From kde-core-devel  Tue Sep 11 20:05:55 2007
From: Fredy Yanardi <fyanardi () gmail ! com>
Date: Tue, 11 Sep 2007 20:05:55 +0000
To: kde-core-devel
Subject: Small Bug in Sonnet::Filter
Message-Id: <200709112005.55355.fyanardi () gmail ! com>
X-MARC-Message: https://marc.info/?l=kde-core-devel&m=118967391526784
MIME-Version: 1
Content-Type: multipart/mixed; boundary="--Boundary-00=_jUv5GVjH1GdO4mi"

--Boundary-00=_jUv5GVjH1GdO4mi
Content-Type: text/plain;
  charset="utf-8"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

Hi all,

I just found a problem in Sonnet::Filter that is causing words not fetched 
correctly in Sonnet's Backgroundchecker. The problem is when iterating a word 
in nextWord() method, if current iterator position exceeds word length, an 
end of word is returned. But I think the method should be returning current 
word. By changing a line in this method, this problem seems to be solved. 
Here I also attach a diff to make it clear what's changed.

Any problem with this?

Best Regards,

Fredy Yanardi

--Boundary-00=_jUv5GVjH1GdO4mi
Content-Type: text/x-diff;
  charset="utf-8";
  name="sonnet_filter.diff"
Content-Transfer-Encoding: 7bit
Content-Disposition: attachment;
	filename="sonnet_filter.diff"

Index: kdecore/sonnet/filter.cpp
===================================================================
--- kdecore/sonnet/filter.cpp	(revision 710627)
+++ kdecore/sonnet/filter.cpp	(working copy)
@@ -126,7 +126,7 @@
         ++m_currentPosition;
         //Test if currentPosition exists, otherwise go out
         if( m_currentPosition >= m_buffer.length())
-            return Filter::end();
+            return Word( foundWord, start );
         currentChar = m_buffer.at( m_currentPosition );
     }
     if ( shouldBeSkipped( allUppercase, runTogether, foundWord ) )

--Boundary-00=_jUv5GVjH1GdO4mi--