[TYPO3-core] RFC: Indexed Search misses OpenDocument support / drop OOoExtract usage

[Michael Stucki]

This is a CVS patch request.

Problem:
Oasis OpenDocument files (OpenOffice.org 2, StarOffice, etc.) are not
supported by indexed_search.

Solution:
Add support (works the same way like existing OpenOffice.org 1 files).

Additionally, I have removed the implementation for using OOoExtract.
Reasons for this are:
- It seems like the splitRegularContent works perfectly fine for XML
formatted documents, so there is no need for using a special tool
- It was too complicated to set paths to a parser AND a script when there
were alternative applications that don't require any special interpreters.

And finally, I have removed the libunzipped usage and replaced it using a
direct "unzip" call. The reason for this is that libunzipped extracts the
whole file (including all images) although it only needs content.xml and
meta.xml. The new behaviour is that only these files are extracted, but
therefore the "unzip" path needs to be adjusted manually.

Compatibility:
This patch removes the options "nativeOOMethod", "OOoExtract" and "ruby".
Instead, a new option "unzip" is added.

Comments:
I think it would be wise to change all default paths from /usr/local/bin
to /usr/bin. Comments about this idea can be discussed of course.

Regards, michael
-- 
Use a newsreader! Check out
http://typo3.org/community/mailing-lists/use-a-news-reader/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: index_oasis.diff
Type: text/x-diff
Size: 12988 bytes
Desc: not available
Url : http://lists.netfielders.de/pipermail/typo3-team-core/attachments/20060203/eda9ba1d/attachment.bin