[cfe-dev] clang-format chromium ternary operator

[Nico Weber]

On Mon, Aug 12, 2013 at 11:07 PM, Peter Kasting <pkasting at chromium.org>wrote:

> On Mon, Aug 12, 2013 at 1:57 PM, Daniel Jasper <djasper at google.com> wrote:
>
>> Finally, regardless of all other arguments, and even ignoring the fact
>>> that the above codesearch links included code they shouldn't: there's a way
>>> of formatting these in Chromium code that is clearly more common even by a
>>> pessimistic search, and is clearly compliant with the style guide.
>>>  Therefore, the auto-formatter for Chromium code should use this pattern.
>>>  It doesn't matter if your way is legal or not, the most important rule in
>>> the whole style guide is "be consistent", which this does not do as well as
>>> it could.
>>>
>>
>> I understand your argument and it is perfectly valid. However, also
>> consider other sorts of consistency: E.g. there are a lot of users that
>> have to develop in both Chromium- and Google-style code. For them, any
>> (unnecessary) inconsistency is harmful.
>>
>
> It seems clearly far more important that Chromium code be internally
> consistent than that the formatter begin preferring a new, less-used style
> over an existing, common, explicitly-valid style regardless of which way is
> common in Google internal code.  This is true within other Google projects
> as well.
>
> I personally don't have any strong feelings about this (I for one would be
>> happy with disallowing all multi-line conditional expressions). The
>> decision to go this way is mostly that complex conditional expressions need
>> as much structure as they can get. I know that we have other Chromium
>> engineers that are happy enough with this. Is there a decision making
>> process for Chromium style?
>>
>
> I don't see what "structure" you're buying with this style; it seems like
> this:
>
> aaaaa ?
>     bbbbb :
>     ccccc;
>
> ...is just as "structured" as what you're asking for, but aligns with the
> common Chromium idiom.
>
> If you are not willing to accept my argument on its face and simply make
> this change, the next step is to escalate to chromium-dev.
>
> But in the case where you "don't have any strong feelings" and I do, and
> I've been working on the codebase for 7 years now, and you accept my
> argument that the style I'm asking for is valid, I don't see what you gain
> by doing so.
>

I told Daniel that we generally follow google-internal style unless there
are strong reasons not to. This issue has been discussed at length for
google-internal style, and I personally don't see a strong reason to
deviate here.

Nico


>
> PK
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/cfe-dev/attachments/20130812/e3be3216/attachment.html>