[FFmpeg-devel] [PATCH]lavf/mpegts: Convert service name and service provider to utf-8

Jan Ekström jeebjp at gmail.com
Fri Feb 8 03:07:10 EET 2019


On Fri, Feb 8, 2019 at 2:44 AM Carl Eugen Hoyos <ceffmpeg at gmail.com> wrote:
>
> 2019-02-08 1:42 GMT+01:00, Jan Ekström <jeebjp at gmail.com>:
> > On Fri, Feb 8, 2019 at 2:38 AM Carl Eugen Hoyos <ceffmpeg at gmail.com> wrote:
> >>
> >> Hi!
> >>
> >> Attached patch fixes ticket #6320, tested with the sample from ticket
> >> #7069.
> >>
> >> Please comment, Carl Eugen
> >
> > Nice!
> >
> > I will try to test this tomorrow. To tell you the truth, I am quite
> > jealous that the text encodings you require are already in upstream
> > libiconv.
>
> The sample you provided (for which it would have been
> extra-nice to tell us where it came from, I found it meanwhile)
> does not allow converting with iconv, or at least I couldn't
> guess the encoding...
>
> Carl Eugen

The samples came from MBS (ARIB captions one) and Tokyo MX (the PID
switch one), if that's what you mean. The channels themselves
shouldn't matter too much as all broadcasts in Japan generally follow
the ARIB specs/recommendations as far as how they're multiplexed.
They're part of various samples I've been capturing in Japan along the
years.

The text coding is ARIB STD-24 and together with the EN300-468 Table00
one I linked both were nack'd at upstream due to not being encodings
utilized in general computing, so the iconv modules now live in the
linuxtv repository*. Which means that they most likely are not
packaged anywhere.
* https://git.linuxtv.org/v4l-utils.git/tree/contrib/gconv

Libaribb24 is another implementation of the ARIB STD-B24 text coding
(in addition to the subtitle part), and I've been so far utilizing it
in my WIP branch.

Jan


More information about the ffmpeg-devel mailing list