[Nepomuk] Review Request: Call AnalysisResult::finishIndexChild everywhere

[Sebastian Trueg]

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
http://git.reviewboard.kde.org/r/102216/
-----------------------------------------------------------

Review request for Nepomuk and Strigi.


Summary
-------

A short while ago I introduced the new method AnalysisResult::finishIndexChild so clients can properly update the indexing status. I figured I would only have to call this method in analysers that afterwards continue to index the main file. However, it turns out that the problem also arises when other analysers are run after the one not calling finishIndexChild. This would for example result in a wrong depth for SHA1 checksums which would then be rejected by Nepomuk.


Diffs
-----

  lib/endanalyzers/arendanalyzer.cpp 2e7c371 
  lib/endanalyzers/bz2endanalyzer.cpp a4b4ba7 
  lib/endanalyzers/cpioendanalyzer.cpp 74e12a4 
  lib/endanalyzers/gzipendanalyzer.cpp 947579e 
  lib/endanalyzers/lzmaendanalyzer.cpp 4515434 
  lib/endanalyzers/odfendanalyzer.cpp 5227925 
  lib/endanalyzers/oleendanalyzer.cpp 7c2105b 
  lib/endanalyzers/pdfendanalyzer.cpp a0e4111 
  lib/endanalyzers/rpmendanalyzer.cpp fe248e7 
  lib/endanalyzers/sdfendanalyzer.cpp bb0bbc4 
  lib/endanalyzers/tarendanalyzer.cpp 4601107 
  lib/endanalyzers/zipendanalyzer.cpp 1ab4d91 

Diff: http://git.reviewboard.kde.org/r/102216/diff


Testing
-------

Indexing works properly with patch. Checksums are saved to Nepomuk again.


Thanks,

Sebastian

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/nepomuk/attachments/20110804/6f416d6a/attachment-0001.html>