[PATCH libreoffice-4-0] Fix crash with fdo60063-1.docx

Hi,

I have submitted a patch for review:

    https://gerrit.libreoffice.org/4093

To pull it, you can do:

    git pull ssh://gerrit.libreoffice.org:29418/core refs/changes/93/4093/1

Fix crash with fdo60063-1.docx

Compared to the original version, this one tries hard
not to produce empty keywords.

Reviewed-on: https://gerrit.libreoffice.org/4091
Reviewed-by: Fridrich Strba <fridrich@documentfoundation.org>
Tested-by: Fridrich Strba <fridrich@documentfoundation.org>
(cherry picked from commit b2f97d77237cde249b4c40acc8ad3f7f8755e768)
(cherry picked from commit 693707e338922d8dc7578975a1990e8a1cba6a53)

Conflicts:
        oox/source/docprop/docprophandler.cxx

Change-Id: I11d036c2d54cde2d3c311a77c115011246f38d0a
---
M oox/source/docprop/docprophandler.cxx
1 file changed, 13 insertions(+), 28 deletions(-)



diff --git a/oox/source/docprop/docprophandler.cxx b/oox/source/docprop/docprophandler.cxx
index 1067315..ff6d170 100644
--- a/oox/source/docprop/docprophandler.cxx
+++ b/oox/source/docprop/docprophandler.cxx
@@ -25,6 +25,9 @@
 
 #include <osl/time.h>
 
+#include <vector>
+#include <boost/algorithm/string.hpp>
+
 #include "oox/helper/attributelist.hxx"
 
 using namespace ::com::sun::star;
@@ -157,39 +160,21 @@
 {
     if ( !aChars.isEmpty() )
     {
-        uno::Sequence< ::rtl::OUString > aResult( 20 );
-        sal_Int32 nCounter = 0;
+        std::string aUtf8Chars = rtl::OUStringToOString( aChars, RTL_TEXTENCODING_UTF8 ).getStr();
+        std::vector<std::string> aUtf8Result;
+        boost::split( aUtf8Result, aUtf8Chars, boost::is_any_of(" ,;:\t"), 
boost::token_compress_on );
 
-        const sal_Unicode* pStr = aChars.getStr();
-        for( sal_Int32 nInd = 0; nInd < aChars.getLength() && pStr[nInd] != 0; nInd++ )
+        if (!aUtf8Result.empty())
         {
-            switch( pStr[nInd] )
-            {
-                case (sal_Unicode)' ':
-                case (sal_Unicode)',':
-                case (sal_Unicode)';':
-                case (sal_Unicode)':':
-                case (sal_Unicode)'\t':
-                    // this is a delimiter
-                    // unfortunately I did not find any specification for the possible delimiters
-                    if ( !aResult[nCounter].isEmpty() )
-                    {
-                        if ( nCounter >= aResult.getLength() )
-                            aResult.realloc( nCounter + 10 );
-                        nCounter++;
-                    }
-                    break;
+            uno::Sequence< rtl::OUString > aResult( aUtf8Result.size() );
+            OUString* pResultValues = aResult.getArray();
+            for ( std::vector< std::string >::const_iterator i = aUtf8Result.begin();
+                  i != aUtf8Result.end(); ++i, ++pResultValues )
+                *pResultValues = rtl::OUString( i->c_str(), static_cast< sal_Int32 >( i->size() 
),RTL_TEXTENCODING_UTF8 );
 
-                default:
-                    // this should be a part of keyword
-                        aResult[nCounter] += ::rtl::OUString( (sal_Unicode)pStr[nInd] );
-            }
+            return aResult;
         }
-
-        aResult.realloc( nCounter + 1 );
-        return aResult;
     }
-
     return uno::Sequence< ::rtl::OUString >();
 }
 // ------------------------------------------------

-- 
To view, visit https://gerrit.libreoffice.org/4093
To unsubscribe, visit https://gerrit.libreoffice.org/settings

Gerrit-MessageType: newchange
Gerrit-Change-Id: I11d036c2d54cde2d3c311a77c115011246f38d0a
Gerrit-PatchSet: 1
Gerrit-Project: core
Gerrit-Branch: libreoffice-4-0
Gerrit-Owner: Fridrich Strba <fridrich@documentfoundation.org>