Re: GZDoomBuilder-Bugfix, a maintenance fork of GZDB
Posted: Fri Jun 14, 2019 3:41 pm
Ah, right, thanks. I knew about DRD being down but forgot GZDB updates came from there.
Thanks again.
Thanks again.
Discussion about ZDoom
https://forum.zdoom.org/
Thanks for working on OBJ support. However, After downloading the latest version, the program crashes when I load a map with a OBJ model defined in MODELDEF.boris wrote:There's now support for OBJ models. I think I got everything right, but who knows with all the MODELDEF weirdness.
I also notices that it's already possible to copy the map analysis results to the clipboard. Just select everything and press Ctrl+C.
Code: Select all
Model HG_Map07_All
{
Model 0 "models/maps/map07abandoneddepot/map07mesh.obj"
SurfaceSkin 0 0 "textures/style1/s1florc.png"
SurfaceSkin 0 1 "textures/style1/s1flora.png"
SurfaceSkin 0 2 "textures/style1/s1walla.png"
SurfaceSkin 0 3 "textures/style1/stairs1.png"
SurfaceSkin 0 4 "textures/style1/s1florb.png"
Scale 1.0 1.0 1.0
ZOffset 0
FrameIndex FRAM A 0 0
}
Code: Select all
***********SYSTEM INFO***********
OS: Microsoft Windows 10 Home
GPU: NVIDIA GeForce GTX 670
GZDB: R3063
Platform: x64
********EXCEPTION DETAILS********
Der Index war außerhalb des Arraybereichs.
bei CodeImp.DoomBuilder.GZBuilder.MD3.ModelReader.ReadOBJModel(BoundingBoxSizes& bbs, Dictionary`2 skins, Stream s, Device device, String name)
bei CodeImp.DoomBuilder.GZBuilder.MD3.ModelReader.LoadModel(ModelData mde, List`1 containers, Device device)
bei CodeImp.DoomBuilder.Data.DataManager.ProcessModel(Int32 type)
bei CodeImp.DoomBuilder.Map.Thing.UpdateCache()
bei CodeImp.DoomBuilder.Data.DataManager.Load(DataLocationList configlist, DataLocationList maplist)
bei CodeImp.DoomBuilder.Data.DataManager.Load(DataLocationList configlist, DataLocationList maplist, DataLocation maplocation)
bei CodeImp.DoomBuilder.MapManager.ReloadResources(Boolean clearerrors)
bei CodeImp.DoomBuilder.MapManager.DoReloadResource()
bei CodeImp.DoomBuilder.Actions.Action.Begin()
bei CodeImp.DoomBuilder.Actions.ActionManager.InvokeAction(String actionname)
bei CodeImp.DoomBuilder.Windows.MainForm.InvokeTaggedAction(Object sender, EventArgs e)
bei System.Windows.Forms.ToolStripItem.RaiseEvent(Object key, EventArgs e)
bei System.Windows.Forms.ToolStripMenuItem.OnClick(EventArgs e)
bei System.Windows.Forms.ToolStripItem.HandleClick(EventArgs e)
bei System.Windows.Forms.ToolStripItem.HandleMouseUp(MouseEventArgs e)
bei System.Windows.Forms.ToolStrip.OnMouseUp(MouseEventArgs mea)
bei System.Windows.Forms.ToolStripDropDown.OnMouseUp(MouseEventArgs mea)
bei System.Windows.Forms.Control.WmMouseUp(Message& m, MouseButtons button, Int32 clicks)
bei System.Windows.Forms.Control.WndProc(Message& m)
bei System.Windows.Forms.ToolStrip.WndProc(Message& m)
bei System.Windows.Forms.ToolStripDropDown.WndProc(Message& m)
bei System.Windows.Forms.NativeWindow.Callback(IntPtr hWnd, Int32 msg, IntPtr wparam, IntPtr lparam)
Let me again say how grateful you are for jumping on this so readily! It's super, super appreciated!boris wrote:There's now support for OBJ models. I think I got everything right, but who knows with all the MODELDEF weirdness.
I also notices that it's already possible to copy the map analysis results to the clipboard. Just select everything and press Ctrl+C.
Code: Select all
MAP01 - Entryway
Unknown top texture 'FAKE3' on first side of linedef 1
Unknown bottom texture 'FAKE4' on first side of linedef 1
Unknown top texture 'FAKE1' on second side of linedef 4
Unknown top texture 'FAKE1' on second side of linedef 5
Unknown top texture 'FAKE1' on second side of linedef 6
Unknown top texture 'FAKE1' on second side of linedef 7
Unknown bottom texture 'FAKE2' on second side of linedef 8
Unknown bottom texture 'FAKE2' on second side of linedef 9
Unknown bottom texture 'FAKE2' on second side of linedef 10
Unknown bottom texture 'FAKE2' on second side of linedef 11
Fixed in 3064. However, the textures won't display correctly. It looks like they are not wrapped, and I'm not sure why. I don't really have a clue about SlimDX/Direct3D, so it's pretty hard for me to figure out what's going on.Cherno wrote:Here is the download link from my DropBox:boris wrote:Can you supply the OBJ?
map07mesh.obj
If you happen to have a GitHub account I'd really appreciate if you opened issues for that. In here it's pretty much bound to get lost if it doesn't get fixed immediately.Enjay wrote:stuff
You probably had "Snap to geometry" turned off. For some disabling it doesn't merge geometry. Without having a look at it I'd say that's a bug. You can merge geometry while having it disabled by holding Ctrl when releasing the RMB. You can stop geometry from being merged while snap to geometry is enabled by holding Ctrl when releasing the RMB, so the behavior seems to be inverted.Enjay wrote:And, on quite a different topic, is there a toggle that prevents geometry being merged? I was working on a map and I'd drawn a little sector inside a bigger sector and I wanted to move some of the sides top of and merge with some of the lines from the bigger sector. However, no matter how I did it (moving vertices, lines and the whole sector) the lines just sat on top of each other and would not merge. I saved the file, quit and restarted and I was then able to merge stuff. I'm guessing that I hit a hotkey of some sort that disabled merging of geometry and that restarting reset things to default, but I have no idea what I hit. I didn't find anything that made the same thing happen when trawling through the menu and the options. It'd be nice to know how to do this in case I do it again by accident.
Aha! That'll be it. I found snapping to geometry could get quite annoying if I zoomed out but still wanted to do fine work. I assume the tolerance for snapping is screen pixels rather than map units because I can move vertices, things (or whatever) very close to lines when zoomed in but if I zoom out, even with the same grid displayed, things start snapping to geometry long before what I would regard as close to the line. It took me ages to figure out what was going on because snap to geometry is not a feature that I am familiar with from DeePsea (it doesn't have it). It's a useful feature though, I just didn't realise what was happening and I couldn't figure out why the mouse seemed so inaccurate when zoomed out.boris wrote:You probably had "Snap to geometry" turned off.