This moves all parts in the Duke/Redneck modules that access ENet into net.cpp/network.cpp and changes global ENet structs to use C++ notation instead of typedefs so that anonymous definitions for them can be added where needed.
These changes are done to avoid including enet.h globally which is a major issue because this file does not work without including windows.h, making it extremely dirty.
For the handful of lookups it got used for the needed data is too large and the lookup not precise enough. ZDoom's BestColor yields better results, does not need any tables and for the small amount of lookups being performed is more than adequate.
The code needs to be refactored anyway to allow better crosshair control but this also seemed to affect global render state, considering how poorly it was all implemented.
All games combined there's 11(!!!) scene render blocks, not counting the sub-blocks for ROR and mirrors.
Does it surprise anyone that most of these sub-blocks do not feature all engine capabilities?
Just replaced with a linear array - for a hash table indexed with a string to be efficient, a much larger amount of data is needed than 20 or 30 entries.
This is also hardly the kind of data where losing a microsecond for performing the search matters.
This eliminates another piece of hideous code.
This commit also moves the memory error handler to the common code, so that all games can call it if triggered.
This needs to be replaced with the game independent ZDoom version and hooked up properly, but it of low priority because it's a multiplayer only feature.
* reverb/echo is not yet implemented, so there's two stub functions for now.
* RTS needs to be done differently, because the sound engine cannot play raw buffers without any control data.
This was only used for displaying the name for user maps, everything else was using other means of getting the data already, and even here currentLevel is better suited.
Reverted this to a sane setting, as it was in the original games and in all other games I have ever seen, i.e. there is a global setting to enable mouse view, and a button to manually trigger it. The toggle can be easily handled by flipping the CVAR directly.
The main problem here was that it triggered a few cases for mouse-less gameplay in the default case with a mouse present, because the mouseaim CVAR was no longer what the game expected.
This misguided change seems to have originated in JFDuke but by now had propagated to all the other games as well, the code was in all 4 frontends.
This was consolidated for both EDuke and RedNukem frontends, put into a class with strict access control and the length limit was lifted.
The new class will eventually allow better localization control.
This was some meticulously preserved relic of bad old DOS times used to block OS facilities to close an app.
Since this has been worked around at a lower level already the variable was essentially without function but some quite bad code depended on it.
* removed some redundant functionality (e.g. Shift-F5 to change - use the console for that!)
* removed a few more leftover parts of the old music system
* savegames should not do more than resuming the music at the point of saving. (DN3D and RR only so far. Blood to be done.)
* handle music enabling/disabling in the backend, which simply knows better what to do. This was only working in the menu, so changing the CVAR had no effect.
I have no idea how this is supposed to work, but all it does is create corrupt images, so for now it reverts to the software renderer which generates working 320x200 images.
* moved the binding commands to osd.cpp. They were in the global namespace already and this way everything to be tossed out is in the same place when the time comes.
* removed support for the OSDs native CVARs. The only ones left were some internal ones I won't need until this code can be replaced.
* same for the custpmization code the games added. Duke Nukem was the only one anyway to have a decent font for it.
- consolidated the 3 identical S_OpenAudio implementations. The replacement code is disabled for the time being because it needs a rewrite. The replacement logic is uses is a bit too volatile.
- removed the old GRP scan code.
* reroute several error conditions to I_Error.
* removed some soon-to-be obsolete GRP loading code.
* explicitly trigger the SetDefaults script events which depended on side effects from the config implementation.
* removed the nonsensical file system switch. All this does is create instabilities because it is non-obvious from where data is loaded. If a resource is mounted, it should be checked for content no matter what. While this may affect the stray weird mod out there it is a necessity if we want to allow transparent project repackaging.
When doing this during startup it can be done by regular cleanup measures.
This also moves two larger chunks of networking code out of game.cpp.
Nevertheless, the fact that enet is a very dirty library which directly depends on Windows types is a big problem because it bleeds Windows definitions everywhere thanks to poor abstraction in all relevant layers.
The Steam/GOG path getters were taken out of the frontends.
This also switches the Windows directory reader touse the wide string version to get Unicode file names.
Some paths were added to the config file instead of hard coding them.
Some part are not done yet, and the file system data is currently ignored - there's no way to properly set this up with the file system code Build came with.
