* Consistently declare Com_Error implementations as noreturn

According to comments in code/game/g_public.h (whose version was
already tagged noreturn), for function pointers this only works with
the gcc/clang-specific spelling __attribute__((noreturn)), and not with
the gcc/clang/MSVC abstraction NORETURN. For function implementations,
we can use NORETURN to get MSVC support too.

This lets the compiler reason about what returns and what doesn't,
and in particular silences gcc warnings about functions that
are declared NORETURN but that did appear to return. They didn't,
because they ended with a call to (a function pointer that
eventually calls) Com_Error, but without these annotations
the compiler couldn't know that.

Signed-off-by: Simon McVittie <smcv@debian.org>

* AI_ClosestGroupEntityNumToPoint: Remove, unused

It issues compiler warnings (or did until recently), and git grep says
nothing actually calls or mentions this function.

Signed-off-by: Simon McVittie <smcv@debian.org>

* Icarus: Squash a compiler warning about possibly uninitialized length

clang can see that BufferRead() does not always assign to length,
issuing this warning:

IcarusImplementation.cpp|601 col 13| warning: ‘length’ may be used uninitialized in this function [-Wmaybe-uninitialized]

(This only happens if the buffer is null, in which case we have
bigger problems.)

Signed-off-by: Simon McVittie <smcv@debian.org>

* Move definition of NORETURN to q_platform.h

Signed-off-by: Simon McVittie <smcv@debian.org>

* Introduce NORETURN_PTR, a version of NORETURN for function pointers

This expands to the same thing as NORETURN on gcc (and clang), but
expands to nothing on MSVC.

Signed-off-by: Simon McVittie <smcv@debian.org>

* Define NORETURN, NORETURN_PTR for non-gcc non-MSVC compilers

Signed-off-by: Simon McVittie <smcv@debian.org>

add cl_drawRecording

This patch comes from the ec-/Quake3e repository.

Cvars marked with CVAR_NODEFAULT explicitly or with CVAR_ARCHIVE_ND (Both) will not be written to config files when they match the default value and are marked for archive.

Also speeds up sorting and listing because sorts now only happen whenever a write would happen instead of at every list and every write.

Several cvars are marked in this commit based on whether they are often set differnetly or not (or if changed values would matter betweeen mods). No cvars in mod code are set as such for the time being.

See original commit: ec-/Quake3e@da8c757
Sorting commit see: ec-/Quake3e@14a5f02

cvar_usercreated - Lists all user created cvars.
unset_usercreated - Destroys all user created cvars. (USE with caution because it will unset valid cvars that are not currently valid)

Scenarios where valid cvars that might be unset:
At main menu and your cg_ g_ ui_ cvars might be not yet active.

This fixes issues with text scaling on Windows.

Note: only works with msvc compiles right now

CMake script comes from ouned/jk2mv and the manifest file is credits to ioquake3.

Fixes Issue #922

Using official libsdl.org bins and libs

Rage force power no longer acted like undying due to a typo in this commit

This reverts commit b5c3f14.

…l angle

Default sv_filterCommands to 1, move ';' filtering to value 2

These two functions existin SP's exe which is where UI resides there.

I am leaving references to the proper syscalls but since they do not get used when hud is normally loaded in SP (commented out), I think it is best to not call them as it is a global state affecting menus as well.

This is mainly for making it similar to the SP renderer. ri. was
preferred over ri-> as there is one less level of indirection (and so
potentially more performant).

The shell command

    sed -i '' 's/ri\->\([A-Z]\)/ri.\1/g' *.cpp

was run to do the majority of the find/replace.

…ixes #943. Ref #939.

Thanks to @peter-kien for pointing out the cause of UB

…mands). Allowed sv_floodProtect to specify flood time in ms

…s (gcc, clang)

…omeone

…g map on local server

…me flag unsetting (EF_SOUNDTRACKER, WPFLAG_GOALPOINT)

This allows the test to match to both Clang and AppleClang, which fixes the build on MacOS systems that use the built in Apple clang compiler.

This commit fixes #875.

G_HeldByMonster was being called on the player and any following spectators every frame. The latter of which is unnecessary.

* Fix Travis CI build

Add missing dependencies - why doesn't aptitude just download them??

* Try apt packages using apt add-on in Travis

* Remove some packages and fix host env variable

* Add libglib2.0-dev dependency for i686

* Maybe Release and Debug flavours?

* Split Debug and Release travis builds

* Add macOS build

* Brew install for OSX build

And reorder the builds

* Try using install script again

* Fix typo when running install script

* Fix install script for macOS

* Whitespace fixes

* Add missing space

…e power now

Previously, the console command "give force" only allowed the amount of force power to go up to 100 and not the actual limit of the current playermodel, because it was set to go until FORCE_POWER_MAX, which was defined as 100. Now, if you are using a playermodel with an increased amount set, that amount will be used as the maximum.

Hopefully this works better than the previous attempt and does not bork things.  CMake 3.4 is required on Windows for this to work. Leaving version as is for now due to other work being done on CMake support elsewhere.

Should fix DPI awareness / scaling on Windows + Have proper version info.

The renderer isn't initialized until the first frame after the restart.
The game frame can run before this happens which will try to do some
operations with the renderer, causing the process to crash.

[SP] Fix vid_restart crash

Visit https://discourse.libsdl.org/t/sdl-2-0-8-released/23957 for release notes on changes

Upgrade to newest version of SDL2 (2.0.8)

Shared: Add missing audio format types for newer SDL

Prevent null pointers reaching the underlying G2 code as is done for cgame.

Also fix minor bug in an unused G2 API UI function.
Also minor code cleanup in cgame G2 API functions.

In original game if the g2 pointer/ref was invalid it would return -1.  In the new modules it seems to have changed to returning to 0 here causing an inconsistency.  UI already returns -1 properly but cgame and server did not.

Fixes some derpiness in cases where handle is invalid and also adds missing calls to check model pointers in some functions.

Fix Ghoul2 API on server and UI for legacy mods

And update all the copyright years in those files

Upgrade minimum CMake to 3.1

Some of the tools were purpose-made for assets to be used on the Xbox.
We don't need these anymore so let's remove them.

This will make the macOS app bundle completely self-contained (as it
should be), i.e. the client will not need to install SDL2 separately.

Remove Xbox-related or unused tools

Remove tools which were moved to OpenJK-Tools

Add SDL2 to macOS app bundle

And remove s_sdlSpeed

Travis-CI's macOS worker image currently has libpng installed at a
slightly outdated version. Unlike "apt install", "brew install"
will not upgrade a package that is already installed, and will exit
unsuccessfully when asked to do so. There does not seem to be a
Homebrew command for "upgrade these packages if installed, or install
them if not installed", so the most straightforward way seems to be
to upgrade everything before invoking brew install.

Fixes: #987

The Allocator is meant to be one that is suitable for allocating
std::pair<key, value> instances, not value instances, and the
Standard C++ library provided by gcc 8 asserts that the types match.
The closest valid allocator here would be
std::allocator<std::pair<key, value>>, but that is the default for
this template anyway, so we can just omit it.

Closes: #985
Bug-Debian: https://bugs.debian.org/905477
Signed-off-by: Simon McVittie <smcv@debian.org>

Use s_khz with SDL sound backend

JK2: Remove mismatched allocator from std::map typedefs

The compressed bone pool of a Ghoul II animation file (MDXA, .gla) is always
stored at the end of the file. It is unnecessary to iterate through all frames
to determine the number of compressed bones present on big-endian machines.

This change is applied to the vanilla renderers of the single- and multiplayer
code of JK:JA.

The original code removed in commit b993a8c is now only used when dealing with
MDXA formats different than the original format (version 6) present in JK:JA.
This ensures the compressed bone count is always calculated properly.

It now contains a three byte array rather than a regular integer. This has been
changed so the G2_GetBonePoolIndex function (and any future implementation) can
convert the three byte integer at such an offset without having to worry about
endianness.

…der than it needed to be. Also removed cheat flag and added archive flag to g_saberRealisticCombat, debug_subdivision, g_dismemberProbabilities

Removed superfluous code to determine the number of compressed bones.

Fix exploit to reset player by sending wrong serverId

If client sends wrong serverId but is already active in the world
(CS_ACTIVE) don't resend initial gamestate for the map. This isn't a
valid situation. The player should be CS_CONNECTED or CS_PRIMED.

Resending gamestate to an active player will cause them to respawn
without dying or disconnecting. If the player had a CTF flag it gets
lost until the map is changed or restarted.

Reported by Ensiform at:

https://bugzilla.icculus.org/show_bug.cgi?id=6324

[MP] Merge ioquake/ioq3@b61e299

* Modified tr_ghoul2.cpp, Lines: 3579-3603 to address enhancment issue #1005.

…messages.

Using proper functions for type at hand.

…ed with if statement using "g_developer->integer"