#!/bin/bash

export LANG="sv_SE.UTF-8"
export LC_CTYPE="sv_SE.UTF-8"
export LC_NUMERIC="sv_SE.UTF-8"
export LC_TIME="sv_SE.UTF-8"
export LC_COLLATE="sv_SE.UTF-8"
export LC_MONETARY="sv_SE.UTF-8"
export LC_MESSAGES="sv_SE.UTF-8"
export LC_PAPER="sv_SE.UTF-8"
export LC_NAME="sv_SE.UTF-8"
export LC_ADDRESS="sv_SE.UTF-8"
export LC_TELEPHONE="sv_SE.UTF-8"
export LC_MEASUREMENT="sv_SE.UTF-8"
export LC_IDENTIFICATION="sv_SE.UTF-8"
export LC_ALL="sv_SE.UTF-8"

SALDO="/home/markus/fm/sblex/bin/saldo"
SALDODICT="/home/markus/fm/dicts/saldo.dict"

DALIN="/home/markus/fm/sblex/bin/dalin"
DALINDICT="/home/markus/fm/dicts/dalin.dict"

FSV="/home/markus/fm/sblex/bin/fsv"
FSVDICT="/home/markus/fm/dicts/fsv.dict"

# make sure that xml files have the correct mime-type
find . -name "*.xml" | grep -v '.svn' | xargs -n 1 svn propset svn:mime-type text/xml
find . -name "*.xml" | grep -v '.svn' | xargs -n 1 svn propset svn:keywords "Id"
# make sure that xml files have ID set
# find . -name "*.xml" | grep -v '.svn' | xargs -n 1 svn propset svn:keywords Id

#lwt  
svn export --force https://svn.spraakdata.gu.se/sb/fnplusplus/lwt/lwt-meanings.txt
cat lwt-meanings.txt | ./skript/_xml.py lwt id +saldo eng def example > lexikon/lwt/lwt.xml

#parole  
svn export --force https://svn.spraakdata.gu.se/sb-arkiv/lexikon/parole/parole.txt
cat parole.txt | ./skript/_xml.py parole gf saldo pos v -id > lexikon/parole/parole.xml

# kelly
#svn export --force https://svn.spraakdata.gu.se/sb-arkiv/lexikon/kelly/kelly_saldo.txt NOT ANYMORE SINCE IT'S LOCALLY CHANGED
cat kelly_saldo.txt | ./skript/_xml.py kelly id raw wpm cefr source grammar gf +saldo pos example > lexikon/kelly/kelly.xml

#simple
svn export --force https://svn.spraakdata.gu.se/sb/fnplusplus/simple_parole/simple.txt
cat simple.txt | ./skript/_xml.py simple gf saldo pos ontology domain class > lexikon/simple/simple.xml

#saldo  
svn export --force https://svn.spraakdata.gu.se/repos/sblex/sal/trunk/saldo.txt
cat saldo.txt | ./skript/_xml.py saldo saldo mf +pf lem gf pos p > lexikon/saldo/saldo.xml

#saldoe
svn export --force https://svn.spraakdata.gu.se/sb-arkiv/lexikon/saldoexempel/saldoexempel.txt
cat saldoexempel.txt saldoexempel_swefn.txt | python skript/remove_example_doubles.py | awk 'BEGIN {FS = "\t"}{if ($6 != "*") print $1 "\t" $6}' | ./skript/_xml.py saldoe saldo example > lexikon/saldoe/saldoe.xml

#swesaurus
svn export --force https://svn.spraakdata.gu.se/sb/fnplusplus/swesaurus/swesaurus.txt
svn export --force https://svn.spraakdata.gu.se/sb/fnplusplus/swesaurus/swesaurus-derived.txt
svn export --force https://svn.spraakdata.gu.se/sb/fnplusplus/wiktionary/swesaurus_wiktionary.txt
svn export --force https://svn.spraakdata.gu.se/sb/fnplusplus/swesaurus/wordnet3/wordnet-saldo-relations.txt
cat swesaurus.txt swesaurus-derived.txt swesaurus_wiktionary.txt wordnet-saldo-relations.txt | ./skript/_xml.py swesaurus saldo saldo type degree source > lexikon/swesaurus/swesaurus.xml

#swefn  
svn export --force https://svn.spraakdata.gu.se/sb/fnplusplus/pub/swefn-db.csv
sed '1d' swefn-db.csv | runhaskell -i./skript ./skript/ShortXML.hs > lexikon/swefn/swefn.xml

#saldom
$SALDO $SALDODICT -p flat_xml > m.xml
cat m.xml | runhaskell -i./skript ./skript/Saldo_senses.hs > lexikon/saldom/saldom.xml

#dalin morphology
svn export --force https://svn.spraakdata.gu.se/repos/sblex/dalin/trunk/material/dalin_saldo.txt
svn export --force https://svn.spraakdata.gu.se/repos/sblex/dalin/trunk/material/dalin_suppl_saldo.txt
cat dalin_saldo.txt dalin_suppl_saldo.txt > dalin.txt
$DALIN $DALINDICT -p flat_xml > d.xml
cat d.xml | runhaskell -i./skript ./skript/Dalin_senses.hs > lexikon/dalin/dalin.xml

#fsv morphology
svn export --force https://svn.spraakdata.gu.se/repos/sblex/fsv/trunk/fsv.txt
$FSV $FSVDICT -p flat_xml > d.xml
cat d.xml | runhaskell -i./skript ./skript/FSV_senses.hs > lexikon/fsv/fsv.xml

#check-in
svn ci -m "automatic lexicon update" lexikon