Since it causes crashes because another thread just accesses the data.
No one has wanted to fix this, so maybe we should just remove it instead
of having clients crash in editor. I'm wondering how much effort it
would be to make this safe. Would we just have to lock for a short part
or is it basically the entire threaded undo operation that has to be
serialized? (No need for a thread then)
4674: Fix evolve limit r=def- a=heinrich5991
Mixed up previous and current. This did not show up in testing, because
these two were always true at the same time. I've now debugged it and
manually made specific conditions false to verify that it's working.
Sorry for the mess.
<!-- 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: heinrich5991 <heinrich5991@gmail.com>
Mixed up previous and current. This did not show up in testing, because
these two were always true at the same time. I've now debugged it and
manually made specific conditions false to verify that it's working.
Sorry for the mess.
4672: Make the evolve limit more visible r=def- a=heinrich5991
This is to make sure that modders don't start to rely on the fact that
we clamp the evolve duration to 3 seconds.
CC #4624
## 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>
4576: HTTP Serverbrowser: Quit faster (fixes#4563) r=heinrich5991 a=def-
```
[2022-01-08 09:57:03][serverbrowse_http]: found master, url='https://master1.ddnet.tw/ddnet/15/servers.json' time=54ms
[2022-01-08 09:57:03][client]: disconnecting. reason='unknown'
[2022-01-08 09:57:03][http]: https://master2.ddnet.tw/ddnet/15/servers.json failed. libcurl error: Callback aborted
[2022-01-08 09:57:03][serverbrowse_http]: master chooser aborted
```
Not instantaneous but much faster, depends on how often curl calls the PROGRESSFUNCTION.
<!-- 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)
4670: Lib updates r=heinrich5991 a=def-
curl 7.79.0 -> 7.81.0
freetype 2.11.0 -> 2.11.1
sqlite 3.36.0 -> 3.37.2
ffmpeg 4.4 -> 5.0
websockets 4.2 -> 4.3
<!-- 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)
4671: Update Chinese translations (by cheeser0613) r=heinrich5991 a=def-
<!-- What is the motivation for the changes of this pull request -->
## 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: Dennis Felsing <dennis@felsin9.de>
4624: Limit evolving to 3 seconds r=heinrich5991 a=ChillerDragon
Took from upstream
5d44714e92 (diff-597779b4eb51af9adfedd04b8a235afff01091a4ba741daa604cb5cbeda4e3daR1306-R1321)
Fixes client freezes if the server sends funny ticks
## 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>
4668: Fix bindings with modifiers r=def- a=Kaffeine
I work on a client for InfClass mod and it is important to have bindings for many situations, so we have to use modifiers (one modifier per bind).
It turned out that bindings (with a modifier) are broken in DDNet since commit 4326d00df9 which changes the implementation to support modifier combinations but it messes up the UI part.
I've considered options to revert the commit or to fix the combined modifiers and bravely 😅 choose the latter way. The code had at least five bugs and fixing e.g. four of them does not help much, making it really hard to say if the fix is correct or not. I had to rename some vars first to make the code readable and now (with the correct names) each fix should seem reasonable.
`Modifier` was confused with `ModifierCombination` in so many places. I do not like the idea of deviating from the `teeworlds` codebase but we already did it here by introducing of combined modifiers. We do not use a stronger type system (enum classes and flag classes) but one thing we can do to give a developer the idea of what is going on is to use the correct variable and argument names at least.
## Checklist
- [x] Tested the change ingame
- [x] 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)
I checked the follow cases:
- Binding of a normal key (e.g. `t`)
- Binding of a sole modifier (e.g. `lctrl`)
- Binding of a key combined with a modifier (e.g. `ctrl+t`)
- Binding of a key combined with many modifier (e.g. `ctrl+alt+t`)
### Before
Bindings with `Shift` modifiers displayed as `Ctrl+`, bindings with `Ctrl` or `Alt` are not displayed at all.
### After
All bindings displayed correctly
but also it is possible to set any crazy binding from UI
Co-authored-by: Alexander Akulich <akulichalexander@gmail.com>
4669: Fix projectiles rendering during demo playback (+ few more small changes) r=def- a=Kaffeine
Due to an upstream bug, some projectiles were not rendered the demo playback.
Rephrasing my in-code comment: "The render timing is usually screwed during demo playback because `s_LastGameTickTime` depends on the system timer, while the other part `Client()->PrevGameTick()-pCurrent->m_StartTick)/(float)SERVER_TICK_SPEED` is virtually constant (for projectiles fired on the current game tick): `(x - (x+2)) / 50 = -0.04`
We have a strict comparison for the passed time being more than the time between ticks `if(CurtickStart > m_Info.m_CurrentTime)` in `CDemoPlayer::Update()` so on the practice the typical value of `s_LastGameTickTime` varies from 0.02386 to 0.03999 which leads to Ct from -0.00001 to -0.01614."
Round up those to 0.0 to fix missing rendering of the projectile.
My merged MR to the upstream: https://github.com/teeworlds/teeworlds/pull/3002
## Checklist
- [x] Tested the change ingame
- [x] 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)
### Before the fix
(demo playback)
### After the fix
All projectiles are rendered correctly
Co-authored-by: Alexander Akulich <akulichalexander@gmail.com>
The modifiers order in the UI and console follows the order
of modifier enum items.
The conventional order of the modifiers is: Ctrl, Alt, Shift, GUI (Win, Meta).
Place Shift after Alt to get proper order, e.g.:
Alt+Shift+R (conventional) instead of Shift+Alt+R (weird)
The first dimention of m_aapKeyBindings is ModifierCombination
since commit 4326d00df9.
Rename the arguments to make it clear when the arg is a combination
(e.g. the arg value '2' refers to '1 << MODIFIER_SHIFT' instead of 'MODIFIER_CTRL')
This is needed to make the code cleaner and the bugs more obvious.
4667: Update Copyright year r=def- a=dragere
Same as last year
## 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: dragere <hjahn25@gmail.com>
4661: Make spawn position independent of players in other teams r=heinrich5991 a=def-
As suggested by Rockus, should help speedrunners who get followed by
other players
<!-- 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>
4666: Fall back to Visual Studio 16 r=heinrich5991 a=def-
since it seems not all github worflow servers are on VS17 yet
<!-- What is the motivation for the changes of this pull request -->
## 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: def <dennis@felsin9.de>
4665: Support saving any size binds (fixes#4660) r=heinrich5991 a=def-
Since we store the bind with malloc, we also need to use malloc here to
allow any size when saving the binds in config file
<!-- 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>
4664: Update translations r=heinrich5991 a=def-
<!-- What is the motivation for the changes of this pull request -->
## 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: def <dennis@felsin9.de>
4663: Update spanish.txt r=def- a=n0Ketchp
<!-- What is the motivation for the changes of this pull request -->
## 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: noKetchup <93238471+n0Ketchp@users.noreply.github.com>
4654: Add a groundjump-only tile r=def- a=heinrich5991
Implements a new mode of jumping: "you only have ground jump". This differs from "you only have one jump" in that it doesn't allow you to air-jump, even if you haven't jumped to get off the ground. To access this mode, it reuses the "air-jump tile" in the switch layer with the number of jumps set to 255.
This tile is planned to be used in tutorial developed by unique2 and Alisa.
## 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>
4655: Add windowed fullscreen ( a special case for Windows ) r=def- a=Jupeyy
fixes#4597
## 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>
4657: Reduce system API calls when listing directory contents r=heinrich5991 a=Robyt3
Optimize `fs_listdir` and `fs_listdir_fileinfo` by reducing nested calls to `fs_is_dir` (and subsequent additional system API calls).
- Windows: check file attributes in `WIN32_FIND_DATAW finddata` instead of calling WinAPI again.
- Others: check file type of `struct dirent *entry` instead of calling `stat`. The file type may not be available on all filesystems, so a call to `fs_is_dir` is used as fallback.
Tested on Windows and Ubuntu.
See #4652. This reduces additional overhead.
## 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: Robert Müller <robert.mueller@uni-siegen.de>
4659: Try using Windows 2022 (fixes#4658) r=Jupeyy a=def-
<!-- What is the motivation for the changes of this pull request -->
## 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: Dennis Felsing <dennis@felsin9.de>
4650: Lower chance of losing written files by syncing them to disk on close, my try r=def- a=heinrich5991
Not sure if it helps, but now it only syncs the config file.
Would supersede #4407.
## Checklist
- [x] Tested the change ingame
- [ ] Provided screenshots if it is a visual change
- [ ] Tested in combination with possibly related configuration options
- [x] 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>
Co-authored-by: heinrich5991 <heinrich5991@gmail.com>
