Value in maintaining separate C++ style guide for Blink and having bot report comments?

[Colin Blundell]

Hi folks,

There's a Blink style guide bot reviewer that kicks in on Blink CLs. It has useful feedback e.g. on which containers to use. It also has specific Blink C++ style guide comments like "accessors in Blink should be named like <x>" and whatnot that are often pre-existing things outside of the scope of the CL itself. I think that as we move toward more automated review comments, we want to be very careful to limit noise. Does anyone have any context on the continued existence of a Blink-specific C++ style guide apart from specific Blink concerns such as using wtf containers when relevant, refcounting idioms, etc?

Thanks,

Colin

[Nate Chapin]

The Blink style I feel most strongly about is our CamelCase rule. Having javascript-exposed methods be lowercase (and nothing else) clearly indicates what is a "public API" where it otherwise isn't obvious at all.

--
You received this message because you are subscribed to the Google Groups "platform-architecture-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to platform-architect...@chromium.org.
To view this discussion visit https://groups.google.com/a/chromium.org/d/msgid/platform-architecture-dev/CAMGE5NEU_G2EC83-UOEcjpuYDdFayf%3DdCmO2t454Q6a2LK-BZQ%40mail.gmail.com.

[Ian Kilpatrick]

+1 to Nate. Also the new/delete GC stuff is important, Create & public ctors etc.

(I'm also partial to the bool rule).

Additionally if we remove rules etc there has to be a plan to migrate existing code to the new way, which is likely more effort than it's worth.

To view this discussion visit https://groups.google.com/a/chromium.org/d/msgid/platform-architecture-dev/CACf%3D2LL24niae4sqW8J6Pg9AZwzwkO5cEEuDgHfmDT%2B6ty0Stg%40mail.gmail.com.

[Colin Blundell]

Thanks for the feedback and the link to the style guide! Looking through it, in my experience the "May leave obvious parameter names out of function declarations" is one that causes the style guide reviewer bot to add a non-trivial amount of review comments. What would folks think of just instructing the style guide reviewer bot not to comment on that rule in particular (given that it's a "may")?

Best,

Colin