Odt2xliff
Revision as of 16:58, 10 January 2008 by Francis Tyers (talk | contribs)
Here is a rough-and-ready script to turn the text from an ODT file into an XLIFF using Apertium and translate toolkit.
#!/bin/sh INPUT_FILE="" OUTPUT_FILE="" function translate_odt { INPUT_TMPDIR=/tmp/$$odtdir export LC_CTYPE=$(locale -a|grep -i "utf[.]*8"|head -1); if [[ LC_CTYPE == "" ]] then echo "Error: Install an UTF-8 locale in your system"; exit 1; fi if [[ $(which zip) == "" ]] then echo "Error: Install 'zip' command in your system"; exit 1; fi if [[ $(which unzip) == "" ]] then echo "Error: Install 'unzip' command in your system"; exit 1; fi if [[ $FICHERO == "" ]] then FICHERO=/tmp/$$odtorig cat > $FICHERO BORRAFICHERO="true" fi unzip -q -o -d $INPUT_TMPDIR $FICHERO find $INPUT_TMPDIR | grep content\\.xml |\ awk '{printf "<file name=\"" $0 "\"/>"; PART = $0; while(getline < PART) printf(" %s", $0); printf("\n");}' |\ apertium-desodt VUELVE=$(pwd) cd $INPUT_TMPDIR cd $VUELVE rm -rf $INPUT_TMPDIR } INPUT_FILE=$1 FICHERO=$INPUT_FILE; TEMPO=/tmp/tem.po translate_odt | sed 's/<\\\/text:p>/]\n\n\n\n[/g' | python txtformat.py | txt2po - > $TEMPO po2xliff $TEMPO rm $TEMPO