new-words

diff new-words.sh @ 16:c65ffd60cc18

ключ -s: Информация об известных словах в тексте

Может использоваться в пакетном режиме,
например, для того чтобы из множества файлов
выбрать тот, в котором процент известных слов наибольший
author Igor Chubin <igor@chub.in>
date Sun Apr 04 19:03:30 2010 +0300 (2010-04-04)
parents 9b18c7efe31c
children 35eeaf2620ce
line diff
     1.1 --- a/new-words.sh	Sun Apr 04 12:54:46 2010 +0300
     1.2 +++ b/new-words.sh	Sun Apr 04 19:03:30 2010 +0300
     1.3 @@ -1,6 +1,15 @@
     1.4  #!/bin/sh
     1.5  cat <<HELP > /dev/null
     1.6  
     1.7 +USAGE: 
     1.8 +
     1.9 +    new-words [ -l lang ] [ -s ] [ ARG ] 
    1.10 +
    1.11 +SWITCHES: 
    1.12 +
    1.13 +    -s          show text statistics and exit
    1.14 +    -l lang     override language settings
    1.15 +
    1.16  Поддержка нескольких языков:
    1.17  
    1.18      new-words -l lang URL
    1.19 @@ -37,6 +46,13 @@
    1.20  VOCABULARY=${LANGUAGE}.txt
    1.21  NOTES_FILE=notes-${LANGUAGE}.txt
    1.22  
    1.23 +STAT_ONLY=NO
    1.24 +if [ "$1" = "-s" ]
    1.25 +then
    1.26 +    STAT_ONLY=YES
    1.27 +    shift
    1.28 +fi
    1.29 +
    1.30  get_words()
    1.31  {
    1.32  tr ' ' '\n' | sed 's/--/ /g' \
    1.33 @@ -59,13 +75,22 @@
    1.34      total_unknown="`cat $after|awk '{s=s+$1}END{print s}'`"
    1.35      total_known="`echo $total-$total_unknown|bc`"
    1.36      percentage="`echo '100*('$total-$total_unknown')'/$total | bc -l | sed 's/\\.\(.\).*/.\1/'`"
    1.37 -    echo "# $LANGUAGE, $percentage, <$total_known/$total>"
    1.38 +    if [ "$STAT_ONLY" = "YES" ]
    1.39 +    then
    1.40 +        echo "LANG  KNOWN%  UNKNOWN%  KNOWN    TOTAL"
    1.41 +        echo "$LANGUAGE    $percentage    `echo \(100-$percentage\) | bc -l`      $total_known    $total"
    1.42 +        return 0
    1.43 +    else 
    1.44 +        echo "# $LANGUAGE, $percentage, <$total_known/$total>"
    1.45 +    fi
    1.46 +
    1.47      PERL_SCRIPT_TEMP_NAME=`mktemp /tmp/perl-grep-v-english-XXXXXXXX`
    1.48      cat <<'PERL_SCRIPT' > $PERL_SCRIPT_TEMP_NAME
    1.49  my $total=shift(@ARGV);
    1.50  my $total_known=shift(@ARGV);
    1.51  my $s=0;
    1.52 -my $mark_line=int($total_known*100/$total/5)*5+5;
    1.53 +my $mark_line=int($total_known*100/$total/5)*5;
    1.54 +if ($mark_line>=90) { $mark_line+=1; } else { $mark_line +=5; };
    1.55  while(<>)
    1.56  {
    1.57      print;
    1.58 @@ -154,7 +179,7 @@
    1.59      s/ing$//;
    1.60  
    1.61      s/ism$//; s/ist$//; s/ful$//; s/able$//; s/ably$//;
    1.62 -    s/ation$//;  s/ness$//; s/ally$//;
    1.63 +    s/ation$//;  s/ness$//; s/ship$//; s/ally$//;
    1.64      s/ment$//; s/ify$//; s/ity$//; s/fy$//; s/ly$//;
    1.65      s/ise$//; s/ize$//;
    1.66  
    1.67 @@ -327,15 +352,20 @@
    1.68     | add_stat ${TEMP1}-full \
    1.69     | tee "$TEMP1" > "$TEMP2"
    1.70  
    1.71 -add_marks "$TEMP2"
    1.72 -if [ "$editor" = vim ]
    1.73 +if [ "$STAT_ONLY" = "YES" ]
    1.74  then
    1.75 -    vim -c 'set keywordprg='"$LANGUAGE" -c 'set iskeyword=@,48-57,/,.,-,_,+,,,#,$,%,~,=' "$TEMP2" < /dev/tty > /dev/tty
    1.76 +    cat "$TEMP1"
    1.77  else
    1.78 -    echo 2
    1.79 -    $editor "$TEMP2"
    1.80 +    add_marks "$TEMP2"
    1.81 +    if [ "$editor" = vim ]
    1.82 +    then
    1.83 +        vim -c 'set keywordprg='"$LANGUAGE" -c 'set iskeyword=@,48-57,/,.,-,_,+,,,#,$,%,~,=' "$TEMP2" < /dev/tty > /dev/tty
    1.84 +    else
    1.85 +        echo 2
    1.86 +        $editor "$TEMP2"
    1.87 +    fi
    1.88 +    remove_marks "$TEMP2"
    1.89  fi
    1.90 -remove_marks "$TEMP2"
    1.91  
    1.92  diff "$TEMP1" "$TEMP2" | awk '{print $3}' | sort -u >> "$VOCABULARY"
    1.93  rm -f "$TEMP1" "$TEMP2" "${TEMP1}-full" "$ORIGINAL_TEXT"