new-words

annotate new-words.sh @ 2:68722cd6faff

find changesets by author, revision, files, or words in the commit message
Добавление пометок к словам.

Справа от слова можно поставить пометку,
которая попадает в файл ~/.new-words/notes-LANG.txt.
Потом, при появлении этого слова,
она вспоминается и выводится вместе с ним.
author igor@book.xt.vpn
date Fri Feb 26 21:41:27 2010 +0200 (2010-02-26)
parents 4c9076f87241
children c703b8898696
rev   line source
igor@0 1 #!/bin/sh
igor@0 2
igor@0 3 cat <<HELP > /dev/null
igor@0 4
igor@0 5 Поддержка нескольких языков:
igor@0 6
igor@0 7 new-words -l lang URL
igor@0 8
igor@0 9 Например, для немецких текстов:
igor@0 10
igor@0 11 new-words -l de URL
igor@0 12
igor@0 13 Или, предварительно создав соответствующую ссылку:
igor@0 14
igor@0 15 de-words URL
igor@0 16
igor@0 17 HELP
igor@0 18
igor@0 19 WORK_DIR=~/.new-words/
igor@0 20 TEMP1=`mktemp /tmp/news-words-XXXXXXXXXX`
igor@0 21 TEMP2=`mktemp /tmp/news-words-XXXXXXXXXX`
igor@0 22 editor=${EDITOR:-vim}
igor@0 23
igor@2 24 LANGUAGE=en
igor@2 25 my_name="`echo $0 | sed s@.*/@@ | sed s/-.*// `"
igor@2 26 [ "${my_name}" = "new" ] || LANGUAGE="$my_name"
igor@0 27 if [ "$1" = "-l" ]
igor@0 28 then
igor@0 29 LANGUAGE="$2"
igor@0 30 VOCABULARY="$LANGUAGE".txt
igor@0 31 shift 2
igor@0 32 fi
igor@2 33 VOCABULARY=${LANGUAGE}.txt
igor@2 34 NOTES_FILE=notes-${LANGUAGE}.txt
igor@0 35
igor@0 36 get_words()
igor@0 37 {
igor@1 38 tr ' ' '\n' | sed 's/--/ /g' \
igor@0 39 | tr -d '*\r,.-:#@()+=—<>$;"?!|·[]^%&'"'" \
igor@0 40 | tr ' ' '\n' | grep_v_english_perl \
igor@1 41 | grep -x '[[:alnum:]]*' \
igor@0 42 | sort | uniq -c | awk '{if ($2!="") print;}' | sort -rn
igor@0 43 }
igor@0 44
igor@0 45 grep_v_english()
igor@0 46 {
igor@0 47 [ -e "$VOCABULARY" ] || touch "$VOCABULARY"
igor@0 48 eval $(cat $VOCABULARY | tr -d "'" | xargs -n10 echo | tr ' ' '|' | sed 's/^/egrep -xv "RRRRRRR|/' | sed 's/$/"/' | tr '\n' '|')cat
igor@0 49 }
igor@0 50
igor@0 51 grep_v_english_perl()
igor@0 52 {
igor@0 53 PERL_SCRIPT_TEMP_NAME=`mktemp /tmp/perl-grep-v-english-XXXXXXXX`
igor@0 54 cat <<'PERL_SCRIPT' > $PERL_SCRIPT_TEMP_NAME
igor@0 55 open(VOC, $ENV{VOCABULARY})
igor@0 56 or die "Can't open VOCABULARY";
igor@0 57 while (<VOC>){
igor@0 58 chomp;
igor@0 59 s/'//g;
igor@0 60 $voc{$_}="1";
igor@0 61 }
igor@0 62 while(<>) {
igor@0 63 chomp;
igor@0 64 if (not defined($voc{$_})) { print "$_\n"; }
igor@0 65 }
igor@0 66 PERL_SCRIPT
igor@0 67 [ -e "$VOCABULARY" ] || touch "$VOCABULARY"
igor@0 68 export VOCABULARY
igor@0 69 perl $PERL_SCRIPT_TEMP_NAME
igor@0 70 rm $PERL_SCRIPT_TEMP_NAME
igor@0 71 }
igor@0 72
igor@0 73 text_from_url()
igor@0 74 {
igor@0 75 lynx -dump "$1" | perl -p -e 's@http://[a-zA-Z&_.:/0-9%?=,#+()\[\]~-]*@@'
igor@0 76 }
igor@0 77
igor@2 78 add_marks()
igor@2 79 {
igor@2 80 PERL_SCRIPT_TEMP_NAME=`mktemp /tmp/perl-grep-v-english-XXXXXXXX`
igor@2 81 cat <<'PERL_SCRIPT' > $PERL_SCRIPT_TEMP_NAME
igor@2 82 $file = $ARGV[0];
igor@2 83 our $dict;
igor@2 84 if (open(NOTES, $ENV{NOTES_FILE})) {
igor@2 85 while(<NOTES>) {
igor@2 86 chomp;
igor@2 87 s/^\s+//;
igor@2 88 my ($a,$b)=split /\s+/,$_,2;
igor@2 89 $dict{$a}=$b;
igor@2 90 }
igor@2 91 }
igor@2 92 if (open(F, $file)) {
igor@2 93 @lines=<F>;
igor@2 94 close(F);
igor@2 95
igor@2 96 if (open(F, ">$file")) {
igor@2 97 for (@lines) {
igor@2 98 m/\s+\S+\s+(\S+)/;
igor@2 99 $name=$1;
igor@2 100 if (defined($dict{$name})) {
igor@2 101 chomp;
igor@2 102 $mark=$dict{$name};
igor@2 103 $space=" "x(30-length($_));
igor@2 104 print F "$_$space$mark\n";
igor@2 105 }
igor@2 106 else {
igor@2 107 print F "$_";
igor@2 108 }
igor@2 109 }
igor@2 110 close(F);
igor@2 111 }
igor@2 112 }
igor@2 113 PERL_SCRIPT
igor@2 114 [ -e "$NOTES_FILE" ] || touch "$NOTES_FILE"
igor@2 115 export NOTES_FILE
igor@2 116 perl $PERL_SCRIPT_TEMP_NAME "$1"
igor@2 117 rm $PERL_SCRIPT_TEMP_NAME
igor@2 118 }
igor@2 119
igor@2 120 remove_marks()
igor@2 121 {
igor@2 122 PERL_SCRIPT_TEMP_NAME=`mktemp /tmp/perl-grep-v-english-XXXXXXXX`
igor@2 123 cat <<'PERL_SCRIPT' > $PERL_SCRIPT_TEMP_NAME
igor@2 124 $file = $ARGV[0];
igor@2 125 our %dict;
igor@2 126 if (open(F, $file)) {
igor@2 127 @lines=<F>;
igor@2 128 close(F);
igor@2 129
igor@2 130 if (open(F, ">$file")) {
igor@2 131 for (@lines) {
igor@2 132 chomp;
igor@2 133 if (m/(\s+)(\S+)(\s+)(\S+)(\s+)(.*)/) {
igor@2 134 my $name=$4;
igor@2 135 my $comment=$6;
igor@2 136 $dict{$name}=$comment;
igor@2 137 print F "$1$2$3$4\n";
igor@2 138 }
igor@2 139 else {
igor@2 140 print F "$_\n";
igor@2 141 }
igor@2 142 }
igor@2 143 }
igor@2 144 }
igor@2 145 if (open(NOTES, $ENV{NOTES_FILE})) {
igor@2 146 @lines=<NOTES>;
igor@2 147 close(NOTES);
igor@2 148
igor@2 149 if (open(NOTES, ">".$ENV{NOTES_FILE})) {
igor@2 150 for (@lines) {
igor@2 151 chomp;
igor@2 152 s/^\s+//;
igor@2 153 my ($a,$b)=split /\s+/,$_,2;
igor@2 154 if (not defined($dict{$a}) || ($dict{$a} eq $b)) {
igor@2 155 print NOTES "$_\n";
igor@2 156 if (defined($dict{$a})) { unset($dict{$a}); }
igor@2 157 }
igor@2 158 }
igor@2 159 for (keys %dict) {
igor@2 160 $mark=$dict{$_};
igor@2 161 $space=" "x(30-length($_));
igor@2 162 print NOTES "$_$space$mark\n";
igor@2 163 }
igor@2 164 }
igor@2 165 }
igor@2 166 PERL_SCRIPT
igor@2 167 [ -e "$NOTES_FILE" ] || touch "$NOTES_FILE"
igor@2 168 export NOTES_FILE
igor@2 169 perl $PERL_SCRIPT_TEMP_NAME "$1"
igor@2 170 rm $PERL_SCRIPT_TEMP_NAME
igor@2 171 }
igor@2 172
igor@0 173 mkdir -p $WORK_DIR
igor@0 174 cd $WORK_DIR
igor@0 175 if echo "$1" | grep -q http:
igor@0 176 then
igor@0 177 text_from_url "$1" | get_words | tee "$TEMP1" > "$TEMP2"
igor@0 178 elif [ "$#" != 0 ]
igor@0 179 then
igor@0 180 cat "$1" | get_words | tee "$TEMP1" > "$TEMP2"
igor@0 181 else
igor@0 182 get_words | tee "$TEMP1" > "$TEMP2"
igor@0 183 fi
igor@0 184
igor@2 185 add_marks "$TEMP2"
igor@0 186 if [ "$editor" = vim ]
igor@0 187 then
igor@0 188 vim -c 'set keywordprg='"$LANGUAGE" -c 'set iskeyword=@,48-57,/,.,-,_,+,,,#,$,%,~,=' "$TEMP2" < /dev/tty > /dev/tty
igor@0 189 else
igor@0 190 echo 2
igor@0 191 $editor "$TEMP2"
igor@0 192 fi
igor@2 193 remove_marks "$TEMP2"
igor@2 194
igor@0 195 diff "$TEMP1" "$TEMP2" | awk '{print $3}' | sort -u >> "$VOCABULARY"
igor@0 196 rm -f "$TEMP1" "$TEMP2"