[4.2.4] GLDEFS: iwad parameter ignored in brightmaps

[NightFright]

It appears that the iwad parameter is not working when used to prevent brightmaps from being applied to PWAD textures.

Example:
Using Brightmaps Plus
Using Sinful Discharge for Heretic

Screenshot from Sinful Discharge E3M9, showing SW2ON replacement with standard brightmaps applied in spite of iwad definitions:


SW2ON/SW2OFF definition (in brightmaps\textures\heretic.bm of Brightmaps Plus):

Code: Select all

brightmap texture sw2off
{
	map "brightmaps/textures/heretic/sw2off.png"
	iwad
}

brightmap texture sw2on
{
	map "brightmaps/textures/heretic/sw2on.png"
	iwad
}

[_mental_]

Is there any reason why lump in build info is assigned after creating a multipatch texture?

Actually, there seems to be two related issues with these lump indices.
The first one is a source lump for multipatch texture is currently set to zero instead of -1. This condition will be always true for such textures.
The second one is the mentioned assignment. It's impossible to determine a WAD with the given texture without setting a source lump.

[Graf Zahl]

How would you determine which file a multipatch texture originates from?
There's several issues at play here:

1. Thanks to the non-cumulative nature of these things in vanilla and classic ports, most of them redefine the entire IWAD content. So what is to be considered the source lump here?
2. Even if the above does not trigger, you can just replace patches and the texture manager will never know.

In short, it is pretty much impossible to restrict brightmaps to composite IWAD textures. Something will always break. Of course the lump assignment here seems to be wrong. For a composite texture the definition lump should qualify as the source lump.

[_mental_]

Exactly, my idea was to use definition lump as a source. So, it's about swapping these two lines.

[Graf Zahl]

Sure, that's how it was supposed to be, the swaparound was an accident.

[_mental_]

Fixed in fd5e8d6.

As for source lump defaulting to zero instead of -1, the given fix doesn't depend on it. And I don't want to change this just in case.