ilovepi wrote: > > @evelez7 am I forgetting some plans issues we've discussed previously? > > Petr mentioned a configuration file before. I generally agree and opened > #181768 just to track it, but I would put it near the bottom of current > clang-doc needs/issues. I think Petr mentioned it because celery-doc has a > lot of options specified that would be nice to have (ignore certain paths, > etc) and that would be nicer to have in a file for reuse. I don't think we > should be making design decisions around it yet or for the near future. >
Right. Thanks for the reminder. That was pretty recent, so not surprised I forgot. Agreed its low priority for quite a while. > > any preferences? > > I'm actually a bit surprised to see that the "one param per line" design is > so common > > https://docs.hdoc.io/hdoc/llvm-project/rD85AF4A2A9884C1C.html#83C1C7AF36ACD866 > > Doxygen does the same thing. I think for simplicity sake going with the > convention would be best. It's an overall improvement compared to what we see > in Fuschia output. > That's an option. I do think its nicer if there's some N that we keep them all on the same line though. Certainly if N == 1, I wouldn't want the one per line formatting (personally). > If this was pure HTML/CSS to determine wrapping, I would've suggested doing > more complicated wrapping. I guess we could always put each param in its own HTML construct and let the layout engine figure it out... that's kind of what its for. Given that we're OK with one per line, lets give it the college try on 0-2 params: same line, 3+: one per line. If that's anything other than trivial, lets just land one per line always for now, and we can figure out a more robust solution later (with tracking bug and FIXME of course 😉 ). does that work for everyone? https://github.com/llvm/llvm-project/pull/181417 _______________________________________________ cfe-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
