[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Emacs-diffs] /srv/bzr/emacs/emacs-24 r111281: * progmodes/python.el

From: Stefan Monnier
Subject: Re: [Emacs-diffs] /srv/bzr/emacs/emacs-24 r111281: * progmodes/python.el (python-info-current-defun): Fix failed
Date: Wed, 20 Feb 2013 09:01:35 -0500
User-agent: Gnus/5.13 (Gnus v5.13) Emacs/24.3.50 (gnu/linux)

>> I suspect this is not The Right Place for the save-match-data, since we
>> should not assume that current-indentation won't also mess up the
>> match-data.
> FWIW this very small change corrected the python-info-current-defun behavior
> and was able to pass the test suite I've been working on.

I don't doubt it works, I'm just pointing out that it would be better to
place the save-match-data elsewhere.  The general rule is: put the
save-match-data between the "match" and the corresponding "get
match-data" rather than around some code that might modify the match-data.

> So far my local test/automated/python-tests.el is covering
> indentation, movement and python-info-* functions completely.  Would it
> be OK to introduce these tests into the emacs-24 branch?

Yes, it's OK to add files to emacs-24/test/...

> Now, I certainly wasn't expecting current-indentation could change
> match-data (could it?).

All functions can do it, except for those where it's blatantly obvious
they can't (e.g. `car', `cdr', `match-beginning').  And yes,
current-indentation probably can in sufficiently weird cases.

> If so, I feel that functions not expected[0] to mess
> with match-data should at least document it.

We should indeed document the small set of functions which are
guaranteed not to mess the match-data.

> In this regard, I have a change in the works that will cause python.el to
> stop messing with match-data so much (use of looking-at-p instead of
> looking-at and such), and document those functions that are intended to.
> If everyone agrees I could get this to the emacs-24 branch.

I generally recommend not to go down that route.  Instead, just make sure
all the match-{beginning/end/string} calls come right after the
corresponding search, and if not, place a save-match-data around the
intermediate code.


reply via email to

[Prev in Thread] Current Thread [Next in Thread]