# HG changeset patch
# User Igor Chubin <igor@chub.in>
# Date 1270397010 -10800
# Node ID c65ffd60cc182d25a5ed8337fcd21cc7f9021f67
# Parent  c6efd17741aaf1761e6a0eb00a28ea4893838624
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

diff -r c6efd17741aa -r c65ffd60cc18 new-words.sh
--- a/new-words.sh	Sun Apr 04 14:12:35 2010 +0300
+++ b/new-words.sh	Sun Apr 04 19:03:30 2010 +0300
@@ -1,6 +1,15 @@
 #!/bin/sh
 cat <<HELP > /dev/null
 
+USAGE: 
+
+    new-words [ -l lang ] [ -s ] [ ARG ] 
+
+SWITCHES: 
+
+    -s          show text statistics and exit
+    -l lang     override language settings
+
 Поддержка нескольких языков:
 
     new-words -l lang URL
@@ -37,6 +46,13 @@
 VOCABULARY=${LANGUAGE}.txt
 NOTES_FILE=notes-${LANGUAGE}.txt
 
+STAT_ONLY=NO
+if [ "$1" = "-s" ]
+then
+    STAT_ONLY=YES
+    shift
+fi
+
 get_words()
 {
 tr ' ' '\n' | sed 's/--/ /g' \
@@ -59,13 +75,22 @@
     total_unknown="`cat $after|awk '{s=s+$1}END{print s}'`"
     total_known="`echo $total-$total_unknown|bc`"
     percentage="`echo '100*('$total-$total_unknown')'/$total | bc -l | sed 's/\\.\(.\).*/.\1/'`"
-    echo "# $LANGUAGE, $percentage, <$total_known/$total>"
+    if [ "$STAT_ONLY" = "YES" ]
+    then
+        echo "LANG  KNOWN%  UNKNOWN%  KNOWN    TOTAL"
+        echo "$LANGUAGE    $percentage    `echo \(100-$percentage\) | bc -l`      $total_known    $total"
+        return 0
+    else 
+        echo "# $LANGUAGE, $percentage, <$total_known/$total>"
+    fi
+
     PERL_SCRIPT_TEMP_NAME=`mktemp /tmp/perl-grep-v-english-XXXXXXXX`
     cat <<'PERL_SCRIPT' > $PERL_SCRIPT_TEMP_NAME
 my $total=shift(@ARGV);
 my $total_known=shift(@ARGV);
 my $s=0;
-my $mark_line=int($total_known*100/$total/5)*5+5;
+my $mark_line=int($total_known*100/$total/5)*5;
+if ($mark_line>=90) { $mark_line+=1; } else { $mark_line +=5; };
 while(<>)
 {
     print;
@@ -154,7 +179,7 @@
     s/ing$//;
 
     s/ism$//; s/ist$//; s/ful$//; s/able$//; s/ably$//;
-    s/ation$//;  s/ness$//; s/ally$//;
+    s/ation$//;  s/ness$//; s/ship$//; s/ally$//;
     s/ment$//; s/ify$//; s/ity$//; s/fy$//; s/ly$//;
     s/ise$//; s/ize$//;
 
@@ -327,15 +352,20 @@
    | add_stat ${TEMP1}-full \
    | tee "$TEMP1" > "$TEMP2"
 
-add_marks "$TEMP2"
-if [ "$editor" = vim ]
+if [ "$STAT_ONLY" = "YES" ]
 then
-    vim -c 'set keywordprg='"$LANGUAGE" -c 'set iskeyword=@,48-57,/,.,-,_,+,,,#,$,%,~,=' "$TEMP2" < /dev/tty > /dev/tty
+    cat "$TEMP1"
 else
-    echo 2
-    $editor "$TEMP2"
+    add_marks "$TEMP2"
+    if [ "$editor" = vim ]
+    then
+        vim -c 'set keywordprg='"$LANGUAGE" -c 'set iskeyword=@,48-57,/,.,-,_,+,,,#,$,%,~,=' "$TEMP2" < /dev/tty > /dev/tty
+    else
+        echo 2
+        $editor "$TEMP2"
+    fi
+    remove_marks "$TEMP2"
 fi
-remove_marks "$TEMP2"
 
 diff "$TEMP1" "$TEMP2" | awk '{print $3}' | sort -u >> "$VOCABULARY"
 rm -f "$TEMP1" "$TEMP2" "${TEMP1}-full" "$ORIGINAL_TEXT"