urllib.parse can't handle URIs with empty #fragments. The fragment is removed and not reconsituted.

http://tools.ietf.org/html/rfc3986#section-3.5 permits empty fragment strings:


      URI-reference = [ absoluteURI | relativeURI ] [ "#" fragment ]
      fragment    = *( pchar / "/" / "?" )

And even specifies component recomposition to distinguish from not being defined and being an empty string:

http://tools.ietf.org/html/rfc3986#section-5.3


   Note that we are careful to preserve the distinction between a
   component that is undefined, meaning that its separator was not
   present in the reference, and a component that is empty, meaning that
   the separator was present and was immediately followed by the next
   component separator or the end of the reference.


This seems to be caused by missing components being represented as '' instead of None.

>>> import urllib.parse
>>> urllib.parse.urlparse("http://example.com/file#")
ParseResult(scheme='http', netloc='example.com', path='/file', params='', query='', fragment='')
>>> urllib.parse.urlunparse(urllib.parse.urlparse("http://example.com/file#"))
'http://example.com/file'

>>> urllib.parse.urlparse("http://example.com/file#").geturl()
'http://example.com/file'

>>> urllib.parse.urlparse("http://example.com/file# ").geturl()
'http://example.com/file# '

>>> urllib.parse.urlparse("http://example.com/file#nonempty").geturl()
'http://example.com/file#nonempty'

>>> urllib.parse.urlparse("http://example.com/file#").fragment
''

The suggested fix is to use None instead of '' to represent missing components, and to check with "if fragment is not None" instead of "if not fragment".


The same issue applies to query and authority. E.g.

http://example.com/file? != http://example.com/file

... but be careful about the implications of

file:///file != file:/file