tris/recoll
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

process other single-stylesheet xml types

Browse Source
This commit is contained in:
Jean-Francois Dockes 2018-12-26 15:51:05 +01:00
parent 00c0c5168b
commit f8a9d9cf7b
8 changed files with 415 additions and 15 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

18
src/Makefile.am
View File

@ -649,6 +649,12 @@ filterdir = $(pkgdatadir)/filters
filter_DATA = \ filter_DATA = \
desktop/hotrecoll.py \ desktop/hotrecoll.py \
filters/abiword.xsl \ filters/abiword.xsl \
filters/fb2.xsl \
filters/gnumeric.xsl \
filters/msodump.zip \
filters/okular-note.xsl \
filters/opendoc-flat.xsl \
filters/ppt-dump.py \
filters/rcl7z \ filters/rcl7z \
filters/rclabw.py \ filters/rclabw.py \
filters/rclaptosidman \ filters/rclaptosidman \
@ -680,19 +686,19 @@ filters/rcllatinstops.zip \
filters/rcllyx \ filters/rcllyx \
filters/rclman \ filters/rclman \
filters/rclmidi.py \ filters/rclmidi.py \
filters/rclpdf.py \
filters/rclps \
filters/rclokulnote.py \ filters/rclokulnote.py \
filters/rclopxml.py \ filters/rclopxml.py \
filters/rclpdf.py \
filters/rclppt.py \ filters/rclppt.py \
filters/rclps \
filters/rclpurple \ filters/rclpurple \
filters/rclpython \ filters/rclpython \
filters/rclrar \ filters/rclrar \
filters/rclrtf.py \ filters/rclrtf.py \
filters/rclscribus \ filters/rclscribus \
filters/rclshowinfo \ filters/rclshowinfo \
filters/rclsoff.py \
filters/rclsoff-flat.py \ filters/rclsoff-flat.py \
filters/rclsoff.py \
filters/rclsvg.py \ filters/rclsvg.py \
filters/rcltar \ filters/rcltar \
filters/rcltex \ filters/rcltex \
@ -706,11 +712,11 @@ filters/rclxmp.py \
filters/rclxslt.py \ filters/rclxslt.py \
filters/rclzip \ filters/rclzip \
filters/recoll-we-move-files.py \ filters/recoll-we-move-files.py \
filters/ppt-dump.py \ filters/recollepub.zip \
filters/svg.xsl \
filters/xls-dump.py \ filters/xls-dump.py \
filters/xlsxmltocsv.py \ filters/xlsxmltocsv.py \
filters/msodump.zip \ filters/xml.xsl \
filters/recollepub.zip \
python/recoll/recoll/conftree.py \ python/recoll/recoll/conftree.py \
python/recoll/recoll/rclconfig.py python/recoll/recoll/rclconfig.py

56
src/filters/fb2.xsl Executable file
View File

@ -0,0 +1,56 @@
<?xml version="1.0"?>
<xsl:stylesheet version="1.0"
xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
xmlns:fb="http://www.gribuser.ru/xml/fictionbook/2.0"
exclude-result-prefixes="fb"
>
<xsl:output method="html" encoding="UTF-8"/>
<xsl:template match="/fb:FictionBook">
<html>
<xsl:apply-templates select="fb:description"/>
<xsl:apply-templates select="fb:body"/>
</html>
</xsl:template>
<xsl:template match="fb:description">
<head>
<xsl:apply-templates select="fb:title-info"/>
</head><xsl:text>
</xsl:text>
</xsl:template>
<xsl:template match="fb:description/fb:title-info">
<xsl:apply-templates select="fb:book-title"/>
<xsl:apply-templates select="fb:author"/>
</xsl:template>
<xsl:template match="fb:description/fb:title-info/fb:book-title">
<title> <xsl:value-of select="."/> </title>
</xsl:template>
<xsl:template match="fb:description/fb:title-info/fb:author">
<meta>
<xsl:attribute name="name">author</xsl:attribute>
<xsl:attribute name="content">
<xsl:value-of select="fb:first-name"/><xsl:text> </xsl:text>
<xsl:value-of select="fb:middle-name"/><xsl:text> </xsl:text>
<xsl:value-of select="fb:last-name"/>
</xsl:attribute>
</meta>
</xsl:template>
<xsl:template match="fb:body">
<body>
<xsl:apply-templates select="fb:section"/>
</body>
</xsl:template>
<xsl:template match="fb:body/fb:section">
<xsl:for-each select="fb:p">
<p><xsl:value-of select="."/></p>
</xsl:for-each>
</xsl:template>
</xsl:stylesheet>

79
src/filters/gnumeric.xsl Executable file
View File

@ -0,0 +1,79 @@
<?xml version="1.0"?>
<xsl:stylesheet version="1.0"
xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
xmlns:office="urn:oasis:names:tc:opendocument:xmlns:office:1.0"
xmlns:xlink="http://www.w3.org/1999/xlink"
xmlns:dc="http://purl.org/dc/elements/1.1/"
xmlns:meta="urn:oasis:names:tc:opendocument:xmlns:meta:1.0"
xmlns:ooo="http://openoffice.org/2004/office"
xmlns:gnm="http://www.gnumeric.org/v10.dtd"
exclude-result-prefixes="office xlink meta ooo dc"
>
<xsl:output method="html" encoding="UTF-8"/>
<xsl:template match="/">
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8"/>
<xsl:apply-templates select="//office:document-meta/office:meta"/>
</head>
<body>
<xsl:apply-templates select="//gnm:Cells"/>
<xsl:apply-templates select="//gnm:Objects"/>
</body>
</html>
</xsl:template>
<xsl:template match="//dc:date">
<meta>
<xsl:attribute name="name">date</xsl:attribute>
<xsl:attribute name="content"><xsl:value-of select="."/></xsl:attribute>
</meta>
</xsl:template>
<xsl:template match="//dc:description">
<meta>
<xsl:attribute name="name">abstract</xsl:attribute>
<xsl:attribute name="content"><xsl:value-of select="."/></xsl:attribute>
</meta>
</xsl:template>
<xsl:template match="//meta:keyword">
<meta>
<xsl:attribute name="name">keywords</xsl:attribute>
<xsl:attribute name="content"><xsl:value-of select="."/></xsl:attribute>
</meta>
</xsl:template>
<xsl:template match="//dc:subject">
<meta>
<xsl:attribute name="name">keywords</xsl:attribute>
<xsl:attribute name="content"><xsl:value-of select="."/></xsl:attribute>
</meta>
</xsl:template>
<xsl:template match="//dc:title">
<title> <xsl:value-of select="."/> </title>
</xsl:template>
<xsl:template match="//meta:initial-creator">
<meta>
<xsl:attribute name="name">author</xsl:attribute>
<xsl:attribute name="content"><xsl:value-of select="."/></xsl:attribute>
</meta>
</xsl:template>
<xsl:template match="office:meta/*"/>
<xsl:template match="gnm:Cell">
<p><xsl:value-of select="."/></p>
</xsl:template>
<xsl:template match="gnm:CellComment">
<blockquote><xsl:value-of select="@Text"/></blockquote>
</xsl:template>
</xsl:stylesheet>

40
src/filters/okular-note.xsl Executable file
View File

@ -0,0 +1,40 @@
<?xml version="1.0"?>
<xsl:stylesheet version="1.0"
xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
<xsl:output method="html" encoding="UTF-8"/>
<xsl:strip-space elements="*" />
<xsl:template match="/">
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8" />
<title>
Okular notes about: <xsl:value-of select="/documentInfo/@url" />
</title>
</head>
<body>
<xsl:apply-templates />
</body>
</html>
</xsl:template>
<xsl:template match="node()">
<xsl:apply-templates select="@* | node() "/>
</xsl:template>
<xsl:template match="text()">
<p><xsl:value-of select="."/></p>
<xsl:text >
</xsl:text>
</xsl:template>
<xsl:template match="@contents|@author">
<p><xsl:value-of select="." /></p>
<xsl:text >
</xsl:text>
</xsl:template>
<xsl:template match="@*"/>
</xsl:stylesheet>

109
src/filters/opendoc-flat.xsl Executable file
View File

@ -0,0 +1,109 @@
<?xml version="1.0"?>
<xsl:stylesheet version="1.0"
xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
xmlns:office="urn:oasis:names:tc:opendocument:xmlns:office:1.0"
xmlns:xlink="http://www.w3.org/1999/xlink"
xmlns:dc="http://purl.org/dc/elements/1.1/"
xmlns:meta="urn:oasis:names:tc:opendocument:xmlns:meta:1.0"
xmlns:ooo="http://openoffice.org/2004/office"
xmlns:text="urn:oasis:names:tc:opendocument:xmlns:text:1.0"
exclude-result-prefixes="office xlink meta ooo dc text"
>
<xsl:output method="html" encoding="UTF-8"/>
<xsl:template match="/">
<html>
<head>
<xsl:apply-templates select="/office:document/office:meta" />
</head>
<body>
<xsl:apply-templates select="/office:document/office:body" />
</body></html>
</xsl:template>
<xsl:template match="/office:document/office:meta">
<xsl:apply-templates select="dc:title"/>
<xsl:apply-templates select="dc:description"/>
<xsl:apply-templates select="dc:subject"/>
<xsl:apply-templates select="meta:keyword"/>
<xsl:apply-templates select="dc:creator"/>
</xsl:template>
<xsl:template match="/office:document/office:body">
<xsl:apply-templates select=".//text:p" />
<xsl:apply-templates select=".//text:h" />
<xsl:apply-templates select=".//text:s" />
<xsl:apply-templates select=".//text:line-break" />
<xsl:apply-templates select=".//text:tab" />
</xsl:template>
<xsl:template match="dc:title">
<title> <xsl:value-of select="."/> </title><xsl:text>
</xsl:text>
</xsl:template>
<xsl:template match="dc:description">
<meta>
<xsl:attribute name="name">abstract</xsl:attribute>
<xsl:attribute name="content">
<xsl:value-of select="."/>
</xsl:attribute>
</meta><xsl:text>
</xsl:text>
</xsl:template>
<xsl:template match="dc:subject">
<meta>
<xsl:attribute name="name">keywords</xsl:attribute>
<xsl:attribute name="content">
<xsl:value-of select="."/>
</xsl:attribute>
</meta><xsl:text>
</xsl:text>
</xsl:template>
<xsl:template match="dc:creator">
<meta>
<xsl:attribute name="name">author</xsl:attribute>
<xsl:attribute name="content">
<xsl:value-of select="."/>
</xsl:attribute>
</meta><xsl:text>
</xsl:text>
</xsl:template>
<xsl:template match="meta:keyword">
<meta>
<xsl:attribute name="name">keywords</xsl:attribute>
<xsl:attribute name="content">
<xsl:value-of select="."/>
</xsl:attribute>
</meta><xsl:text>
</xsl:text>
</xsl:template>
<xsl:template match="office:body//text:p">
<p><xsl:apply-templates/></p><xsl:text>
</xsl:text>
</xsl:template>
<xsl:template match="office:body//text:h">
<p><xsl:apply-templates/></p><xsl:text>
</xsl:text>
</xsl:template>
<xsl:template match="office:body//text:s">
<xsl:text> </xsl:text>
</xsl:template>
<xsl:template match="office:body//text:line-break">
<br />
</xsl:template>
<xsl:template match="office:body//text:tab">
<xsl:text> </xsl:text>
</xsl:template>
</xsl:stylesheet>

76
src/filters/svg.xsl Executable file
View File

@ -0,0 +1,76 @@
<?xml version="1.0"?>
<xsl:stylesheet version="1.0"
xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
xmlns:svg="http://www.w3.org/2000/svg"
xmlns:dc="http://purl.org/dc/elements/1.1/"
exclude-result-prefixes="svg"
>
<xsl:output method="html" encoding="UTF-8"/>
<xsl:template match="/">
<html>
<head>
<xsl:apply-templates select="svg:svg/svg:title"/>
<xsl:apply-templates select="svg:svg/svg:desc"/>
<xsl:apply-templates select="svg:svg/svg:metadata/descendant::dc:creator"/>
<xsl:apply-templates select="svg:svg/svg:metadata/descendant::dc:subject"/>
<xsl:apply-templates select="svg:svg/svg:metadata/descendant::dc:description"/>
</head>
<body>
<xsl:apply-templates select="//svg:text"/>
</body>
</html>
</xsl:template>
<xsl:template match="svg:desc">
<meta>
<xsl:attribute name="name">keywords</xsl:attribute>
<xsl:attribute name="content">
<xsl:value-of select="."/>
</xsl:attribute>
</meta><xsl:text>
</xsl:text>
</xsl:template>
<xsl:template match="dc:creator">
<meta>
<xsl:attribute name="name">author</xsl:attribute>
<xsl:attribute name="content">
<xsl:value-of select="."/>
</xsl:attribute>
</meta><xsl:text>
</xsl:text>
</xsl:template>
<xsl:template match="dc:subject">
<meta>
<xsl:attribute name="name">keywords</xsl:attribute>
<xsl:attribute name="content">
<xsl:value-of select="."/>
</xsl:attribute>
</meta><xsl:text>
</xsl:text>
</xsl:template>
<xsl:template match="dc:description">
<meta>
<xsl:attribute name="name">description</xsl:attribute>
<xsl:attribute name="content">
<xsl:value-of select="."/>
</xsl:attribute>
</meta><xsl:text>
</xsl:text>
</xsl:template>
<xsl:template match="svg:title">
<title><xsl:value-of select="."/></title><xsl:text>
</xsl:text>
</xsl:template>
<xsl:template match="svg:text">
<p><xsl:value-of select="."/></p><xsl:text>
</xsl:text>
</xsl:template>
</xsl:stylesheet>

34
src/filters/xml.xsl Executable file
View File

@ -0,0 +1,34 @@
<?xml version="1.0"?>
<xsl:stylesheet version="1.0"
xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
<xsl:output method="html" encoding="UTF-8"/>
<xsl:template match="/">
<html>
<head>
<xsl:if test="//*[local-name() = 'title']">
<title>
<xsl:value-of select="//*[local-name() = 'title'][1]"/>
</title>
</xsl:if>
</head>
<body>
<xsl:apply-templates/>
</body>
</html>
</xsl:template>
<xsl:template match="text()">
<xsl:if test="string-length(normalize-space(.)) &gt; 0">
<p><xsl:value-of select="."/></p>
<xsl:text>
</xsl:text>
</xsl:if>
</xsl:template>
<xsl:template match="*">
<xsl:apply-templates/>
</xsl:template>
</xsl:stylesheet>

18
src/sampleconf/mimeconf
View File

@ -84,9 +84,9 @@ application/vnd.oasis.opendocument.text-template = execm rclsoff.py
application/vnd.oasis.opendocument.presentation = execm rclsoff.py application/vnd.oasis.opendocument.presentation = execm rclsoff.py
application/vnd.oasis.opendocument.spreadsheet = execm rclsoff.py application/vnd.oasis.opendocument.spreadsheet = execm rclsoff.py
application/vnd.oasis.opendocument.graphics = execm rclsoff.py application/vnd.oasis.opendocument.graphics = execm rclsoff.py
application/vnd.oasis.opendocument.presentation-flat-xml = execm rclsoff-flat.py application/vnd.oasis.opendocument.presentation-flat-xml = internal xsltproc opendoc-flat.xsl
application/vnd.oasis.opendocument.text-flat-xml = execm rclsoff-flat.py application/vnd.oasis.opendocument.text-flat-xml = internal xsltproc opendoc-flat.xsl
application/vnd.oasis.opendocument.spreadsheet-flat-xml = execm rclsoff-flat.py application/vnd.oasis.opendocument.spreadsheet-flat-xml = internal xsltproc opendoc-flat.xsl
application/vnd.openxmlformats-officedocument.wordprocessingml.document = \ application/vnd.openxmlformats-officedocument.wordprocessingml.document = \
execm rclopxml.py execm rclopxml.py
application/vnd.openxmlformats-officedocument.wordprocessingml.template = \ application/vnd.openxmlformats-officedocument.wordprocessingml.template = \
@ -110,7 +110,7 @@ application/vnd.sun.xml.writer = execm rclsoff.py
application/vnd.sun.xml.writer.global = execm rclsoff.py application/vnd.sun.xml.writer.global = execm rclsoff.py
application/vnd.sun.xml.writer.template = execm rclsoff.py application/vnd.sun.xml.writer.template = execm rclsoff.py
application/vnd.wordperfect = exec wpd2html;mimetype=text/html application/vnd.wordperfect = exec wpd2html;mimetype=text/html
application/x-abiword = internal xsltproc abw.xsl application/x-abiword = internal xsltproc abiword.xsl
application/x-awk = internal text/plain application/x-awk = internal text/plain
application/x-chm = execm rclchm application/x-chm = execm rclchm
application/x-dia-diagram = execm rcldia;mimetype=text/plain application/x-dia-diagram = execm rcldia;mimetype=text/plain
@ -118,12 +118,12 @@ application/x-dvi = exec rcldvi
application/x-flac = execm rclaudio application/x-flac = execm rclaudio
application/x-gnote = execm rclxml.py application/x-gnote = execm rclxml.py
application/x-gnuinfo = execm rclinfo application/x-gnuinfo = execm rclinfo
application/x-gnumeric = execm rclgnm.py application/x-gnumeric = internal xsltproc gnumeric.xsl
application/x-kword = exec rclkwd application/x-kword = exec rclkwd
application/x-lyx = exec rcllyx application/x-lyx = exec rcllyx
application/x-mimehtml = internal message/rfc822 application/x-mimehtml = internal message/rfc822
#application/x-mobipocket-ebook = execm rclmobi #application/x-mobipocket-ebook = execm rclmobi
application/x-okular-notes = execm rclokulnote.py application/x-okular-notes = internal xsltproc okular-note.xsl
application/x-perl = internal text/plain application/x-perl = internal text/plain
# Returned by xdg-mime for .php. Future-proofing # Returned by xdg-mime for .php. Future-proofing
application/x-php = internal text/plain application/x-php = internal text/plain
@ -149,7 +149,7 @@ image/jpeg = execm rclimg
image/png = execm rclimg image/png = execm rclimg
image/tiff = execm rclimg image/tiff = execm rclimg
image/vnd.djvu = execm rcldjvu.py image/vnd.djvu = execm rcldjvu.py
image/svg+xml = execm rclsvg.py image/svg+xml = internal xsltproc svg.xsl
image/x-xcf = execm rclimg image/x-xcf = execm rclimg
image/x-nikon-nef = execm rclimg image/x-nikon-nef = execm rclimg
inode/symlink = internal inode/symlink = internal
@ -168,7 +168,7 @@ text/css = internal text/plain
application/javascript = internal text/plain application/javascript = internal text/plain
text/x-bibtex = exec rclbibtex.sh ; mimetype = text/plain text/x-bibtex = exec rclbibtex.sh ; mimetype = text/plain
text/x-csv = internal text/plain text/x-csv = internal text/plain
text/x-fictionbook = execm rclfb2.py text/x-fictionbook = internal xsltproc fb2.xsl
text/x-gaim-log = exec rclgaim text/x-gaim-log = exec rclgaim
text/x-html-aptosid-man = exec rclaptosidman text/x-html-aptosid-man = exec rclaptosidman
text/x-lua = internal text/x-lua = internal
@ -190,7 +190,7 @@ text/x-tex = exec rcltex
# All parameter and tag names, attribute values etc, are indexed as # All parameter and tag names, attribute values etc, are indexed as
# text. rclxml.py tries to just index the text content. # text. rclxml.py tries to just index the text content.
#application/xml = execm rclxml.py #application/xml = execm rclxml.py
#text/xml = execm rclxml.py #text/xml = internal xsltproc xml.xsl
application/xml = internal text/plain application/xml = internal text/plain
text/xml = internal text/plain text/xml = internal text/plain