Possibly not a bug but perhaps an area for tightening up?
Just out of interest, I loaded up a project into GZDoomBuilder and it flagged up a genuine error with my PK3 but GZDoom had been working with the mod anyway.
When the files were loaded into GZDB, I got an error message saying that one of my models could not be found. On checking the MODELDEF, sure enough, the "Path" statement was missing from the model definition. However, the model had been working in GZDoom (possibly because the previous actor in the MODELDEF had a model in the same folder and the path was correctly entered for that one, so perhaps the path just carried across to the next actor). I put the Path statement in to the MODELDEF and GZDB stopped giving the error message.
I just wondered if this is something that GZDoom should have been flagging up?
My model worked... but I don't think it should have. (Paths)
Moderator: Graf Zahl
- Enjay
- Developer
- Posts: 4748
- Joined: Tue Aug 30, 2005 23:19
- Location: Scotland
- Contact:
- Graf Zahl
- GZDoom Developer
- Posts: 7148
- Joined: Wed Jul 20, 2005 9:48
- Location: Germany
- Contact:
Re: My model worked... but I don't think it should have. (Pa
Ouch.
But...
Do you know the shitstorm that might ensue if I CHANGED it???
But...
Do you know the shitstorm that might ensue if I CHANGED it???
- Enjay
- Developer
- Posts: 4748
- Joined: Tue Aug 30, 2005 23:19
- Location: Scotland
- Contact:
Re: My model worked... but I don't think it should have. (Pa
Yeah, could be nasty.Graf Zahl wrote:Do you know the shitstorm that might ensue if I CHANGED it???

Perhaps don't change the behaviour (unless it is otherwise undefined/problematic) but allow GZDoom to print a console warning?
- NeuralStunner
- Posts: 253
- Joined: Tue Dec 29, 2009 3:46
- Location: IN SPACE
- Contact:
Re: My model worked... but I don't think it should have. (Pa
If my opinion amounts to anything: Fix it anyway. If someone can come up with a compelling case of "mod is doing it wrong and we can't fix the mod instead for whatever reason", then it can probably be "demoted" to a warning. Until then, it's a laughably small thing for anyone to be that upset over.
Maybe it's just me, but where the documentation indicates a path attribute for each model definition, I would consider that the standard that should be followed. But then, I'm more careful than a lot of modders about making sure the behavior I'm getting is expected.
Maybe it's just me, but where the documentation indicates a path attribute for each model definition, I would consider that the standard that should be followed. But then, I'm more careful than a lot of modders about making sure the behavior I'm getting is expected.

Dean Koontz wrote:Human beings can always be relied upon to exert, with vigor, their God-given right to be stupid.
Spoiler: System Specs
-
- Developer
- Posts: 197
- Joined: Sun Nov 29, 2009 16:36
Re: My model worked... but I don't think it should have. (Pa
Please ensure that the model name indicated in MODELDEF has the same case as the file itself, or else the users with case-sensitive filename (for example linux) would have problems when loading the data (or is it already checked with a case insensitive string comparison?)
Last edited by Edward-san on Tue Jul 29, 2014 22:01, edited 1 time in total.
- Graf Zahl
- GZDoom Developer
- Posts: 7148
- Joined: Wed Jul 20, 2005 9:48
- Location: Germany
- Contact:
Re: My model worked... but I don't think it should have. (Pa
The entire lump directory is case insensitive.
- Graf Zahl
- GZDoom Developer
- Posts: 7148
- Joined: Wed Jul 20, 2005 9:48
- Location: Germany
- Contact: