Thanks for those corrections.
How about this? https://github.com/OpenImageIO/oiio/pull/446
I took it even further -- there were only a couple spots where errors could
occur, so I just addressed those directly and eliminated the 'ok' variable
completely.
On Oct 29, 2012, at 3:15 AM, Michel Lerenard wrote:
> Hi,
>
> two points i'd like to share:
>
> 1. as I stated in the message I sent friday, there is a problem with maketx
> source code. In the write_mipmap function a boolean is defined twice. It
> causes no problem for GCC, but Visual crashes in a particular case.
>
> line 1224
>
> bool ok = true;
>
> then line 1306
>
> bool ok = false;
>
> The second declaration doesn't seem to be useful at all, so i removed it and
> it works.
>
> 2. in case an error occurs before writing mimaps, nothing is done, here's a
> small modification to the code, starting line 1224.
>
> bool ok = true;
> ok &= img.write (out);
> stat_writetime += writetimer();
> if (!ok) {
> std::cerr << "maketx ERROR: Write failed \" : " << out->geterror() <<
> " img.geterror: " << img.geterror() << "\n";
> }
>
> if (ok && mipmap) { // Mipmap levels:
>
> This way we can properly ouptput the error if one occured.
>
>
> Does this seem ok ?
> _______________________________________________
> Oiio-dev mailing list
> [email protected]
> http://lists.openimageio.org/listinfo.cgi/oiio-dev-openimageio.org
--
Larry Gritz
[email protected]
_______________________________________________
Oiio-dev mailing list
[email protected]
http://lists.openimageio.org/listinfo.cgi/oiio-dev-openimageio.org
