On Sat, 12 Sep 2020 at 07:38, Scott Kostyshak <skost...@lyx.org> wrote:
> On Fri, Sep 11, 2020 at 02:04:31PM -0400, Scott Kostyshak wrote: > > > >> By the way, what would you prefer I do in this > > > > case where I'm not sure the fix is correct? Since we're on a branch, > I > > > > just committed so that way the tests work for me and Kornel (who I > > > > imagine is also affected by this configure issue). But if you prefer > I > > > > post the patch to the list instead of pushing, that's fine too. In my > > > > opinion, since it's your branch you should choose the workflow. > > > > > > > > > > I have no problem with you pushing commits onto that branch. Just make > > > sure to send the hash on the list, in case I make one force push too > many > > > :). > > > > Sounds good! I will try to remember to always specify the hash (please > > let me know if I forget). > > I just pushed commit 7dc72ef3, which cherry-picked a commit from master > that fixes a path in an ru example file. After this, all docbook tests > pass for me. > > I get a lot of failures from the xhtml tests. I think most of these > failures are due to xmllint not being happy. I can give you the log that > has these errors (and xmllint output) if that would be helpful. > Otherwise, would it just be better to disable xmllint checking in the > tests for now? If not, it seems we will have a lot of test failures and > it will be difficult to catch more serious regressions (e.g., crashes). > That said, I don't know anything about this stuff so whatever you and > Kornel prefer is fine with me. > Kornel (in CC) said somewhere else he'd prefer to wait before merging because of this very issue. I suppose the new failures are related to the (quite recent) fix for <section> tags in XHTML, that's the only XHTML-only change in my branch (plus a few things in XML that are shared). I'd appreciate at least a list of tests that fail, if that's possible :)!
-- lyx-devel mailing list lyx-devel@lists.lyx.org http://lists.lyx.org/mailman/listinfo/lyx-devel
