# HG changeset patch
# User Igor Chubin <igor@chub.in>
# Date 1272477638 -10800
# Node ID 23f949c03f95a727aff931c93c66e2030a5f910c
# Parent  416394a87d9f93e0e1da87fceabbbaace37d3015
Usage information, oneliners prototype script, autoraising of the words that are similar to known

diff -r 416394a87d9f -r 23f949c03f95 new-words.sh
--- a/new-words.sh	Tue Apr 20 21:15:19 2010 +0400
+++ b/new-words.sh	Wed Apr 28 21:00:38 2010 +0300
@@ -1,5 +1,8 @@
 #!/bin/sh
-cat <<HELP > /dev/null
+
+show_usage()
+{
+cat <<HELP > /dev/stderr
 
 USAGE: 
 
@@ -7,22 +10,26 @@
 
 SWITCHES: 
 
-    -s          show text statistics and exit
+    -h          print this screen
+    -k          put higher words that are similar to the known words (only for English)
     -l lang     override language settings
+    -s          show the text statistics (percentage of known words and so on) and exit
 
-Поддержка нескольких языков:
+The language of the text can be specified also
+by name of the program new-words (correspondent link must be created before).
+For example, these calls are equivalent:
 
-    new-words -l lang URL
-
-Например, для немецких текстов:
-
+    de-words URL
     new-words -l de URL
 
-Или, предварительно создав соответствующую ссылку:
+HELP
+}
 
-    de-words URL
-
-HELP
+if [ "$1" = "-h" ]
+then
+    show_usage
+    exit 0
+fi
 
 WORK_DIR=~/.new-words/
 TEMP1=`mktemp /tmp/new-words-XXXXXXXXXX-temp1`
@@ -53,6 +60,13 @@
     shift
 fi
 
+NEED_TO_USE_VOCABULARY_WHEN_SORT=NO
+if [ "$1" = "-k" ]
+then
+    NEED_TO_USE_VOCABULARY_WHEN_SORT=YES
+    shift
+fi
+
 get_words()
 {
 tr ' ' '\n' | sed 's/--/ /g' \
@@ -219,7 +233,17 @@
     }
 }
 
-@lines=<>;
+our %Vocabulary;
+open(VOC, $ENV{VOCABULARY})
+ or die "Can't open VOCABULARY";
+while (<VOC>){
+    chomp;
+    #s/'//g;
+    $Vocabulary{normalize($_)}="1";
+}
+close(VOC);
+
+@lines=<STDIN>;
 for $L (@lines) {
     chomp($L);
     $l=$L;
@@ -227,11 +251,20 @@
     my ($a, $b)=split(/\s+/,$l,2);
     $group_weight{normalize($b)}+=$a;
 }
+if ($ENV{NEED_TO_USE_VOCABULARY_WHEN_SORT} eq "YES") {
+    for $k (keys %group_weight) {
+        if (defined($Vocabulary{$k})) {
+            $group_weight{$k} *= 2;
+        }
+    }
+}
 @lines2 = sort { compare($b,$a) } @lines;
 for $l (@lines2) {
     print "$l\n";
 }
 PERL_SCRIPT
+    export VOCABULARY
+    export NEED_TO_USE_VOCABULARY_WHEN_SORT
     perl $PERL_SCRIPT_TEMP_NAME
     rm $PERL_SCRIPT_TEMP_NAME
 }
diff -r 416394a87d9f -r 23f949c03f95 oneliners.sh
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/oneliners.sh	Wed Apr 28 21:00:38 2010 +0300
@@ -0,0 +1,11 @@
+temp=/tmp/oneliner$$
+while read l
+do
+    dict -h dictd.xdsl.by -d mueller24 "$l" 2> /dev/null | perl -n -e 'print if ($y and not /^\s*$/) ; $y=1 if /'$l'/;' | grep -v _Syn | grep -v _Ant > $temp
+    if [ `wc -l $temp | awk '{print $1}'` = 1 ]
+    then
+        echo -n $l
+        cat $temp
+    fi
+done
+