Why are tuples immutable?

Gerrit gerrit at nl.linux.org
Tue Dec 14 17:16:43 CET 2004


Fredrik Lundh wrote:
> > I could imagine that anything accepting numerical values for __getitem__
> > (foo[0], foo[1], ...) or that is iterable (foo.next(), foo.next()) could
> > be sensibly used as a formatting rhs. Of course, it is not compatible
> > because "foo %s" % [2, 4] is correct and "foo %s" % (2, 4) is not, but
> > are there other reasons that it has been chosen like this?
> 
> so what should
> 
>     "foo %s bar %s" % "24"
> 
> do?

Hmm. Good question (-:

Not possible before Python 3.0, but could __rmod__ play a role here?
str.__mod__ would be something like:
try:
    # use other.__rmod__(...) result
except AttributeError:
    try:
        # use it as an array
    except TypeError:
        # use string representation

Hm, something else I don't understand in current behaviour:
>>> "foo".__rmod__("testing %s")
'testing foo'
>>> (55).__rmod__("testing %s")
NotImplemented
>>> print (55).__rmod__.__doc__
x.__rmod__(y) <==> y%x

...apparantly not? How should I interpret this?

To answer your question:

"foo %s bar %s" % "24" would do "24".__rmod__("foo %s bar %s") resulting in:

>>> "24".__rmod__("foo %s bar %s")
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
TypeError: not enough arguments for format string

Would it be possible to use something like this?

regards,
Gerrit.

-- 
Weather in Lulea / Kallax, Sweden 14/12 16:20:
	2.0°C Broken clouds mostly cloudy wind 7.6 m/s S (34 m above NAP)
-- 
In the councils of government, we must guard against the acquisition of
unwarranted influence, whether sought or unsought, by the
military-industrial complex. The potential for the disastrous rise of
misplaced power exists and will persist.
    -Dwight David Eisenhower, January 17, 1961



More information about the Python-list mailing list