Ways to get number of seconds from Epoch

M.-A. Lemburg mal at lemburg.com
Sat May 4 14:05:49 EDT 2002


Mark McEahern wrote:
> 
> Mark,
> 
> Regarding this format:
> 
>   [15/Apr/2002:08:28:56 +0200]
> 
> I wrote:
> 
> > That format seems a little odd:
> 
> [brueckd at tbye.com] replied:
> > It is a little odd, but it's also how Common Log Format does timestamps:
> >
> > http://www.bacuslabs.com/WsvlCLF.html

It's an odd format indeed... I wonder why people have to reinvent
new date formats all the time, the ISO format is both eas to read
and easy to parse.
 
> When you use mx.DateTime.DateTimeFrom on that format, here's what happens:
> 
> >>> from mx.DateTime import DateTimeFrom
> >>> s = "[15/Apr/2002:08:28:56 +0200]"
> >>> dt = DateTimeFrom(s)
> >>> str(dt)
> '2002-04-01 02:08:28.00'
> 
> So my question is whether that's a format mx.DateTime should support?  It's
> easy enough to massage the data so that it's in a more recognizable format
> before passing it to mx.DateTime.
> 
> I'm not personally invested in the answer, I just wanted to alert you to a
> question that came up on c.l.py.

I'll see whether the parse can be extended to support this 
format as well. The usual problem with this is that it support
*very* many different formats (including partially broken ones),
so extending it is not as easy as you might expect.

-- 
Marc-Andre Lemburg
CEO eGenix.com Software GmbH
______________________________________________________________________
Company & Consulting:                           http://www.egenix.com/
Python Software:                   http://www.egenix.com/files/python/





More information about the Python-list mailing list