2000.12.19 05:36 "Re: Stripes in thumbnail", by Steve Underwood
No, I'm not going to take up the responsability to prove that point. For starters, I did not say for a fact that it behaves differently. I said it has been reported to behave differently. Which is something quite different. Which refers to a number of postings on this mailing list reporting so, from different independent sources, and you must have read them and know what I'm talking about.
Over the years I've seen rumours started (or perpetuated) by postings like yours. While I'm not about to rule out bugs in the software the library has been around for a long time and incorporated in many many products. Given that exposure I would have expected to hear of any problem of the sort you alluded to.
I've seen postings from folks suggesting bugs for >10 years. Some have been correct. Most have not. I used to follow up on all of them. I don't anymore 'cuz the software isn't mine to maintain. Nonetheless the issue is still the same: if you've found a problem provide a way to reproduce it.
The only problem I have seen with libtiff's G4 handling is when processing TIFF files produced by some of the computer FAX packages and hardware. Almost all of them are buggy, and produce screwy TIFF files. The problem is that libtiff doesn't deal with some of them very gracefully. As an example, take the puzzlingly popular, overpriced and very buggy GammaFAX cards. They set the image length to zero, and this makes libtiff seg fault. I know they are doing the wrong thing, but perhaps libtiff could do sanity checks on the data, and behave more gracefully.