Bug#618923: ITP: libguess -- Character-set detection library
On Sun, Mar 20, 2011 at 01:03, Bilal Akhtar <bilalakhtar@ubuntu.com> wrote:
> On 03/19/2011 07:53 PM, Aron Xu wrote:
>>
>> Firstly, I'm looking forward to such an implemention that could work
>> well. But I have to say there are so many attempts to deal with GBK and
>> UTF-8 detections but none of them do the exactly right thing, wish this
>> one helps, :-)
>
> The primary reason why I felt this to be good was that I have a need for
> it in my package: audacious. Audacious upstream (also part of atheme)
> uses libguess. We have it disabled in the debian package for now. I
> wanted to enable that. Hence I wanted to maintain this package.
> Moreover, I am quite good in maintaining shared libs, and I just
> completed the packaging fully lintian-clean, with symbols control files,
> etc etc.
>
> So if you feel doubtful about its future in the case of charset
> detection, feel free to hand over the ITP to me and I'll get the package
> into the NEW queue within a week.
>
> If not, then just mention that you are willing to continue with your ITP
> and I'll point you out to a few sponsors.
>
> Regards,
>
> Bilal Akhtar.
>
>
>
> --
> Bilal Akhtar - Ubuntu Developer <bilalakhtar@ubuntu.com>
> IRC nick: cdbs
>
>
Sorry if my previous mail has lead to some misunderstanding...
I just want to leave a comment about my thoughts on charset detection,
no other meanings.
What I've said might be useful for you (as you are the maintainer of
audacious package) is that auto-detection might lead to troubles in
some cases.
For example there are many GBK/UTF-8 auto-detection algorithms but all
of them have the chance to claim the charset is GBK but it is UTF-8
actually. This is bad for people who originally don't have problems
with their UTF-8 encoded files (tags) and the poor algorithm might
break their hearts.
--
Regards,
Aron Xu
Reply to: