Hello,

I would need your help for the following problem:

In milestome m83, we introduced (and we are deeply sorry for it) a regression with our WordPerfect import filter. Due to some flaws in our handling of OLE2 documents, the WordPerfect TypeDetection was throwing an NotConnectedException (http://www.openoffice.org/issues/show_bug.cgi?id=43995).

We produced a patch that fixes this issue. It was reported as solving a problem by a person who was building m79 with CWS libwpdupgrade merged in. I has a funny behaviour on m82 with the same CWS merged. The WP TypeDetection did not throw any exception and actually returned empty string since the document (a MS Word document without extension) was detected as not being WordPerfect document. Immediatelly after this, we receive an pop-up window saying "Read-Error. Cannot read the file".

During week-end I rebuilt the m83 milestone (with my patch) with debug and tried today whether the problem was just my hoaked build system or not. I run the soffice.bin in gdb and have the output.

My questions are:
1) Is there something substantial that changed between m79 and m83 concerning the typedetection that could be reason of the difference of behaviour between the two builds?
2) Would someone be interested in examining the output of soffice.bin running in gdb? It is 19kB.


Thanks in advance for your help

Fridrich

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



Reply via email to