| Age | Commit message (Collapse) | Author |
|---|
|
|
|
- A bunch of stuff is now defined in one header, engineapi.h
- engineapi.c is responsible for setting up any interfaces/stuff that's
used in more than one place
- mkgamedata is pretty much rewritten and now supports nested
conditionals
- gamedata variables no longer have the gamedata_ prefix because it was
just annoyingly long all the time
- vcall macros are somewhat revamped and support dynamic (gamedata)
indices
- Portal 1 FOV can be set anywhere from 75-120 using fov_desired -
tested in both the main versions currently used by runners
- A few typos were also fixed ("intput," "writeable," "indexes")
|
|
It's amazing how long these things can go unnoticed.
|
|
|
|
I'd noticed this at some point but it didn't seem to matter, and now all
of sudden it does matter because magic. The hook just stopped working
suddenly.
|
|
|
|
This was dumb.
|
|
Default gamedata values actually work the way they're supposed to now.
|
|
|
|
|
|
Thanks Aciidz for helping test this.
|
|
|
|
- Demo recording can now be started before connecting to a server
- The overall demo-number-preserving logic is a lot simpler and should
work even if the plugin is reloaded or something
- As an added feature, recording to a nonexistent directory will produce
an error instead of silently failing
|
|
No idea when the next release will happen yet, but this might as well
happen now to avoid forgetting later.
|
|
They're literally always defined thusfar, and it's always possible to
include specific headers for other stuff. So, rather than litter the
source with defines (which might not even work in the likes of os.h if
some other system header already decided to define/include stuff), let's
just define it at the build script level.
|
|
This brings the DLL size down to a quarter of what it was. Dunno why
Clang doesn't do this by default, but apparently it doesn't!
|
|
hook_inline() uses the new x86_len() function to get instruction lengths
instead of doing full-blown disassembly, which should be a tiny bit
quicker, and also removes the next for about 90KiB of lookup tables and
such in the final binary. The code-digging logic in demorecord is also
rewritten to be opcode-based rather than mnenmonic based. In general,
going forward the plan is to always rely on opcodes and thus avoid a
bunch of disassembly work every plugin load.
udis86 is still in the tree for now to provide dbg_asmdump(), but it's
only compiled into debug builds and left out of releases completely. As
such, the whole BSD licence statement is also gone from the distribution
LICENCE files. There's now also a dbg_toghidra() which spits out a
rebased address to look stuff up for proper reverse engineering, which
might be more useful than dbg_asmdump() anyway. If nobody ends up using
the latter ever again, udis86 could get chucked completely. We'll see.
Also shoehorned into this commit are a couple more forgotten copyright
year bumps and some general minor cleanup here and there, because I
couldn't be bothered wading through all the diff hunks. Oh, and
makebindist.bat now makes an effort to make the zip file timestamps
predictable/reproducible. That should be a different commit for sure,
but oh well too bad.
|
|
|
|
Thanks Aciidz and Bill for spotting these.
Lesson (ostensibly) learned: Don't program while sleep-deprived!
|
|
They're still temporary, I promise, but this will make tweaking the
toolchain or turning debug builds on and off a lot simpler.
|
|
Chalk this up to improper testing and being tired and not using brain.
|
|
Breaks the Linux build because I can't file paths brain I need sleeping
|
|
- Just ask the engine for the game directory instead of doing the stupid
argv sniffing hacks from the early days of trying to get the damn
thing working.
- Also add some other path variables, functions and whatnot, and do some
other minor tidying up.
- Also also, another damn copyright year, somebody please help me.
Unfortunate negative effect off this change: con_init() no longer
reports the game name, because it has to happen before gameinfo_init().
I've decided I don't really care, though.
|
|
This also introduces some stuff for interacting with the current plugin
list. Other plugin management utilies are Coming Soon...
|
|
Turns out creating the input window on the main thread means it actually
gets events from the game's event loop on the main thread which means we
don't have to do our own event loop at all and in fact doing so is kind
of broken and causes thread lockups and leaks and other such nonsense.
|
|
It's late, okay?
|
|
It's archive so you can set m_rawinput 1, load SST via VDF and then
never think about it again.
|
|
|
|
This allows stuff to be registered conditionally.
Unfortunately cmeta is now truly the worst thing of all time, but
cleaning it up isn't a huge priority. On the plus side, codegen actually
got simpler.
|
|
1. Why is it so hard to remember to bump the years aaaaaaaaaaaaaaaaaaaa
2. Aciidz wrote the mat_monitorgamma patch a while ago and it turns out
he since found out it's not version-specific, so the comment was
confusing.
|
|
I had plans to use these at some point but never did. Maybe I still will
later but listing them otherwise is a bit confusing and weird, and stems
from an oversight when I initially imported all the early dev stuff into
the public Git tree.
|
|
Committer note: I, mike, am a big dumb idiot. Thanks, Bill.
|
|
|
|
Also unhide demo_interpolateview because yes.
|
|
Sometimes, you just want to Unicode.
|
|
- Implement conditionals in the lexer and reject or ignore them in
callbacks. This will allow something to use them later if needed.
- Make error handling less stupid (return a bool instead of using the
state struct).
|
|
More handwavey Linux prep, nothing too significant.
|
|
This was a smaller oversight in the crash fix commit. I'd thought about
it, but forgotten to do it.
|
|
|
|
|
|
Mostly L4D leaderboard related stuff, but shouldn't really hurt anyone
in other games.
|
|
I hadn't realised this was all that newer versions did.
Thanks Aciidz for testing the fix on earlier versions too and
confirming that it really is this simple.
|
|
|
|
The format string one is cool because it implies the Windows code was
also always wrong but the Windows headers aren't annotated for Clang so
I guess that means it doesn't warn. At least for the nonstandard Windows
wide character versions of things. Cool!
|
|
Important note: it doesn't WORK on Linux, and there's tons of warnings
and stuff, but it's easier to work on when all the compiler output and
whatnot is there.
|
|
|
|
SST itself doesn't build, but it's a start!
Really interesting is the declaration after a switch label. That's of
course invalid, but Clang on Windows never complained. I guess it's an
MSVC extension, eh? How annoying.
Also, haha yes the script wasn't even executable. Forgot to
update-index when I remade the repo, I guess...
|
|
Preparing for the January beta release.
|
|
|
|
|
