-
-
Notifications
You must be signed in to change notification settings - Fork 307
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
docs: RFC on C++17 minimum standard support #4566
base: main
Are you sure you want to change the base?
Conversation
Is there a way to write it such that the RFC doesn't have to be updated at each version? Such as the conditions that allow us to say: these are the supported ones without having to think more about it? For Python, the case is quite easy. |
I can only repeat what I said before on this matter: #4040 (comment). |
@echoix I would prefer the Python way as well, but I'm afraid @nilason is right that it is just more complicated for C and C++. Maybe it will change it in the future, but the nature of the ecosystem is just very different now. One thing which we could do differently is updating the same RFC instead of creating a new one. I don't know if that's a good idea here or in general. For things like release schedule it seems to make sense, though. |
It is probable we will want to update C and C++ minimum standard only every 5–10 years, so a new RFC weighing in all factors is not such a big deal I think. The current standard should be mentioned in doc/development/style_guide.md. For reference eg. GDAL RFCs work in the same way, without "updating". Each RFC is a document in itself, with its history and adoption procedure. |
@nilason, do you want to make any changes to tested C and C++ version in relation to this? The gcc.yml workflow is already using |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I have some wording suggestions, but otherwise we are only testing C++17 in gcc.yml and our dependencies already require C++17, so I don't think there is much to discuss.
Co-authored-by: Vaclav Petras <[email protected]>
Thanks for the good suggestions. In practice, also the mac runner is C++17, as that is the default with Clang 19. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
If I recall correctly, the procedure is that the PSC needs to vote here by approving the PR. Please, check PSC mailing list for the last RFC voting.
I guess we could merge the draft and then only PSC-vote to "un-draft" it. I don't see much advantage in that.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks @nilason!
Co-authored-by: Veronica Andreo <[email protected]>
Addresses #4040.
For reference:
GDAL RFC98
PDAL 2.4
GEOS#1144