Steve Holden wrote:
> These versions differ with respect to treatment of blank lines, which
> indicates how easy it is to go astray in this kind of semantic
> optimization. Your example simply wouldn't work (though you could patch
> it up using "if line is None". (despite the use of short-circuiting
> predicates).
> > both examples show reduction by 3 lines.
> >
> Perhaps so, but you have to realise that Python has never valued code
> compactness over clarity.
> I'm not sure that your democratic wish to ensure fairness to sequences
> will garner much support, interesting though it is in an academic sense.
Thank you for the patch. My incentives are not academic though.
I convinced that this
line = self._buffer.get(self._bufindex)
if line is None:
self._bufindex += 1
self._lineno += 1
self._filelineno += 1
return line
line = self.readline()
is more clear than
try:
line = self._buffer[self._bufindex]
except IndexError:
pass
else:
self._bufindex += 1
self._lineno += 1
self._filelineno += 1
return line
line = self.readline()
I mentioned 3 lines reduction just because it is
objective, while 'more clear' is subjective.
Code snippets are again not equivalent, e.g. in the
case when self._buffer[self._bufindex] exists and is
equal to None. Although the author could probably
make a guarantee that is never the case.
--
http://mail.python.org/mailman/listinfo/python-list
