Various translation checks for LibreOffice

* gsicheck
* readmecheck which is a modified gsicheck
* doublecheck from Gregor Hartmann (OOo) see also #i112478#

2 files changed, 198 insertions, 0 deletions

diff --git a/doublecheck.pl b/doublecheck.pl
new file mode 100755
index 0000000..12a8e97
--- /dev/null
+++ b/doublecheck.pl
@@ -0,0 +1,176 @@
+:
+eval 'exec perl -wS $0 ${1+"$@"}'
+    if 0;
+#*************************************************************************
+#
+# DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
+# 
+# Copyright 2000, 2010 Oracle and/or its affiliates.
+#
+# OpenOffice.org - a multi-platform office productivity suite
+#
+# This file is part of OpenOffice.org.
+#
+# OpenOffice.org is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Lesser General Public License version 3
+# only, as published by the Free Software Foundation.
+#
+# OpenOffice.org is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU Lesser General Public License version 3 for more details
+# (a copy is included in the LICENSE file that accompanied this code).
+#
+# You should have received a copy of the GNU Lesser General Public License
+# version 3 along with OpenOffice.org.  If not, see
+# <http://www.openoffice.org/license.html>
+# for a copy of the LGPLv3 License.
+#
+#*************************************************************************
+#
+# doublecheck - Checks sdf files for identical translations where they have to be unique.
+#
+
+use strict;
+
+my ($source);
+my (%languages, %languages_seen, %Stores, $smodule, $help, $lineOK, $module, $gid, $lid, $lang, $found, $text, $Store_ref, $error, $file, $line, $sfile, $sgid, $slid, $goodlangs );
+
+# this string should contain only languages that are known to be well translated i.e. no newly translated languages or even in pregress
+$goodlangs = "de fr ja es ko";
+
+%languages = ();
+%languages_seen = ();
+%Stores = ();
+
+if ( !exists $ENV{ "WITH_LANG" } || !$ENV{ "WITH_LANG" } )
+{
+    print "\$WITH_LANG not set or empty: nothing to do\n";
+    exit 0;
+}
+
+foreach ( split( " ", $ENV{ "WITH_LANG" } ) )
+{
+    $languages{ $_ } = 1;
+}
+
+$smodule = $sfile = $sgid = $slid = ".*";
+
+
+&get_options;
+
+if ( ! $source )
+{
+    print "Error: No sdf_filename defined\n\n";
+    $help = "Yes";
+}
+if ( !$help )
+{
+
+    $found = 0;
+    open SOURCE, "<$source";
+    while ( <SOURCE> )
+    {
+        analyze_line( $_ );
+
+        $languages_seen{ $lang } = 1 if ( $lineOK );
+
+        if ( $lineOK && $module =~ $smodule && $file =~ $sfile && $gid =~ $sgid && $lid =~ $slid )
+        {
+            $found += 1;
+            if ( exists $languages{ $lang } )
+            {
+				if ( !exists $Stores{ $lang } )
+				{
+				    my %S = ();
+					$Stores{ $lang } = \%S;
+				}
+				$Store_ref = $Stores{ $lang };
+				if ( exists $$Store_ref{ $text } )
+				{
+					print "Entry double for LANG=$lang in file $source\n";
+					print $$Store_ref{ $text };
+					print $line;
+					$error = 1;
+				}
+				else
+				{
+					$$Store_ref{ $text } = $line;
+				}
+            }
+        }
+    }	
+    close SOURCE;
+
+#    print "$found strings found in $source for -m $smodule -f $sfile -g $sgid -l $slid\n";
+    foreach my $goodlang ( split( " ", $goodlangs ) )
+    {
+        if ( exists $languages_seen{ $goodlang } && exists $languages{ $goodlang } && !exists $Stores{ $goodlang } )
+        {
+            die "no sting found to check for language $lang in $source for -m $smodule -f $sfile -g $sgid -l $slid. Maybe the strings have been relocated.";
+        }
+    }
+    die "errors found" if ( $error );
+
+}
+else
+{
+    print "\n\n";
+    print "doubleckeck v.1.0\n";
+    print "---------------------\n";
+    print "Usage:\n";
+    print "doubleckeck [-m <module>] [-f <filepath>] [-g <GID>] [-l <LID>] <sdf_filename>\n";
+    print "doubleckeck --help\n";
+    print "\n";
+    print "------------------------------------------------------------------------------\n";
+    print "Checks sdf files for identical translations where they have to be unique.\n";
+    print "The switches -m, -f, -g and -l are treated as perl regular expressions.\n";
+    print "They default to .*\n";
+    print "All Languages in \$WITH_LANG are checked.\n";
+    print "\n\n";
+}
+
+
+
+sub get_options {
+	my ($arg,$has_source);
+	$has_source = 0;
+	while ($arg = shift @ARGV) {
+		$arg =~ /^-m$/  and $smodule = shift @ARGV and next;
+		$arg =~ /^-f$/  and $sfile = shift @ARGV and next;
+		$arg =~ /^-g$/  and $sgid = shift @ARGV and next;
+		$arg =~ /^-l$/  and $slid = shift @ARGV and next;
+		$arg =~ /^--help$/  and $help = 1 and next; #show help
+		
+		if ( !$has_source )
+		{
+		    $source = $arg;
+		    $has_source = 1;
+		}
+		else
+		{
+		    die "only one file can be parsed";
+		}
+	}
+}
+
+
+sub analyze_line {
+
+    ($line) = @_;
+    
+    $lineOK = 0;
+    if ( $line =~ /^([^\t]*)\t([^\t]*)\t[^\t]*\t[^\t]*\t([^\t]*)\t([^\t]*)\t[^\t]*\t[^\t]*\t[^\t]*\t([^\t]*)\t([^\t]*)\t[^\t]*\t[^\t]*\t[^\t]*\t[^\t]*/ )
+    {
+        $module = $1;
+        $file = $2;
+        $gid = $3;
+        $lid = $4;
+        
+        $lang = $5;
+        $text = $6;
+        
+        $lineOK = 1;
+    }
+}
+
diff --git a/translationscheck.sh b/translationscheck.sh
new file mode 100755
index 0000000..5373e5e
--- /dev/null
+++ b/translationscheck.sh
@@ -0,0 +1,22 @@
+#!/bin/bash 
+
+OOBUILDDIR='/home/timar/libreoffice-3-4/'
+MY_PATH=`pwd`
+. $OOBUILDDIR/LinuxX86Env.Set.sh 
+
+base=`dirname $0`
+
+#for i in `ls -1 $SRC_ROOT/translations/source`
+for i in hu
+do
+    po2oo -i $SRC_ROOT/translations/source/$i -t $SRC_ROOT/translations/unxlngi6.pro/misc/sdf-template/en-US.sdf -o $i.sdf -l $i
+    gsicheck -c -wef $i.err $i.sdf >$i.log
+    READMETMP=$(mktemp)
+    grep "^readlic" $i.sdf | grep -v "	en-US	" | sed -e "s/</\\\</g" -e "s/>/\\\>/g" -e 's/"/\\\"/g' > $READMETMP
+    gsicheck -c -l "" $READMETMP | sed -e "s/\\\>/>/g" -e "s/\\\</</g" -e 's/\\\"/"/g' >> $i.log
+    WITH_LANG=$i $base/doublecheck.pl -m formula -f source.core.resource.core_resource.src -g RID_STRLIST_FUNCTION_NAMES $i.sdf >>$i.log || exit 1 
+    WITH_LANG=$i $base/doublecheck.pl -m scaddins -f source.analysis.analysis_funcnames.src -g RID_ANALYSIS_FUNCTION_NAMES $i.sdf >>$i.log || exit 1
+    WITH_LANG=$i $base/doublecheck.pl -m scaddins -f source.datefunc.datefunc.src -g RID_DATE_FUNCTION_NAMES $i.sdf >> $i.log || exit 1
+    WITH_LANG=$i $base/doublecheck.pl -m sw -f source.ui.utlui.poolfmt.src -g "STR_POOLNUMRULE_.*" $i.sdf >> $i.log || exit 1
+    WITH_LANG=$i $base/doublecheck.pl -m sw -f source.ui.utlui.poolfmt.src -g "STR_POOLCOLL_.*" $i.sdf >> $i.log || exit 1
+done