prepare (1290B)
1 #!/bin/bash 2 # Check common misspellings 3 # input file format: 4 # word->word1, ... 5 # Source: http://en.wikipedia.org/wiki/Wikipedia:Lists_of_common_misspellings/For_machines 6 7 hunspell=../../src/tools/hunspell 8 hlang=${HUNSPELL:-en_US} 9 alang=${ASPELL:-en_US} 10 input=${INPUT:-List_of_common_misspellings.txt} 11 12 # remove bad words recognised by Hunspell as good 13 cat $input | sed 's/[-]>/ /' | $hunspell -d $hlang -1 -L | 14 15 # remove items with dash for Aspell 16 grep '^[^-]* ' | 17 18 # remove spaces from end of lines 19 sed 's/ *$//' >$input.1 20 21 # remove bad words recognised by Aspell as good 22 cut -f 1 -d ' ' $input.1 | aspell -l $alang --list | 23 awk 'FILENAME=="-"{a[$1]=1;next}a[$1]{print$0}' - $input.1 | 24 25 # change commas with tabs 26 sed 's/, */ /g' >$input.2 27 28 # remove lines with unrecognised suggestions (except suggestion with spaces) 29 cut -d ' ' -f 2- $input.2 | tr "\t" "\n" | grep -v ' ' >x.1 30 cat x.1 | $hunspell -l -d $hlang >x.2 31 cat x.1 | aspell -l $alang --list >>x.2 32 cat x.2 | awk 'BEGIN{FS="\t"} 33 FILENAME=="-"{a[$1]=1;next}a[$2]!=1 && a[$3]!=1{print $0}' - $input.2 >$input.3 34 35 cut -f 1 -d ' ' $input.3 | aspell -l $alang -a | grep -v ^$ | sed -n '2,$p' | 36 sed 's/^.*: //;s/, / /g' >$input.4 37 38 cat $input.3 | $hunspell -d $hlang -a -1 | grep -v ^$ | sed -n '2,$p' | 39 sed 's/^.*: //;s/, / /g' >$input.5 40