Thanks for the update.
The first problem I hit was git pull from the src directory comes back with 'You are not currently on a branch'. I tixed that by doing a git pull . v0.3.7, which fastforwarded my local image correctly, so I skipped the checkout step.
Then my system didn't have sdl2_mixer installed, but that was an easy fix. Finally, running it it built, created a window, hung about for a few seconds not updating it, then crashed with './run: line 53: 6011 Segmentation fault (core dumped) GALLIUM_HUD=".dfps;cpu0+cpu1+cpu2+cpu3;GPU-load;VRAM-usage" ./pyragles'.
If I run it manually by going to the build folder and running .pyragles, I get the following out:
Code:
Initialised 0 gamepad(s)
Changing game state to PLAY
Segmentation fault (core dumped)
Running the whole kaboolde through gdb, I see it crashes on line 26 of this file:
Code:
hread 1 "pyragles" received signal SIGSEGV, Segmentation fault.
0x004d0ac5 in Gltf::init (this=0x679ea0, filename="ship.glb")
at /home/williain/devt/projects/gles2.0/pyragles/src/gltf.cpp:26
26 std::string j;
(gdb) l
21 fs_.read( ( char* )&jsonChunkType_ , 4 );
22
23 binStartByte_ = jsonChunkDataStartByte_ + jsonChunkLength_; // start of binary buffer
24
25 char jsonBuffer[ jsonChunkLength_ ];
26 std::string j; <--
27 fs_.read( ( char* )&jsonBuffer, jsonChunkLength_ );
28
29 for( unsigned int i = 0; i < jsonChunkLength_; i++ ) {
30 j += jsonBuffer[ i ];
I'm not sure why a simple std:string definition on the heap could cause a segfault, especially when a declaration on the heap successfully happens on line 12 of that file apparently, so I think the signal is not actually indicating a fault here. I've never debugged anything multithreaded before so I don't know what happens there.