Proposal: allow #elifdef and #elifndef

[Victor Vianna]

Hi cxx,

The C++23 change to the styleguide just landed, but ~all features are in the "TBD" pool right now. I would like to move some of them to the "allowed" pool before announcing cpp23 availability to chromium-dev@ (otherwise the announcement is likely to just be ignored).

I propose to allow #elifdef and #elifndef (docs). An example usage can be seen here.

-- Victor

[Joe Mason]

SGTM!

--
You received this message because you are subscribed to the Google Groups "cxx" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cxx+uns...@chromium.org.
To view this discussion visit https://groups.google.com/a/chromium.org/d/msgid/cxx/CANFyMwxtVuv4jAFA7aYpc4XB8xsRDuEr3muE9pHk3Y8nHAxfcg%40mail.gmail.com.

[David Benjamin]

SGTM as well

To view this discussion visit https://groups.google.com/a/chromium.org/d/msgid/cxx/CAH%3DT95Rkx_2HA2F3BTFuKjXgV3EdaX7ZHEjYNmGKebMxjAr8eg%40mail.gmail.com.

[Mark Mentovai]

I’m OK with this in general.

Be aware that we may have tools (PRESUBMITs, maybe?) with baked-in patterns to match preprocessor directives like these, and we should make sure that they handle the new tokens correctly before changing the rule.

To view this discussion visit https://groups.google.com/a/chromium.org/d/msgid/cxx/CAF8qwaBrrdVJi5M8kobfQP9ULZuHc%3DciNR%3Dz1os5NL71xb6kUw%40mail.gmail.com.

[Avi Drissman]

SGTM2

To view this discussion visit https://groups.google.com/a/chromium.org/d/msgid/cxx/CAHk1GDwfp2rLPQbawO%3DsnZc45T1j837TdeGYt0AQuT%3DUDwbuiQ%40mail.gmail.com.

[Victor Vianna]

> Be aware that we may have tools (PRESUBMITs, maybe?) with baked-in patterns to match preprocessor directives like these, and we should make sure that they handle the new tokens correctly before changing the rule.

Good point. With this query, the closest I could find was this list, but it doesn't require updating because WebGL has its own set of preprocessor directives IIUC.

I also checked that `git cl format` works correctly with #elifdef, so I will go ahead and land crrev.com/c/7362614

[Mark Mentovai]

Victor Vianna wrote:

> Be aware that we may have tools (PRESUBMITs, maybe?) with baked-in patterns to match preprocessor directives like these, and we should make sure that they handle the new tokens correctly before changing the rule.

Good point. With this query, the closest I could find was this list, but it doesn't require updating because WebGL has its own set of preprocessor directives IIUC.

I also checked that `git cl format` works correctly with #elifdef, so I will go ahead and land crrev.com/c/7362614

I was thinking about things like PRESUBMIT.py _CheckForInvalidIfDefinedMacrosInFile, which your query didn’t seem to turn up, because it didn’t mention elif in so many wordsbytes.

[David Benjamin]

On Tue, Jan 6, 2026 at 12:07 PM Mark Mentovai <ma...@chromium.org> wrote:

Victor Vianna wrote:

> Be aware that we may have tools (PRESUBMITs, maybe?) with baked-in patterns to match preprocessor directives like these, and we should make sure that they handle the new tokens correctly before changing the rule.

Good point. With this query, the closest I could find was this list, but it doesn't require updating because WebGL has its own set of preprocessor directives IIUC.

I also checked that `git cl format` works correctly with #elifdef, so I will go ahead and land crrev.com/c/7362614

I was thinking about things like PRESUBMIT.py _CheckForInvalidIfDefinedMacrosInFile, which your query didn’t seem to turn up, because it didn’t mention elif in so many wordsbytes.

At a glance, that one might actually be fine with the .*, though adding a test is probably prudent. :-)

[Victor Vianna]

> At a glance, that one might actually be fine with the .*, though adding a test is probably prudent. :-)

It is, adding the test in crrev.com/c/7390218

I replaced elif with ifdef in my original query and got this. PRESUBMIT.py _CheckForInvalidIfDefinedMacrosInFile was the only new case I found, so hopefully we're good now