Armin K. wrote: > On 06/29/2013 09:50 PM, Bruce Dubbs wrote: >> I'm about to update to gtk+-3.8.2. The build and install are fine (with >> a few warnings). The issue I have is make check. There are three >> failures. I can remove two by deleting a couple of files: >> >> tests/a11y/pickers.ui >> tests/reftests/image-load-from-file.ui >> >> but I was only able to skip the third failure by editing (commenting >> out) some lines in gtk/tests/filechooser.c for one sub-test. That would >> need a patch for the book. >> >> The question is whether to address this at all. We can use a generic >> statement like "Some tests fail" or "Three tests are known to fail" or >> we can try to skip the failing tests. >> >> I'm inclined to just mention the failing tests and press on, but I'd >> like some other opinions.
> I've personaly never cared about testsuites. Do as you may please. I do think it's useful for the book's users to know about the testsuites and what we have found when we run them. One extreme case would be to remove all mention of them. The other extreme is to try to go into each issue and fix it, but that's really the developer's job. A user, especially in LFS, needs to do some tests to see if the install is OK. After all, some mistakes lead to a huge numbers of errors. That said, in my last build of BLFS, I've skipped the tests when using the version in the book because I'm reasonably confident that the tests have been run. Building BLFS certainly goes a lot faster when you skip the tests. I think I'll just mention the known failures when updating the book. -- Bruce -- http://linuxfromscratch.org/mailman/listinfo/blfs-dev FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
