igor@0: #!/bin/sh
igor@0: cat <<HELP > /dev/null
igor@0: 
igor@0: Поддержка нескольких языков:
igor@0: 
igor@0:     new-words -l lang URL
igor@0: 
igor@0: Например, для немецких текстов:
igor@0: 
igor@0:     new-words -l de URL
igor@0: 
igor@0: Или, предварительно создав соответствующую ссылку:
igor@0: 
igor@0:     de-words URL
igor@0: 
igor@0: HELP
igor@0: 
igor@0: WORK_DIR=~/.new-words/
igor@0: TEMP1=`mktemp /tmp/news-words-XXXXXXXXXX`
igor@0: TEMP2=`mktemp /tmp/news-words-XXXXXXXXXX`
igor@0: editor=${EDITOR:-vim}
igor@0: 
igor@2: LANGUAGE=en
igor@2: my_name="`echo $0 | sed s@.*/@@ | sed s/-.*// `"
igor@3: if echo "$1" | grep -q http://...wikipedia.org/wiki/
igor@3: then
igor@3: LANGUAGE="`echo $1 | sed s@http://@@ | sed s@.wikipedia.*@@`"
igor@3: fi
igor@2: [ "${my_name}" = "new" ] || LANGUAGE="$my_name"
igor@0: if [ "$1" = "-l" ]
igor@0: then
igor@0:     LANGUAGE="$2"
igor@0:     VOCABULARY="$LANGUAGE".txt
igor@0:     shift 2
igor@0: fi
igor@2: VOCABULARY=${LANGUAGE}.txt
igor@2: NOTES_FILE=notes-${LANGUAGE}.txt
igor@0: 
igor@0: get_words()
igor@0: {
igor@1: tr ' ' '\n' | sed 's/--/ /g' \
igor@3: | sed "s/'/__APOSTROPHE__/g" \
igor@3: | tr  '—·-' '-----' \
igor@3: | tr '*\r,.:#@()+=—<>$;"?!|·[]^%&' '                           ' \
igor@3: | tr ' ' '\n' \
igor@3: | grep -x '[[:alpha:]'"'"'-]*' \
igor@3: | tee "$1" \
igor@3: | grep_v_english_perl \
igor@0: | sort | uniq -c | awk '{if ($2!="") print;}' | sort -rn
igor@0: }
igor@0: 
igor@3: add_stat()
igor@3: {
igor@3:     before="$1"
igor@3:     after=${before}2
igor@3:     cat > "$after"
igor@3:     total="`wc -w $1 | awk '{print $1}'`"
igor@3:     total_unknown="`cat $after|awk '{s=s+$1}END{print s}'`"
igor@3:     total_known="`echo $total-$total_unknown|bc`"
igor@3:     percentage="`echo '100*('$total-$total_unknown')'/$total | bc -l | sed 's/\\.\(.\).*/.\1/'`"
igor@3:     echo "# $LANGUAGE, $percentage, <$total_known/$total>"
igor@3:     cat "$after"
igor@3:     rm $after
igor@3: }
igor@3: 
igor@0: grep_v_english()
igor@0: {
igor@0: [ -e "$VOCABULARY" ] || touch "$VOCABULARY"
igor@0: eval $(cat $VOCABULARY | tr -d "'" | xargs -n10 echo | tr ' ' '|' | sed 's/^/egrep -xv "RRRRRRR|/' | sed 's/$/"/' | tr '\n' '|')cat
igor@0: }
igor@0: 
igor@0: grep_v_english_perl()
igor@0: {
igor@0:     PERL_SCRIPT_TEMP_NAME=`mktemp /tmp/perl-grep-v-english-XXXXXXXX`
igor@0:     cat <<'PERL_SCRIPT' > $PERL_SCRIPT_TEMP_NAME
igor@0: open(VOC, $ENV{VOCABULARY})
igor@0:  or die "Can't open VOCABULARY";
igor@0: while (<VOC>){
igor@0:     chomp;
igor@3:     #s/'//g;
igor@0:     $voc{$_}="1";
igor@0: }
igor@0: while(<>) {
igor@0:     chomp;
igor@0:     if (not defined($voc{$_})) { print "$_\n"; }
igor@0: }
igor@0: PERL_SCRIPT
igor@0:     [ -e "$VOCABULARY" ] || touch "$VOCABULARY"
igor@0:     export VOCABULARY 
igor@0:     perl $PERL_SCRIPT_TEMP_NAME
igor@0:     rm $PERL_SCRIPT_TEMP_NAME
igor@0: }
igor@0: 
igor@0: text_from_url()
igor@0: {
igor@0: lynx -dump "$1" | perl -p -e 's@http://[a-zA-Z&_.:/0-9%?=,#+()\[\]~-]*@@'
igor@0: }
igor@0: 
igor@2: add_marks()
igor@2: {
igor@2:     PERL_SCRIPT_TEMP_NAME=`mktemp /tmp/perl-grep-v-english-XXXXXXXX`
igor@2:     cat <<'PERL_SCRIPT' > $PERL_SCRIPT_TEMP_NAME
igor@2: $file = $ARGV[0];
igor@2: our $dict;
igor@2: if (open(NOTES, $ENV{NOTES_FILE})) {
igor@2:     while(<NOTES>) {
igor@2:         chomp;
igor@2:         s/^\s+//;
igor@2:         my ($a,$b)=split /\s+/,$_,2;
igor@2:         $dict{$a}=$b;
igor@2:     }
igor@2: }
igor@2: if (open(F, $file)) {
igor@2:     @lines=<F>;
igor@2:     close(F);
igor@2: 
igor@2:     if (open(F, ">$file")) {
igor@2:         for (@lines) {
igor@2:             m/\s+\S+\s+(\S+)/;
igor@2:             $name=$1;
igor@3:             if (not /^#/ and defined($dict{$name})) {
igor@2:                 chomp;
igor@2:                 $mark=$dict{$name};
igor@2:                 $space=" "x(30-length($_));
igor@2:                 print F "$_$space$mark\n";
igor@2:             }
igor@2:             else {
igor@2:                 print F "$_";
igor@2:             }
igor@2:         }
igor@2:         close(F);
igor@2:     }
igor@2: }
igor@2: PERL_SCRIPT
igor@2:     [ -e "$NOTES_FILE" ] || touch "$NOTES_FILE"
igor@2:     export NOTES_FILE
igor@2:     perl $PERL_SCRIPT_TEMP_NAME "$1"
igor@2:     rm $PERL_SCRIPT_TEMP_NAME
igor@2: }
igor@2: 
igor@2: remove_marks()
igor@2: {
igor@2:     PERL_SCRIPT_TEMP_NAME=`mktemp /tmp/perl-grep-v-english-XXXXXXXX`
igor@2:     cat <<'PERL_SCRIPT' > $PERL_SCRIPT_TEMP_NAME
igor@2: $file = $ARGV[0];
igor@2: our %dict;
igor@2: if (open(F, $file)) {
igor@2:     @lines=<F>;
igor@2:     close(F);
igor@2: 
igor@2:     if (open(F, ">$file")) {
igor@2:         for (@lines) {
igor@2:             chomp;
igor@3:             if (not /^#/ and m/(\s+)(\S+)(\s+)(\S+)(\s+)(.*)/) {
igor@2:                 my $name=$4;
igor@2:                 my $comment=$6;
igor@2:                 $dict{$name}=$comment;
igor@2:                 print F "$1$2$3$4\n";
igor@2:             }
igor@2:             else {
igor@2:                 print F "$_\n";
igor@2:             }
igor@2:         }
igor@2:     }
igor@2: }
igor@2: if (open(NOTES, $ENV{NOTES_FILE})) {
igor@2:     @lines=<NOTES>;
igor@2:     close(NOTES);
igor@2: 
igor@2:     if (open(NOTES, ">".$ENV{NOTES_FILE})) {
igor@2:         for (@lines) {
igor@2:             chomp;
igor@2:             s/^\s+//;
igor@2:             my ($a,$b)=split /\s+/,$_,2;
igor@2:             if (not defined($dict{$a}) || ($dict{$a} eq $b)) {
igor@2:                 print NOTES "$_\n";
igor@2:                 if (defined($dict{$a})) { unset($dict{$a}); }
igor@2:             }
igor@2:         }
igor@2:         for (keys %dict) {
igor@2:             $mark=$dict{$_};
igor@2:             $space=" "x(30-length($_));
igor@2:             print NOTES "$_$space$mark\n";
igor@2:         }
igor@2:     }
igor@2: }
igor@2: PERL_SCRIPT
igor@2:     [ -e "$NOTES_FILE" ] || touch "$NOTES_FILE"
igor@2:     export NOTES_FILE
igor@2:     perl $PERL_SCRIPT_TEMP_NAME "$1"
igor@2:     rm $PERL_SCRIPT_TEMP_NAME
igor@2: }
igor@2: 
igor@0: mkdir -p $WORK_DIR
igor@0: cd $WORK_DIR
igor@0: if echo "$1" | grep -q http: 
igor@0: then 
igor@3:     text_from_url "$1" | get_words ${TEMP1}-full | add_stat ${TEMP1}-full| tee "$TEMP1" > "$TEMP2"
igor@0: elif [ "$#" != 0 ]
igor@0: then
igor@3:     cat "$1" | get_words ${TEMP1}-full | add_stat ${TEMP1}-full | tee "$TEMP1" > "$TEMP2"
igor@0: else 
igor@3:     get_words ${TEMP1}-full| add_stat ${TEMP1}-full | tee "$TEMP1" > "$TEMP2"
igor@0: fi
igor@0: 
igor@2: add_marks "$TEMP2"
igor@0: if [ "$editor" = vim ]
igor@0: then
igor@0:     vim -c 'set keywordprg='"$LANGUAGE" -c 'set iskeyword=@,48-57,/,.,-,_,+,,,#,$,%,~,=' "$TEMP2" < /dev/tty > /dev/tty
igor@0: else
igor@0:     echo 2
igor@0:     $editor "$TEMP2"
igor@0: fi
igor@2: remove_marks "$TEMP2"
igor@2: 
igor@0: diff "$TEMP1" "$TEMP2" | awk '{print $3}' | sort -u >> "$VOCABULARY"
igor@3: rm -f "$TEMP1" "$TEMP2" "${TEMP1}-full"