Re: YADTR (Yet Another DateTime Rant)

From	Ben Finney <ben+python@benfinney.id.au>
Subject	Re: YADTR (Yet Another DateTime Rant)
Date	2014-03-28 11:43 +1100
References	(3 earlier) <lh0u51$dj8$1@news.albasani.net> <mailman.8613.1395917059.18130.python-list@python.org> <roy-E22E44.08522427032014@news.panix.com> <5334b747$0$29994$c3e8da3$5496439d@news.astraweb.com> <roy-E6F0E7.20291727032014@news.panix.com>
Newsgroups	comp.lang.python
Message-ID	<mailman.8643.1395967408.18130.python-list@python.org> (permalink)

Show all headers | View raw

Roy Smith <roy@panix.com> writes:

> In article <5334b747$0$29994$c3e8da3$5496439d@news.astraweb.com>,
>  Steven D'Aprano <steve+comp.lang.python@pearwood.info> wrote:
>
> > On Thu, 27 Mar 2014 08:52:24 -0400, Roy Smith wrote:
> > > Give ma a real-life situation where you would want such behavior
> > > [the ‘datetime.timedelta.__str__’ method returning a string
> > > showing some portions negative and others positive].
> > 
> > Easy -- I'm debugging timedelta routines, and I want to easily see
> > that the timedeltas calculated match what I expect them to be when I
> > print them. The quickest, easiest and simplest way is for
> > str(timedelta) to follow the internal representation.
>
> That's what __repr__() is for.

Indeed, that is why Python's data model defines separate ‘__str__’ and
‘__repr__’ methods.

    object.__repr__(self)

    […] If at all possible, this should look like a valid Python
    expression that could be used to recreate an object with the same
    value (given an appropriate environment). If this is not possible, a
    string of the form <...some useful description...> should be
    returned.

    <URL:http://docs.python.org/3/reference/datamodel.html#object.__repr__>

    object.__str__(self)

    […] compute the “informal” or nicely printable string representation
    of an object. […] a more convenient or concise representation can be
    used.

    <URL:http://docs.python.org/3/reference/datamodel.html#object.__str__>

If you're a programmer wanting to know about the internal representation
of an object, call its ‘__repr__’ method (by calling ‘repr(foo)’).

If you're wanting to see a string representation that follows the
*semantics* of the object, call its ‘__str__’ method (by calling
‘str(foo)’).

If the ‘__str__’ method follows the internal representation at the
expense of the user's conceptual model, that's terrible (which I think
is what Roy means by “braindead”).

I wouldn't go as far as that insult, because it could be mere omission:

    The default implementation [of ‘__str__’] defined by the built-in
    type object calls object.__repr__().

So, where ‘str(foo)’ is returning an unhelpful representation that
follows the implementation, it could simply be that there is no
‘__str__’ defined for that type.

> > Oh look, that's exactly what the docs say:
> > 
> > "String representations of timedelta objects are normalized
> > similarly to their internal representation. This leads to somewhat
> > unusual results for negative timedeltas."
>
> Yes, I know that's what the docs say.  That's why it's not an 
> implementation bug.  It's a design bug :-)

One which to some extent violates the defined data model for Python
objects, and hence should be fixed.

-- 
 \      “I don't know half of you half as well as I should like, and I |
  `\   like less than half of you half as well as you deserve.” —Bilbo |
_o__)                                                          Baggins |
Ben Finney

Thread

YADTR (Yet Another DateTime Rant) Roy Smith <roy@panix.com> - 2014-03-25 20:58 -0400
  Re: YADTR (Yet Another DateTime Rant) Ethan Furman <ethan@stoneleaf.us> - 2014-03-25 18:19 -0700
  Re: YADTR (Yet Another DateTime Rant) Mark Lawrence <breamoreboy@yahoo.co.uk> - 2014-03-26 09:27 +0000
  Re: YADTR (Yet Another DateTime Rant) Skip Montanaro <skip@pobox.com> - 2014-03-26 08:37 -0500
    Re: YADTR (Yet Another DateTime Rant) Roy Smith <roy@panix.com> - 2014-03-26 08:04 -0700
      Re: YADTR (Yet Another DateTime Rant) Skip Montanaro <skip@pobox.com> - 2014-03-26 10:39 -0500
        Re: YADTR (Yet Another DateTime Rant) Marko Rauhamaa <marko@pacujo.net> - 2014-03-26 17:58 +0200
          Re: YADTR (Yet Another DateTime Rant) Skip Montanaro <skip@pobox.com> - 2014-03-26 11:34 -0500
            Re: YADTR (Yet Another DateTime Rant) Marko Rauhamaa <marko@pacujo.net> - 2014-03-26 19:13 +0200
            Re: YADTR (Yet Another DateTime Rant) Grant Edwards <invalid@invalid.invalid> - 2014-03-26 19:12 +0000
              Re: YADTR (Yet Another DateTime Rant) Ben Finney <ben+python@benfinney.id.au> - 2014-03-27 07:06 +1100
      Re: YADTR (Yet Another DateTime Rant) Roy Smith <roy@panix.com> - 2014-03-26 11:50 -0400
  Re: YADTR (Yet Another DateTime Rant) Antoon Pardon <antoon.pardon@rece.vub.ac.be> - 2014-03-26 15:02 +0100
    Re: YADTR (Yet Another DateTime Rant) Marko Rauhamaa <marko@pacujo.net> - 2014-03-26 16:43 +0200
  Re: YADTR (Yet Another DateTime Rant) Chris Angelico <rosuav@gmail.com> - 2014-03-27 01:12 +1100
    Re: YADTR (Yet Another DateTime Rant) Gregory Ewing <greg.ewing@canterbury.ac.nz> - 2014-03-27 11:08 +1300
  Re: YADTR (Yet Another DateTime Rant) Dennis Lee Bieber <wlfraed@ix.netcom.com> - 2014-03-26 19:25 -0400
    Re: YADTR (Yet Another DateTime Rant) Steven D'Aprano <steve+comp.lang.python@pearwood.info> - 2014-03-27 00:16 +0000
      Re: YADTR (Yet Another DateTime Rant) Chris Angelico <rosuav@gmail.com> - 2014-03-27 12:28 +1100
        Re: YADTR (Yet Another DateTime Rant) Roy Smith <roy@panix.com> - 2014-03-26 21:38 -0400
          Re: YADTR (Yet Another DateTime Rant) Mark Lawrence <breamoreboy@yahoo.co.uk> - 2014-03-27 08:40 +0000
      Re: YADTR (Yet Another DateTime Rant) Dan Sommers <dan@tombstonezero.net> - 2014-03-27 03:56 +0000
      Re: YADTR (Yet Another DateTime Rant) Johannes Bauer <dfnsonfsduifb@gmx.de> - 2014-03-27 11:22 +0100
        Re: YADTR (Yet Another DateTime Rant) Chris Angelico <rosuav@gmail.com> - 2014-03-27 21:44 +1100
          Re: YADTR (Yet Another DateTime Rant) Johannes Bauer <dfnsonfsduifb@gmx.de> - 2014-03-27 12:05 +0100
            Re: YADTR (Yet Another DateTime Rant) Chris Angelico <rosuav@gmail.com> - 2014-03-27 22:33 +1100
          Re: YADTR (Yet Another DateTime Rant) Johannes Bauer <dfnsonfsduifb@gmx.de> - 2014-03-27 12:25 +0100
            Re: YADTR (Yet Another DateTime Rant) Chris Angelico <rosuav@gmail.com> - 2014-03-27 22:42 +1100
            Re: YADTR (Yet Another DateTime Rant) Skip Montanaro <skip@pobox.com> - 2014-03-27 06:56 -0500
              Re: YADTR (Yet Another DateTime Rant) Marko Rauhamaa <marko@pacujo.net> - 2014-03-27 16:04 +0200
            Re: YADTR (Yet Another DateTime Rant) Chris Angelico <rosuav@gmail.com> - 2014-03-27 23:20 +1100
            Re: YADTR (Yet Another DateTime Rant) Skip Montanaro <skip@pobox.com> - 2014-03-27 08:01 -0500
            Re: YADTR (Yet Another DateTime Rant) Terry Reedy <tjreedy@udel.edu> - 2014-03-27 17:10 -0400
            Re: YADTR (Yet Another DateTime Rant) Ethan Furman <ethan@stoneleaf.us> - 2014-03-27 17:04 -0700
            Re: YADTR (Yet Another DateTime Rant) Terry Reedy <tjreedy@udel.edu> - 2014-03-27 21:46 -0400
              Re: YADTR (Yet Another DateTime Rant) Roy Smith <roy@panix.com> - 2014-03-27 21:59 -0400
                Re: YADTR (Yet Another DateTime Rant) Terry Reedy <tjreedy@udel.edu> - 2014-03-27 22:24 -0400
                Re: YADTR (Yet Another DateTime Rant) Chris Angelico <rosuav@gmail.com> - 2014-03-28 14:03 +1100
          Re: YADTR (Yet Another DateTime Rant) Roy Smith <roy@panix.com> - 2014-03-27 08:52 -0400
            Re: YADTR (Yet Another DateTime Rant) Antoon Pardon <antoon.pardon@rece.vub.ac.be> - 2014-03-27 15:12 +0100
            Re: YADTR (Yet Another DateTime Rant) Chris Angelico <rosuav@gmail.com> - 2014-03-28 01:15 +1100
            Re: YADTR (Yet Another DateTime Rant) Steven D'Aprano <steve+comp.lang.python@pearwood.info> - 2014-03-27 23:41 +0000
              Re: YADTR (Yet Another DateTime Rant) Roy Smith <roy@panix.com> - 2014-03-27 20:29 -0400
                Re: YADTR (Yet Another DateTime Rant) Ben Finney <ben+python@benfinney.id.au> - 2014-03-28 11:43 +1100
                Re: YADTR (Yet Another DateTime Rant) Steven D'Aprano <steve+comp.lang.python@pearwood.info> - 2014-03-28 11:11 +0000
                Re: YADTR (Yet Another DateTime Rant) Roy Smith <roy@panix.com> - 2014-03-28 08:30 -0400
                Re: YADTR (Yet Another DateTime Rant) Marko Rauhamaa <marko@pacujo.net> - 2014-03-28 15:10 +0200
                Re: YADTR (Yet Another DateTime Rant) Chris Angelico <rosuav@gmail.com> - 2014-03-29 00:19 +1100
                Re: YADTR (Yet Another DateTime Rant) Roy Smith <roy@panix.com> - 2014-03-28 09:22 -0400
                Re: YADTR (Yet Another DateTime Rant) Rustom Mody <rustompmody@gmail.com> - 2014-03-28 06:30 -0700
                Re: YADTR (Yet Another DateTime Rant) Chris Angelico <rosuav@gmail.com> - 2014-03-29 00:32 +1100
                Re: YADTR (Yet Another DateTime Rant) Roy Smith <roy@panix.com> - 2014-03-28 09:20 -0400
                Re: YADTR (Yet Another DateTime Rant) Marko Rauhamaa <marko@pacujo.net> - 2014-03-28 16:34 +0200
                Re: YADTR (Yet Another DateTime Rant) Steven D'Aprano <steve+comp.lang.python@pearwood.info> - 2014-03-28 14:35 +0000
                Re: YADTR (Yet Another DateTime Rant) Chris Angelico <rosuav@gmail.com> - 2014-03-29 02:25 +1100
  Re: YADTR (Yet Another DateTime Rant) Ethan Furman <ethan@stoneleaf.us> - 2014-03-26 16:33 -0700

csiph-web