3772: Add client-side HTTP server info r=def- a=heinrich5991
Summary
=======
The idea of this is that clients will not have to ping each server for
server infos which takes long, leaks the client's IP address even to
servers the user does not join and is a DoS vector of the game servers
for attackers.
For the Internet, DDNet and KoG tab, the server list is entirely fetched
from the master server, filtering out servers that don't belong into the
list.
The favorites tab is also supposed to work that way, except for servers
that are marked as "also ping this server if it's not in the master
server list".
The LAN tab continues to broadcast the server info packet to find
servers in the LAN.
How does it work?
=================
The client ships with a list of master server list URLs. On first start,
the client checks which of these work and selects the fastest one.
Querying the server list is a HTTP GET request on that URL. The
response is a JSON document that contains server infos, server addresses
as URLs and an approximate location.
It can also contain a legacy server list which is a list of bare IP
addresses similar to the functionality the old master servers provided
via UDP. This allows us to backtrack on the larger update if it won't
work out.
Lost functionality
==================
(also known as user-visible changes)
Since the client doesn't ping each server in the list anymore, it has no
way of knowing its latency to the servers.
This is alleviated a bit by providing an approximate location for each
server (continent) so the client only has to know its own location for
approximating pings.
## Checklist
- [x] Tested the change ingame
- [ ] Provided screenshots if it is a visual change
- [ ] Tested in combination with possibly related configuration options
- [ ] Written a unit test if it works standalone, system.c especially
- [ ] Considered possible null pointers and out of bounds array indexing
- [ ] Changed no physics that affect existing maps
- [ ] Tested the change with [ASan+UBSan or valgrind's memcheck](https://github.com/ddnet/ddnet/#using-addresssanitizer--undefinedbehavioursanitizer-or-valgrinds-memcheck) (optional)
Co-authored-by: heinrich5991 <heinrich5991@gmail.com>
3844: Add UI Ex class to share common ui implementations r=def- a=Jupeyy
Between menus and editor for example
Currently only did DoEditBox, as this created crashes in the editor
fixes#3395
It's copy and paste from the implementation of CMenus, except for 4 lines, which i'll mark below
## Checklist
- [x] Tested the change ingame
- [ ] Provided screenshots if it is a visual change
- [ ] Tested in combination with possibly related configuration options
- [ ] Written a unit test if it works standalone, system.c especially
- [ ] Considered possible null pointers and out of bounds array indexing
- [ ] Changed no physics that affect existing maps
- [ ] Tested the change with [ASan+UBSan or valgrind's memcheck](https://github.com/ddnet/ddnet/#using-addresssanitizer--undefinedbehavioursanitizer-or-valgrinds-memcheck) (optional)
Co-authored-by: Jupeyy <jupjopjap@gmail.com>
3826: Respect TuneZone for characters r=def- a=TsFreddie
*I might be totally insane for doing this.
*The effect is only truly noticeable if you have high ping AND dropping packets.*
*clang-tidy's warning doesn't seems to relate to anything I've changed*
Now characters will try to respect TuneZones when ~~evolving and~~ predicting (like projectiles do). Since it is really important to keep TuningList as accurate as possible, I made a TuningList update logic so we try our best to update tunings only for the correct TuneZone (instead of constantly shoving current tuning into TuningList)
And each CCharacterCore now has its own Tunings to avoid any characters influencing GameWorld's tuning. Some server code has to be updated accordingly. I checked the code path, didn't notice anything that would make our server behave differently.
And I guess I should tag @trml myself this time.
## Checklist
- [x] Tested the change ingame
- [ ] Provided screenshots if it is a visual change
- [ ] Tested in combination with possibly related configuration options
- [ ] Written a unit test if it works standalone, system.c especially
- [ ] Considered possible null pointers and out of bounds array indexing
- [x] Changed no physics that affect existing maps (as far as I could tell)
- [ ] Tested the change with [ASan+UBSan or valgrind's memcheck](https://github.com/ddnet/ddnet/#using-addresssanitizer--undefinedbehavioursanitizer-or-valgrinds-memcheck) (optional)
Co-authored-by: Freddie Wang <tsfreddiewang@gmail.com>
Co-authored-by: def <dennis@felsin9.de>
3841: Add better support for compose binds in demo menu r=def- a=ChillerDragon
I assume shift is a common compose bind. So users might have bound
shift+scroll or shift+arrowkeys to something else and do not want to
change the playback speed while triggering that compose bind.
I have to admit this is something I personally need and I am not sure what the demand for it is.
But for example my zoom bind is shift+mousewheelup and shift+mousewheeldown and I change that a lot while watching demos but it keeps affecting the playback speed. Maybe someone else has a compose bind using the alt key I do not know. But this solves my issue :)
## Checklist
- [X] Tested the change ingame
- [ ] Provided screenshots if it is a visual change
- [ ] Tested in combination with possibly related configuration options
- [ ] Written a unit test if it works standalone, system.c especially
- [ ] Considered possible null pointers and out of bounds array indexing
- [ ] Changed no physics that affect existing maps
- [ ] Tested the change with [ASan+UBSan or valgrind's memcheck](https://github.com/ddnet/ddnet/#using-addresssanitizer--undefinedbehavioursanitizer-or-valgrinds-memcheck) (optional)
Co-authored-by: ChillerDragon <ChillerDragon@gmail.com>
3804: HiDPI-aware Resolution List r=def- a=TsFreddie
This is a version of https://github.com/teeworlds/teeworlds/pull/2827
Which gives you a resolution list that allows you to select appropriate window size (which is saved in config and used to initialize videos), while showing the canvas size (which is the drawing resolution), at least on MacOS.
This also fixes incorrect clips that happens without HiDPI enabled, because `Graphics()->ScreenWidth()` (in canvas size) was reported incorrectly before, which is now fixed.
Also, I noticed that `m_DesktopScreenWidth` and `m_DesktopScreenHeight` (in window size in MacOS) is used incorrectly or differently from what it was used during initialization:
* During initialization, DesktopScreenWidth is the "screen width of your desktop", but it was used as the "game window size on your desktop" when checking whether to resize window. I added a separate window size calculation to use during resize.
Also, god, I hope SDL's different sizes and stuff is reported consistently at least in Wayland or X, so please test. I know in Windows they are always in pixel/canvas size so it might not be a huge issue there.
## Checklist
- [x] Tested the change ~~ingame~~ on MacOS
- [ ] Provided screenshots if it is a visual change
- [x] Tested in combination with possibly related configuration options (again, on MacOS, with both HiDPI and non-HiDPI settings)
- [ ] Written a unit test if it works standalone, system.c especially
- [ ] Considered possible null pointers and out of bounds array indexing
- [ ] Changed no physics that affect existing maps
- [ ] Tested the change with [ASan+UBSan or valgrind's memcheck](https://github.com/ddnet/ddnet/#using-addresssanitizer--undefinedbehavioursanitizer-or-valgrinds-memcheck) (optional)
Co-authored-by: Freddie Wang <tsfreddiewang@gmail.com>
I assume shift is a common compose bind. So users might have bound
shift+scroll or shift+arrowkeys to something else and do not want to
change the playback speed while triggering that compose bind.
3815: Remove "Show only supported" switch from graphics tab r=def- a=ardadem
gfx_display_all_modes can be used instead
## Checklist
- [X] Tested the change ingame
- [ ] Provided screenshots if it is a visual change
- [ ] Tested in combination with possibly related configuration options
- [ ] Written a unit test if it works standalone, system.c especially
- [ ] Considered possible null pointers and out of bounds array indexing
- [ ] Changed no physics that affect existing maps
- [ ] Tested the change with [ASan+UBSan or valgrind's memcheck](https://github.com/ddnet/ddnet/#using-addresssanitizer--undefinedbehavioursanitizer-or-valgrinds-memcheck) (optional)
Co-authored-by: Arda Demir <ddmirarda@gmail.com>
3834: Remove alpha clamp r=def- a=Jupeyy
Was still from the first version of the HSL sliders, which didn't had a border around the selector and made the alpha slider invisible(for menu colors slider), but thats not needed anymore(bcs borders are not transparent).
## Checklist
- [x] Tested the change ingame
- [ ] Provided screenshots if it is a visual change
- [ ] Tested in combination with possibly related configuration options
- [ ] Written a unit test if it works standalone, system.c especially
- [ ] Considered possible null pointers and out of bounds array indexing
- [ ] Changed no physics that affect existing maps
- [ ] Tested the change with [ASan+UBSan or valgrind's memcheck](https://github.com/ddnet/ddnet/#using-addresssanitizer--undefinedbehavioursanitizer-or-valgrinds-memcheck) (optional)
Co-authored-by: Jupeyy <jupjopjap@gmail.com>
Summary
=======
The idea of this is that clients will not have to ping each server for
server infos which takes long, leaks the client's IP address even to
servers the user does not join and is a DoS vector of the game servers
for attackers.
For the Internet, DDNet and KoG tab, the server list is entirely fetched
from the master server, filtering out servers that don't belong into the
list.
The favorites tab is also supposed to work that way, except for servers
that are marked as "also ping this server if it's not in the master
server list".
The LAN tab continues to broadcast the server info packet to find
servers in the LAN.
How does it work?
=================
The client ships with a list of master server list URLs. On first start,
the client checks which of these work and selects the fastest one.
Querying the server list is a HTTP GET request on that URL. The
response is a JSON document that contains server infos, server addresses
as URLs and an approximate location.
It can also contain a legacy server list which is a list of bare IP
addresses similar to the functionality the old master servers provided
via UDP. This allows us to backtrack on the larger update if it won't
work out.
Lost functionality
==================
(also known as user-visible changes)
Since the client doesn't ping each server in the list anymore, it has no
way of knowing its latency to the servers.
This is alleviated a bit by providing an approximate location for each
server (continent) so the client only has to know its own location for
approximating pings.
3798: Add GLES support r=def- a=Jupeyy
I know i *****ed up, that i didnt move the files in a different commit, but when i noticed it was already too late.
Its mostly disabling stuff that GLES doesn't support. Global LOD bias or glOrtho for examlpe.
~***Strong note***:
This will move ALL linux implementation to GLES instead of GL, since GLES is supported by X11 and wayland, we probably want some testing for this.
This also means, that the default for Linux gets GLES 3.0 which is equivalent to OpenGL 3.3, not OpenGL 3.0!!!~
~If you wonder "why":
Gles just works differently to OGL, it knows about all functions that exists, and e.g. doesn't require GLEW at all, but that also means, that GLES does NOT work while we use OGL, because GLEW would kill GLES.~
GLEW in EGL mode works fine with wayland, just requires X11 to use EGL too(SDL_VIDEO_X11_FORCE_EGL)
_Additionally, for GLES only, only 3.0 and 1.0 context are supported._(doesn't matter anymore for linux)
probably fixes:
~not_really_fixes #3797~
fixes#2885
And would also allow android support, if i ever get these build scripts upto date, or smby is interested into fixing them for me :D
## Checklist
- [x] Tested the change ingame
- [ ] Provided screenshots if it is a visual change
- [ ] Tested in combination with possibly related configuration options
- [ ] Written a unit test if it works standalone, system.c especially
- [ ] Considered possible null pointers and out of bounds array indexing
- [ ] Changed no physics that affect existing maps
- [ ] Tested the change with [ASan+UBSan or valgrind's memcheck](https://github.com/ddnet/ddnet/#using-addresssanitizer--undefinedbehavioursanitizer-or-valgrinds-memcheck) (optional)
Co-authored-by: Jupeyy <jupjopjap@gmail.com>
3800: Get rid of current tuning in predicted world r=Jupeyy a=TsFreddie
Missed a spot where the predicted world still returns current tuning.
closes#3799
## Checklist
- [x] Tested the change ingame
- [ ] Provided screenshots if it is a visual change
- [ ] Tested in combination with possibly related configuration options
- [ ] Written a unit test if it works standalone, system.c especially
- [ ] Considered possible null pointers and out of bounds array indexing
- [ ] Changed no physics that affect existing maps
- [ ] Tested the change with [ASan+UBSan or valgrind's memcheck](https://github.com/ddnet/ddnet/#using-addresssanitizer--undefinedbehavioursanitizer-or-valgrinds-memcheck) (optional)
Co-authored-by: Freddie Wang <tsfreddiewang@gmail.com>
3792: Add hammer_hit_fire_delay tune r=heinrich5991 a=def-
as requested by J$ON for new map with Genex
the current hammer_fire_delay is only used when no other tee has been
hit. In order not to change any existing behaviour I have added a new
tune for the case of hitting, instead of making the existing one control
both.
<!-- What is the motivation for the changes of this pull request -->
## Checklist
- [x] Tested the change ingame
- [ ] Provided screenshots if it is a visual change
- [ ] Tested in combination with possibly related configuration options
- [ ] Written a unit test if it works standalone, system.c especially
- [ ] Considered possible null pointers and out of bounds array indexing
- [ ] Changed no physics that affect existing maps
- [ ] Tested the change with [ASan+UBSan or valgrind's memcheck](https://github.com/ddnet/ddnet/#using-addresssanitizer--undefinedbehavioursanitizer-or-valgrinds-memcheck) (optional)
Co-authored-by: def <dennis@felsin9.de>
as requested by J$ON for new map with Genex
the current hammer_fire_delay is only used when no other tee has been
hit. In order not to change any existing behaviour I have added a new
tune for the case of hitting, instead of making the existing one control
both.
3784: Fix serverbrowser crash r=heinrich5991 a=Jupeyy
Putting a search string that doesnt find any server, or hold Up key while loading the client, or switching tabs can cause a crash, bcs the server list is empty, but input signals, it was changed
## Checklist
- [ ] Tested the change ingame
- [ ] Provided screenshots if it is a visual change
- [ ] Tested in combination with possibly related configuration options
- [ ] Written a unit test if it works standalone, system.c especially
- [ ] Considered possible null pointers and out of bounds array indexing
- [ ] Changed no physics that affect existing maps
- [ ] Tested the change with [ASan+UBSan or valgrind's memcheck](https://github.com/ddnet/ddnet/#using-addresssanitizer--undefinedbehavioursanitizer-or-valgrinds-memcheck) (optional)
Co-authored-by: Jupeyy <jupjopjap@gmail.com>
3778: Fix composed binds involving F-keys r=def- a=Learath2
<!-- What is the motivation for the changes of this pull request -->
This fixes the issues for the meanwhile, but I think I want to rewrite this. We are kind of wasting memory and it behaves really oddly with multiple binds that match the chord progression. The bug was introduced in #2675Fixed#3771 for now. @fokkonaut please check
## Checklist
- [x] Tested the change ingame
- [ ] Provided screenshots if it is a visual change
- [ ] Tested in combination with possibly related configuration options
- [ ] Written a unit test if it works standalone, system.c especially
- [x] Considered possible null pointers and out of bounds array indexing
- [ ] Changed no physics that affect existing maps
- [ ] Tested the change with [ASan+UBSan or valgrind's memcheck](https://github.com/ddnet/ddnet/#using-addresssanitizer--undefinedbehavioursanitizer-or-valgrinds-memcheck) (optional)
Co-authored-by: Learath2 <learath2@gmail.com>
3788: Allow unpredicted TuneZone settings for projectiles and lasers r=def- a=TsFreddie
Video demo: https://cdn.discordapp.com/attachments/293493549758939136/834971531464671252/2021-04-23_09-57-35.mp4
(notice the speed difference between player's shotgun bullets and the bouncing ones)
Video demo2: https://cdn.discordapp.com/attachments/293493549758939136/834990657147764736/demotopleftisnew.mp4
(Right: before, Left: after)
Since TuneZones are known (baked into the map), and entities' position received are known. We have a pretty good reason, if the server is believed to make use of it (GAMEINFOFLAG_PREDICT_DDRACE_TILES), to make entities always reference TuneZone settings, even if they are not predicted.
Introduced a new worldconfig, `m_UseTuneZones`, that ignores antiping settings and apply TuneZone where applicable if the server is believed to support TuneZone.
When antiping is off, predictedworld is still inactive and only the rendering are affected, making them still **unpredicted** but now with correct intra-tick interpolations.
## Checklist
- [x] Tested the change ingame
- [x] Provided screenshots if it is a visual change
- [x] Tested in combination with possibly related configuration options (with different antiping settings and ping)
- [ ] Written a unit test if it works standalone, system.c especially
- [ ] Considered possible null pointers and out of bounds array indexing
- [x] Changed no physics that affect existing maps (client-side only)
- [ ] Tested the change with [ASan+UBSan or valgrind's memcheck](https://github.com/ddnet/ddnet/#using-addresssanitizer--undefinedbehavioursanitizer-or-valgrinds-memcheck) (optional)
Co-authored-by: Freddie Wang <tsfreddiewang@gmail.com>
3707: Set search field active when opening vote menu ingame r=Jupeyy a=def-
As suggested by RafaelFF
<!-- What is the motivation for the changes of this pull request -->
## Checklist
- [x] Tested the change ingame
- [ ] Provided screenshots if it is a visual change
- [ ] Tested in combination with possibly related configuration options
- [ ] Written a unit test if it works standalone, system.c especially
- [ ] Considered possible null pointers and out of bounds array indexing
- [ ] Changed no physics that affect existing maps
- [ ] Tested the change with [ASan+UBSan or valgrind's memcheck](https://github.com/ddnet/ddnet/#using-addresssanitizer--undefinedbehavioursanitizer-or-valgrinds-memcheck) (optional)
3709: Switch back to using a stable sort r=Jupeyy a=def-
so that for example demos stay sorted the same way when sorting by demo
markers and removing a demo, otherwise they get totally reordered every
time.
as reported by Shyzo
<!-- What is the motivation for the changes of this pull request -->
## Checklist
- [x] Tested the change ingame
- [ ] Provided screenshots if it is a visual change
- [ ] Tested in combination with possibly related configuration options
- [ ] Written a unit test if it works standalone, system.c especially
- [ ] Considered possible null pointers and out of bounds array indexing
- [ ] Changed no physics that affect existing maps
- [ ] Tested the change with [ASan+UBSan or valgrind's memcheck](https://github.com/ddnet/ddnet/#using-addresssanitizer--undefinedbehavioursanitizer-or-valgrinds-memcheck) (optional)
Co-authored-by: def <dennis@felsin9.de>
ChillerDragon
