.first <<
                         " to  byte " << m_curfrag.second << endl);
                 LOGDEB1("FRAGMENT TEXT [" << m_rawtext.substr(
-                            m_curfrag.first, m_curfrag.second-m_curfrag.first)
+                            m_curfrag.first, m_curfrag.second-m_curfrag.first) << "]\n");
-                        << "]\n");
+                // We used to not push weak fragments if we had a lot already. This can cause
-                // We used to not push weak fragments if we had a lot
+                // problems if the fragments we drop are actually group fragments (which have not
-                // already. This can cause problems if the fragments
+                // got their boost yet). The right cut value is difficult to determine, because the
-                // we drop are actually group fragments (which have
+                // absolute values of the coefs depend on many things (index size, etc.) The old
-                // not got their boost yet). The right cut value is
+                // test was if (m_totalcoef < 5.0 || m_curfragcoef >= 1.0) We now just avoid
-                // difficult to determine, because the absolute values
+                // creating a monster by testing the current fragments count at the top of the
-                // of the coefs depend on many things (index size,
+                // function
-                // etc.) The old test was if (m_totalcoef < 5.0 ||
-                // m_curfragcoef >= 1.0) We now just avoid creating a
-                // monster by testing the current fragments count at
-                // the top of the function
                 m_fragments.push_back(MatchFragment(m_curfrag.first,
                                                     m_curfrag.second,
                                                     m_curfragcoef,
@@ -298,8 +289,7 @@ public:
                 m_curtermcoef = 0.0;
         }
 
-        LOGDEB("TextSplitABS: stored total " << m_fragments.size() <<
+        LOGDEB("TextSplitABS: stored total " << m_fragments.size() << " fragments" << endl);
-               " fragments" << endl);
         vector<GroupMatchEntry> tboffs;
 
         // Look for matches to PHRASE and NEAR term groups and finalize
@@ -340,9 +330,8 @@ public:
         }
         auto fragit = m_fragments.begin();
         for (const auto& grpmatch : tboffs) {
-            LOGDEB2("LOOKING FOR FRAGMENT: group: " << grpmatch.offs.first <<
+            LOGDEB2("LOOKING FOR FRAGMENT: group: " << grpmatch.offs.first << "-" <<
-                    "-" << grpmatch.offs.second << " curfrag " <<
+                    grpmatch.offs.second<<" curfrag "<<fragit->start<<"-"<<fragit->stop<<"\n");
-                    fragit->start << "-" << fragit->stop << endl);
             while (fragit->stop < grpmatch.offs.first) {
                 fragit++;
                 if (fragit == m_fragments.end()) {
@@ -417,21 +406,19 @@ int Query::Native::abstractFromText(
     bool sortbypage
     )
 {
-    (void)chron;
+    PRETEND_USE(chron);
     LOGABS("abstractFromText: entry: " << chron.millis() << "mS\n");
     string rawtext;
     if (!ndb->getRawText(docid, rawtext)) {
         LOGDEB0("abstractFromText: can't fetch text\n");
         return ABSRES_ERROR;
     }
-    LOGABS("abstractFromText: got raw text: size " << rawtext.size() << " " <<
+    LOGABS("abstractFromText: got raw text: size "<<rawtext.size()<<" "<<chron.millis()<<"mS\n");
-           chron.millis() << "mS\n");
 
-#if 0 && ! (XAPIAN_MAJOR_VERSION <= 1 && XAPIAN_MINOR_VERSION <= 2)  && \
+#if 0 && XAPIAN_AT_LEAST(1,3,5)
-    (defined(RAWTEXT_IN_DATA))
     // Tryout the Xapian internal method.
-    string snippet = xmset.snippet(rawtext);
+    string snippet = xmset.snippet(rawtext, 60);
-    LOGDEB("SNIPPET: [" << snippet << "] END SNIPPET\n");
+    std::cerr << "XAPIAN SNIPPET: [" << snippet << "] END SNIPPET\n";
 #endif
 
     // We need the q coefs for individual terms
@@ -452,8 +439,7 @@ int Query::Native::abstractFromText(
     }
     LOGABS("abstractFromText: getterms: " << chron.millis() << "mS\n");
 
-    TextSplitABS splitter(rawtext, matchTerms, hld, wordcoefs, ctxwords,
+    TextSplitABS splitter(rawtext, matchTerms, hld, wordcoefs, ctxwords, TextSplit::TXTS_NONE,
-                          TextSplit::TXTS_NONE,
                           m_q->m_snipMaxPosWalk);
     splitter.text_to_words(rawtext);
     LOGABS("abstractFromText: text_to_words: " << chron.millis() << "mS\n");
@@ -484,8 +470,7 @@ int Query::Native::abstractFromText(
     // main term and the page positions. 
     unsigned int count = 0;
     for (const auto& entry : result) {
-        string frag(
+        string frag(fixfrag(rawtext.substr(entry.start, entry.stop - entry.start)));
-            fixfrag(rawtext.substr(entry.start, entry.stop - entry.start)));
 
 #ifdef COMPUTE_HLZONES
         // This would need to be modified to take tag parameters
@@ -506,8 +491,7 @@ int Query::Native::abstractFromText(
             if (page < 0)
                 page = 0;
         }
-        LOGDEB0("=== FRAGMENT: p. " << page << " Coef: " << entry.coef <<
+        LOGDEB0("=== FRAGMENT: p. " << page << " Coef: " << entry.coef << ": " << frag << endl);
-                ": " << frag << endl);
         vabs.push_back(Snippet(page, frag).setTerm(entry.term));
         if (count++ >= maxtotaloccs)
             break;
@@ -515,4 +499,45 @@ int Query::Native::abstractFromText(
     return ABSRES_OK | splitter.getretflags();
 }
 
+class TermLineSplitter : public TextSplit {
+public:
+    TermLineSplitter(const std::string& term)
+        : TextSplit(TextSplit::TXTS_NOSPANS), m_term(term) {
+    }
+    bool takeword(const std::string& _term, int, int, int) override {
+        std::string term;
+        if (o_index_stripchars) {
+            if (!unacmaybefold(_term, term, "UTF-8", UNACOP_UNACFOLD)) {
+                LOGINFO("PlainToRich::takeword: unac failed for [" << term << "]\n");
+                return true;
+            }
+        }
+        if (term == m_term) {
+            return false;
+        }
+        return true;
+    }
+    void newline(int) override {
+        m_line++;
+    }
+    int getline() {
+        return m_line;
+    }
+private:
+    int m_line{1};
+    std::string m_term;
+};
+
+int Query::getFirstMatchLine(const Doc &doc, const std::string& term)
+{
+    int line = 1;
+    TermLineSplitter splitter(term);
+    bool ret = splitter.text_to_words(doc.text);
+    // The splitter takeword() breaks by returning false as soon as the term is found
+    if (ret == false) {
+        line = splitter.getline();
+    }
+    return line;
+}
+
 }

src/rcldb/rclabstract.cpp

@@ -254,7 +254,7 @@ double Query::Native::qualityTerms(Xapian::docid docid,
 }
 
 
-// Return page number for first match of "significant" term.
+// Choose most interesting term and return the page number for its first match
 int Query::Native::getFirstMatchPage(Xapian::docid docid, string& term)
 {
     LOGDEB("Query::Native::getFirstMatchPage\n");
@@ -286,9 +286,7 @@ int Query::Native::getFirstMatchPage(Xapian::docid docid, string& term)
     qualityTerms(docid, terms, byQ);
 
     for (auto mit = byQ.rbegin(); mit != byQ.rend(); mit++) {
-        for (vector<string>::const_iterator qit = mit->second.begin();
+        for (const auto& qterm : mit->second) {
-             qit != mit->second.end(); qit++) {
-            string qterm = *qit;
             Xapian::PositionIterator pos;
             string emptys;
             try {
@@ -619,9 +617,8 @@ int Query::Native::abstractFromIndex(
 // possibly retried by our caller.
 //
 // @param[out] vabs the abstract is returned as a vector of snippets.
-int Query::Native::makeAbstract(Xapian::docid docid,
+int Query::Native::makeAbstract(
-                                vector<Snippet>& vabs, 
+    Xapian::docid docid, vector<Snippet>& vabs, int imaxoccs, int ictxwords, bool sortbypage)
-                                int imaxoccs, int ictxwords, bool sortbypage)
 {
     chron.restart();
     LOGDEB("makeAbstract: docid " << docid << " imaxoccs " <<

src/rcldb/rcldb.cpp

@@ -96,10 +96,13 @@ const string pathelt_prefix = "XP";
 static const string udi_prefix("Q");
 static const string parent_prefix("F");
 
-// Special terms to mark begin/end of field (for anchored searches), and
+// Special terms to mark begin/end of field (for anchored searches).
-// page breaks
 string start_of_field_term;
 string end_of_field_term;
+
+// Special term for page breaks. Note that we use a complicated mechanism for multiple page
+// breaks at the same position, when it would have been probably simpler to use XXPG/n terms
+// instead (did not try to implement though). A change would force users to reindex.
 const string page_break_term = "XXPG/";
 
 // Special term to mark documents with children.
@@ -1846,16 +1849,14 @@ bool Db::addOrUpdate(const string &udi, const string &parent_udi, Doc &doc)
             }
         }
 
-        // If empty pages (multiple break at same pos) were recorded, save
+        // If empty pages (multiple break at same pos) were recorded, save them (this is
-        // them (this is because we have no way to record them in the
+        // because we have no way to record them in the Xapian list)
-        // Xapian list
         if (!tpidx.m_pageincrvec.empty()) {
             ostringstream multibreaks;
             for (unsigned int i = 0; i < tpidx.m_pageincrvec.size(); i++) {
                 if (i != 0)
                     multibreaks << ",";
-                multibreaks << tpidx.m_pageincrvec[i].first << "," << 
+                multibreaks << tpidx.m_pageincrvec[i].first << "," << tpidx.m_pageincrvec[i].second;
-                    tpidx.m_pageincrvec[i].second;
             }
             RECORD_APPEND(record, string(cstr_mbreaks), multibreaks.str());
         }

src/rcldb/rclquery.cpp

@@ -360,7 +360,6 @@ int Query::getFirstMatchPage(const Doc &doc, string& term)
     return m_reason.empty() ? pagenum : -1;
 }
 
-
 // Mset size
 // Note: times for retrieving (multiple times)all docs from a sample
 // 25k docs db (q: mime:*)
@@ -511,8 +510,7 @@ vector<string> Query::expand(const Doc &doc)
             Xapian::ESet eset = m_nq->xenquire->get_eset(20, rset, false);
             LOGDEB("ESet terms:\n");
             // We filter out the special terms
-            for (Xapian::ESetIterator it = eset.begin(); 
+            for (Xapian::ESetIterator it = eset.begin(); it != eset.end(); it++) {
-                 it != eset.end(); it++) {
                 LOGDEB(" [" << (*it) << "]\n");
                 if ((*it).empty() || has_prefix(*it))
                     continue;

src/rcldb/rclquery.h

@@ -115,10 +115,18 @@ public:
     // Returned as a vector of pair<page,snippet> page is 0 if unknown
     int makeDocAbstract(const Doc &doc, std::vector<Snippet>& abst, 
                         int maxoccs= -1, int ctxwords= -1,bool sortbypage=false);
-    /** Retrieve page number for first match for "significant" query term 
+
-     *  @param term returns the chosen term */
+    /** Choose most interesting term and return the page number for its first match
+     *  @param term returns the chosen term 
+     *  @return page number or -1 if term not found or other issue
+     */
     int getFirstMatchPage(const Doc &doc, std::string& term);
 
+    /** Compute line number for first match of term. Only works if doc.text has text.
+     * This uses a text split. Both this and the above getFirstMaxPage() could be done and saved
+     * while we compute the abstracts, quite a lot of waste here. */
+    int getFirstMatchLine(const Doc &doc, const std::string& term);
+    
     /** Retrieve a reference to the searchData we are using */
     std::shared_ptr<SearchData> getSD() {
         return m_sd;

src/sampleconf/mimeconf

@@ -225,6 +225,7 @@ text/x-lua = internal
 text/x-mail = internal
 text/x-man = exec rclman;maxseconds=30
 text/x-orgmode = execm rclorgmode.py
+text/x-orgmode-sub = internal text/plain
 text/x-perl = internal text/plain
 text/x-purple-html-log = internal text/html
 text/x-purple-log = exec rclpurple
@@ -359,6 +360,7 @@ text/x-java = source
 text/x-lua = source
 text/x-mail = message
 text/x-man = document
+text/x-orgmode = document
 text/x-perl = source
 text/x-php = source
 text/x-purple-html-log = pidgin
@@ -443,6 +445,7 @@ text = \
       text/x-ini \
       text/x-java \
       text/x-man \
+      text/x-orgmode \
       text/x-perl \
       text/x-php \
       text/x-python \

src/sampleconf/recoll.conf

@@ -214,12 +214,13 @@ nomd5types = rclaudio
 # files.</brief><descr>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.</descr></var>
+# processing of any compressed file. Default 100 MB.</descr></var>
 compressedfilemaxkbs = 100000
 
-# <var name="textfilemaxmbs" type="int"><brief>Size limit for text
+# <var name="textfilemaxmbs" type="int">
-# files.</brief><descr>Mostly for skipping monster
+# <brief>Size limit for text files.</brief>
-# logs. Default 20 MB.</descr></var> 
+# <descr>Mostly for skipping monster logs. Default 20 MB. Use a value of -1 to
+# disable.</descr></var>
 textfilemaxmbs = 20
 
 # <var name="indexallfilenames" type="bool"><brief>Index the file names of

src/utils/rclutil.cpp

@@ -619,11 +619,17 @@ static const string& thumbnailsdir()
     return thumbnailsd;
 }
 
+// Place for 1024x1024 files
+static const string thmbdirxxlarge = "xx-large";
+// Place for 512x512 files
+static const string thmbdirxlarge = "x-large";
 // Place for 256x256 files
 static const string thmbdirlarge = "large";
 // 128x128
 static const string thmbdirnormal = "normal";
 
+static const vector<string> thmbdirs{thmbdirxxlarge, thmbdirxlarge, thmbdirlarge, thmbdirnormal};
+
 static void thumbname(const string& url, string& name)
 {
     string digest;
@@ -635,26 +641,47 @@ static void thumbname(const string& url, string& name)
 
 bool thumbPathForUrl(const string& url, int size, string& path)
 {
-    string name;
+    string name, path128, path256, path512, path1024;
     thumbname(url, name);
     if (size <= 128) {
         path = path_cat(thumbnailsdir(), thmbdirnormal);
         path = path_cat(path, name);
+        path128 = path;
+    } else if (size <= 256) {
+        path = path_cat(thumbnailsdir(), thmbdirlarge);
+        path = path_cat(path, name);
+        path256 = path;
+    } else if (size <= 512) {
+        path = path_cat(thumbnailsdir(), thmbdirxlarge);
+        path = path_cat(path, name);
+        path512 = path;
+    } else {
+        path = path_cat(thumbnailsdir(), thmbdirxxlarge);
+        path = path_cat(path, name);
+        path1024 = path;
+    }
+    if (access(path.c_str(), R_OK) == 0) {
+        return true;
+    }
+
+    // Not found in requested size. Try to find any size and return it. Let the client scale.
+    for (const auto& tdir : thmbdirs) {
+        path = path_cat(thumbnailsdir(), tdir);
+        path = path_cat(path, name);
         if (access(path.c_str(), R_OK) == 0) {
             return true;
         }
     }
-    path = path_cat(thumbnailsdir(), thmbdirlarge);
-    path = path_cat(path, name);
-    if (access(path.c_str(), R_OK) == 0) {
-        return true;
-    }
 
-    // File does not exist. Path corresponds to the large version at this point,
+    // File does not exist. Return appropriate path anyway.
-    // fix it if needed.
     if (size <= 128) {
-        path = path_cat(path_home(), thmbdirnormal);
+        path = path128;
-        path = path_cat(path, name);
+    } else if (size <= 256) {
+        path = path256;
+    } else if (size <= 512) {
+        path = path512;
+    } else {
+        path = path1024;
     }
     return false;
 }

src/utils/smallut.cpp

@@ -318,7 +318,7 @@ template <class T> void stringsToCSV(const T& tokens, string& s, char sep)
         s.append(1, sep);
     }
     // Remove last separator.
-    if (s.size())
+    if (!s.empty())
         s.pop_back();
 }
 
@@ -951,7 +951,7 @@ bool parsedateinterval(const string& s, DateInterval *dip)
         return false;
     }
 
-    vector<string>::const_iterator it = vs.begin();
+    auto it = vs.cbegin();
     if (*it == "P" || *it == "p") {
         it++;
         if (!parseperiod(it, vs.end(), &p1)) {
@@ -1221,7 +1221,7 @@ std::string SimpleRegexp::simpleSub(
     const std::string& in, const std::string& repl)
 {
     if (!ok()) {
-        return std::string();
+        return {};
     }
 
     int err;
@@ -1256,7 +1256,7 @@ bool SimpleRegexp::simpleMatch(const string& val) const
 string SimpleRegexp::getMatch(const string& val, int i) const
 {
     if (i > m->nmatch) {
-        return string();
+        return {};
     }
     return val.substr(m->matches[i].rm_so,
                       m->matches[i].rm_eo - m->matches[i].rm_so);

src/windows/mimeconf

@@ -187,6 +187,7 @@ text/x-csv = internal text/plain
 text/x-fictionbook = internal xsltproc fb2.xsl
 text/x-ini = internal text/plain
 text/x-mail = internal
+text/x-orgmode = execm python rclorgmode.py
 text/x-perl = internal text/plain
 text/x-python = execm python rclpython.py
 text/x-shellscript = internal text/plain
@@ -291,6 +292,7 @@ text/x-html-sidux-man = sidux-book
 text/x-ini = txt
 text/x-mail = message
 text/x-man = document
+text/x-orgmode = document
 text/x-perl = source
 text/x-purple-html-log = pidgin
 text/x-purple-log = pidgin
@@ -359,6 +361,7 @@ text = \
       text/x-html-sidux-man \
       text/x-ini \
       text/x-man \
+      text/x-orgmode \
       text/x-perl \
       text/x-python \
       text/x-shellscript