Re: CFD: eXIf 2017-05-31

Cosmin Truta <[email protected]> Thu, 8 Jun 2017 01:10:43 -0400
Newsgroups gmane.comp.graphics.png.general
Message-ID <CAAoVtZxyw2mjhbLnHvrBQXi1EmNDEnzSTPPaRTrXZnYq7KrsQQ@mail.gmail.com>
Speaking of RFC-2119: should the draft not use "should" instead of
"must" in the paragraph that mentions the 64KB size limit?

The 64KB size limit is correct if it comes to converting PNG to JPEG.
But that sort of conversion is not intended, the applications should
otherwise not be restricted by this limit.

"While the PNG specification allows the chunk size to be as large as
2^31-1 bytes, [...] the total length of the Exif data [...] _should_
not exceed 64 kbytes, because it _should_ fit into a JPEG APP1 marker.

I'm not sweating too much over this. If we can replace this word from
the CFD to the CFV (without going through another CFD round), great. I
personally find it useful to have EXIF profiles larger than 64KB, for
example, to be able to store the preview image (which is different
than, and larger than the thumbnail image) inside EXIF, rather than
outside. Storing the preview outside EXIF leaves a dangling pointer
inside EXIF. Not a problem with TIFF, but a problem with JPEG. By
imposing this limit in PNG, then PNG would inherit the same problem as
JPEG. Changing the above two words from "must" to "should" would
resolve this problem.

Sincerely,
Cosmin

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot