Re: Sort of a not really a potential bug report. -ish.
- From: Andre Majorel <amajorel@xxxxxxxxx>
- To: yadex@xxxxxxxxxxxxx
- Date: Sun, 4 Jan 2004 13:25:06 +0100
On 2004-01-03 17:04 +0100, ras2 wrote:
> > If you drag vertices, and there are more than two superimposed
> > linedefs in the whole level and you reply "yes" to "Linedefs x
> > and y are superimposed (and perhaps other too) Merge them ?",
> > most of the times random linedefs will disappear.
>
> Maybe the superimposed ones were the reason I couldn't reproduce it.
> Are you certain it takes two of them?
If by "two of them" you mean a total of three or more
superimposed linedefs, yes, I'm pretty sure of that. The bug
occurs when more than one linedef is deleted (and even then, not
always). When you have just two superimposed linedefs, only one is
deleted.
> > It's funny that no-one has reported it so far because this bug
> > has been here forever (I introduced it in 1997). I guess that
> > most people don't drag that many linedefs at once.
>
> mh. I've /suspected/ a bug for years, but it's occured so randomly
> that there was a possibility that I was just hallucinating or
> something. With an Undo function, I would probably have been
> more certain much earlier (not a complaint, I just usually don't
> make backup copies of my levels very often, so it's nearly
> impossible to bring one back to the state it was in when the bug
> was triggered).
Sometimes, I wish Yadex automatically recorded user input to a log
file for replay on another machine.
> > I'm working on it. It's not trivial to fix, but I'm seizing the
> > opportunity to rewrite the auto-merging code to ignore linedefs
> > that were already superimposed before dragging. And if I'm on a
> > roll, do the same for vertices as I said I would for 1.6.
>
> Lovely. But the error checker will still report these? Otherwise
> they can be a tad difficult to find.
Yes. The idea is: after dragging vertices, only what has changed
is checked. When explicitly asking for a check (f10 menu),
everything is checked.
--
André Majorel <amajorel@xxxxxxxxx>
http://www.teaser.fr/~amajorel/
Other related posts:
