[issue25652] collections.UserString.__rmod__() raises NameError

[Raymond Hettinger]

Raymond Hettinger added the comment:

The UserString1 patch is incorrect as it leads to infinite recursion because the __rmod__ operation only gets called when the other argument doesn't have __mod__.

One possible fix is to coerce the template to a regular str: 

    def __rmod__(self, template):
        return self.__class__(str(template) % self)

That would get the following test to pass:

    class S:
        'strlike class without a __mod__'
        def __init__(self, value):
            self.value = value
        def __str__(self):
            return str(self.value)

    assert S('say %s') % UserString('hello') == 'say hello'

That said, the goal of UserString is to parallel what a regular string would do.  In this case, a TypeError is raised:

    >>> print(S('answer is %s') % 'hello')
    Traceback (most recent call last):
      ...
    TypeError: unsupported operand type(s) for %: 'S' and 'str'

Serhiy, what do you think should be done, coerce to a str or remove the __rmod__ method entirely?

----------
assignee: rhettinger -> serhiy.storchaka

_______________________________________
Python tracker <report at bugs.python.org>
<http://bugs.python.org/issue25652>
_______________________________________