New test to automatically importing all bugzilla documents

[Markus Mohrhard]

Hey,

so after much work the idea to automatically import all bugzilla files
was succesful. An initial test run with all docx files in Release
configuration showed that it is quite stable and I'm currently
rebuilding with dbgutil (to get the safe iterators) and will run an
initial test of all Writer documents (rtf, fodt, odt, doc, docx). The
output of the docx run is attached to this mail.

Currently I plan to run the script every 2 or 3 weeks on a TDF server
and report the documents back. Hopefully some of these tasks can be
automated by time and we only need someone to start the build. There
are several open tasks and some things I noticed would be nice to
have. I list them here to make is easier for others to step in and
implement some of them or add their own ideas:

TODOS:
* Calc and Draw/Impress document import
* Script allowing to manage running several scripts in parallel: In
therory this is a task that can be done in perfect parallelisation and
it would help with the time needed as we have alone more than 15000
writer documents, at least more than 8000 calc documents ...
* Automatically mail a set of people with the result (ML, ESC, ...)
* Automatically store the results in a git repository (git diff
showing which files regressed)
* A lot of clean-up

Nice to have:
* easy way to plug-in custom checkers that report all documents
containing a specific feature: we can use any uno expression to check
the document for a specific feature if we need all documents
containing a feature
* reporting how long a document needs for import and log it into a
data structure ( simple way to find performance regressions )

Additionally I saw some comments like "report this document to OD,
might miss a graphic" or similar. I realised that this new script
provides a nice way to find such problems and if someone has such a
case we can easily add an assertion and find the documents containing
this error. If anyone is interested in such an action please write me
a mail.

I would like to thank a few people who made this possible. Norbert for
providing the help with the TDF server, Stephan for all my dumb
questions about uno, pyuno, process handling in LibO and last but not
least Joren who wrote some parts of the script and forced me by taking
this Easy Hack to create this script after more than 1 year.
Additionaly Michael Stahl for convwatch, which was the base for this
script and Caolan who wrote the get-bugzilla-attachments script.

Please feel free to comment and make suggestions how to improve this idea.

Regards,
Markus
-------------- next part --------------
A non-text attachment was scrubbed...
Name: logs.tar.gz
Type: application/x-gzip
Size: 3586 bytes
Desc: not available
URL: <http://lists.freedesktop.org/archives/libreoffice/attachments/20130218/5d697b5f/attachment-0001.bin>