diff options
Diffstat (limited to 'tools/yelp-check.in')
| -rwxr-xr-x | tools/yelp-check.in | 2438 |
1 files changed, 1227 insertions, 1211 deletions
diff --git a/tools/yelp-check.in b/tools/yelp-check.in index d46e004..2578800 100755 --- a/tools/yelp-check.in +++ b/tools/yelp-check.in @@ -1,8 +1,7 @@ -#!/bin/sh -# -*- indent-tabs-mode: nil -*- +#!/bin/python3 # # yelp-check -# Copyright (C) 2011-2015 Shaun McCance <shaunm@gnome.org> +# Copyright (C) 2011-2020 Shaun McCance <shaunm@gnome.org> # # 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 @@ -18,1214 +17,1231 @@ # along with this program; if not, write to the Free Software # Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. -xsl_mal_link='@XSL_MAL_LINK@' -xsl_mal_license='@DATADIR@/xslt/mal-license.xsl' -xsl_mal_status='@DATADIR@/xslt/mal-status.xsl' -xsl_mal_rng='@DATADIR@/xslt/mal-rng.xsl' -xsl_comments='@DATADIR@/xslt/comments.xsl' -xsl_media='@DATADIR@/xslt/media.xsl' - -yelp_check_retval="0" - -urlencode () { - # We usually don't want to urlencode slashes, because we're - # usually converting file paths to URIs. But we do want to - # urlencode slases for names of RNG files in validate_page. - if [ "x$2" = "x/" ]; then - urlencode_slash='' - else - urlencode_slash='\/' - fi - echo "$1" | LC_ALL=C awk ' -BEGIN { - for (i = 1; i <= 255; i++) chars[sprintf("%c", i)] = i; -} -{ - ret = ""; - for (i = 1; i <= length($0); i++) { - c = substr($0, i, 1); - if (c ~ /['$urlencode_slash'a-zA-Z0-9._-]/) - ret = ret c; - else - ret = ret sprintf("%%%X%X", int(chars[c] / 16), chars[c] % 16); - } - print ret; -}' -} - -urldecode () { - echo "$1" | LC_ALL=C awk ' -BEGIN { - for(i = 0; i < 10; i++) hex[i] = i; - hex["A"] = hex["a"] = 10; - hex["B"] = hex["b"] = 11; - hex["C"] = hex["c"] = 12; - hex["D"] = hex["d"] = 13; - hex["E"] = hex["e"] = 14; - hex["F"] = hex["f"] = 15; -} -{ - ret = ""; - for (i = 1; i <= length($0); i++) { - c = substr($0, i, 1); - if (c == "+") { - ret = ret " "; - } - else if (c == "%") { - c = sprintf("%c", hex[substr($0, i + 1, 1)] * 16 + hex[substr($0, i + 2, 1)]); - ret = ret c; - i += 2; - } - else { - ret = ret c; +import configparser +import lxml.etree +import os +import sys +import urllib.request +import shutil +import subprocess +import tempfile +import textwrap + + +DATADIR = '@DATADIR@' + +XML_ID = '{http://www.w3.org/XML/1998/namespace}id' +NAMESPACES = { + 'mal': 'http://projectmallard.org/1.0/', + 'cache': 'http://projectmallard.org/cache/1.0/', + 'db': 'http://docbook.org/ns/docbook', + 'e': 'http://projectmallard.org/experimental/', + 'ui': 'http://projectmallard.org/ui/1.0/', + 'uix': 'http://projectmallard.org/experimental/ui/', + 'xlink': 'http://www.w3.org/1999/xlink' } - } - print ret; -}' -} - -docbook_version=' -<xsl:stylesheet - xmlns:xsl="http://www.w3.org/1999/XSL/Transform" - xmlns:db="http://docbook.org/ns/docbook" - version="1.0"> -<xsl:output method="text"/> -<xsl:template match="/"> - <xsl:choose> - <xsl:when test="/db:*/@version"> - <xsl:value-of select="/db:*/@version"/> - </xsl:when> - <xsl:when test="/db:*"> - <xsl:text>5.0</xsl:text> - </xsl:when> - <xsl:otherwise> - <xsl:text>4</xsl:text> - </xsl:otherwise> - </xsl:choose> -</xsl:template> -</xsl:stylesheet> -' - -mallard_style=' -<xsl:stylesheet - xmlns:xsl="http://www.w3.org/1999/XSL/Transform" - xmlns:mal="http://projectmallard.org/1.0/" - version="1.0"> -<xsl:output method="text"/> -<xsl:template match="/mal:page"><xsl:value-of select="@style"/></xsl:template> -</xsl:stylesheet>' - -yelp_usage() { - ( - echo "Usage: yelp-check <COMMAND> [OPTIONS] [FILES]" - echo "" - echo "Commands:" - echo " comments Print the editorial comments in a document" - echo " hrefs Find broken external links in a document" - echo " ids Find Mallard page IDs that do not match file names" - echo " license Report the license of Mallard pages" - echo " links Find broken xref or linkend links in a document" - echo " media Find broken references to media files" - echo " orphans Find orphaned pages in a Mallard document" - echo " status Report the status of Mallard pages" - echo " style Report the style attribute of Mallard pages" - echo " validate Validate files against a DTD or RNG" - ) 1>&2 -} -yelp_usage_hrefs () { - ( - echo "Usage: yelp-check hrefs <FILES>" - echo "" - echo " Find broken href links in FILES in a Mallard document, or" - echo " broken ulink or XLink links in FILES in a DocBook document." - echo "" - echo "Options:" - echo " -s Treat pages as belonging to a Mallard site" - ) 1>&2 -} -yelp_usage_ids () { - ( - echo "Usage: yelp-check ids <FILES>" - echo "" - echo " Find pages in a Mallard document whose page ID does not match" - echo " the base file name of the page file." - echo "" - echo "Options:" - echo " -s Treat pages as belonging to a Mallard site" - ) 1>&2 -} -yelp_usage_links () { - ( - echo "Usage: yelp-check links <FILES>" - echo "" - echo " Find broken xref links in FILES in a Mallard document," - echo " or broken linkend links in FILES in a DocBook document." - echo "" - echo "Options:" - echo " -c CACHE Use the existing Mallard cache CACHE" - echo " -s Treat pages as belonging to a Mallard site" - echo " -i Ignore xrefs where href is present" - ) 1>&2 -} -yelp_usage_media () { - ( - echo "Usage: yelp-check media <FILES>" - echo "" - echo " Find broken references to media files. In Mallard, this" - echo " checks media and thumb elements. In DocBook, this checks" - echo " audiodata, imagedata, and videodata elements." - echo "" - echo "Options:" - echo " -s Treat pages as belonging to a Mallard site" - ) 1>&2 -} -yelp_usage_orphans () { - ( - echo "Usage: yelp-check orphans <FILES>" - echo "" - echo " Locate orphaned pages among FILES in a Mallard document." - echo " Orphaned pages are any pages that cannot be reached by" - echo " topic links alone from the index page." - echo "" - echo "Options:" - echo " -c CACHE Use the existing Mallard cache CACHE" - echo " -s Treat pages as belonging to a Mallard site" - ) 1>&2 -} -yelp_usage_comments () { - ( - echo "Usage: yelp-check comments <FILES>" - echo "" - echo " Print the editorial comments in the files FILES, using the" - echo " comment element in Mallard and the remark element in DocBook." - echo "" - echo "Options:" - echo " -s Treat pages as belonging to a Mallard site" - ) 1>&2 -} -yelp_usage_license () { - ( - echo "Usage: yelp-check license <FILES>" - echo "" - echo " Report the license of the Mallard page files FILES. Each" - echo " matching page is reporting along with its license, reported" - echo " based on the href attribute of the license element. Common" - echo " licenses use a shortened identifier. Pages with multiple" - echo " licenses have the identifiers separated by spaces. Pages" - echo " with no license element report 'none'. Licenses with no" - echo " href attribute are reported as 'unknown'." - echo "" - echo "Options:" - echo " -s Treat pages as belonging to a Mallard site" - echo " --only LICENSES Only show pages whose license is in LICENSES" - echo " --except LICENSES Exclude pages whose license is in LICENSES" - echo " --totals Show total counts for each license" - echo "LICENSES may be a comma- and/or space-separated list." - ) 1>&2 -} -yelp_usage_style () { - ( - echo "Usage: yelp-check style <FILES>" - echo "" - echo " Report the page style attribute of the Mallard page files" - echo " FILES. Each matching page is reporting along with its status." - echo "" - echo "Options:" - echo " -s Treat pages as belonging to a Mallard site" - echo " --only STYLES Only show pages whose style is in STATUSES" - echo " --except STYLES Exclude pages whose style is in STATUSES" - echo " --totals Show total counts for each style" - echo "STYLES may be comma- and/or space-separated lists." - ) 1>&2 -} -yelp_usage_status () { - ( - echo "Usage: yelp-check status <FILES>" - echo "" - echo " Report the status of the Mallard page files FILES. Each" - echo " matching page is reporting along with its status." - echo "" - echo "Options:" - echo " -s Treat pages as belonging to a Mallard site" - echo " --version VER Select revisions with the version attribute VER" - echo " --docversion VER Select revisions with the docversion attribute VER" - echo " --pkgversion VER Select revisions with the pkgversion attribute VER" - echo " --older DATE Only show pages older than DATE" - echo " --newer DATE Only show pages newer than DATE" - echo " --only STATUSES Only show pages whose status is in STATUSES" - echo " --except STATUSES Exclude pages whose status is in STATUSES" - echo " --totals Show total counts for each status" - echo "VER and STATUSES may be comma- and/or space-separated lists." - ) 1>&2 -} -yelp_usage_validate () { - ( - echo "Usage: yelp-check validate <FILES>" - echo "" - echo " Validate FILES against the appropriate DTD or RNG." - echo " For Mallard pages, perform automatic RNG merging" - echo " based on the version attribute." - echo "" - echo "Options:" - echo " -s Treat pages as belonging to a Mallard site" - echo " --strict Disallow unknown namespaces" - echo " --allow NS Explicitly allow namespace NS in strict mode" - echo " --jing Use jing instead of xmllint for RNG validation" - ) 1>&2 -} - -if [ $# = 0 ]; then - yelp_usage - exit 1 -fi - -yelp_check_iter_site () { - for dir in "$1"/*; do - if [ -d "$dir" ]; then - if [ $(basename "$dir") != "__pintail__" ]; then - yelp_check_iter_site "$dir" - fi - fi - done - for page in "$1"/*.page; do - if [ -e "$page" ]; then - $check_page "$page" || yelp_check_retval="$?" - fi - done -} - -yelp_check_iter_args () { - for arg in "$@"; do - ext=$(echo "$arg" | sed -e 's/.*\.//') - if [ -d "$arg" ]; then - if [ "x$check_site" = "x1" ]; then - yelp_check_iter_site "$arg" - else - for page in "${arg%%/}"/*.page; do - if [ -e "$page" ]; then - $check_page "$page" - fi - done - fi - elif [ "x$ext" = "xpage" -o "x$ext" = "xstub" -o "x$ext" = "xcache" ]; then - $check_page "$arg" || yelp_check_retval="$?" - elif [ "x$check_db" != "x" -a \( "x$ext" = "xdocbook" -o "x$ext" = "xxml" \) ]; then - $check_db "$arg" || yelp_check_retval="$?" - else - echo "Unrecognized page $arg" 1>&2 - exit 1 - fi - done - return $yelp_check_retval -} - -yelp_hrefs_page () { - base=$(dirname "$1") - if [ "x$check_site" = "x1" ]; then - sdir=$(cd $(dirname "$1") && pwd) - sdir=${sdir##${check_site_root}}/ - fi - ( - echo '<xsl:stylesheet' - echo ' xmlns:xsl="http://www.w3.org/1999/XSL/Transform"' - echo ' xmlns:mal="http://projectmallard.org/1.0/"' - echo ' xmlns:db="http://docbook.org/ns/docbook"' - echo ' xmlns:xlink="www.w3.org/1999/xlink"' - echo ' version="1.0">' - echo '<xsl:output method="text"/>' - echo '<xsl:template match="/mal:page">' - echo ' <xsl:for-each select="//*[@href]">' - echo ' <xsl:if test="not(starts-with(@href, '\''mailto:'\''))">' - echo ' <xsl:value-of select="/mal:page/@id"/>' - echo ' <xsl:text> </xsl:text>' - echo ' <xsl:value-of select="@href"/>' - echo ' <xsl:text>
</xsl:text>' - echo ' </xsl:if>' - echo ' </xsl:for-each>' - echo '</xsl:template>' - echo '<xsl:template match="/*[namespace-uri(.) = '\'\''] | /db:*">' - echo ' <xsl:for-each select="//ulink/@url | //*/xlink:href">' - echo ' <xsl:if test="not(starts-with(string(.), '\''mailto:'\''))">' - echo ' <xsl:value-of select="(ancestor-or-self::*/@id | ancestor-or-self::*/@xml:id)[last()]"/>' - echo ' <xsl:text> </xsl:text>' - echo ' <xsl:value-of select="string(.)"/>' - echo ' <xsl:text>
</xsl:text>' - echo ' </xsl:if>' - echo ' </xsl:for-each>' - echo '</xsl:template>' - echo '</xsl:stylesheet>' - ) | xsltproc --xinclude - "$1" | sort | uniq | \ - while read id url; do - colon=`echo "$url" | cut -d: -f1` - if [ "x$colon" = "x$url" ]; then - test -f "$base/"$(urldecode "$url") || echo "$sdir$id: $url" - else - status=$(cat "$check_href_cache" | while read trystatus tryurl; do - if [ "x$tryurl" = "x$url" ]; then echo "$trystatus"; break; fi - done) - if [ "x$status" = "x1" ]; then - true - elif [ "x$status" = "x0" ]; then - echo "$sdir$id: $url" - else - (curl -s -I -L "$url" | \ - grep '^HTTP/' | tail -n 1 | head -n 1 | \ - grep -q 'HTTP/.\.. 200 .*') \ - && (echo "1 $url" >> "$check_href_cache") \ - || (echo "0 $url" >> "$check_href_cache"; echo "$sdir$id: $url") - fi - fi - done -} - -yelp_hrefs () { - if [ "$#" = "0" -o "x$1" = "x--help" -o "x$1" = "x-h" ]; then - yelp_usage_hrefs - exit 1 - fi - while [ "$#" != "0" ]; do - case "$1" in - "-s") - check_site="1" - check_site_root=$(pwd) - shift - ;; - *) - break - ;; - esac - done - if [ "$#" = "0" -o "x$1" = "x--help" -o "x$1" = "x-h" ]; then - yelp_usage_hrefs - exit 1 - fi - check_out_file=`mktemp "${TMPDIR:-/tmp}"/yelp-XXXXXXXX` - check_href_cache=`mktemp "${TMPDIR:-/tmp}"/yelp-XXXXXXXX` - echo > "$check_href_cache" - check_db=yelp_hrefs_page - check_page=yelp_hrefs_page - yelp_check_iter_args "$@" > "$check_out_file" - yelp_check_retval=$(wc -l < "$check_out_file") - if test "x$yelp_check_retval" != "x0"; then - yelp_check_retval=1 - fi - cat "$check_out_file" - rm "$check_out_file" - rm "$check_href_cache" - exit $yelp_check_retval -} - -yelp_ids_page () { - pageid=$( - ( - echo '<xsl:stylesheet' - echo ' xmlns:xsl="http://www.w3.org/1999/XSL/Transform"' - echo ' xmlns:mal="http://projectmallard.org/1.0/"' - echo ' version="1.0">' - echo '<xsl:output method="text"/>' - echo '<xsl:template match="/mal:page">' - echo '<xsl:value-of select="@id"/>' - echo '</xsl:template>' - echo '</xsl:stylesheet>' - ) | xsltproc --xinclude - "$1") - dname=$(dirname "$1") - bname=$(basename "$1") - if [ "x$pageid.page" != "x$bname" ]; then - if [ "x$check_site" = "x1" ]; then - sdir=$(cd $(dirname "$1") && pwd) - sdir=${sdir##${check_site_root}}/ - echo $sdir$(basename "$1")": $pageid" - elif [ "x$dname" = 'x.' ]; then - echo "$bname: $pageid" - else - echo "$1: $pageid" - fi - yelp_check_retval=1 - fi -} - -yelp_ids () { - if [ "$#" = "0" -o "x$1" = "x--help" -o "x$1" = "x-h" ]; then - yelp_usage_ids - exit 1 - fi - while [ "$#" != "0" ]; do - case "$1" in - "-s") - check_site="1" - check_site_root=$(pwd) - shift - ;; - *) - break - ;; - esac - done - if [ "$#" = "0" -o "x$1" = "x--help" -o "x$1" = "x-h" ]; then - yelp_usage_ids - exit 1 - fi - check_db= - check_page=yelp_ids_page - yelp_check_iter_args "$@" - exit $yelp_check_retval -} - -yelp_links_db () { - ( - echo '<xsl:stylesheet' - echo ' xmlns:xsl="http://www.w3.org/1999/XSL/Transform"' - echo ' xmlns:db="http://docbook.org/ns/docbook"' - echo ' xmlns:exsl="http://exslt.org/common"' - echo ' extension-element-prefixes="exsl"' - echo ' version="1.0">' - echo '<xsl:output method="text"/>' - echo '<xsl:key name="idkey" match="*[@id or @xml:id]" use="@id | @xml:id"/>' - echo '<xsl:template match="/">' - echo ' <xsl:for-each select="//*[@linkend]">' - echo ' <xsl:if test="not(key('"'idkey'"', @linkend))">' - echo ' <xsl:value-of select="(ancestor-or-self::*/@id | ancestor-or-self::*/@xml:id)[last()]"/>' - echo ' <xsl:text>: </xsl:text>' - echo ' <xsl:value-of select="@linkend"/>' - echo ' <xsl:text>
</xsl:text>' - echo ' </xsl:if>' - echo ' </xsl:for-each>' - echo '</xsl:template>' - echo '</xsl:stylesheet>' - ) | xsltproc --xinclude - "$1" -} - -yelp_links_page () { - if [ "x$check_site" = "x1" ]; then - sdir=$(cd $(dirname "$1") && pwd) - sdir=${sdir##${check_site_root}}/ - fi - ( - echo '<xsl:stylesheet' - echo ' xmlns:xsl="http://www.w3.org/1999/XSL/Transform"' - echo ' xmlns:mal="http://projectmallard.org/1.0/"' - echo ' xmlns:site="http://projectmallard.org/site/1.0/"' - echo ' xmlns:exsl="http://exslt.org/common"' - echo ' extension-element-prefixes="exsl"' - echo ' version="1.0">' - xsl='file://'`urlencode "$xsl_mal_link"` - echo '<xsl:import href="'"$xsl"'"/>' - check_cache_url='file://'`urlencode "$check_cache_file"` - echo '<xsl:param name="mal.cache.file" select="'"'$check_cache_url'"'"/>' - echo '<xsl:variable name="site.dir" select="'"'$sdir'"'"/>' - echo '<xsl:output method="text"/>' - echo '<xsl:key name="__site.cache.key" match="mal:page | mal:section"' - echo ' use="concat(ancestor-or-self::mal:page/@site:dir, @id)"/>' - echo '<xsl:template match="/mal:page">' - echo ' <xsl:variable name="page" select="@id"/>' - if [ "x$check_links_ignore" = "x1" ]; then - echo ' <xsl:for-each select="//*[@xref][not(@href)]">' - else - echo ' <xsl:for-each select="//*[@xref]">' - fi - echo ' <xsl:variable name="xref" select="@xref"/>' - echo ' <xsl:variable name="linkid">' - echo ' <xsl:call-template name="mal.link.xref.linkid"/>' - echo ' </xsl:variable>' - echo ' <xsl:for-each select="$mal.cache">' - echo ' <xsl:if test="count(key('"'mal.cache.key'"', $linkid) | ' - echo ' key('"'__site.cache.key'"', $linkid)) = 0">' - echo ' <xsl:value-of select="$site.dir"/>' - echo ' <xsl:value-of select="$page"/>' - echo ' <xsl:text>: </xsl:text>' - echo ' <xsl:value-of select="$xref"/>' - echo ' <xsl:text>
</xsl:text>' - echo ' </xsl:if>' - echo ' </xsl:for-each>' - echo ' </xsl:for-each>' - echo '</xsl:template>' - echo '</xsl:stylesheet>' - ) | xsltproc --xinclude - "$1" -} - -yelp_links () { - if [ "$#" = "0" -o "x$1" = "x--help" -o "x$1" = "x-h" ]; then - yelp_usage_links - exit 1 - fi - while [ "$#" != "0" ]; do - case "$1" in - "-c") - shift - check_cache_file="$1" - shift - ;; - "-s") - check_site="1" - check_site_root=$(pwd) - shift - ;; - "-i") - shift - check_links_ignore="1" - ;; - *) - break - ;; - esac - done - if [ "$#" = "0" -o "x$1" = "x--help" -o "x$1" = "x-h" ]; then - yelp_usage_links - exit 1 - fi - if [ "x$check_cache_file" != "x" ]; then - check_cache_dir=$(dirname "$check_cache_file") - check_cache_dir=$(cd "$check_cache_dir" && pwd) - check_cache_file="$check_cache_dir/"$(basename "$check_cache_file") - elif [ -d "$1" ]; then - check_cache_file=1 - else - case "$1" in - *.page | *.stub | *.cache) - check_cache_file=1 - ;; - *) - break - ;; - esac - fi - if [ "x$check_cache_file" = "x1" ]; then - check_cache_file_is_tmp="yes" - check_cache_file=$(mktemp "${TMPDIR:-/tmp}"/yelp-XXXXXXXX) - if [ "x$check_site" = "x1" ]; then - yelp-build cache -s -o "$check_cache_file" "$@" - else - yelp-build cache -o "$check_cache_file" "$@" - fi - fi - - check_out_file=`mktemp "${TMPDIR:-/tmp}"/yelp-XXXXXXXX` - check_db=yelp_links_db - check_page=yelp_links_page - yelp_check_iter_args "$@" > "$check_out_file" - yelp_check_retval=$(wc -l < "$check_out_file") - if test "x$yelp_check_retval" != "x0"; then - yelp_check_retval=1 - fi - cat "$check_out_file" - rm "$check_out_file" - if [ "x$check_cache_file_is_tmp" = "xyes" ]; then - rm "$check_cache_file" - fi - exit $yelp_check_retval -} - -yelp_media_page () { - ext=$(echo "$1" | sed -e 's/.*\.//') - bname=$(basename "$1" ".$ext") - dname=$(dirname "$1") - if [ "x$dname" = "x." ]; then - dname="" - else - dname="$dname"/ - fi; - if [ "x$check_site" = "x1" ]; then - sdir=$(cd "$dname" && pwd) - sdir=${sdir##${check_site_root}}/ - else - sdir="$dname" - fi - xsltproc "$xsl_media" "$1" | \ - sort | uniq | \ - while read line; do - src=$(urldecode "$line") - if [ ! -f "$dname$src" ]; then - echo "$sdir$bname: $line" - fi - done -} - -yelp_media () { - if [ "$#" = "0" -o "x$1" = "x--help" -o "x$1" = "x-h" ]; then - yelp_usage_media - exit 1 - fi - while [ "$#" != "0" ]; do - case "$1" in - "-s") - check_site="1" - check_site_root=$(pwd) - shift - ;; - *) - break - ;; - esac - done - if [ "$#" = "0" -o "x$1" = "x--help" -o "x$1" = "x-h" ]; then - yelp_usage_media - exit 1 - fi - check_out_file=`mktemp "${TMPDIR:-/tmp}"/yelp-XXXXXXXX` - check_db=yelp_media_page - check_page=yelp_media_page - yelp_check_iter_args "$@" > "$check_out_file" - yelp_check_retval=$(wc -l < "$check_out_file") - if test "x$yelp_check_retval" != "x0"; then - yelp_check_retval=1 - fi - cat "$check_out_file" - rm "$check_out_file" - exit $yelp_check_retval -} - -yelp_orphans_page () { - if [ "x$check_site" = "x1" ]; then - sdir=$(cd $(dirname "$1") && pwd) - sdir=${sdir##${check_site_root}}/ - fi - ( - echo '<xsl:stylesheet' - echo ' xmlns:xsl="http://www.w3.org/1999/XSL/Transform"' - echo ' xmlns:mal="http://projectmallard.org/1.0/"' - echo ' xmlns:exsl="http://exslt.org/common"' - echo ' extension-element-prefixes="exsl"' - echo ' version="1.0">' - xsl='file://'`urlencode "$xsl_mal_link"` - echo '<xsl:import href="'"$xsl"'"/>' - check_cache_url='file://'`urlencode "$check_cache_file"` - echo '<xsl:param name="mal.cache.file" select="'"'$check_cache_url'"'"/>' - echo '<xsl:variable name="site.dir" select="'"'$sdir'"'"/>' - echo '<xsl:output method="text"/>' - echo '<xsl:template match="/mal:page">' - echo ' <xsl:variable name="trails">' - echo ' <xsl:call-template name="mal.link.linktrails"/>' - echo ' </xsl:variable>' - echo ' <xsl:if test="@id != '"'index'"' and count(exsl:node-set($trails)/*) = 0">' - echo ' <xsl:value-of select="$site.dir"/>' - echo ' <xsl:value-of select="@id"/>' - echo ' <xsl:text>
</xsl:text>' - echo ' </xsl:if>' - echo '</xsl:template>' - echo '</xsl:stylesheet>' - ) | xsltproc --xinclude - "$1" -} - -yelp_orphans () { - if [ "$#" = "0" -o "x$1" = "x--help" -o "x$1" = "x-h" ]; then - yelp_usage_orphans - exit 1 - fi - while [ "$#" != "0" ]; do - case "$1" in - "-s") - check_site="1" - check_site_root=$(pwd) - shift - ;; - "-c") - shift - check_cache_file="$1" - shift - ;; - *) - break - ;; - esac - done - if [ "$#" = "0" -o "x$1" = "x--help" -o "x$1" = "x-h" ]; then - yelp_usage_orphans - exit 1 - fi - if [ "x$check_cache_file" != "x" ]; then - check_cache_dir=$(dirname "$check_cache_file") - check_cache_dir=$(cd "$check_cache_dir" && pwd) - check_cache_file="$check_cache_dir/"$(basename "$check_cache_file") - elif [ -d "$1" ]; then - check_cache_file=1 - else - case "$1" in - *.page | *.stub | *.cache) - check_cache_file=1 - ;; - *) - break - ;; - esac - fi - if [ "x$check_cache_file" = "x1" ]; then - check_cache_file_is_tmp="yes" - check_cache_file=$(mktemp "${TMPDIR:-/tmp}"/yelp-XXXXXXXX) - if [ "x$check_site" = "x1" ]; then - yelp-build cache -s -o "$check_cache_file" "$@" - else - yelp-build cache -o "$check_cache_file" "$@" - fi - fi - - check_out_file=`mktemp "${TMPDIR:-/tmp}"/yelp-XXXXXXXX` - check_db= - check_page=yelp_orphans_page - yelp_check_iter_args "$@" > "$check_out_file" - yelp_check_retval=$(wc -l < "$check_out_file") - if test "x$yelp_check_retval" != "x0"; then - yelp_check_retval=1 - fi - cat "$check_out_file" - rm "$check_out_file" - if [ "x$check_cache_file_is_tmp" = "xyes" ]; then - rm "$check_cache_file" - fi - exit $yelp_check_retval -} - -yelp_comments_page () { - ext=$(echo "$1" | sed -e 's/.*\.//') - bname=$(basename "$1" ".$ext") - if [ "x$check_site" = "x1" ]; then - sdir=$(cd $(dirname "$1") && pwd) - sdir=${sdir##${check_site_root}}/ - fi - xsltproc --stringparam basename "$bname" \ - --stringparam site.dir "$sdir" \ - "$xsl_comments" "$1" -} - -yelp_comments () { - if [ "$#" = "0" -o "x$1" = "x--help" -o "x$1" = "x-h" ]; then - yelp_usage_comments - exit 1 - fi - while [ "$#" != "0" ]; do - case "$1" in - "-s") - check_site="1" - check_site_root=$(pwd) - shift - ;; - *) - break - ;; - esac - done - if [ "$#" = "0" -o "x$1" = "x--help" -o "x$1" = "x-h" ]; then - yelp_usage_comments - exit 1 - fi - check_db=yelp_comments_page - check_page=yelp_comments_page - yelp_check_iter_args "$@" - exit $yelp_check_retval -} - -yelp_license_page () { - if [ "x$check_site" = "x1" ]; then - sdir=$(cd $(dirname "$1") && pwd) - sdir=${sdir##${check_site_root}}/ - fi - xsltproc --xinclude \ - --stringparam only "$check_only" \ - --stringparam except "$check_except" \ - --stringparam totals "$check_totals" \ - --stringparam site.dir "$sdir" \ - "$xsl_mal_license" "$1" -} - -yelp_license () { - if [ "$#" = "0" -o "x$1" = "x--help" -o "x$1" = "x-h" ]; then - yelp_usage_license - exit 1 - fi - while [ "$#" != "0" ]; do - case "$1" in - "-s") - check_site="1" - check_site_root=$(pwd) - shift - ;; - "--only") - shift - check_only="$1" - shift - ;; - "--except") - shift - check_except="$1" - shift - ;; - "--totals") - check_totals="1" - shift - ;; - *) - break - ;; - esac - done - if [ "$#" = "0" -o "x$1" = "x--help" -o "x$1" = "x-h" ]; then - yelp_usage_license - exit 1 - fi - check_db= - check_page=yelp_license_page - if [ "x$check_totals" = "x1" ]; then - yelp_check_iter_args "$@" | \ - sort | uniq -c | sed -e 's/^ *//' | awk '{print $2 ": " $1}' - else - yelp_check_iter_args "$@" | sort - fi -} - -yelp_style_page () { - ext=$(echo "$1" | sed -e 's/.*\.//') - bname=$(basename "$1" ".$ext") - if [ "x$check_site" = "x1" ]; then - sdir=$(cd $(dirname "$1") && pwd) - sdir=${sdir##${check_site_root}}/ - fi - style=$(echo "$mallard_style" | xsltproc - "$1") - output=1 - if [ "x$check_only_defined" = "x1" ]; then - output=0 - if [ "x$check_only" = "x" ]; then - # We treat an empty --only '' as requesting pages with no style - if [ "x$style" = "x" ]; then output=1; fi - else - for pstyle in "$style"; do - for sstyle in $(echo "$check_only" | sed -e 's/,/ /g'); do - if [ "$pstyle" = "$sstyle" ]; then - output=1 + +def _stringify(el): + ret = el.text or '' + for ch in el: + ret = ret + _stringify(ch) + if el.tail is not None: + ret = ret + el.tail + return ret + +def get_format(node): + ns = lxml.etree.QName(node).namespace + if ns in (NAMESPACES['mal'], NAMESPACES['cache']): + return 'mallard' + elif ns == NAMESPACES['db']: + return 'docbook5' + elif ns is None: + # For now, just assume no ns means docbook4 + return 'docbook4' + else: + return None + +class InputFile: + def __init__(self, filepath, filename, sitedir=None): + self.filepath = filepath + self.filename = filename + self.absfile = os.path.join(filepath, filename) + self.absdir = os.path.dirname(self.absfile) + self.sitedir = sitedir or '' + self.sitefilename = self.sitedir + self.filename + + +class Checker: + name = None + desc = None + blurb = None + formats = [] + arguments = [] + postblurb = None + xinclude = True + config = None + + def __init__(self, yelpcheck): + self.yelpcheck = yelpcheck + self.options = {} + self.fileargs = [] + self.tmpdir = None + + def __del__(self): + if self.tmpdir is not None: + shutil.rmtree(self.tmpdir) + self.tmpdir = None + + def parse_args(self, args): + while len(args) > 0: + argdef = None + if args[0].startswith('--'): + for arg_ in self.arguments: + if args[0] == '--' + arg_[0]: + argdef = arg_ + break + if argdef is None: + self.print_help() + return 1 + elif args[0].startswith('-'): + for arg_ in self.arguments: + if args[0] == arg_[1]: + argdef = arg_ + break + if argdef is None: + self.print_help() + return 1 + if argdef is not None: + takesarg = (argdef[2] is not None) + if takesarg: + if len(args) < 2: + self.print_help() + return 1 + self.options.setdefault(argdef[0], []) + self.options[argdef[0]].append(args[1]) + args = args[2:] + else: + self.options[argdef[0]] = True + args = args[1:] + else: + self.fileargs.append(args[0]) + args = args[1:] + cfgfile = None + if len(self.fileargs) > 0: + cfgfile = os.path.join(os.path.dirname(self.fileargs[0]), '.yelp-tools.cfg') + if not os.path.exists(cfgfile): + cfgfile = None + if cfgfile is None: + cfgfile = os.path.join(os.getcwd(), '.yelp-tools.cfg') + if os.path.exists(cfgfile): + self.config = configparser.ConfigParser() + try: + self.config.read(cfgfile) + except Exception as e: + print(e, file=sys.stderr) + sys.exit(1) + return 0 + + def get_option_bool(self, arg): + if arg in self.options: + return self.options[arg] == True + if self.config is not None: + val = self.config.get('check:' + self.name, arg, fallback=None) + if val is not None: + return (val == 'true') + val = self.config.get('check', arg, fallback=None) + if val is not None: + return (val == 'true') + val = self.config.get('default', arg, fallback=None) + if val is not None: + return (val == 'true') + return False + + def get_option_str(self, arg): + if arg in self.options: + if isinstance(self.options[arg], list): + return self.options[arg][-1] + if self.config is not None: + val = self.config.get('check:' + self.name, arg, fallback=None) + if val is not None: + return val + val = self.config.get('check', arg, fallback=None) + if val is not None: + return val + val = self.config.get('default', arg, fallback=None) + if val is not None: + return val + return None + + def get_option_list(self, arg): + if arg in self.options: + if isinstance(self.options[arg], list): + ret = [] + for opt in self.options[arg]: + ret.extend(opt.replace(',', ' ').split()) + return ret + if self.config is not None: + val = self.config.get('check:' + self.name, arg, fallback=None) + if val is not None: + return val.replace(',', ' ').split() + val = self.config.get('check', arg, fallback=None) + if val is not None: + return val.replace(',', ' ').split() + val = self.config.get('default', arg, fallback=None) + if val is not None: + return val.replace(',', ' ').split() + return None + + def iter_files(self, sitedir=None): + issite = self.get_option_bool('site') + if len(self.fileargs) == 0: + self.fileargs.append('.') + for filearg in self.fileargs: + if os.path.isdir(filearg): + if issite: + for infile in self.iter_site(filearg, '/'): + yield infile + else: + for fname in os.listdir(filearg): + if fname.endswith('.page'): + yield InputFile(filearg, fname) + else: + if issite: + # FIXME: should do some normalization here, I guess. + # It's hard to get this perfect without a defined start dir + yield InputFile(os.getcwd(), filearg, '/' + os.path.dirname(filearg)) + else: + yield InputFile(os.getcwd(), filearg) + + def iter_site(self, filepath, sitedir): + for fname in os.listdir(filepath): + newpath = os.path.join(filepath, fname) + if os.path.isdir(newpath): + # FIXME https://github.com/projectmallard/pintail/issues/36 + if fname == '__pintail__': + continue + for infile in self.iter_site(newpath, sitedir + fname + '/'): + yield infile + elif fname.endswith('.page'): + yield InputFile(filepath, fname, sitedir) + + def get_xml(self, xmlfile): + # FIXME: we can cache these if we add a feature to run multiple + # checkers at once + tree = lxml.etree.parse(xmlfile.absfile) + if self.xinclude: + lxml.etree.XInclude()(tree.getroot()) + return tree + + def create_tmpdir(self): + if self.tmpdir is None: + self.tmpdir = tempfile.mkdtemp() + + def print_help(self): + print('Usage: yelp-check ' + self.name + ' [OPTIONS] [FILES]') + print('Formats: ' + ' '.join(self.formats) + '\n') + #FIXME: prettify names of formats + if self.blurb is not None: + print(self.blurb + '\n') + print('Options:') + maxarglen = 2 + args = [] + for arg in self.arguments: + argkey = '--' + arg[0] + if arg[1] is not None: + argkey = arg[1] + ', ' + argkey + if arg[2] is not None: + argkey = argkey + ' ' + arg[2] + args.append((argkey, arg[3])) + for arg in args: + maxarglen = max(maxarglen, len(arg[0]) + 1) + for arg in args: + print(' ' + (arg[0]).ljust(maxarglen) + ' ' + arg[1]) + if self.postblurb is not None: + print(self.postblurb) + + def main(self, args): + pass + + +class HrefsChecker (Checker): + name = 'hrefs' + desc = 'Find broken external links in a document' + blurb = ('Find broken href links in FILES in a Mallard document, or\n' + + 'broken ulink or XLink links in FILES in a DocBook document.') + formats = ['docbook4', 'docbook5', 'mallard'] + arguments = [ + ('help', '-h', None, 'Show this help and exit'), + ('site', '-s', None, 'Treat pages as belonging to a Mallard site'), + ('allow', None, 'URL', 'Allow URL or list of URLs without checking') + ] + postblurb = 'URL may be a comma- and/or space-separated list, or specified\nmultiple times.' + + def main(self, args): + if self.parse_args(args) != 0: + return 1 + if 'help' in self.options: + self.print_help() + return 0 + + # safelisting URLs that we use as identifiers + hrefs = { + 'http://creativecommons.org/licenses/by-sa/3.0/': True, + 'https://creativecommons.org/licenses/by-sa/3.0/': True, + 'http://creativecommons.org/licenses/by-sa/3.0/us/': True, + 'https://creativecommons.org/licenses/by-sa/3.0/us/': True + } + allow = self.get_option_list('allow') + if allow is not None: + for url in allow: + hrefs[url] = True + retcode = 0 + + for infile in self.iter_files(): + xml = self.get_xml(infile) + for el in xml.xpath('//*[@href | @xlink:href | self::ulink/@url]', + namespaces=NAMESPACES): + href = el.get('href', None) + if href is None: + href = el.get('{www.w3.org/1999/xlink}href') + if href is None: + href = el.get('url') + if href is None: + continue + if href.startswith('mailto:'): + continue + if href not in hrefs: + try: + req = urllib.request.urlopen(href) + hrefs[href] = (req.status == 200) + except Exception as e: + hrefs[href] = False + if not hrefs[href]: + retcode = 1 + print(infile.sitefilename + ': ' + href) + + return retcode + + +class IdsChecker (Checker): + name = 'ids' + desc = 'Find Mallard page IDs that do not match file names' + blurb = ('Find pages in a Mallard document whose page ID does not match\n' + + 'the base file name of the page file.') + formats = ['mallard'] + arguments = [ + ('help', '-h', None, 'Show this help and exit'), + ('site', '-s', None, 'Treat pages as belonging to a Mallard site') + ] + + def main(self, args): + if self.parse_args(args) != 0: + return 1 + if 'help' in self.options: + self.print_help() + return 0 + + retcode = 0 + + for infile in self.iter_files(): + xml = self.get_xml(infile) + isok = False + pageid = None + if infile.filename.endswith('.page'): + try: + pageid = xml.getroot().get('id') + isok = (pageid == os.path.basename(infile.filename)[:-5]) + except: + isok = False + if not isok: + retcode = 1 + print(infile.sitefilename + ': ' + (pageid or '')) + + return retcode + + +class LinksChecker (Checker): + name = 'links' + desc = 'Find broken xref or linkend links in a document' + blurb = ('Find broken xref links in FILES in a Mallard document,\n' + + 'or broken linkend links in FILES in a DocBook document.') + formats = ['docbook4', 'docbook5', 'mallard'] + arguments = [ + ('help', '-h', None, 'Show this help and exit'), + ('site', '-s', None, 'Treat pages as belonging to a Mallard site'), + ('cache', '-c', 'CACHE', 'Use the existing Mallard cache CACHE'), + ('ignore', '-i', None, 'Ignore xrefs where href is present') + ] + + def __init__(self, yelpcheck): + super().__init__(yelpcheck) + self.idstoxrefs = {} + self.idstolinkends = {} + + def _accumulate_mal(self, node, pageid, sectid, xrefs, sitedir=None): + thisid = node.get('id') + if thisid is not None: + if node.tag == '{' + NAMESPACES['mal'] + '}page': + pageid = thisid + else: + sectid = thisid + curid = pageid + ignore = self.get_option_bool('ignore') + if curid is not None: + if sectid is not None: + # id attrs in cache files are already fully formed + if '#' in sectid: + curid = sectid + else: + curid = curid + '#' + sectid + if sitedir is not None: + # id attrs in cache files already have sitedir prefixed + if curid[0] != '/': + curid = sitedir + curid + self.idstoxrefs.setdefault(curid, []) + if xrefs: + xref = node.get('xref') + if xref is not None: + if not (ignore and (node.get('href') is not None)): + self.idstoxrefs[curid].append(xref) + for child in node: + self._accumulate_mal(child, pageid, sectid, xrefs, sitedir) + + def _accumulate_db(self, node, nodeid): + thisid = node.get('id') + if thisid is None: + thisid = node.get(XML_ID) + if thisid is not None: + nodeid = thisid + self.idstolinkends.setdefault(nodeid, []) + if nodeid is not None: + linkend = node.get('linkend') + if linkend is not None: + self.idstolinkends[nodeid].append(linkend) + for child in node: + self._accumulate_db(child, nodeid) + + def main(self, args): + if self.parse_args(args) != 0: + return 1 + if 'help' in self.options: + self.print_help() + return 0 + + retcode = 0 + + cachefile = self.get_option_str('cache') + if cachefile is not None: + xml = self.get_xml(InputFile(os.getcwd(), cachefile)) + self._accumulate_mal(xml.getroot(), None, None, False) + + for infile in self.iter_files(): + xml = self.get_xml(infile) + format = get_format(xml.getroot()) + if format == 'mallard': + self._accumulate_mal(xml.getroot(), None, None, True, infile.sitedir) + elif format in ('docbook4', 'docbook5'): + # For DocBook, we assume each filearg is its own document, so + # we reset the dict each time and only check within the file. + # Note that XInclude and SYSTEM includes DO happen first. + self.idstolinkends = {} + self._accumulate_db(xml.getroot(), None) + for curid in self.idstolinkends: + for linkend in self.idstolinkends[curid]: + if linkend not in self.idstolinkends: + print(curid + ': ' + linkend) + retcode = 1 + + for curid in self.idstoxrefs: + for xref in self.idstoxrefs[curid]: + checkref = xref + if checkref[0] == '#': + checkref = curid.split('#')[0] + checkref + if curid[0] == '/' and checkref[0] != '/': + checkref = curid[:curid.rfind('/')+1] + checkref + if checkref not in self.idstoxrefs: + print(curid + ': ' + xref) + retcode = 1 + + return retcode + + +class MediaChecker (Checker): + name = 'media' + desc = 'Find broken references to media files' + blurb = ('Find broken references to media files. In Mallard, this\n' + + 'checks media and thumb elements. In DocBook, this checks\n' + + 'audiodata, imagedata, and videodata elements.') + formats = ['docbook4', 'docbook5', 'mallard'] + arguments = [ + ('help', '-h', None, 'Show this help and exit'), + ('site', '-s', None, 'Treat pages as belonging to a Mallard site') + ] + + def main(self, args): + if self.parse_args(args) != 0: + return 1 + if 'help' in self.options: + self.print_help() + return 0 + + retcode = 0 + + for infile in self.iter_files(): + xml = self.get_xml(infile) + format = get_format(xml.getroot()) + srcs = [] + if format == 'mallard': + for el in xml.xpath('//mal:media[@src] | //uix:thumb | //ui:thumb | //e:mouseover', + namespaces=NAMESPACES): + srcs.append(el.get('src')) + elif format == 'docbook5': + # FIXME: do we care about entityref? + for el in xml.xpath('//db:audiodata | //db:imagedata | //db:videodata', + namespaces=NAMESPACES): + srcs.append(el.get('fileref')) + elif format == 'docbook4': + for el in xml.xpath('//audiodata | //imagedata | //videodata'): + srcs.append(el.get('fileref')) + for src in srcs: + fsrc = os.path.join(infile.absdir, src) + if not os.path.exists(fsrc): + print(infile.sitefilename + ': ' + src) + retcode = 1 + + return retcode + + +class OrphansChecker (Checker): + name = 'orphans' + desc = 'Find orphaned pages in a Mallard document' + blurb = ('Locate orphaned pages among FILES in a Mallard document.\n' + + 'Orphaned pages are any pages that cannot be reached by\n' + + 'topic links alone from the index page.') + formats = ['mallard'] + arguments = [ + ('help', '-h', None, 'Show this help and exit'), + ('site', '-s', None, 'Treat pages as belonging to a Mallard site'), + ('cache', '-c', 'CACHE', 'Use the existing Mallard cache CACHE') + ] + + def __init__(self, yelpcheck): + super().__init__(yelpcheck) + self.guidelinks = {} + self.sitesubdirs = set() + + def _collect_links(self, node, sitedir): + pageid = node.get('id') + if pageid[0] != '/': + # id attrs in cache files already have sitedir prefixed + pageid = sitedir + pageid + else: + sitedir = pageid[:pageid.rfind('/')+1] + self.guidelinks.setdefault(pageid, set()) + # For the purposes of finding orphans, we'll just pretend that + # all links to or from sections are just to or from pages. + for el in node.xpath('//mal:info/mal:link[@type="guide"]', + namespaces=NAMESPACES): + xref = el.get('xref') + if xref is None or xref == '': + continue + if xref[0] == '#': + continue + if '#' in xref: + xref = xref[:xref.find('#')] + if sitedir is not None and sitedir != '': + if xref[0] != '/': + xref = sitedir + xref + self.guidelinks[pageid].add(xref) + for el in node.xpath('//mal:info/mal:link[@type="topic"]', + namespaces=NAMESPACES): + xref = el.get('xref') + if xref is None or xref == '': + continue + if xref[0] == '#': + continue + if '#' in xref: + xref = xref[:xref.find('#')] + if sitedir is not None and sitedir != '': + if xref[0] != '/': + xref = sitedir + xref + self.guidelinks.setdefault(xref, set()) + self.guidelinks[xref].add(pageid) + for el in node.xpath('//mal:links[@type="site-subdirs" or @type="site:subdirs"]', + namespaces=NAMESPACES): + self.sitesubdirs.add(pageid) + + def main(self, args): + if self.parse_args(args) != 0: + return 1 + if 'help' in self.options: + self.print_help() + return 0 + + retcode = 0 + + cachefile = self.get_option_str('cache') + if cachefile is not None: + xml = self.get_xml(InputFile(os.getcwd(), cachefile)) + for page in xml.getroot(): + if page.tag == '{' + NAMESPACES['mal'] + '}page': + pageid = page.get('id') + if pageid is None or pageid == '': + continue + self._collect_links(page, page.get('{http://projectmallard.org/site/1.0/}dir', '')) + + pageids = set() + for infile in self.iter_files(): + xml = self.get_xml(infile) + pageid = xml.getroot().get('id') + if pageid is None: + continue + pageids.add(infile.sitedir + pageid) + self._collect_links(xml.getroot(), infile.sitedir) + + siteupdirs = {} + for pageid in self.sitesubdirs: + dirname = pageid[:pageid.rfind('/')+1] + for subid in self.guidelinks: + if subid.startswith(dirname): + if subid.endswith('/index'): + mid = subid[len(dirname):-6] + if mid != '' and '/' not in mid: + siteupdirs[subid] = pageid + + if self.get_option_bool('site'): + okpages = set(['/index']) + else: + okpages = set(['index']) + for pageid in sorted(pageids): + if pageid in okpages: + isok = True + else: + isok = False + guides = [g for g in self.guidelinks[pageid]] + if pageid in siteupdirs: + updir = siteupdirs[pageid] + if updir not in guides: + guides.append(updir) + cur = 0 + while cur < len(guides): + if guides[cur] in okpages: + isok = True break - fi - done - if [ "x$output" = "x1" ]; then break; fi - done - fi - fi - if [ "x$check_except_defined" = "x1" ]; then - if [ "x$check_except" = "x" ]; then - # We treat an empty --except '' as excluding pages with no style - if [ "x$style" = "x" ]; then output=0; fi - else - for pstyle in "$style"; do - for sstyle in $(echo "$check_except" | sed -e 's/,/ /g'); do - if [ "$pstyle" = "$sstyle" ]; then - output=0 + if guides[cur] in self.guidelinks: + for guide in self.guidelinks[guides[cur]]: + if guide not in guides: + guides.append(guide) + cur += 1 + if isok: + okpages.add(pageid) + else: + print(pageid) + retcode = 1 + + return retcode + + +class ValidateChecker (Checker): + name = 'validate' + desc = 'Validate files against a DTD or RNG' + blurb = ('Validate FILES against the appropriate DTD or RNG.\n' + + 'For Mallard pages, perform automatic RNG merging\n' + + 'based on the version attribute.') + formats = ['docbook4', 'docbook5', 'mallard'] + arguments = [ + ('help', '-h', None, 'Show this help and exit'), + ('site', '-s', None, 'Treat pages as belonging to a Mallard site'), + ('strict', None, None, 'Disallow unknown namespaces'), + ('allow', None, 'NS', 'Explicitly allow namespace NS in strict mode'), + ('jing', None, None, 'Use jing instead of xmllint for RNG validation') + ] + postblurb = 'NS may be a comma- and/or space-separated list, or specified\nmultiple times.' + + def main(self, args): + if self.parse_args(args) != 0: + return 1 + if 'help' in self.options: + self.print_help() + return 0 + + retcode = 0 + + for infile in self.iter_files(): + xml = self.get_xml(infile) + format = get_format(xml.getroot()) + command = None + if format == 'mallard': + version = xml.getroot().get('version') + if version is None or version == '': + tag = xml.getroot().tag + if tag == '{' + NAMESPACES['mal'] + '}stack': + # 1.2 isn't final yet as of 2020-01-09. Stacks will + # likely be in 1.2, so we can assume at least that. + version = '1.2' + elif tag == '{' + NAMESPACES['cache'] + '}cache': + version = 'cache/1.0' + else: + version = '1.0' + self.create_tmpdir() + rng = os.path.join(self.tmpdir, + version.replace('/', '__').replace(' ', '__')) + if not os.path.exists(rng): + strict = 'true()' if self.get_option_bool('strict') else 'false()' + allow = self.get_option_list('allow') + if allow is None: + allow = '' + else: + allow = ' '.join(allow) + subprocess.call(['xsltproc', '-o', rng, + '--param', 'rng.strict', strict, + '--stringparam', 'rng.strict.allow', allow, + os.path.join(DATADIR, 'xslt', 'mal-rng.xsl'), + infile.absfile]) + if self.get_option_bool('jing'): + command = ['jing', '-i', rng, infile.filename] + else: + command = ['xmllint', '--noout', '--xinclude', '--noent', + '--relaxng', rng, infile.filename] + elif format == 'docbook4': + if xml.docinfo.doctype.startswith('<!DOCTYPE'): + command = ['xmllint', '--noout', '--xinclude', '--noent', + '--postvalid', infile.filename] + else: + command = ['xmllint', '--noout', '--xinclude', '--noent', + '--dtdvalid', + 'http://www.oasis-open.org/docbook/xml/4.5/docbookx.dtd', + infile.filename] + elif format == 'docbook5': + version = xml.getroot().get('version') + if version is None or version == '': + version = '5.0' + # Canonical URIs are http, but they 301 redirect to https. jing + # can handle https fine, but not the redirect. And jing doesn't + # look at catalogs. So just always feed jing an https URI. + rnghttp = 'http://docbook.org/xml/' + version + '/rng/docbook.rng' + rnghttps = 'https://docbook.org/xml/' + version + '/rng/docbook.rng' + if self.get_option_bool('jing'): + command = ['jing', '-i', rnghttps, infile.filename] + else: + # xmllint, on the other hand, does support catalogs. It also + # doesn't do the redirect, but it wouldn't matter if it did + # because it doesn't do https. So if the schema is available + # locally in the catalog, hand xmllint the http URI so it + # can use the local copy. Otherwise, we have to get curl + # involved to do https. + try: + catfile = subprocess.check_output(['xmlcatalog', + '/etc/xml/catalog', + rnghttp], + stderr=subprocess.DEVNULL, + text=True) + for catline in catfile.split('\n'): + if catline.startswith('file://'): + command = ['xmllint', '--noout', '--xinclude', '--noent', + '--relaxng', rnghttp, infile.filename] + except: + pass + if command is None: + self.create_tmpdir() + rngfile = os.path.join(self.tmpdir, 'docbook-' + version + '.rng') + if not os.path.exists(rngfile): + urllib.request.urlretrieve(rnghttps, rngfile) + command = ['xmllint', '--noout', '--xinclude', '--noent', + '--relaxng', rngfile, infile.filename] + if command is not None: + try: + subprocess.check_output(command, + cwd=infile.filepath, + stderr=subprocess.STDOUT, + text=True) + except subprocess.CalledProcessError as e: + retcode = e.returncode + print(e.output) + else: + retcode = 1 + + return retcode + + +class CommentsChecker (Checker): + name = 'comments' + desc = 'Print the editorial comments in a document' + blurb = ('Print the editorial comments in the files FILES, using the\n' + + 'comment element in Mallard and the remark element in DocBook.') + formats = ['docbook4', 'docbook5', 'mallard'] + arguments = [ + ('help', '-h', None, 'Show this help and exit'), + ('site', '-s', None, 'Treat pages as belonging to a Mallard site') + ] + + def main(self, args): + if self.parse_args(args) != 0: + return 1 + if 'help' in self.options: + self.print_help() + return 0 + + for infile in self.iter_files(): + xml = self.get_xml(infile) + format = get_format(xml.getroot()) + if format == 'mallard': + for el in xml.xpath('//mal:comment', namespaces=NAMESPACES): + thisid = xml.getroot().get('id') + par = el + while par is not None: + if par.tag == '{' + NAMESPACES['mal'] + '}section': + sectid = par.get('id') + if sectid is not None: + thisid = thisid + '#' + sectid + break + par = par.getparent() + print('Page: ' + infile.sitedir + thisid) + for ch in el.xpath('mal:cite[1]', namespaces=NAMESPACES): + name = _stringify(ch).strip() + href = ch.get('href') + if href is not None and href.startswith('mailto:'): + name = name + ' <' + href[7:] + '>' + print('From: ' + name) + date = ch.get('date') + if date is not None: + print('Date: ' + date) + print('') + for ch in el: + if isinstance(ch, lxml.etree._ProcessingInstruction): + continue + elif ch.tag == '{' + NAMESPACES['mal'] + '}cite': + continue + elif ch.tag in ('{' + NAMESPACES['mal'] + '}p', + '{' + NAMESPACES['mal'] + '}title'): + for s in _stringify(ch).strip().split('\n'): + print(' ' + s.strip()) + print('') + else: + name = lxml.etree.QName(ch).localname + print(' <' + name + '>...</' + name + '>\n') + elif format in ('docbook4', 'docbook5'): + if format == 'docbook4': + dbxpath = '//remark' + else: + dbxpath = '//db:remark' + for el in xml.xpath(dbxpath, namespaces=NAMESPACES): + thisid = infile.filename + par = el + while par is not None: + sectid = par.get('id') + if sectid is None: + sectid = par.get(XML_ID) + if sectid is not None: + thisid = thisid + '#' + sectid + break + par = par.getparent() + print('Page: ' + thisid) + flag = el.get('revisionflag') + if flag is not None: + print('Flag: ' + flag) + print('') + for s in _stringify(el).strip().split('\n'): + print(' ' + s.strip()) + print('') + + return 0 + + +class LicenseChecker (Checker): + name = 'license' + desc = 'Report the license of Mallard pages' + blurb = ('Report the license of the Mallard page files FILES. Each\n' + + 'matching page is reporting along with its license, reported\n' + + 'based on the href attribute of the license element. Common\n' + + 'licenses use a shortened identifier. Pages with multiple\n' + + 'licenses have the identifiers separated by spaces. Pages\n' + + 'with no license element report \'none\'. Licenses with no\n' + + 'href attribute are reported as \'unknown\'') + formats = ['mallard'] + arguments = [ + ('help', '-h', None, 'Show this help and exit'), + ('site', '-s', None, 'Treat pages as belonging to a Mallard site'), + ('only', None, 'LICENSES', 'Only show pages whose license is in LICENSES'), + ('except', None, 'LICENSES', 'Exclude pages whose license is in LICENSES'), + ('totals', None, None, 'Show total counts for each license') + ] + postblurb = 'LICENSES may be a comma- and/or space-separated list, or specified\nmultiple times.' + + def get_license(self, href): + if href is None: + return 'unknown' + elif (href.startswith('http://creativecommons.org/licenses/') or + href.startswith('https://creativecommons.org/licenses/')): + return 'cc-' + '-'.join([x for x in href.split('/') if x][3:]) + elif (href.startswith('http://www.gnu.org/licenses/') or + href.startswith('https://www.gnu.org/licenses/')): + return href.split('/')[-1].replace('.html', '') + else: + return 'unknown' + + def main(self, args): + if self.parse_args(args) != 0: + return 1 + if 'help' in self.options: + self.print_help() + return 0 + + totals = {} + + for infile in self.iter_files(): + xml = self.get_xml(infile) + thisid = xml.getroot().get('id') or infile.filename + licenses = [] + for el in xml.xpath('/mal:page/mal:info/mal:license', + namespaces=NAMESPACES): + licenses.append(self.get_license(el.get('href'))) + if len(licenses) == 0: + licenses.append('none') + + only = self.get_option_list('only') + if only is not None: + skip = True + for lic in licenses: + if lic in only: + skip = False + if skip: + continue + cept = self.get_option_list('except') + if cept is not None: + skip = False + for lic in licenses: + if lic in cept: + skip = True + if skip: + continue + + if self.get_option_bool('totals'): + for lic in licenses: + totals.setdefault(lic, 0) + totals[lic] += 1 + else: + print(infile.sitedir + thisid + ': ' + ' '.join(licenses)) + + if self.get_option_bool('totals'): + for lic in sorted(totals): + print(lic + ': ' + str(totals[lic])) + + return 0 + + +class StatusChecker (Checker): + name = 'status' + desc = 'Report the status of Mallard pages' + blurb = ('Report the status of the Mallard page files FILES. Each\n' + + 'matching page is reporting along with its status.') + formats = ['mallard'] + arguments = [ + ('help', '-h', None, 'Show this help and exit'), + ('site', '-s', None, 'Treat pages as belonging to a Mallard site'), + ('version', None, 'VER', 'Select revisions with the version attribute VER'), + ('docversion', None, 'VER', 'Select revisions with the docversion attribute VER'), + ('pkgversion', None, 'VER', 'Select revisions with the pkgversion attribute VER'), + ('older', None, 'DATE', 'Only show pages older than DATE'), + ('newer', None, 'DATE', 'Only show pages newer than DATE'), + ('only', None, 'STATUSES', 'Only show pages whose status is in STATUSES'), + ('except', None, 'STATUSES', 'Exclude pages whose status is in STATUSES'), + ('totals', None, None, 'Show total counts for each status') + ] + postblurb = 'VER and STATUSES may be comma- and/or space-separated lists, or specified\nmultiple times.' + + def main(self, args): + if self.parse_args(args) != 0: + return 1 + if 'help' in self.options: + self.print_help() + return 0 + + totals = {} + + checks = [] + ver = self.get_option_list('version') + if ver is not None: + checks.append(ver) + ver = self.get_option_list('docversion') + if ver is not None: + checks.append(['doc:' + v for v in ver]) + ver = self.get_option_list('pkgversion') + if ver is not None: + checks.append(['pkg:' + v for v in ver]) + + for infile in self.iter_files(): + xml = self.get_xml(infile) + pageid = xml.getroot().get('id') + bestrev = None + for rev in xml.xpath('/mal:page/mal:info/mal:revision', namespaces=NAMESPACES): + revversion = (rev.get('version') or '').split() + docversion = rev.get('docversion') + if docversion is not None: + revversion.append('doc:' + docversion) + pkgversion = rev.get('pkgversion') + if pkgversion is not None: + revversion.append('pkg:' + pkgversion) + revok = True + for check in checks: + checkok = False + for v in check: + if v in revversion: + checkok = True + break + if not checkok: + revok = False break - fi - done - #if [ "x$output" = "x0" ]; then break; fi - done - fi - fi - if [ "x$output" = "x1" ]; then - echo "$sdir$bname: $style" - fi -} - -yelp_style () { - if [ "$#" = "0" -o "x$1" = "x--help" -o "x$1" = "x-h" ]; then - yelp_usage_style - exit 1 - fi - while [ "$#" != "0" ]; do - case "$1" in - "-s") - check_site="1" - check_site_root=$(pwd) - shift - ;; - "--only") - shift - check_only_defined=1 - check_only="$1" - shift - ;; - "--except") - shift - check_except_defined=1 - check_except="$1" - shift - ;; - "--totals") - check_totals="1" - shift - ;; - *) - break - ;; - esac - done - if [ "$#" = "0" -o "x$1" = "x--help" -o "x$1" = "x-h" ]; then - yelp_usage_style - exit 1 - fi - check_db= - check_page=yelp_style_page - if [ "x$check_totals" = "x1" ]; then - yelp_check_iter_args "$@" | \ - while read line; do - styles=$(echo "$line" | sed -e 's/^[^:]*://') - if [ "x$styles" = "x" ]; then - echo "" - else - for style in $styles; do - echo "$style" - done - fi - done | \ - sort | uniq -c | sed -e 's/^ *//' | awk '{print $2 ": " $1}' - else - yelp_check_iter_args "$@" | sort - fi -} - -yelp_status () { - if [ "$#" = "0" -o "x$1" = "x--help" -o "x$1" = "x-h" ]; then - yelp_usage_status - exit 1 - fi - while [ "$#" != "0" ]; do - case "$1" in - "-s") - check_site="1" - check_site_root=$(pwd) - shift - ;; - "--version") - shift - check_version="$1" - shift - ;; - "--docversion") - shift - check_docversion="$1" - shift - ;; - "--pkgversion") - shift - check_pkgversion="$1" - shift - ;; - "--older") - shift - check_older="$1" - shift - ;; - "--newer") - shift - check_newer="$1" - shift - ;; - "--only") - shift - check_only="$1" - shift - ;; - "--except") - shift - check_except="$1" - shift - ;; - "--totals") - check_totals="1" - shift - ;; - *) - break - ;; - esac - done - if [ "$#" = "0" -o "x$1" = "x--help" -o "x$1" = "x-h" ]; then - yelp_usage_status - exit 1 - fi - check_cache_file=`mktemp "${TMPDIR:-/tmp}"/yelp-XXXXXXXX` - if [ "x$check_site" = "x1" ]; then - yelp-build cache -s -o "$check_cache_file" "$@" - else - yelp-build cache -o "$check_cache_file" "$@" - fi - xsltproc \ - --stringparam version "$check_version" \ - --stringparam docversion "$check_docversion" \ - --stringparam pkgversion "$check_pkgversion" \ - --stringparam newer "$check_newer" \ - --stringparam older "$check_older" \ - --stringparam only "$check_only" \ - --stringparam except "$check_except" \ - --stringparam totals "$check_totals" \ - "$xsl_mal_status" "$check_cache_file" - rm "$check_cache_file" - return 0 -} - -yelp_validate_db () { - version=$(echo "$docbook_version" | xsltproc - "$1") - major=$(echo "$version" | cut -c1) - if [ "x$major" = "x5" ]; then - check_out_file=`mktemp "${TMPDIR:-/tmp}"/yelp-XXXXXXXX` - # Canonical URIs are http, but they 301 redirect to https. jing can handle - # https fine, but not the redirect. And jing doesn't look at catalogs. So - # just always feed jing an https URI. - if [ "x$check_jing" = "x1" ]; then - rng_uri="https://docbook.org/xml/$version/rng/docbook.rng" - jing -i "$rng_uri" "$1" > "$check_out_file" 2>&1 - else - # xmllint, on the other hand, does support catalogs. It also doesn't - # do the redirect, but it wouldn't matter if it did because it doesn't - # do https. So if the schema is available locally in the catalog, hand - # xmllint the http URI so it can use the local copy. Otherwise, we have - # to get curl involved to do https. - rng_uri="http://docbook.org/xml/$version/rng/docbook.rng" - incat=$(xmlcatalog /etc/xml/catalog "$rng_uri" | grep -c '^file:') - if [ "x$incat" != "x0" ]; then - xmllint --noout --xinclude --noent --relaxng "$rng_uri" "$1" > "$check_out_file" 2>&1 - else - rng_uri="https://docbook.org/xml/$version/rng/docbook.rng" - check_rng_file=`mktemp "${TMPDIR:-/tmp}"/yelp-XXXXXXXX` - curl -sL -o "$check_rng_file" "$rng_uri" - xmllint --noout --xinclude --noent --relaxng "$check_rng_file" "$1" > "$check_out_file" 2>&1 - rm "$check_rng_file" - fi - fi - yelp_check_retval="$?" - cat "$check_out_file" | grep -v 'validates$' - rm "$check_out_file" - elif xmllint --nocdata "$1" | grep -q '<!DOCTYPE'; then - xmllint --noout --xinclude --noent --postvalid "$1" || yelp_check_retval="$?" - else - dtd_uri='http://www.oasis-open.org/docbook/xml/4.5/docbookx.dtd' - xmllint --noout --xinclude --noent --dtdvalid "$dtd_uri" "$1" || yelp_check_retval="$?" - fi -} - -yelp_validate_page () { - # Using temp files because pipes create subshells, making it really - # hard to return the right exit status in a portable way. - if [ "x$check_rng_dir" = "x" ]; then - check_rng_dir=`mktemp -d "${TMPDIR:-/tmp}"/yelp-XXXXXXXX` - fi - check_out_file=`mktemp "${TMPDIR:-/tmp}"/yelp-XXXXXXXX` - check_rng_file=`( - echo '<xsl:stylesheet' - echo ' xmlns:cache="http://projectmallard.org/cache/1.0/"' - echo ' xmlns:xsl="http://www.w3.org/1999/XSL/Transform"' - echo ' version="1.0">' - echo '<xsl:output method="text"/>' - echo '<xsl:template match="/*">' - echo '<xsl:choose>' - echo '<xsl:when test="string(@version) != '"''"'">' - echo '<xsl:value-of select="@version"/>' - echo '</xsl:when>' - echo '<xsl:when test="/cache:cache">' - echo '<xsl:text>cache/1.0 1.0</xsl:text>' - echo '</xsl:when>' - echo '<xsl:otherwise>' - echo '<xsl:text>1.0</xsl:text>' - echo '</xsl:otherwise>' - echo '</xsl:choose>' - echo '</xsl:template>' - echo '</xsl:stylesheet>' - ) | xsltproc - "$1"` - check_rng_file=`urlencode "$check_rng_file" /`.rng - if [ ! -f "$check_rng_dir/$check_rng_file" ]; then - # If we've already made an RNG file for this version string, don't - # do it again. We've urlencoded the file name + slashes, because - # version strings often contain slashes. But xsltproc treats the - # -o option as a URL and urldecodes, so doubly urlencode, because - # we want the urlencoded string to be the on-disk name. - xsltproc -o "$check_rng_dir/"`urlencode "$check_rng_file"` \ - --param rng.strict "$check_strict" \ - --stringparam rng.strict.allow "$check_strict_allow" \ - "$xsl_mal_rng" "$1" - fi - if [ "x$check_jing" = "x1" ]; then - jing -i "$check_rng_dir/$check_rng_file" "$1" > "$check_out_file" 2>&1 - else - xmllint --noout --xinclude --noent --relaxng "$check_rng_dir/$check_rng_file" "$1" > "$check_out_file" 2>&1 - fi - ret="$?" - cat "$check_out_file" | grep -v 'validates$' - rm "$check_out_file" - return $ret; -} - -yelp_validate () { - if [ "$#" = "0" -o "x$1" = "x--help" -o "x$1" = "x-h" ]; then - yelp_usage_validate - exit 1 - fi - check_strict="false()" - check_strict_allow="" - while [ "$#" != "0" ]; do - case "$1" in - "-s") - check_site="1" - check_site_root=$(pwd) - shift - ;; - "--strict") - check_strict="true()" - shift - ;; - "--allow") - shift - check_strict_allow="$check_strict_allow $1" - shift - ;; - "--jing") - check_jing="1" - shift - ;; - *) - break - ;; - esac - done - if [ "$#" = "0" -o "x$1" = "x--help" -o "x$1" = "x-h" ]; then - yelp_usage_validate - exit 1 - fi - check_db=yelp_validate_db - check_page=yelp_validate_page - yelp_check_iter_args "$@" - if [ "x$check_rng_dir" != "x" ]; then - rm "$check_rng_dir"/*.rng - rmdir "$check_rng_dir" - fi - exit $yelp_check_retval -} - -cmd="$1" -shift -case "x$cmd" in - "xcomments") - yelp_comments "$@" - ;; - "xhrefs") - yelp_hrefs "$@" - ;; - "xids") - yelp_ids "$@" - ;; - "xlinks") - yelp_links "$@" - ;; - "xorphans") - yelp_orphans "$@" - ;; - "xlicense") - yelp_license "$@" - ;; - "xmedia") - yelp_media "$@" - ;; - "xstatus") - yelp_status "$@" - ;; - "xstyle") - yelp_style "$@" - ;; - "xvalidate") - yelp_validate "$@" - ;; - *) - yelp_usage - ;; -esac + if revok: + if bestrev is None: + bestrev = rev + continue + bestdate = bestrev.get('date') + thisdate = rev.get('date') + if bestdate is None: + bestrev = rev + elif thisdate is None: + pass + elif thisdate >= bestdate: + bestrev = rev + if bestrev is not None: + status = bestrev.get('status') or 'none' + date = bestrev.get('date') or None + else: + status = 'none' + date = None + older = self.get_option_str('older') + if older is not None: + if date is None or date >= older: + continue + newer = self.get_option_str('newer') + if newer is not None: + if date is None or date <= newer: + continue + only = self.get_option_list('only') + if only is not None: + if status not in only: + continue + cept = self.get_option_list('except') + if cept is not None: + if status in cept: + continue + if self.get_option_bool('totals'): + totals.setdefault(status, 0) + totals[status] += 1 + else: + print(infile.sitedir + pageid + ': ' + status) + + if self.get_option_bool('totals'): + for st in sorted(totals): + print(st + ': ' + str(totals[st])) + + return 0 + + +class StyleChecker (Checker): + name = 'style' + desc = 'Report the style attribute of Mallard pages' + blurb = ('Report the page style attribute of the Mallard page files\n' + + 'FILES. Each matching page is reporting along with its status.') + formats = ['mallard'] + arguments = [ + ('help', '-h', None, 'Show this help and exit'), + ('site', '-s', None, 'Treat pages as belonging to a Mallard site'), + ('only', None, 'STYLES', 'Only show pages whose style is in STATUSES'), + ('except', None, 'STYLES', 'Exclude pages whose style is in STATUSES'), + ('totals', None, None, 'Show total counts for each style') + ] + postblurb = 'STYLES may be comma- and/or space-separated lists, or specified\nmultiple times.' + + def main(self, args): + if self.parse_args(args) != 0: + return 1 + if 'help' in self.options: + self.print_help() + return 0 + + totals = {} + + for infile in self.iter_files(): + xml = self.get_xml(infile) + thisid = xml.getroot().get('id') + style = xml.getroot().get('style') + if style is None: + style = 'none' + styles = style.split() + # We'll set style to None if it doesn't meat the criteria + only = self.get_option_list('only') + if only is not None: + if len(only) == 0: + # We treat a blank --only as requesting pages with no style + if style != 'none': + style = None + else: + allow = False + for st in styles: + if st in only: + allow = True + break + if not allow: + style = None + cept = self.get_option_list('except') + if cept is not None: + for st in styles: + if st in cept: + style = None + break + if self.get_option_bool('totals'): + if style is not None: + for st in styles: + totals.setdefault(st, 0) + totals[st] += 1 + else: + if style is not None: + print(infile.sitedir + thisid + ': ' + style) + + if self.get_option_bool('totals'): + for st in sorted(totals): + print(st + ': ' + str(totals[st])) + + return 0 + + +class CustomChecker(Checker): + formats = ['docbook4', 'docbook5', 'mallard'] + arguments = [ + ('help', '-h', None, 'Show this help and exit'), + ('site', '-s', None, 'Treat pages as belonging to a Mallard site') + ] + + def __init__(self, name, yelpcheck): + super().__init__(yelpcheck) + self.name = name + + def main(self, args): + if self.parse_args(args) != 0: + return 1 + + sect = 'check:' + self.name + if self.config is None or (sect not in self.config.sections()): + print('Unrecognized command: ' + self.name, file=sys.stderr) + return 1 + self.blurb = self.config.get(sect, 'blurb', fallback=None) + if self.blurb is not None: + self.blurb = '\n'.join(textwrap.wrap(self.blurb)) + + if 'help' in self.options: + self.print_help() + return 0 + + assertexpr = self.config.get(sect, 'assert', fallback=None) + if assertexpr is not None: + return self.run_assert(assertexpr) + + print('No action found for command: ' + self.name, file=sys.stderr) + return 1 + + def run_assert(self, assertexpr): + sect = 'check:' + self.name + selectexpr = self.config.get(sect, 'select', fallback='/') + message = self.config.get(sect, 'message', fallback='Assertion failed') + self.xinclude = self.config.get(sect, 'xinclude', fallback='true') != 'false' + + namespaces = {} + if 'namespaces' in self.config.sections(): + for ns in self.config.options('namespaces'): + namespaces[ns] = self.config.get('namespaces', ns) + + for infile in self.iter_files(): + xml = self.get_xml(infile) + thisid = xml.getroot().get('id') or infile.filename + # FIXME check these expressions and give better errors + for root in xml.xpath(selectexpr, namespaces=namespaces): + if not bool(root.xpath(assertexpr, namespaces=namespaces)): + print(infile.sitedir + thisid + ': ' + message) + # FIXME are these comments outdated? remove? + # check if self.config has section check:self.name + # check if section has select, assert, message + + +class YelpCheck: + def __init__(self): + pass + + def main(self): + if len(sys.argv) < 2: + self.print_usage() + return 1 + + checker = None + for cls in Checker.__subclasses__(): + if sys.argv[1] == cls.name: + checker = cls(self) + + if checker is None: + checker = CustomChecker(sys.argv[1], self) + + return checker.main(sys.argv[2:]) + + def print_usage(self): + print('Usage: yelp-check <COMMAND> [OPTIONS] [FILES]') + namelen = 2 + checks = [] + reports = [] + others = [] + for cls in sorted(Checker.__subclasses__(), key=(lambda cls: cls.name or '')): + if cls is CustomChecker: + continue + namelen = max(namelen, len(cls.name) + 2) + if cls in (HrefsChecker, IdsChecker, LinksChecker, + MediaChecker, OrphansChecker, ValidateChecker): + checks.append(cls) + elif cls in (CommentsChecker, LicenseChecker, StatusChecker, + StyleChecker): + reports.append(cls) + else: + others.append(cls) + if len(checks) > 0: + print('\nCheck commands:') + for cls in checks: + print(' ' + cls.name.ljust(namelen) + cls.desc) + if len(reports) > 0: + print('\nReport commands:') + for cls in reports: + print(' ' + cls.name.ljust(namelen) + cls.desc) + if len(others) > 0: + print('\nOther commands:') + for cls in others: + print(' ' + cls.name.ljust(namelen) + cls.desc) + config = configparser.ConfigParser() + try: + config.read('.yelp-tools.cfg') + except: + return + customs = [] + for sect in config.sections(): + if sect.startswith('check:'): + name = sect[6:] + skip = False + for cls in Checker.__subclasses__(): + if name == cls.name: + skip = True + break + if skip: + continue + if config.get(sect, 'assert', fallback=None) == None: + continue + desc = config.get(sect, 'desc', fallback='') + namelen = max(namelen, len(name) + 2) + customs.append((name, desc)) + if len(customs) > 0: + print('\nCustom commands:') + for name, desc in customs: + print(' ' + name.ljust(namelen) + desc) + + +if __name__ == '__main__': + try: + sys.exit(YelpCheck().main()) + except KeyboardInterrupt: + sys.exit(1) |