llvm dev - Jul 2018 - RFC: should we spell lambdas like functions?

> On Jul. 5, 2018, at 18:12, Bruce Hoult via llvm-dev <llvm-dev at
lists.llvm.org> wrote:
> 
> 
> 
> On Thu, Jul 5, 2018 at 5:30 PM, Adrian Prantl via llvm-dev <llvm-dev at
lists.llvm.org <mailto:llvm-dev at lists.llvm.org>> wrote:
> 
> 
> > On Jul 5, 2018, at 1:09 PM, Rui Ueyama via llvm-dev <llvm-dev at
lists.llvm.org <mailto:llvm-dev at lists.llvm.org>> wrote:
> > 
> >> On Thu, Jul 5, 2018 at 12:57 PM Duncan P. N. Exon Smith
<dexonsmith at apple.com <mailto:dexonsmith at apple.com>> wrote:
> >> I argue we should spell C++ lambdas (and other function-like
variables) like functions, not like variables.
> >> 
> >> - Use verbs, not nouns.
> > 
> > I think I agree with this.
> >  
> >> - Use lowerCamelCase.
> > 
> > In lld we use UpperCamelCase, as they are technically not function
names but variable names. Using lowerCamelCase for function pointers feels
really weird to me, as they are really variables than functions to me.
> 
> I also agree that this is weird either way because they are variables when
declared and passed around and functions when invoked.
> 
> How is this different to function pointers in traditional C?
>  
It's not different.  The patch proposes to spell C++ lambdas *and other
function-like variables* like functions.  Essentially, anything that's
callable.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20180710/fad91007/attachment.html>

On Tue, Jul 10, 2018 at 4:21 PM Duncan P. N. Exon Smith <
dexonsmith at apple.com> wrote:
>
>
> On Jul. 5, 2018, at 18:12, Bruce Hoult via llvm-dev <
> llvm-dev at lists.llvm.org> wrote:
>
>
>
> On Thu, Jul 5, 2018 at 5:30 PM, Adrian Prantl via llvm-dev <
> llvm-dev at lists.llvm.org> wrote:
>
>>
>>
>> > On Jul 5, 2018, at 1:09 PM, Rui Ueyama via llvm-dev <
>> llvm-dev at lists.llvm.org> wrote:
>> >
>> >> On Thu, Jul 5, 2018 at 12:57 PM Duncan P. N. Exon Smith <
>> dexonsmith at apple.com> wrote:
>> >> I argue we should spell C++ lambdas (and other function-like
>> variables) like functions, not like variables.
>> >>
>> >> - Use verbs, not nouns.
>> >
>> > I think I agree with this.
>> >
>> >> - Use lowerCamelCase.
>> >
>> > In lld we use UpperCamelCase, as they are technically not function
>> names but variable names. Using lowerCamelCase for function pointers
feels
>> really weird to me, as they are really variables than functions to me.
>>
>> I also agree that this is weird either way because they are variables
>> when declared and passed around and functions when invoked.
>>
>
> How is this different to function pointers in traditional C?
>
>
>
> It's not different.  The patch proposes to spell C++ lambdas *and other
> function-like variables* like functions.  Essentially, anything that's
> callable.
>
That seems a bit more awkward/not-clearly-good to me (though I do
appreciate that the standard C++ naming convention, for all that it doesn't
really differentiate between anything (except macros) by having everything
lower_case, does use the same convention between variables and functions so
you don't get that disconnect) - ending up with function-named variables
that don't behave like functions (get used in assignments, have member
functions called on them, compared to nullptr (weak functions
notwithstanding), etc).

Seems more like a narrow kind of hungarian notation to me - to designate
that this variable can be called. Which doesn't seem important enough to
callout with a separate naming convention.

But I'm not too fussed about it if others feel strongly.

- Dave
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20180710/9e99ed47/attachment.html>

I wrote that I prefer CamelCase over camelCase for callable objects, but I
don't have a strong opinion. If the majority prefers camelCase, I'm
happy
to update the existing lld's code to conform to the new coding style.

On Tue, Jul 10, 2018 at 4:25 PM David Blaikie via llvm-dev <
llvm-dev at lists.llvm.org> wrote:
>
> On Tue, Jul 10, 2018 at 4:21 PM Duncan P. N. Exon Smith <
> dexonsmith at apple.com> wrote:
>
>>
>>
>> On Jul. 5, 2018, at 18:12, Bruce Hoult via llvm-dev <
>> llvm-dev at lists.llvm.org> wrote:
>>
>>
>>
>> On Thu, Jul 5, 2018 at 5:30 PM, Adrian Prantl via llvm-dev <
>> llvm-dev at lists.llvm.org> wrote:
>>
>>>
>>>
>>> > On Jul 5, 2018, at 1:09 PM, Rui Ueyama via llvm-dev <
>>> llvm-dev at lists.llvm.org> wrote:
>>> >
>>> >> On Thu, Jul 5, 2018 at 12:57 PM Duncan P. N. Exon Smith
<
>>> dexonsmith at apple.com> wrote:
>>> >> I argue we should spell C++ lambdas (and other
function-like
>>> variables) like functions, not like variables.
>>> >>
>>> >> - Use verbs, not nouns.
>>> >
>>> > I think I agree with this.
>>> >
>>> >> - Use lowerCamelCase.
>>> >
>>> > In lld we use UpperCamelCase, as they are technically not
function
>>> names but variable names. Using lowerCamelCase for function
pointers feels
>>> really weird to me, as they are really variables than functions to
me.
>>>
>>> I also agree that this is weird either way because they are
variables
>>> when declared and passed around and functions when invoked.
>>>
>>
>> How is this different to function pointers in traditional C?
>>
>>
>>
>> It's not different.  The patch proposes to spell C++ lambdas *and
other
>> function-like variables* like functions.  Essentially, anything
that's
>> callable.
>>
>
> That seems a bit more awkward/not-clearly-good to me (though I do
> appreciate that the standard C++ naming convention, for all that it
doesn't
> really differentiate between anything (except macros) by having everything
> lower_case, does use the same convention between variables and functions so
> you don't get that disconnect) - ending up with function-named
variables
> that don't behave like functions (get used in assignments, have member
> functions called on them, compared to nullptr (weak functions
> notwithstanding), etc).
>
> Seems more like a narrow kind of hungarian notation to me - to designate
> that this variable can be called. Which doesn't seem important enough
to
> callout with a separate naming convention.
>
> But I'm not too fussed about it if others feel strongly.
>
> - Dave
> _______________________________________________
> LLVM Developers mailing list
> llvm-dev at lists.llvm.org
> http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20180710/9079f1a5/attachment-0001.html>