<div dir="ltr"><div>Jean-Marc,</div><div><br></div><div>Clearly I haven't thought these changes through properly and was only really considering common desktop systems. I was originally fixing an issue when compiling my code as 64bit where my use of int64_t (defined as long on my platform) was conflicting with the opusfile stream callbacks (e.g. op_seek_func) which were using the opus_int64 type which was defined as long long. When I looked at the header I thought it could do with some work, so instead of putting casts in my code, I modified the header and sent a patch. I didn't even consider building for architectures which could have a 40bit long etc. and am obviously a bit out of my depth in this case.</div><div><br></div><div>Sorry for the bad patch, please just disregard it. It might be worth updating the header based on the ogg one though. Even without the breaking changes I made, it seems to be much more update in some areas.</div><div><br></div><div>Simon</div></div><div class="gmail_extra"><br><div class="gmail_quote">On 3 June 2015 at 17:59, Ralph Giles <span dir="ltr"><<a href="mailto:giles@thaumas.net" target="_blank">giles@thaumas.net</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><span>On 2015-06-03 6:21 AM, Simon Jackson wrote:<br>
<br>
> - Replaced blanket #define of 8 & 64 bit types with typedefs for each<br>
> platform to match 16 & 32 bit types.<br>
<br>
</span>You didn't say why you need these additional types?<br>
<span class="HOEnZb"><font color="#888888"><br>
-r<br>
</font></span></blockquote></div><br></div>