missing ifdef for g_object_ref_sink call

Ethan Blanton elb at pidgin.im
Fri Jul 4 12:47:40 EDT 2008

Sadrul Habib Chowdhury spake unto us the following wisdom:
> On a somewhat related note: in a situation like this, do we want to
> disapprove the revision and commit the correct fix, or just fix the
> change and make a new commit? I have done the latter, in this case,
> since that involves less revision(s). I think that's a reasonable
> practice for such small changesets. But I would like to know if someone
> has any thoughts on this.

If an entire revision is erroneous, the "right" thing to do is
disapprove it, then commit a correct revision in its place.  Probably
like this, even:

A = Ancestor revision of erroneous revision
E = Erroneous revision
D = Disapproval of erroneous revision
F = "Correct" fix
H = Current (pre-disapproval) head
M = post-fix merges

  / \
 E   F
 |\  |
 : D |
 | | |
 H | |
  \| |
   M |

This provides the most information about what actually happened, and
why.  Of course, if the fix F is not really in any way related to A
(that is, E just *happened* to be committed onto A, and wasn't
otherwise logically tied to it), something like this is easier (?) and
just as meaningful:

: D
| |
H |

If E is not completely erroneous, or if for some other reason the
above doesn't seem appetizing, a simple "Daggy Fix" is the next most
appropriate action:

: F
| |
H |

Either way, a fix committed is the best kind of fix, so if you're
morally opposed to, or not inclined to figure out, or don't have time
for one of the above methods, whatever kind of commit you like is
fine.  ;-)

Sorry for the late reply, I've been overseas.


The laws that forbid the carrying of arms are laws [that have no remedy
for evils].  They disarm only those who are neither inclined nor
determined to commit crimes.
		-- Cesare Beccaria, "On Crimes and Punishments", 1764
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
URL: <http://pidgin.im/pipermail/devel/attachments/20080704/27669e52/attachment.sig>

More information about the Devel mailing list