[Nepomuk] Review Request: Walkaround for bug in glibc 2.14 cause strigi index hang on UTF-16 encoded mp3 files

[Sebastian Trueg]

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
http://git.reviewboard.kde.org/r/103259/#review8500
-----------------------------------------------------------


Two things: 1. please add "strigi" as the review group, 2. please add comments into the code. Apart from that it looks ok to me - but it is really up to Jos to give the final ok.
Also it seems that Qt might need patching, too.

- Sebastian Trueg


On Nov. 26, 2011, 9:45 p.m., Xuetian Weng wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> http://git.reviewboard.kde.org/r/103259/
> -----------------------------------------------------------
> 
> (Updated Nov. 26, 2011, 9:45 p.m.)
> 
> 
> Review request for Nepomuk.
> 
> 
> Description
> -------
> 
> The upstream bug report, though unconfirmed.
> More glibc 2.14 test result is welcome. (Tested on Chakra and Arch).
> http://sourceware.org/bugzilla/show_bug.cgi?id=13439
> 
> UTF-16 endianness will be remember once glibc 2.14 iconv is called, which will cause QString toLocal8Bit behaviour incorrect if strigi resolve a big endian encoded UTF-16 string first on little endian (x86) machine. So this patch ignore the endian test provided by iconv and do it by hand.
> 
> 
> This addresses bug 274251.
>     http://bugs.kde.org/show_bug.cgi?id=274251
> 
> 
> Diffs
> -----
> 
>   lib/endanalyzers/id3endanalyzer.cpp a3b33ae 
> 
> Diff: http://git.reviewboard.kde.org/r/103259/diff/diff
> 
> 
> Testing
> -------
> 
> Test on my machine, index no longer hangs.
> 
> 
> Thanks,
> 
> Xuetian Weng
> 
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/nepomuk/attachments/20111127/9f3beced/attachment.html>