On Wed, Feb 02, 2022 at 03:23:34PM +0100, Mark Gaiser wrote:
> On Wed, Feb 2, 2022 at 2:29 PM Michael Niedermayer <michael@niedermayer.cc>
> wrote:
> 
> > On Tue, Feb 01, 2022 at 10:58:30PM +0100, Mark Gaiser wrote:
[...]
> >
> >
> > > +    AVClass *class;
> > > +    URLContext *inner;
> > > +    char *gateway;
> > > +} Context;
> > > +
> > > +// A best-effort way to find the IPFS gateway.
> > > +// Only the most appropiate gateway is set. It's not actually requested
> > (http call) to prevent
> > > +// a potential slowdown in startup. A potential timeout is handled by
> > the HTTP protocol.
> > > +//
> > > +// When done and the return value is 1, a potential IPFS Gateway is set
> > in the gateway
> > > +// variable in the inner URLContext.
> >
> > > +static int ff_populate_ipfs_gateway(URLContext *h)
> >
> > ff_ prefix seems not needed
> >
> 
> Done. Removed the prefix from both functions.
> For the record, I thought it was mandatory to prefix internal functions
> with "ff_"?

only non static ones, these prefixes are to avoid
conflicing with other things in the global namespace


[...]

> 
> 
> >
> >
> > > +    }
> > > +
> >
> > > +    fscanf(gateway_file, "%[^\n]", gateway_file_data);
> >
> > how large is the array ?
> > how large is the data written into the array ?
> > what is the relation between the 2 ?
> >
> 
> Yeah... I'm going to need more information on that one.
> What do you mean?
> Is this wrong?

the array is fixed length, the string written into it is from
some external file IIRC. 
Maybe i missed a check but to me it seemed like the array end can
be overwritten which is bad


thx

[...]

-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

Awnsering whenever a program halts or runs forever is
On a turing machine, in general impossible (turings halting problem).
On any real computer, always possible as a real computer has a finite number
of states N, and will either halt in less than N cycles or never halt.