[patch] LWG 2019 - std::isblank<C>(C, const std::locale&)

[Jonathan Wakely]

Re-sending as plain text ...

On 22 October 2014 00:15, Jonathan Wakely <jwakely.gcc@gmail.com> wrote:
>
> On Oct 21, 2014 6:33 PM, "Tim Shen" <timshen@google.com> wrote:
>>
>> On Tue, Oct 21, 2014 at 5:26 AM, Jonathan Wakely <jwakely@redhat.com>
>> wrote:
>> > I don't think we can remove _RegexMask entirely, as we still need
>> > special handling for the "w" class, so maybe something like the
>> > attached.
>> >
>> > Alternatively we could replace _M_extended with bool _M_under and not
>> > need to do any bitwise operations like _M_extended & _S_valid_mask.
>>
>> I didn't take a deep look, so I'm probably wrong, but
>> <http://cplusplus.github.io/LWG/lwg-defects.html#2018> says that "w"
>> 's Corresponding ctype_base::mask value is ctype_base::alnum and,
>> everyone in Table X has its own builtin mask value.
>
> The example  implementation shows that "w" needs special handling to match
> underscore as well as the members of the alnum class.
>
>> Also, I use n3376 when I implemented this. I suppose we should update
>> it to the latest version?
>
> I don't think there are many changes in clause 28 but it would be good to
> check.