| AWARE [SYSTEMS] | Imaging expertise for the Delphi developer | |||||||
![]() |
TIFF and LibTiff Mailing List Archive | |||||||
LibTiff Mailing List
TIFF and LibTiff Mailing List Archive Contact
The TIFF Mailing List Homepage |
Thread1999.10.15 17:20 "Re: TIFF Version validation", by Thomas E Deweese>>>>> "TL" == Tom Lane <tgl@sss.pgh.pa.us> writes:
TL> Bill Comstock <comstock@fas.harvard.edu> writes:
>> I'm interested in putting a large collection of TIFF images through
>> a validation process, one that tests each image against a series of
>> rules and reports whether it conforms to the TIFF standard [...]
TL> A more useful way to look at things is a profile of features
TL> required to decode any particular TIFF --- for example, if your
TL> program said "tiled TIFF using LZW compression and containing
TL> 24-bit RGB data", that would convey helpful information. You
TL> can't usefully boil it down to "version 5.0" (or whichever spec
TL> release was the first one to mention all of those features),
TL> because almost no decoder supports the *whole* spec of any
TL> release.
Admittedly talking about 'versions' of Tiff is not real
correct (I prefer to talk about 'flavors', but profiles also works).
But a good tool to at least validate baseline Tiff files would, I
think, be very useful. If the tool could check the validity some of
the more common profiles, tiled, alpha channels, extra samples,
alternate color spaces - 'inks support', various compression sets,
would be really welcome.
I think that many of us have had the experience of reading the
TIFF spec. and doing our darn best to implement to what it says, but
over looking this or that little detail, and always wondering if we
were writing out subtlety 'corrupt' Tiff files.
--
Thomas DeWeese
deweese@kodak.com
"The only difference between theory and practice is
that in theory there isn't any." -- unknown
|
|||||||