# HG changeset patch
# User Igor Chubin <igor@chub.in>
# Date 1274096914 -10800
# Node ID d1eb7dc37feb7fb9b4b4e2a42176b9a71ffa6e7e
# Parent  1318aa5898ee54beafb49fe2e5096f1aec4d5215
minifix: -p pages

diff -r 1318aa5898ee -r d1eb7dc37feb new-words.sh
--- a/new-words.sh	Mon May 17 12:14:44 2010 +0300
+++ b/new-words.sh	Mon May 17 14:48:34 2010 +0300
@@ -15,6 +15,7 @@
     -l lang     override language settings
     -n          non-interactive mode (don't run vi)
     -m          don't add marks (and don't save marks added by user)
+    -p pages    work with specified pages only (pages = start-stop/total )
     -s          show the text statistics (percentage of known words and so on) and exit
 
 The language of the text can be specified also
@@ -59,7 +60,8 @@
 NEED_TO_USE_VOCABULARY_WHEN_SORT=NO
 DONT_ADD_MARKS=NO
 NON_INTERACTIVE_MODE=NO
-while getopts l:skmnp opt
+PART_TO_PROCESS=''
+while getopts l:skmnp: opt
 do
     case "$opt" in
       s)  STAT_ONLY=YES;;
@@ -67,7 +69,7 @@
       l)  LANGUAGE="$OPTARG";;
       m)  DONT_ADD_MARKS=YES;;
       n)  NON_INTERACTIVE_MODE=YES;;
-      p)  PAGES="$OPTARG";;
+      p)  PART_TO_PROCESS="$OPTARG";;
       \?)       # unknown flag
           show_usage
           exit 1;;
@@ -424,6 +426,49 @@
     rm $PERL_SCRIPT_TEMP_NAME
 }
 
+part()
+{
+    PERL_SCRIPT_TEMP_NAME=`mktemp /tmp/perl-part-XXXXXXXX`
+    cat <<'PERL_SCRIPT' > $PERL_SCRIPT_TEMP_NAME
+#!/usr/bin/perl
+
+my @lines=<STDIN>;
+my $lines=$#lines;
+my $interval=$ARGV[0];
+if (not $interval) {
+    print @lines;
+}
+else {
+    my ($start,$stop,$total);
+    if ($interval =~ m@(.*)/(.*)@) {
+        $start = $1;
+        $total = $2;
+    }
+    else {
+        $start=$interval;
+        $total=0;
+    }
+    if ($start =~ m@(.*)-(.*)@) {
+        $start = $1;
+        $stop = $2;
+    }
+    if ($start =~ m@(.*)\+(.*)@) {
+        $start = $1;
+        $stop = $start+$2;
+    }
+
+    $start=int($lines/$total*$start);
+    $stop=int($lines/$total*$stop);
+
+    for($i=$start;$i<$stop;$i++){
+        print $lines[$i];
+    }
+}
+PERL_SCRIPT
+    perl $PERL_SCRIPT_TEMP_NAME "$1"
+    rm $PERL_SCRIPT_TEMP_NAME
+}
+
 mkdir -p $WORK_DIR
 oldpwd="$PWD"
 cd $WORK_DIR
@@ -441,6 +486,7 @@
 else 
     cat
 fi \
+   | part $PART_TO_PROCESS \
    | tee $ORIGINAL_TEXT \
    | get_words ${TEMP1}-full \
    | group_words \
diff -r 1318aa5898ee -r d1eb7dc37feb part.pl
--- a/part.pl	Mon May 17 12:14:44 2010 +0300
+++ /dev/null	Thu Jan 01 00:00:00 1970 +0000
@@ -1,39 +0,0 @@
-#!/usr/bin/perl
-
-my @lines=<STDIN>;
-my $lines=$#lines;
-my $interval=$ARGV[0];
-my ($start,$stop,$total);
-if ($interval =~ m@(.*)/(.*)@) {
-    $start = $1;
-    $total = $2;
-}
-else {
-    $start=$interval;
-    $total=0;
-}
-if ($start =~ m@(.*)-(.*)@) {
-    $start = $1;
-    $stop = $2;
-}
-if ($start =~ m@(.*)\+(.*)@) {
-    $start = $1;
-    #if ($2 =~ /%$/) {
-    #}
-    #else {
-    $stop = $start+$2;
-    #}
-}
-#print "start = $start\n";
-#print "stop = $stop\n";
-#print "total = $total\n";
-
-$start=int($lines/$total*$start);
-$stop=int($lines/$total*$stop);
-
-#print "start = $start\n";
-#print "stop = $stop\n";
-#print "total = $total\n";
-for($i=$start;$i<$stop;$i++){
-    print $lines[$i];
-}