po: Modularisation of the diff_po.pl script

diff --git a/po/diff_po.pl b/po/diff_po.pl
index 30ca28dcf2800be2c47df295aab9521714d5c617..d5dec02839b3a0c4849947d0b0a20e943f3fb497 100755 (executable)
--- a/po/diff_po.pl
+++ b/po/diff_po.pl
@@ -24,14 +24,22 @@
 # License along with this software; if not, write to the Free Software
 # Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
 #
-# Copyright (c) 1010-2011 Kornel Benko, kornel@lyx.org
+# Copyright (c) 2010-2013 Kornel Benko, kornel@lyx.org
 #
 # TODO:
 # 1.) Check for ".git" or ".svn" to decide about revisioning
 # 2.) Search for good correlations of deleted <==> inserted string
 #     using Text::Levenshtein or Algorithm::Diff
 
+BEGIN {
+    use File::Spec;
+    my $p = File::Spec->rel2abs( __FILE__ );
+    $p =~ s/[\/\\]?diff_po\.pl$//;
+    unshift(@INC, "$p");
+}
+
 use strict;
+use parsePoLine;
 use Term::ANSIColor qw(:constants);
 use File::Temp;
 
@@ -74,6 +82,7 @@ if ($ARGV[0] =~ /^-r(.*)/) {
        print $tmpfile $l;
       }
       close(FI);
+      $tmpfile->seek( 0, SEEK_END );           # Flush()
       push(@args, $tmpfile->filename, $argf);
       print "===================================================================\n";
       &diff_po(@args);
@@ -241,74 +250,6 @@ sub printIfDiff($$$)
   }
 }
 
-sub parse_po_file($$)
-{
-  my ($file, $rMessages) = @_;
-  if (open(FI, '<', $file)) {
-    $status = "normal";
-    $fuzzy = 0;
-    my $lineno = 0;
-    while (my $line = <FI>) {
-      $lineno++;
-      &parse_po_line($line, $lineno, $rMessages);
-    }
-    &parse_po_line("", $lineno + 1, $rMessages);
-    close(FI);
-  }
-}
-
-sub parse_po_line($$$)
-{
-  my ($line, $lineno, $rMessages) = @_;
-  chomp($line);
-
-  if ($status eq "normal") {
-    if ($line =~ /^#, fuzzy/) {
-      $fuzzy = 1;
-    }
-    elsif ($line =~ s/^msgid\s+//) {
-      $foundline = $lineno;
-      $status = "msgid";
-      $msgid = "";
-      &parse_po_line($line);
-    }
-  }
-  elsif ($status eq "msgid") {
-    if ($line =~ /^\s*"(.*)"\s*/) {
-      $msgid .= $1;
-    }
-    elsif ($line =~ s/^msgstr\s+//) {
-      $status = "msgstr";
-      $msgstr = "";
-      &parse_po_line($line);
-    }
-  }
-  elsif ($status eq "msgstr") {
-    if ($line =~ /^\s*"(.*)"\s*/) {
-      $msgstr .= $1;
-    }
-    else {
-      if ($msgid ne "") {
-       $rMessages->{$msgid}->{line} = $foundline;
-       $rMessages->{$msgid}->{fuzzy} = $fuzzy;
-       $rMessages->{$msgid}->{msgstr} = $msgstr;
-      }
-      $fuzzy = 0;
-      $status = "normal";
-    }
-  }
-  else {
-    die("invalid status");
-  }
-}
-
-sub getLineSortedKeys($)
-{
-  my ($rMessages) = @_;
-
-  return sort {$rMessages->{$a}->{line} <=> $rMessages->{$b}->{line};} keys %{$rMessages};
-}
-
 sub printExtraMessages($$)
 {
   my ($type, $rExtra) = @_;

diff --git a/po/parsePoLine.pm b/po/parsePoLine.pm
new file mode 100644 (file)
index 0000000..fff1418
--- /dev/null
+++ b/po/parsePoLine.pm
@@ -0,0 +1,186 @@
+package parsePoLine;
+
+use strict;
+
+our(@EXPORT, @ISA);
+
+BEGIN {
+    use Exporter   ();
+    @ISA        = qw(Exporter);
+    @EXPORT     = qw(parse_po_file getLineSortedKeys);
+}
+
+my ($status, $foundline, $msgid, $msgstr, $fuzzy);
+
+
+my $alternative = 0;
+my @entry = ();
+my %entries = ();
+
+sub parse_po_file($$)
+{
+  $alternative = 0;
+  @entry = ();
+  %entries = ();
+
+  my @result = ();
+  my $resindex = 0;
+  my ($file, $rMessages) = @_;
+  if (open(FI, '<', $file)) {
+    $status = "normal";
+    $fuzzy = 0;
+    my $lineno = 0;
+    while (my $line = <FI>) {
+      $lineno++;
+      &parse_po_line($line, $lineno, $rMessages, \@result, \$resindex);
+      push(@entry, $line);
+
+    }
+    &parse_po_line("", $lineno + 1, $rMessages, \@result, \$resindex);
+    my @entr1 = @entry;
+    $result[$resindex] = ["zzzzzzzzzzzz", \@entr1];
+    close(FI);
+  }
+  return(@result);
+}
+
+sub  parse_po_line($$$$$)
+{
+  my ($line, $lineno, $rMessages, $rresult, $rresindex) = @_;
+  chomp($line);
+
+  if ($status eq "normal") {
+    if ($line =~ /^#, fuzzy/) {
+      $fuzzy = 1;
+    }
+    elsif ($line =~ s/^msgid\s+//) {
+      die("line alternate") if ($alternative);
+      $foundline = $lineno;
+      $status = "msgid";
+      $msgid = "";
+      &parse_po_line($line, $lineno, $rMessages, $rresult, $rresindex);
+    }
+    elsif ($line =~ s/^\#\~ msgid\s+//) {
+      $alternative = 1;
+      $foundline = $lineno;
+      $status = "msgid";
+      $msgid = "";
+      &parse_po_line($line, $lineno, $rMessages, $rresult, $rresindex);
+    }
+  }
+  elsif ($status eq "msgid") {
+    if ($line =~ /^\s*"(.*)"\s*/) {
+      $msgid .= $1;
+    }
+    elsif ($line =~ /^\#\~\s*"(.*)"\s*/) {
+      die("line not alternate") if (! $alternative);
+      $msgid .= $1;
+    }
+    elsif ($line =~ s/^msgstr\s+//) {
+      $alternative = 0;
+      $status = "msgstr";
+      $msgstr = "";
+      &parse_po_line($line, $lineno, $rMessages, $rresult, $rresindex);
+    }
+    elsif ($line =~ s/^\#\~ msgstr\s+//) {
+      $alternative = 1;
+      $status = "msgstr";
+      $msgstr = "";
+      &parse_po_line($line, $lineno, $rMessages, $rresult, $rresindex);
+    }
+  }
+  elsif ($status eq "msgstr") {
+    if ($line =~ /^\s*"(.*)"\s*/) {
+      $msgstr .= $1;
+    }
+    elsif ($line =~ /^\#\~\s+"(.*)"\s*/) {
+      die("line not alternate") if (! $alternative);
+      $msgstr .= $1;
+    }
+    else {
+      if (!defined($entries{$msgid})) {
+       my @entr1 = @entry;
+       $rresult->[${$rresindex}] = [$msgid, \@entr1];
+       $entries{$msgid} = $msgstr;
+      }
+      else {
+       if ($alternative) {
+         print "duplicated alternate entry: \"$msgid\"\n";
+       }
+       else {
+         print "duplicated entry: \"$msgid\"\n";
+       }
+       print " on line:          $foundline\n";
+       print " original on line: $rMessages->{$msgid}->{line}\n\n";
+      }
+      @entry = ();
+      $rMessages->{$msgid}->{line} = $foundline;
+      $rMessages->{$msgid}->{fuzzy} = $fuzzy;
+      $rMessages->{$msgid}->{msgstr} = $msgstr;
+      $rMessages->{$msgid}->{alternative} = $alternative;
+      $rMessages->{$msgid}->{entryidx} = ${$rresindex};
+      ${$rresindex} = ${$rresindex}+1;
+      $fuzzy = 0;
+      $status = "normal";
+    }
+  }
+  else {
+    die("invalid status");
+  }
+}
+
+sub getLineSortedKeys($)
+{
+  my ($rMessages) = @_;
+
+  return sort {$rMessages->{$a}->{line} <=> $rMessages->{$b}->{line};} keys %{$rMessages};
+}
+
+1;
+
+
+__END__
+
+=pod
+
+=encoding utf8
+
+=head1 NAME
+
+parsePoLine
+
+=head1 SYNOPSIS
+
+  use parsePoLine; #imports functions 'parse_po_file() and getLineSortedKeys()'
+
+  my %Messages = ();
+  my @entries = parse_po_file("sk.po", \%Messages);
+
+=head1 DESCRIPTION
+
+This is used to parse a single po-file.
+
+Results:
+    %Messages The keys in this hash are the msgid-strings
+              the value is reference to a hash wit following values:
+        msgstr:      the translated string
+        line:        the line-no in the po-file
+        fuzzy:       boolean, if the string is fuzzy
+        alternative: if set, so this entry is part of help-strings
+        entryidx:    The index in the correspondig @entries array
+    @entries  List of references to 2-valued arrays
+        [0]:         msgid-string
+        [1]:         The sequence of lines from the po-file
+                     belonging to this entry.
+
+To print the whole po-file:
+    for my $entry (@entries) {
+      print @{$entry->[1]};
+    }
+
+To get the index to a known $msgid:
+    my $entriesidx = $Messages{$msgid}->{entryidx};
+
+=head1 AUTHOR
+
+Kornel Benko <Kornel.Benko@berlin.de>