Re: [PATCH v2 1/2] ctime.3: mktime() may modify tm_hour due to tm_isdst

["Alejandro Colomar (man-pages)" <alx.manpages@gmail.com>]

Hi Paul,

On 10/11/21 5:36 PM, Paul Eggert wrote:
> On 10/11/21 4:12 AM, Alejandro Colomar wrote:
>> +If the initial value of
>> +.I tm_isdst
>> +is inconsistent with the one set by
>> +.BR mktime (),
>> +.I tm_hour
>> +(and possibly other fields)
>> +will be modified to normalize the time to the correct DST.
> 
> I don't see why this change is necessary. mktime normalizes all its 
> input fields: there's nothing special about tm_isdst and tm_hour.
> 
> If normalization isn't explained clearly enough elsewhere in the man 
> page, that explanation should be fixed; there shouldn't be special-case 
> wording that implies that this is the only special case.

Hmm, you're right.  I think I was misled by the wording "if structure 
members are outside their valid interval", which led me to think that 
since 08h was between 0 and 23 it shouldn´t be affected, but the 
normalization goes beyond that and interprets "valid range" as a more 
general concept so that the full time has to correspond to a valid time 
for a given timezone.

I'm not sure how to reword it.  I'll keep it with the original text, and 
keep thinking about it.

Thanks,

Alex


-- 
Alejandro Colomar
Linux man-pages comaintainer; https://www.kernel.org/doc/man-pages/
http://www.alejandro-colomar.es/