Re: Fw: [PATCH 1/2] strftime.c(__strftime): add %i, %q, %v, tests; tweak %Z docs

On 2022-10-21 06:12, Corinna Vinschen wrote:
> On Oct 20 16:07, Brian Inglis wrote:
>> On 2022-10-18 08:03, Corinna Vinschen wrote:
>>> On Sep 19 11:51, C Howland wrote:
>>>> On Saturday, September 17, 2022 1:00 AM, Brian Inglis wrote:
>>>>> newlib/libc/time/strftime.c(__strftime):
>>>>> %i year in century [00..99] Synonym for "%y". Non-POSIX extension.
>>>>> [tm_year]
>>>>> %q GNU quarter of the year (from `<<1>>' to `<<4>>') [tm_mon]
>>>>> %v OSX/Ruby VMS/Oracle date "%d-%b-%Y". Non-POSIX extension. [tm_mday,
>>>>> tm_mon, tm_year]
>>>>> add %i %q %v tests
>>>>> %Z clarify current time zone *abbreviation* not "name" [tm_isdst]
>>>>> ---
>>>>>    newlib/libc/time/strftime.c | 67 +++++++++++++++++++++++++++++++++++--
>>>>>    1 file changed, 64 insertions(+), 3 deletions(-)
>>
>>> - %i: Where is used and documented?  I don't see this in glibc, not even
>>>     in the latest from the glibc git repo.
>>
>> Some portable language implementations for non-Unix platforms that I came
>> across and documented but did not keep track of, and now can't find!
> 
> So we don't have that in either BSD, nor GNU, nor POSIX, but only
> in other non-POSIX platforms.  Hmm.  Along these lines, we could also
> make a case for supporting Microsoft's %I64 printf format and I'm
> not so hot on that...
> 
> For the time being, please let's drop %i.  Defining it now may result in
> accidental collisions with the POSIX standard, should it get extended.
> 
>>> - %q: Ditto. For a GNU extension, it's surprisingly absent from the
>>>     most recent glibc code, or do I miss something?
>>
>> Looks like this may have been lost in the discussion:
>>
>> https://inbox.sourceware.org/libc-alpha/[email protected]/
>>
>> but as he says, it was already in gnulib, and made it into date the same month:
>>
>> https://git.savannah.gnu.org/cgit/coreutils.git/diff/?id=30012b290facf66551cdf395ace397903d00483d
> 
> Ok, that's fine then.
> 
>>> - %v: OSX/Ruby?  Isn't that already gone?  Also, it introduces another
>>>     ambiguous date format where %F or equivalent should be used instead.
>>
>> All the BSDs, Darwin, Oracle, and Ruby support it: it is localized, but not
>> ambiguous, as it's dd-Mon-yyyy.
> 
> I found it in all three BSDs with a comment
> 
>      /*
>      ** From Arnold Robbins' strftime version 3.0:
>      ** "date as dd-bbb-YYYY"
>      ** (ado, 1993-05-24)
>      */
> 
> Also, it's defined as a recursion with the format "%e-%b-%Y",
> unconditionally (no padding, etc).  Maybe we should replicate that
> verbatim (including the comment) to keep in line with BSD.
> 
> Also, the BSDs do *not* define %v in strptime, so I think we shouldn't
> do this either.  It may accidentally collide with a standards extension,
> just as %i.

Consensus seems to be that, as it's been claimed by BSDs, it may be standardized 
or will have to be avoided: but no problem with your decision to drop from 
strptime.

Noticed "%+" in man-pages-linux and BSD strftime(3) defined as default locale 
(C/POSIX) date(1) format like as/ctime with %Z - "%a %b %e %T %Z %Y".
Presumably added so that the previous default locale output could easily be 
provided for backward compatibility once %c %x %X could be localized.
Should I also add this to our newlib strftime patch?

Are there pointers to that default locale and "%c" format in newlib and Cygwin?
I will search and check but could miss something defined away in the wild.

Presumably I should also drop %i and %v from Cygwin strptime.cc patch on 
cygwin-patches?

-- 
Take care. Thanks, Brian Inglis			Calgary, Alberta, Canada

La perfection est atteinte			Perfection is achieved
non pas lorsqu'il n'y a plus rien à ajouter	not when there is no more to add
mais lorsqu'il n'y a plus rien à retirer	but when there is no more to cut
			-- Antoine de Saint-Exupéry