| 1 |
#!/usr/bin/env bash |
| 2 |
|
| 3 |
ASPELL=aspell |
| 4 |
HUNSPELL=hunspell |
| 5 |
: ${SCOWL:=..} |
| 6 |
SPELLER="$SCOWL/speller" |
| 7 |
: ${UNIX2DOS:=unix2dos} |
| 8 |
|
| 9 |
set -e |
| 10 |
|
| 11 |
export LANG=C |
| 12 |
export LC_ALL=C |
| 13 |
export LC_CTYPE=C |
| 14 |
export LC_COLLATE=C |
| 15 |
|
| 16 |
SIZE=60 |
| 17 |
|
| 18 |
mk-list() { $SCOWL/mk-list -d $SCOWL/final "$@"; } |
| 19 |
|
| 20 |
prep() { |
| 21 |
echo prep |
| 22 |
|
| 23 |
cat $SCOWL/misc/{offensive.1,offensive.2,profane.1} | sort -u > nosug |
| 24 |
} |
| 25 |
|
| 26 |
doit() { |
| 27 |
echo creating $1.dic |
| 28 |
|
| 29 |
eval $2 | sort -u > $1.0 |
| 30 |
|
| 31 |
comm -12 $1.0 nosug > $1-nosug.1 |
| 32 |
|
| 33 |
comm -23 $1.0 nosug > $1.1 |
| 34 |
|
| 35 |
$SPELLER/munch-list munch $SPELLER/en.aff < $1-nosug.1 | $SPELLER/add-no-suggest > $1.2 |
| 36 |
|
| 37 |
$SPELLER/munch-list munch $SPELLER/en.aff < $1.1 >> $1.2 |
| 38 |
|
| 39 |
cat $SPELLER/en.dic.supp >> $1.2 |
| 40 |
|
| 41 |
wc -l < $1.2 | tr -d '[:blank:]' > $1.dic |
| 42 |
cat $1.2 | sort | iconv -f iso-8859-1 -t utf-8 >> $1.dic |
| 43 |
|
| 44 |
cp $SPELLER/en.aff $1.aff |
| 45 |
|
| 46 |
if [ "$SCOWL_VERSION" ]; then |
| 47 |
fn="$1-$SCOWL_VERSION" |
| 48 |
else |
| 49 |
fn="$1" |
| 50 |
fi |
| 51 |
|
| 52 |
WHAT="$1 Hunspell Dictionary" sh $SPELLER/README_en.txt.sh > README_$1.txt |
| 53 |
if [ -z "$3" ]; then |
| 54 |
echo "Wordlist Command: $2" >> README_$1.txt |
| 55 |
else |
| 56 |
cat $3 >> README_$1.txt |
| 57 |
fi |
| 58 |
|
| 59 |
rm -f hunspell-$fn.zip |
| 60 |
zip -9 hunspell-$fn.zip README_$1.txt $1.dic $1.aff |
| 61 |
|
| 62 |
if [ -z "$3" ]; then |
| 63 |
mkdir -p hunspell |
| 64 |
cp hunspell-$fn.zip hunspell/ |
| 65 |
|
| 66 |
#echo check |
| 67 |
|
| 68 |
cat $1-nosug.1 $1.1 | sort -u > $1.tocheck |
| 69 |
|
| 70 |
#hunspell -l -d ./$1 < $1.dic.tocheck > misspelled |
| 71 |
|
| 72 |
cat $1.tocheck | iconv -f iso-8859-1 -t utf-8 | $UNIX2DOS > $1.txt |
| 73 |
cat <<EOF > README.txt |
| 74 |
This zip file contains the words found in the corresponding Hunspell |
| 75 |
dictionary. See the file README_$1.txt. |
| 76 |
EOF |
| 77 |
zip -9 hunspell/wordlist-$fn.zip README.txt README_$1.txt $1.txt |
| 78 |
fi |
| 79 |
} |
| 80 |
|
| 81 |
prep |
| 82 |
|
| 83 |
if [ "$1" = "-all" ] |
| 84 |
then |
| 85 |
|
| 86 |
doit en_US "mk-list --accents=strip en_US $SIZE" |
| 87 |
doit en_CA "mk-list --accents=strip en_CA $SIZE" |
| 88 |
doit en_GB-ize "mk-list --accents=strip en_GB-ize $SIZE" |
| 89 |
doit en_GB-ise "mk-list --accents=strip en_GB-ise $SIZE" |
| 90 |
doit en_AU "mk-list --accents=strip en_AU $SIZE" |
| 91 |
|
| 92 |
doit en_US-large "mk-list -v1 --accents=both en_US 70" |
| 93 |
doit en_CA-large "mk-list -v1 --accents=both en_CA 70" |
| 94 |
doit en_GB-large "mk-list -v1 --accents=both en_GB-ize en_GB-ise 70" |
| 95 |
doit en_AU-large "mk-list -v1 --accents=both en_AU 70" |
| 96 |
|
| 97 |
sh $SPELLER/README_en.txt.sh > hunspell/README |
| 98 |
|
| 99 |
elif [ "$1" = "-one" -a -n "$2" -a -n "$3" ] |
| 100 |
then |
| 101 |
|
| 102 |
doit $2 "cat" $3 |
| 103 |
|
| 104 |
else |
| 105 |
|
| 106 |
echo "usage: $0 -all | -one <dict-name> <parms file>" |
| 107 |
|
| 108 |
fi |
| 109 |
|
| 110 |
#rm eng*.dat nosug en_US*.? en_CA*.? |