init

Telegram/ThirdParty/hunspell/tests/suggestiontest/.gitignore

@@ -0,0 +1,4 @@
+result.aspell
+result.hunspell
+List_of_common_misspellings.txt.*
+x.*

Telegram/ThirdParty/hunspell/tests/suggestiontest/Makefile.am

@@ -0,0 +1,6 @@
+EXTRA_DIST= \
+List_of_common_misspellings.txt \
+Makefile.orig \
+prepare \
+README \
+test 

Telegram/ThirdParty/hunspell/tests/suggestiontest/Makefile.orig

@@ -0,0 +1,11 @@
+all:
+	./prepare
+	./test
+
+single:
+	./prepare2
+	./test
+
+clean:
+	rm *.[1-5] result.*
+

Telegram/ThirdParty/hunspell/tests/suggestiontest/README

@@ -0,0 +1,16 @@
+source of text data: Wikipedia
+http://en.wikipedia.org/wiki/Wikipedia:Lists_of_common_misspellings/For_machines
+
+For testing Hunspell you need the extended en_US dictionary with phonetic table:
+http://hunspell.sourceforge.net/en_US.zip
+
+test:
+make -f Makefile.orig
+
+test only with Hunspell:
+
+make -f Makefile.orig single
+
+test with different input file and dictionaries:
+
+INPUT=dutchlist.txt HUNSPELL=nl_NL ASPELL=nl make -f Makefile.orig

Telegram/ThirdParty/hunspell/tests/suggestiontest/prepare

@@ -0,0 +1,40 @@
+#!/bin/bash
+# Check common misspellings
+# input file format:
+# word->word1, ...
+# Source: http://en.wikipedia.org/wiki/Wikipedia:Lists_of_common_misspellings/For_machines
+
+hunspell=../../src/tools/hunspell
+hlang=${HUNSPELL:-en_US}
+alang=${ASPELL:-en_US}
+input=${INPUT:-List_of_common_misspellings.txt}
+
+# remove bad words recognised by Hunspell as good
+cat $input | sed 's/[-]>/	/' | $hunspell -d $hlang -1 -L |
+
+# remove items with dash for Aspell
+grep '^[^-]*	' |
+
+# remove spaces from end of lines
+sed 's/ *$//' >$input.1
+
+# remove bad words recognised by Aspell as good
+cut -f 1 -d '	' $input.1 | aspell -l $alang --list |
+awk 'FILENAME=="-"{a[$1]=1;next}a[$1]{print$0}' - $input.1 |
+
+# change commas with tabs
+sed 's/, */	/g' >$input.2
+
+# remove lines with unrecognised suggestions (except suggestion with spaces)
+cut -d '	' -f 2- $input.2 | tr "\t" "\n" | grep -v ' ' >x.1
+cat x.1 | $hunspell -l -d $hlang >x.2
+cat x.1 | aspell -l $alang --list >>x.2
+cat x.2 | awk 'BEGIN{FS="\t"}
+FILENAME=="-"{a[$1]=1;next}a[$2]!=1 && a[$3]!=1{print $0}' - $input.2 >$input.3
+
+cut -f 1 -d '	' $input.3 | aspell -l $alang -a | grep -v ^$ | sed -n '2,$p' |
+sed 's/^.*: //;s/, /	/g' >$input.4
+
+cat $input.3 | $hunspell -d $hlang -a -1 | grep -v ^$ | sed -n '2,$p' |
+sed 's/^.*: //;s/, /	/g' >$input.5
+

Telegram/ThirdParty/hunspell/tests/suggestiontest/prepare2

@@ -0,0 +1,30 @@
+#!/bin/bash
+# Check common misspellings
+# input file format:
+# word->word1, ...
+# Source: http://en.wikipedia.org/wiki/Wikipedia:Lists_of_common_misspellings/For_machines
+
+hunspell=../../src/tools/hunspell
+hlang=${HUNSPELL:-en_US}
+input=${INPUT:-List_of_common_misspellings.txt}
+
+# remove bad words recognised by Hunspell as good
+cat $input | sed 's/[-]>/	/' | $hunspell -d $hlang -1 -L |
+
+# remove spaces from end of lines
+sed 's/ *$//' >$input.1
+
+# change commas with tabs
+cat $input.1 | sed 's/, */	/g' >$input.2
+
+# remove lines with unrecognised suggestions (except suggestion with spaces)
+cut -d '	' -f 2- $input.2 | tr "\t" "\n" | grep -v ' ' >x.1
+cat x.1 | $hunspell -l -d $hlang >x.2
+cat x.2 | awk 'BEGIN{FS="\t"}
+FILENAME=="-"{a[$1]=1;next}a[$2]!=1 && a[$3]!=1{print $0}' - $input.2 >$input.3
+
+test -f $input.4 && rm $input.4
+
+cat $input.3 | $hunspell -d $hlang -a -1 | grep -v ^$ | sed -n '2,$p' |
+sed 's/^.*: //;s/, /	/g' >$input.5
+

Telegram/ThirdParty/hunspell/tests/suggestiontest/test

@@ -0,0 +1,25 @@
+#!/bin/bash
+# Check common misspellings
+# input file format:
+# word->word1, ...
+# Source: http://en.wikipedia.org/wiki/Wikipedia:Lists_of_common_misspellings/For_machines
+
+input=${INPUT:-List_of_common_misspellings.txt}
+
+function check() {
+cat $1 | awk 'BEGIN{maxord=0;FS="\t"}FILENAME=="-"{for (i=1; i<=NF; i++){a[NR,$(i)]=i};max=NR;next}{x1=a[NR-max,$2];x2=a[NR-max,$3];sug++;if($3)sug++;if (!x1&&!x2){mis2++;misrow=misrow"\n"$0};if(!x1||($3 && !x2))mis++;ord+=x1+x2;}END{
+print "Missed rows", misrow;
+print "======================================="
+print maxord, "max. suggestion for a word";
+print max, "input rows";
+print mis2, "missing rows";
+print sug, "expected suggestions";
+print mis, "missing suggestions";
+print ord/(sug-mis), "average ranking";
+}' - $2
+}
+
+test -f $input.4 && check $input.4 $input.3 >result.aspell
+check $input.5 $input.3 >result.hunspell
+test -f result.aspell && tail -6 result.aspell
+tail -6 result.hunspell