Re: RFE: printf '%(fmt)T' prints current time by default
I think the ksh behavior is makes more sense so can we use the current time as the default? -Clark I agree that a null or empty argument as equivalent to -1 is a better default. 0 is identical to the current behavior for empty/unset, so no functionality is lost. That's not unreasonable. The current default is what Posix specifies for printf: Any extra c or s conversion specifiers shall be evaluated as if a null string argument were supplied; other extra conversion specifications shall be evaluated as if a zero argument were supplied. Additionally, an empty format in ksh is equivalent to the date(1) default for the current locale. So, LC_TIME=C; [[ $(printf '%()T') == $(date) ]] is true. Bash uses the strftime '%X' conversion, which is described as the locale's appropriate time representation. Chet -- ``The lyf so short, the craft so long to lerne.'' - Chaucer ``Ars longa, vita brevis'' - Hippocrates Chet Ramey, ITS, CWRUc...@case.eduhttp://cnswww.cns.cwru.edu/~chet/
Re: RFE: printf '%(fmt)T' prints current time by default
On Friday, December 14, 2012 09:57:11 AM Chet Ramey wrote: I think the ksh behavior is makes more sense so can we use the current time as the default? -Clark I agree that a null or empty argument as equivalent to -1 is a better default. 0 is identical to the current behavior for empty/unset, so no functionality is lost. That's not unreasonable. The current default is what Posix specifies for printf: Any extra c or s conversion specifiers shall be evaluated as if a null string argument were supplied; other extra conversion specifications shall be evaluated as if a zero argument were supplied. Ooh ok... hrm I didn't consider it's actually consistent with everything else this way. -- Dan Douglas
Re: RFE: printf '%(fmt)T' prints current time by default
On 12/14/12 12:03 PM, Dan Douglas wrote: On Friday, December 14, 2012 09:57:11 AM Chet Ramey wrote: I think the ksh behavior is makes more sense so can we use the current time as the default? -Clark I agree that a null or empty argument as equivalent to -1 is a better default. 0 is identical to the current behavior for empty/unset, so no functionality is lost. That's not unreasonable. The current default is what Posix specifies for printf: Any extra c or s conversion specifiers shall be evaluated as if a null string argument were supplied; other extra conversion specifications shall be evaluated as if a zero argument were supplied. Ooh ok... hrm I didn't consider it's actually consistent with everything else this way. I'm not saying I can't change the %T default -- I probably will, and it will be documented as an exception in the man page. The above is the explanation for the status quo. Chet -- ``The lyf so short, the craft so long to lerne.'' - Chaucer ``Ars longa, vita brevis'' - Hippocrates Chet Ramey, ITS, CWRUc...@case.eduhttp://cnswww.cns.cwru.edu/~chet/
