On 2025/09/14 14:57, Jan Klemkow wrote:
> ping?
> 
> On Tue, Aug 12, 2025 at 06:29:56PM +0200, Jan Klemkow wrote:
> > ping?
> > 
> > On Fri, Jul 25, 2025 at 07:31:47PM +0200, Jan Klemkow wrote:
> > > The following diff replaces the internals of parse8601() by an
> > > equivalent call of strptime(3) to parse the time string.
> > > 
> > > The only functional difference is the 2-digit year handling:
> > > Values in the range 69-99 refer to years 1969 to 1999 and
> > > values in the range 00-68 refer to years 2000 to 2068.
> > > In the old version, 69-99 also refer to the 2069 to 2099 range.

I think this approach makes sense, having fewer places in the tree where
we might need to update 2-digit year handling in the future is useful.

> > > +	if (s != t) {
> > > +		switch (t == NULL ? strlen(s) : t - s) {
> > > +		case 8: strlcat(format, "%C", sizeof format);
> > > +		case 6: strlcat(format, "%y", sizeof format);
> > > +		case 4: strlcat(format, "%m", sizeof format);
> > > +		case 2: strlcat(format, "%d", sizeof format);
> > >  		case 0:

style(9) asks for /* FALLTHROUGH */ comments and I think that makes
sense.

otherwise ok

> > >  			break;
> > >  		default:
> > > -			return (-1);
> > > +			return -1;
> > >  		}
> > > +	}
> > >  
> > > -		/* sanity check */
> > > -		if (tm.tm_sec < 0 || tm.tm_sec > 60 || tm.tm_min < 0 ||
> > > -		    tm.tm_min > 59 || tm.tm_hour < 0 || tm.tm_hour > 23)
> > > -			return (-1);
> > > +	if (t != NULL) {
> > > +		strlcat(format, "T", sizeof format);
> > > +
> > > +		switch (strlen(t)) {
> > > +		case 7: strlcat(format, "%H", sizeof format);
> > > +		case 5: strlcat(format, "%M", sizeof format);
> > > +		case 3: strlcat(format, "%S", sizeof format);
> > > +		case 1:
> > > +			break;
> > > +		default:
> > > +			return -1;
> > > +		}
> > >  	}
> > > -	return (mktime(&tm));
> > > +
> > > +	if (strptime(s, format, tm) == NULL)
> > > +		return -1;
> > > +
> > > +	return mktime(tm);
> > >  }
> > >  
> > >  /*-
> > > 
> > 
>