Re: [PATCH 06/11] gdb: Use std::string_view instead of gdb::string_view

[Simon Marchi <simark@simark.ca>]

On 10/17/23 10:13, Lancelot SIX wrote:
> Hi,
> 
> On 16/10/2023 22:51, Pedro Alves wrote:
> 
>>> Lancelot> +         std::string_view name;
>>> Lancelot> +         if (const char *name_str = keyname (ch);
>>> Lancelot> +             name_str != nullptr)
>>>
>>> We should probably discuss whether to allow this construct before
>>> letting it in.
>>>
>>> Personally I'm ok with it.
>>
>> I think it may look a little weird if you're not used to it (like I am not), but
>> OTOH it isn't really different from introducing a variable to scope with "for", like:
>>
>>   for (const char *name_str = keyname (ch);
>>        name_str != nullptr; ....)
>>     ....
>>
>> so I suspect we'll all just get used to it and think nothing of it in a few years.
>>
>> So am I fine with it too.
>>
> 
> I don't mind not using this for now.  The "name_str" will be declared outside of the if scope for now.
> 
> If this construct is comes in, there will also be the question of coding style (i.e. single line if line length permits or multiple lines as I did here).  I guess we could go the same way we do for "for" and prefer single line.

I didn't even know this construct existed, I was a bit shocked at first
:P.  But once you said "it's just like for", it looked just fine to me.

Declaring variables in the narrowest scope possible has its technical
merits, so I think we should allow that construct.  I think it's fine to
put it on one line if it fits.

Simon