[MPlayer-dev-eng] [PATCH] subreader: recode the whole file to utf-8 before parsing

[Evgeniy Stepanov]

On Tuesday 29 May 2007, Evgeniy Stepanov wrote:
> On Tuesday 29 May 2007, Zuxy Meng wrote:
> > Hi,
> >
> > 2007/5/29, Evgeniy Stepanov <eugeni.stepanov at gmail.com>:
> > > Hi,
> > >
> > > this patch changes the way subtitle recoding takes place. Instead of
> > > recoding each 'struct subtitle', the whole file is read into memory,
> > > recoded to utf-8, and a stream is created with new_memory_stream.
> > >
> > > This patch was created as an attempt to resolve the problem with BOM in
> > > the beginning of a unicode subtitle file. The problem is not solved
> > > yet, but it is easy now, and will be done in a separate patch. As a
> > > side effect, which is arguably more important, it allows to read UTF-16
> > > encoded subtitles.
> > >
> > > Subtitle related code at the beginning of subreader.c is unused now and
> > > can be removed.
> > >
> > > All in all, it seems like the right thing to do. Opinions ?
> >
> > I've tested your patch with enca and ucs-2le/ucs-2be encoded subtitles
> > and it works great! Thanks!
> >
> > BTW is libass the right place to put the code?
>
> The patch simply splits ass_read_file() in 2 functions, no new code is
> added to libass.

Applied.