doc: Revert back to using breathe instead of docleaf#61339
doc: Revert back to using breathe instead of docleaf#61339fabiobaltieri merged 5 commits intozephyrproject-rtos:mainfrom
Conversation
|
@michaeljones please provide your feedback here |
I think it's been provided in #61284 (comment) already |
|
Yep, seems to be a necessary step given the license. Sorry that it wasn't clearer from the start. If there is the interest and the funding behind Zephyr for such things then I'd be happy to discuss a license exception that could also apply to Zephyr derivatives. That said, it sounds like you have a path forward exploring a Doxygen focused solution. |
|
@michaeljones is breathe going to be actively maintained now that docleaf exists? |
|
@gmarull - There are two partially active maintainers on Breathe. I won't be actively working on it (and honestly haven't meaningfully in a long time) but they might. Development is slow and primarily responds to changes in Sphinx to my knowledge. They would benefit from funding if that is an option. |
Mmm they're not that recent actually, so not sure why they are now failing (@gmarull @michaeljones any brilliant idea as to why this could be the case?). I guess I could add these as known warning but... Also, not sure they are meant to be internal either (although they might better off handled using helper functions just like other Marking as DNM for now and will try to investigate more. |
include/zephyr/net/wifi.h
Outdated
There was a problem hiding this comment.
actually... why do we have static arrays defined in headers? This should be in a C file...
There was a problem hiding this comment.
+1, I actually raised the issue to @krish2718 in this discussion #60686 (comment)
There was a problem hiding this comment.
This originally was an inline function but was later changed to static arrays, I can move to wifi_shell.c and as extern (As they might be used by others).
There was a problem hiding this comment.
If it's is desirable it can be used by others, and for consistency, would it make sense to turn it into a wifi_ps_txt() helper function, similar to how it's done for e.g.
zephyr/include/zephyr/net/wifi.h
Lines 52 to 75 in 4cbb9d8
Thanks!
There was a problem hiding this comment.
wifi_security_txt this was the original code, and wifi_ps_txt was also added but due to a review comment went with static array. So, will take care of converting this to static array and moving to C file.
There was a problem hiding this comment.
So, will take care of converting this to static array and moving to C file.
Thanks! Please ping me when done so that I can update this PR accordingly.
There was a problem hiding this comment.
Sure, but lets not wait for that PR to be ready soon :), it might be a while.
|
FWIW I am waiting for #60686 to be merged so that the change from 195c01be62012811aaade92e703066fed268c5dd can be taken care off there rather than in this PR. |
This reverts commit 88c1519. Signed-off-by: Benjamin Cabé <benjamin@zephyrproject.org>
This reverts commit a3b7102. Signed-off-by: Benjamin Cabé <benjamin@zephyrproject.org>
This reverts commit 10cb87d. Signed-off-by: Benjamin Cabé <benjamin@zephyrproject.org>
This reverts commit 30c3ce4. Signed-off-by: Benjamin Cabé <benjamin@zephyrproject.org>
This reverts commit 20986ec. Signed-off-by: Benjamin Cabé <benjamin@zephyrproject.org>
|
@gmarull @jhedberg please revisit, I've now rebased to incorporate the latest changes in wifi.h that make breathe happy again. |
This reverts the subset of commits from #59570 that introduced docleaf and fixes #61284.
Also added a commit marking some apparently unsupported and recently added constructs in
wifi.h(namely, arrays with designated initializers) as INTERNAL_HIDDEN as they seem to be causing an issue (and are AFAICT meant to be internal anyway).CI-built doc website: https://builds.zephyrproject.io/zephyr/pr/61339/docs/