Re: Looking for comments on changing strptime(3) %p behavior

[Mouse <mouse%Rodents-Montreal.ORG@localhost>]

> b. strptime(str, "%p", &tm)

>    Currently if str contains a am/pm string and no hour is specified
>    then tm_hour is adjusted as per above or an error is returned
>    based on tm_hour.  This would seem to be a bug on both counts.

>    The %p should be parsed and only applied to a valid 12-hour clock
>    value obtained via the same parse.

I think I disagree.  What you suggest leaves no way for someone to
parse a 12-hour time and then, later, correct for an AM/PM indicator,
short of either (1) doing one or both of those operations some way than
strptime or (2) assembling a single string containing (at least) the
hour and the AM/PM indicator and strptime()ing it.  This seems
suboptimal to me, especially since the current implementation supports
that operation just fine.

Does XPG specify what happens when %p is used without an hour spec, or
with %H but not %I?  (And what if both %H and %I are used?)

> c. strptime("12", "%I", &tm)

>    Currently the 12 is converted tm_hour = 0.  This seems rather odd
>    since %I must be in the range of 1-12.

Well, tm_hour is a 24-hour hour figure; I'm not sure what else you
think a 12-hour "12" should convert into.

Unfortunately, fixing this right seems to me to require changing struct
tm so it contains a 12-hour hour, a 24-hour hour, and an AM/PM
indicator.  But that would be a significant API (not to mention ABI)
change, and may well conflict with XPG.

/~\ The ASCII				  Mouse
\ / Ribbon Campaign
 X  Against HTML		mouse%rodents-montreal.org@localhost
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B