Handle parsing errors #711
Conversation
|
Any chance this PR will be reviewed? merged? released? |
Note, both of these PRs are on top of the changes introduced by jakobandersen#698. Links to PRs for future reference: - breathe-doc#698 - breathe-doc#711 - breathe-doc#723 Signed-off-by: Abrar Rahman Protyasha <abrar@openrobotics.org>
Note, both of these PRs are on top of the changes introduced by upstream PR breathe-doc#698. Links to PRs for future reference: - breathe-doc#698 - breathe-doc#711 - breathe-doc#723 Signed-off-by: Abrar Rahman Protyasha <abrar@openrobotics.org>
However, don't use 4.1, as it will need a new release of breathe to work (e.g. see breathe-doc/breathe#711).
jakobandersen
force-pushed
the
handle_parsing_errors
branch
2 times, most recently
from
c0ea7f3
to
fe8ddb7
Compare
jakobandersen
force-pushed
the
handle_parsing_errors
branch
from
fe8ddb7
to
d5b1b13
Compare
|
@michaeljones, unless you object, I'll merge this later today or tomorrow. |
|
Thank you @michaeljones and @jakobandersen ! Can we have a release now, please? |
|
@vermeeren is the release manager at the moment and has been a little busy. I'm not sure when he might be able to do it. It seems reasonable to release though. |
|
@michaeljones @0x8000-0000 I can definitely make a new release when/if that is desired. Past months there have been some irl issues causing an unfortunate lack of activity from my end, I really do appreciate everyone's work on this (including also @jakobandersen of course). Definitely ok to ping me when it's time, perhaps after #735 is merged to avoid conflicts? I'll at the minimum finalise the changelog and make a formal release, not yet sure when personal issues are resolved fully after which I can get back to more proper maintainership. Thanks! |
|
Thank you for the reply @vermeeren and for sharing about your current situation. I hope you're doing ok and that the personal issues are manageable and go as well as can be. I'm happy to merge #735. It is a bit opinionated and I feel a bit out of place having wandered back in after such a time away but I think it has merit if only a small amount. I can merge in a day or two if no one has objections. @jakobandersen - do you have anything else you would like to see merged for a release? |
I think merging #735 and then updating with whatever actually changed since the last release is the only thing I think is needed before a release. Unfortunately I haven't had a habit of adding changelog entries so I'm not sure what is missing. |
|
I think this is the syntax for finding merges since the approximate time of the last release: https://github.com/michaeljones/breathe/pulls?q=is%3Apr+merged%3A%3E2021-05-05+ I'm happy to look through them and dig some stuff out. We could add a PR template with a checklist for updating the CHANGELOG if desirable. |
|
I've created PR with changelog & minor readme changes: #739 Hopefully when that is merged we'll be in a good place to release? |
|
I'd love to know how far are we from a release. |
|
@vermeeren - I think we're in a place to release now if you have the time at some point? |
|
@michaeljones The job is done, apologies for delay. Did some minor changelog update and also fixup CICD quick in #741. Thanks to all for the hard work! |
|
Thank you @vermeeren. Kind of you to take the time when you've got other things going on. |
|
Thank you @vermeeren and @michaeljones ! |
(This depends on / includes #698)
In a few places the C++ domain in Sphinx is used to parse fragments, but potential errors are not handled. This tries to fix that.
One case is the parsing of the function parameters in
doxygenfunction, so.. doxygenfunction:: f(would crash Breathe.The second case is parsing the function parameters in the XML, I don't know how to trigger that directly, but it should be handled now.
The third case is in the hax to insert the name in template parameters (#681). For parameters
typename ...Argsit will trigger a parsing error (observed in the pybind11 project).Note, the third case is only enabled when using Sphinx 4.1 which is still not released, but probably will soon'ish.