On Thu, 16 Jun 2011, Ronald S. Bultje wrote:
> On Thu, Jun 16, 2011 at 4:00 AM, Martin Storsjö <[email protected]> wrote:
> > From: Kirill Zorin <[email protected]>
> >
> > Signed-off-by: Martin Storsjö <[email protected]>
> > ---
> > libavformat/mmsh.c | 2 +-
> > 1 files changed, 1 insertions(+), 1 deletions(-)
> >
> > diff --git a/libavformat/mmsh.c b/libavformat/mmsh.c
> > index b19973e..af040e2 100644
> > --- a/libavformat/mmsh.c
> > +++ b/libavformat/mmsh.c
> > @@ -231,7 +231,7 @@ static int mmsh_open(URLContext *h, const char *uri,
> > int flags)
> > host, sizeof(host), &port, path, sizeof(path), location);
> > if (port<0)
> > port = 80; // default mmsh protocol port
> > - ff_url_join(httpname, sizeof(httpname), "http", NULL, host, port,
> > path);
> > + ff_url_join(httpname, sizeof(httpname), "http", NULL, host, port,
> > "%s", path);
>
> OK, nice catch.
Ok'd for push during freeze by Reinhard (together with the one for rtsp),
and pushed.
> For the future, do we need to make the API more resilient to this by
> e.g. having two functions, one w/o printf-style formatting and one
> with?
I think that's overkill - instead we could add gcc specific attributes to
allow the compiler to check for these kinds of issues, we do that for
av_log already.
// Martin
_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel
