bzrlib.urlutils

[Andrew Bennetts]

Andrew Bennetts wrote:
[...]
> Where in urlparse are alternatives to urlutils.join, 

I should elaborate on this one.  urlparse.urljoin is pretty similar[1], but
there's at least one crucial difference in behaviour for bzrlib:

    >>> from urlparse import urljoin
    >>> urljoin('http://example.com/foo', '../../newpath')
    'http://example.com/../../newpath'
    
    >>> from bzrlib.urlutils import join
    >>> join('http://example.com/foo', '../../newpath')
    Traceback (most recent call last):
      File "<stdin>", line 1, in <module>
      File "bzrlib/urlutils.py", line 168, in join
        path = joinpath(path, arg)
      File "bzrlib/urlutils.py", line 207, in joinpath
        base, args)
    bzrlib.errors.InvalidURLJoin: Invalid URL join request: Cannot go above root: '/foo' + ('../../newpath',)

-Andrew.

[1] although the documentation at
    <http://docs.python.org/library/urlparse.html#urlparse.urljoin> isn't very
    clear that it can take both “absolute” URLs like http://example.com/foo as
    well as a “relative” URL like /foo.