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

adapt rcllyx to newer lyx which apparently always export utf-8

Browse Source
This commit is contained in:
Jean-Francois Dockes 2010-09-30 17:49:38 +02:00
parent 7b803170c2
commit 910f995530
2 changed files with 72 additions and 57 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

118
src/filters/rcllyx
View File

@ -139,62 +139,78 @@ textfile=$workdir/`basename "$binfile" .lyx`.txt
#echo binfile: $binfile;echo lyxfile: $lyxfile ; ls -l $lyxfile; echo textfile: $textfile #echo binfile: $binfile;echo lyxfile: $lyxfile ; ls -l $lyxfile; echo textfile: $textfile
# Run lyx --export # Run lyx --export
lyx --export text "$lyxfile" lyx --export text "$lyxfile" || senderror "lyx --export not working"
# Charset and language # Need the lyx version. After some point -export prints utf-8,
formatline=`egrep '^\\\lyxformat ' "$lyxfile"` # whatever the input version
if test -n "$formatline" ; then LYXOUTPUTUTF=No
set $formatline vline=`lyx --version 2>&1 | head -1 | tr '.' ' '`
format=$2 set $vline
maj=`expr $2 '*' 10000` || senderror "Bad lyx version string $vline"
med=`expr $3 '*' 100` || senderror "Bad lyx version string $vline"
min=`expr $4 '*' 1` || senderror "Bad lyx version string $vline"
version=`expr $maj + $med + $min` || senderror "Bad lyx version string $vline"
if test $version -ge 10607 ; then
LYXOUTPUTUTF=Yes
fi fi
charsetline=`egrep '^\\\inputencoding ' "$lyxfile"` if test X$LYXOUTPUTUTF = XNo ; then
if test -n "$charsetline" ; then echo "OLD VERSION"
set $charsetline # Charset and language
charset=$2 formatline=`egrep '^\\\lyxformat ' "$lyxfile"`
fi if test -n "$formatline" ; then
langline=`egrep '^\\\language ' "$lyxfile"` set $formatline
if test -n "$langline" ; then format=$2
set $langline fi
lang=$2 charsetline=`egrep '^\\\inputencoding ' "$lyxfile"`
fi if test -n "$charsetline" ; then
#echo format: [$format] charset: [$charset] lang [$lang] set $charsetline
charset=$2
if test "$format" -ge 249 ; then fi
charset=utf-8 langline=`egrep '^\\\language ' "$lyxfile"`
else if test -n "$langline" ; then
# try to guess the charset from the language: this is in no way guaranteed set $langline
# to work, the logic has built-in inconsistencies even beyond the numerous lang=$2
# external ones (what if the ukrainian writer prefers koi8-r ?). This is a fi
# joke. #echo format: [$format] charset: [$charset] lang [$lang]
if test "$format" -ge 249 ; then
charset=utf-8
else
# try to guess the charset from the language: this is in no way guaranteed
# to work, the logic has built-in inconsistencies even beyond the numerous
# external ones (what if the ukrainian writer prefers koi8-r ?). This is a
# joke.
if test -z "$charset" -o "$charset" = default -o "$charset" = auto ; then if test -z "$charset" -o "$charset" = default -o "$charset" = auto ; then
case "$lang" in case "$lang" in
american|afrikaans|basque|catalan|danish|dutch|english|faeroese|finnish|french|galician|german|icelandic|irish|italian|norwegian|portuguese|spanish|swedish) american|afrikaans|basque|catalan|danish|dutch|english|faeroese|finnish|french|galician|german|icelandic|irish|italian|norwegian|portuguese|spanish|swedish)
charset=iso-8859-1;; charset=iso-8859-1;;
czech|german|hungarian|polish|romanian|croatian|slovak|slovene) czech|german|hungarian|polish|romanian|croatian|slovak|slovene)
charset=iso-8859-2;; charset=iso-8859-2;;
esperanto|galician|maltese|Turkish) esperanto|galician|maltese|Turkish)
charset=iso-8859-3;; charset=iso-8859-3;;
estonian|latvian|lithuanian) estonian|latvian|lithuanian)
charset=iso-8859-4;; charset=iso-8859-4;;
bulgarian|byelorussian|macedonian|russian|serbian|ukrainian) bulgarian|byelorussian|macedonian|russian|serbian|ukrainian)
charset=iso-8859-5;; charset=iso-8859-5;;
arabic) arabic)
charset=iso-8859-6;; charset=iso-8859-6;;
greek) greek)
charset=iso-8859-7;; charset=iso-8859-7;;
hebrew) hebrew)
charset=iso-8859-8;; charset=iso-8859-8;;
#ISO-8859-9 - Latin 5 Same as 8859-1 except for Turkish instead of #ISO-8859-9 - Latin 5 Same as 8859-1 except for Turkish instead of
#Icelandic. ? What is one to do :) #Icelandic. ? What is one to do :)
#ISO-8859-10 - Latin 6 #ISO-8859-10 - Latin 6
lappish|nordic|eskimo|inuit|sami) lappish|nordic|eskimo|inuit|sami)
charset=iso-8859-10;; charset=iso-8859-10;;
albanian|german|english|basque|breton|catalan|danish|spanish|estonian|esthonian|faeroese|faroese|finnish|french|frisian|friesian|scottish|goidelic|irish|gaelic|galician|welsh|greenlandic|inuit|icelandic|italian|latin|dutch|norvegian|portuguese|romansch|romansh|friulian|ladin|swedish) albanian|german|english|basque|breton|catalan|danish|spanish|estonian|esthonian|faeroese|faroese|finnish|french|frisian|friesian|scottish|goidelic|irish|gaelic|galician|welsh|greenlandic|inuit|icelandic|italian|latin|dutch|norvegian|portuguese|romansch|romansh|friulian|ladin|swedish)
charset=iso-8859-15;; charset=iso-8859-15;;
*) *)
charset=iso-8859-1;; charset=iso-8859-1;;
esac esac
fi
fi fi
# End Old lyx needing output tweaking
fi fi
if test -n "$charset" ; then if test -n "$charset" ; then

11
tests/lyx/lyx.sh
View File

@ -5,12 +5,11 @@ topdir=`dirname $0`/..
initvariables $0 initvariables $0
recollq 'Bienvenue Dans Univers De Lyx' 2> $mystderr | (
egrep -v '^Recoll query: ' > $mystdout recollq 'Bienvenue Dans Univers De Lyx'
recollq 'Welcome To Lyx' 2>> $mystderr | recollq 'Welcome To Lyx'
egrep -v '^Recoll query: ' >> $mystdout recollq 'Udvozli Ont A LyX'
recollq 'Udvozli Ont A LyX' 2>> $mystderr | ) 2> $mystderr | egrep -v '^Recoll query: ' > $mystdout
egrep -v '^Recoll query: ' >> $mystdout
diff -w ${myname}.txt $mystdout > $mydiffs 2>&1 diff -w ${myname}.txt $mystdout > $mydiffs 2>&1