Re: strptime() inconsistency

[Ben Abbott]

On Mar 3, 2010, at 3:44 PM, Ben Abbott wrote:

> I've noticed that behavior of strptime() is not consistent for me on MacOS X.
> 
> Running the MacOS X binary for 3.2.3 from OctaveForge
> 
> fmt = "%d-%b-%Y";
> str = "17-Sep-1789";
> [tm, nc] = strptime (str, fmt)
> 
> tm =
> {
>  usec = 0
>  sec = 0
>  min = 0
>  hour = 0
>  mday =  17
>  mon =  8
>  year =  89
>  wday = 0
>  yday =  259
>  isdst = 0
>  zone = JST
> }
> 
> nc =  12
> 
> However, the 3.2.3 binary I built using gcc 4.4.2, I obtain the result below. 
> When I build the current sources using gcc 4.2.1, I also see the result below.
> 
> tm =
> {
>  usec = 0
>  sec = 0
>  min = 0
>  hour = 0
>  mday =  17
>  mon =  8
>  year = 0
>  wday = 0
>  yday = 0
>  isdst = 0
>  zone = unknown
> }
> 
> nc = 0
> 
> This problem only occurs for dates prior to 01-Jan-1900.
> 
> Does anyone have an idea what might be the cause of this inconsistent 
> behavior?
> 
> Ben

I noticed the MacOS X has its own strptime.

        
http://developer.apple.com/mac/library/documentation/Darwin/Reference/ManPages/man3/strptime.3.html

Can someone tell me how I may determine if Octave is calling the version from 
gnulib or the one included with MacOS X?

Ben