Red Hat Bugzilla – Bug 469321
[fix available] base cannot open open new format TT_hsqldb.odb from from qatesttools
Last modified: 2009-09-02 05:08:36 EDT
Description of problem:
Use automation testtool to run "f_java_upgrade.bas" will get following segmentation fault error messages:
/usr/lib/openoffice.org/program/soffice: line 147: 28361 段错误 "$sd_prog/$sd_binary" "$@"
Version-Release number of selected component (if applicable):
Steps to Reproduce:
1. setup OOO automation test env
2. run f_java_upgrade.bas script
One of the problems I have here is that I don't know for sure what qatesttool contents you have. I'm been assuming that it is a head checkout e.g. the contents of cvs -d:pserver:firstname.lastname@example.org:/cvs co qatesttool
Looking at the tests there I believe that the actual crash is the same crash as rhbz#468136 i.e. a modal dialog of "Filter Dialog" appeared unexpectedly and then shutting down OOo ran into the intractable problem of http://www.openoffice.org/issues/show_bug.cgi?id=95654
So the question is why did the unexpected dialog appear. I think it is because the version of OOo we are testing is too old for some of the tests. Maybe/Maybe not. But If you take the attachment I will put here and overwrite qatesttool/base/input/hsql_datasource/TT_hsqldb.odb with it then I believe that will workaround the problem.
The original TT_hsqldb.odb *can* be opened with 3.0.0 and with 2.4.1, but not apparently with 2.3.0
Created attachment 322100 [details]
replacement .odb for test-case
Created attachment 322102 [details]
and the original that it cannot open
So I believe the question becomes how come 2.3.0 cannot open the original TT_hsqldb.odb leading to the wrong dialog getting presented to the testtool.
Created attachment 322295 [details]
patch to enable 2.3.0 to open those new .odbs
We can patch it up to be able to open those new format .odbs. And we would get this for free on any rebase to >= 2.4.0
dtardon->yzhou: You _really_ should use ooo230 revision of qatesttool for testing OOo 2.3.0.; what's the logic in checking for features which simply aren't there?
committed to >= 2.3.0-6.11
An advisory has been issued which should help the problem
described in this bug report. This report is therefore being
closed with a resolution of ERRATA. For more information
on therefore solution and/or where to find the updated files,
please follow the link below. You may reopen this bug report
if the solution does not work for you.