new-words: c3a50c0d2400 new-words.sh

new-words

view new-words.sh @ 40:c3a50c0d2400

Functions for adding/removing notes + statistics now implemented in Python.

Option -O (old-style) is not supported anymore. If you need old-style new-words use new-words.sh

author	Igor Chubin <igor@chub.in>
date	Sun Jan 23 17:09:44 2011 +0100 (2011-01-23)
parents	3827cce83602
children	4e931db74618

line source

1 #!/bin/bash

3 show_usage()

4 {

5 cat <<HELP > /dev/stderr

7 USAGE:

9 new-words [ -l lang ] [ -s ] [ ARG ]

11 SWITCHES:

13 -h print this screen

14 -c show compressed wordlist: one word per group

15 -k put higher words that are similar to the known words (only for English)

16 -l lang override language settings

17 -n non-interactive mode (don't run vi)

18 -N turn off known words filtering

19 -a don't add marks (and don't save marks added by user)

20 -p pages work with specified pages only (pages = start-stop/total )

21 -s show the text statistics (percentage of known words and so on) and exit

22 -S show your vocabulary statistics (number of words and word groups)

23 -t tag tag known words with tag

24 -T show list of active tags

25 -m tag merge the words tagged with "tag" into the main vocabulary

26 -M merge the words tagged with any tag into the main vocabulary

27 -r tag remove subvocabulary for the "tag"

28 -2 -3 find 2 and 3 words' sequences

30 The language of the text can be specified also

31 by name of the program new-words (correspondent link must be created before).

32 For example, these calls are equivalent:

34 de-words URL

35 new-words -l de URL

37 HELP

38 }

40 if [ "$1" = "-h" ]

41 then

42 show_usage

43 exit 0

44 fi

46 WORK_DIR=~/.new-words/

47 TEMP1=`mktemp /tmp/new-words-temp1.XXXXXXXXXX`

48 TEMP2=`mktemp /tmp/new-words-temp2.XXXXXXXXXX`

49 export ORIGINAL_TEXT=`mktemp /tmp/new-words-orig.XXXXXXXXXX`

50 editor=${EDITOR:-vim}

52 # language detection

54 LANGUAGE=en

55 my_name="`echo $0 | sed s@.*/@@ | sed s/-.*// `"

56 for arg

57 do

58 if echo "$arg" | grep -q http://...wikipedia.org/wiki/

59 then

60 LANGUAGE="`echo $arg | sed s@http://@@ | sed s@.wikipedia.*@@`"

61 fi

62 done

63 [ "${my_name}" = "new" ] || LANGUAGE="$my_name"

65 #----------------------------------------------------

66 # command line options processing

68 STAT_ONLY=NO

69 NEED_TO_USE_VOCABULARY_WHEN_SORT=NO

70 DONT_ADD_MARKS=NO

71 NON_INTERACTIVE_MODE=NO

72 PART_TO_PROCESS=''

73 GROUP_WORDS_BY_THREE=NO

74 GROUP_WORDS_BY_TWO=NO

75 TAG_NAME=''

76 MERGE_THIS_TAGS=''

77 TAGS_LIST_ONLY=NO

78 MERGE_TAGGED_WORDS=NO

79 MERGE_ALL_TAGGED=NO

80 DONT_ADD_MARKLINES=NO

81 FILTER_WORDS=YES

82 SHOW_VOC_STAT=NO

83 COMPRESSED_WORDLIST=NO

84 while getopts cl:sSkanNp:t:Tm:Mr:23 opt

85 do

86 case "$opt" in

87 c) COMPRESSED_WORDLIST=YES;;

88 s) STAT_ONLY=YES;;

89 S) SHOW_VOC_STAT=YES;;

90 k) NEED_TO_USE_VOCABULARY_WHEN_SORT=YES;;

91 l) LANGUAGE="$OPTARG";;

92 a) DONT_ADD_MARKS=YES;;

93 n) NON_INTERACTIVE_MODE=YES;;

94 N) FILTER_WORDS=NO;;

95 p) PART_TO_PROCESS="$OPTARG";;

96 t) TAG_NAME="$OPTARG";;

97 T) TAGS_LIST_ONLY="YES";;

98 m) DONT_ADD_MARKLINES="YES"; MERGE_TAGGED_WORDS="YES"; MERGE_THIS_TAGS="$TAG_NAME $OPTARG";;

99 M) DONT_ADD_MARKLINES="YES"; MERGE_ALL_TAGGED="YES";;

100 r) REMOVE_TAG="YES"; TAG_NAME="$TAG_NAME $OPTARG";;

101 2) GROUP_WORDS_BY_TWO=YES;;

102 3) GROUP_WORDS_BY_THREE=YES;;

103 \?) # unknown flag

104 show_usage

105 exit 1;;

106 esac

107 done

108 shift `expr $OPTIND - 1`

109

110 if [ "$1" = "-l" ]

111 then

112 LANGUAGE="$2"

113 shift 2

114 fi

115

116 VOCABULARY=${LANGUAGE}.txt

117 NOTES_FILE=notes-${LANGUAGE}.txt

118

119 if [ "${SHOW_VOC_STAT}" = "YES" ]

120 then

121 $0 -l "${LANGUAGE}" -N -n ${WORK_DIR}/${VOCABULARY} | head -1 | awk '{print $5}' | tr -d "<>"

122 exit 0

123 fi

124

125 #----------------------------------------------------

126

127 get_words()

128 {

129 export FILTER_WORDS

130 tr ' ' '\n' | sed 's/--/ /g' \

131 | sed "s/'/__APOSTROPHE__/g" \

132 | perl -MEncode -Mutf8 -n -e '$_ = decode( "utf8", $_);y/*\r,.:#@()+=—<>$;"?!|·[]^%&/ /; binmode STDOUT, ":utf8"; print if /^[[:alpha:] '"'"'_-]*$/'\

133 | sed "s/__APOSTROPHE__/'/g" \

134 | tr ' ' '\n' \

135 | tee "$1" \

136 | grep_v_english_perl \

137 | sort | uniq -c | awk '{if ($2!="") print;}' | sort -rn

138 }

139

140 add_stat()

141 {

142 if [ "$DONT_ADD_MARKLINES" = "YES" ]

143 then

144 cat

145 return

146 fi

147 before="$1"

148 after=${before}2

149 cat > "$after"

150 total="`wc -w $1 | awk '{print $1}'`"

151 total_unknown="`cat $after|awk '{s=s+$1}END{print s}'`"

152 total_known="`echo $total-$total_unknown|bc`"

153 percentage="`echo '100*('$total-$total_unknown')'/$total | bc -l | sed 's/\\.$.$.*/.\1/'`"

154 #sentences="`cat $after | perl -e 'local $/; $_=<>; s@http://[a-zA-Z&_.:/0-9%?=,\#+()\[\]~-]*@@g; s@\n@@g; s@(Mr|Mrs)\.@\1POINT@g; @sentences=split /\\./;print $#sentences;'`"

155 sentences="`cat $ORIGINAL_TEXT | perl -e 'local $/; $_=<>; s/[^.]//msg; print length($_);'`"

156

157

158 if [ "$STAT_ONLY" = "YES" ]

159 then

160 echo "LANG KNOWN% UNKNOWN% KNOWN TOTAL WPS UWPS*10"

161 echo "$LANGUAGE $percentage `echo $100-$percentage$ | bc -l` $total_known $total `echo $total/$sentences|bc` `echo 10*$total_unknown/$sentences|bc` "

162 rm $after

163 return 0

164 else

165 groups="`echo $(grep '# groups' $after | awk '{print $3}')`"

166 words="`echo $(grep -v '^#' $after | wc -l)`"

167 echo "# $LANGUAGE, $percentage, <$total_known/$total>, <$groups/$words>"

168 fi

169

170 PERL_SCRIPT_TEMP_NAME=`mktemp /tmp/perl-grep-v-english-XXXXXXXX`

171 cat <<'PERL_SCRIPT' > $PERL_SCRIPT_TEMP_NAME

172 my $total=shift(@ARGV);

173 my $total_known=shift(@ARGV);