Do not warn if half.h has already being included

[jmaselbas]

The issue with the warning of halfLimits.h being deprecated is that it
is very tempting to simply remove the include of halfLimits.h in favor
of simply half.h. However this change does depends on the version of
Imath used, and halfLimits.h should be kept for compatibility with
older version. Simply do not warn if short.h is already included.

[linux-foundation-easycla]

The committers listed above are authorized under a signed CLA.

• ✅ login: jmaselbas / name: Jules Maselbas (c304ed2)

[lgritz]

I'm ok with this, but I think at some point we really do want to warn about using the halfLimits.h header, which we would eventually like to remove entirely if it serves no purpose. We wouldn't want to remove headers except at break points when back compatibility can be fully broken (that would be 4.0, by convention), so to give time for people to get the warnings and adjust, there should be a 3.x in which we do want warnings to show up. Maybe it's not 3.1, per this PR. But at some point.

[cary-ilm]

Thanks, this looks good to me, and it's indeed a better solution when needing to support older versions of Imath/OpenEXR. The ASWF policy is to require signed Contributor License Agreements for all contributions, no matter how small. If you'd prefer to avoid the hassle of signing the form via the big red button above, I'm happy to submit a similar PR myself, let me know.

[alvinhochun]

Friendly ping, the author signed the CLA.

I also commented on #241 (comment). I am not entirely against letting users know the include is now "redundant", but I think Imath may at least give an example on how to correctly exclude the include only on supported versions. Having a way to suppress the warning (which is what this PR does) will also be nice.

[cary-ilm]

Agreed, I filed #258 as a reminder to straighten this out before the next release.