How did this not blow up all day long for years?!
Found by UBSAN:
/home/teeworlds/src/master/src/game/server/entities/dragger.cpp:32:29: runtime error: member call on address 0x0000013319b8 which does not point to an object of type 'CCharacter'
0x0000013319b8: note: object has invalid vptr
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
^~~~~~~~~~~~~~~~~~~~~~~
invalid vptr
#0 0x5501de in CDragger::Move() /home/teeworlds/src/master/src/game/server/entities/dragger.cpp:32:29
#1 0x5522c2 in CDragger::Tick() /home/teeworlds/src/master/src/game/server/entities/dragger.cpp:153:3
#2 0x5eb11e in CGameWorld::Tick() /home/teeworlds/src/master/src/game/server/gameworld.cpp:261:11
#3 0x58d660 in CGameContext::OnTick() /home/teeworlds/src/master/src/game/server/gamecontext.cpp:751:10
#4 0x4afefe in CServer::Run() /home/teeworlds/src/master/src/engine/server/server.cpp:2558:19
#5 0x4c26fd in main /home/teeworlds/src/master/src/engine/server/server.cpp:3551:21
#6 0x7f2c8cc9309a in __libc_start_main (/lib/x86_64-linux-gnu/libc.so.6+0x2409a)
#7 0x451509 in _start (/home/teeworlds/servers/DDNet-Server-ubsan+0x451509)
3237: Create fallback wordlist in sourcecode (fixes#3206) r=Learath2 a=def-
<!-- 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
- [x] Considered possible null pointers and out of bounds array indexing
- [ ] Changed no physics that affect existing maps
- [x] 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>
3222: Kick-voting someone from team doesn't kill locked team (fixes#3207) r=heinrich5991 a=def-
<!-- 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
- [x] Changed no physics that affect existing maps
- [x] 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>
3233: Fix clang-tidy in text.cpp 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>
3117: Enable further clang-tidy checks to modernize the code r=heinrich5991 a=def-
https://clang.llvm.org/extra/clang-tidy/checks/modernize-use-auto.html
And fix the code:
cmake -GNinja -DWEBSOCKETS=OFF -DMYSQL=ON -DCMAKE_BUILD_TYPE=RelWithDebInfo -DVIDEORECORDER=ON -DAUTOUPDATE=OFF -DANTIBOT=OFF -DUPNP=OFF -DPREFER_BUNDLED_LIBS=OFF -DINFORM_UPDATE=OFF -DCMAKE_CXX_CLANG_TIDY="clang-tidy;--fix-errors" -DCMAKE_C_CLANG_TIDY="clang-tidy;--fix-errors" .
ninja -j 1
Manually fixed sizeof(auto) in a few occurences (i.e. server.cpp:1359)
First steps for #3088
Co-authored-by: def <dennis@felsin9.de>
3231: Fix skin index r=def- a=Jupeyy
Beside that this was wrong anyway, i hoped i could atleast get the index from the sorted list insert, but that returns the last index in the list, instead of the new inserted index(just like rn), so just revert what I added in 8d6148b20b
## 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>
3228: Add m_PosEnv (of sound sources) as index that must be offset when appending a map r=heinrich5991 a=Patiga
If you append a map to another, all indices of it must be offset. This doesn't happen with the position envelope index of sound sources, meaning it would (if the original maps had at least one envelope) point to the wrong envelope.
## Checklist
- [x] Tested the change ingame
- [ ] Provided screenshots if it is a visual change
- [x] 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
- [x] 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: Patiga <dev@patiga.eu>
3220: Fix overlapping of combined binds with modifiers [Fixes#3182] r=heinrich5991 a=Banana090
Just fixes https://github.com/ddnet/ddnet/issues/3182
Works as it used to work, but does not call for binds with no modifier when the same key was used already for modified bind
Co-authored-by: Дядя Женя <spy090@yandex.ru>
3221: Revert "Show halloween skins only on Halloween" r=Jupeyy a=def-
People are clearly unhappy with this event, so let's revert it. On
servers I've set events 0 already.
This reverts commit ae90bdf3f9.
## 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>
3204: Much more readable Call Vote menu and code r=def- a=Banana090
I find "Call Vote" menu useless af right now, because it's simply impossible to read it. So when you want to call vote for a map you open website and find the one you want, after that making /map skipping the "Call Vote" menu completely. It's useful for calling for random map btw, but anyway too ugly to look at.
**OLD LOOK**
**NEW LOOK**
**NEW LOOK (RconAuted)**
Changed nothing, but row and text size + View rect size, because for some reason it was small for preserving space for RCON Auted users. Why?
Also changed the name of rect "Extended" to "RconExtension" to make it more clear what is it for.
Would be cool also to remove all that junk that is not actual votes from DDNet server vote list (finishers, average time) and merge it into one line with map's name.
## 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
- [x] Considered possible null pointers and out of bounds array indexing
- [x] 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: Дядя Женя <spy090@yandex.ru>
3203: Fix style in render.cpp 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>
3197: 0.7 map converter - Stop embedding rgb images r=heinrich5991 a=Patiga
DDNet doesn't CMapItemTypeImage version 2, meaning it doesn't have
support yet for having rgb images in maps. Currently, trying that would result in
invalid map files.
## 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
- [x] Considered possible null pointers and out of bounds array indexing
- [x] 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: Patiga <dev@patiga.eu>
DDNet doesn't CMapItemTypeImage version 2, meaning it doesn't have
support yet for having rgb images in maps. Currently, trying that would result in
invalid map files.
3195: Use str_truncate more often 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>
3128: Add add_map_votes command for easier local map testing r=heinrich5991 a=def-
No need to tell new players how to manually add a vote or load a map from rcon
Co-authored-by: def <dennis@felsin9.de>
Co-authored-by: Dennis Felsing <dennis@felsin9.de>
3124: Remove option to make map sounds external r=def- a=Patiga
Right now it is impossible to have a working external sound. External
sounds render unusable the moment the map is saved and given to another
player. For the mapper it seems like everything is working since the
mapper has the sound file. For anyone else it simply won't work. Since
1. there is no use for it and
2. it adds unnecessary complexity and confusion,
I propose removing it entirely.
None of the ddnet maps use external sounds, only 25 maps from the map archive have external sounds
Please have a look over the changed ui code.
Co-authored-by: Patiga <dev@patiga.eu>
3077: UI Elements r=def- a=Jupeyy
Build will fail bcs of clang-format,
i'll update this comment to explain.
Co-authored-by: Jupeyy <jupjopjap@gmail.com>
3107: Log close error to console r=Jupeyy a=def-
3155: Add memcheck/asan/ubsan to readme r=Jupeyy a=def-
To be linked from https://github.com/ddnet/ddnet/pull/3133
Co-authored-by: def <dennis@felsin9.de>
3184: Add non-space blanks: Hangul Filler characters r=Jupeyy a=def-
Taken from https://en.wikipedia.org/wiki/Whitespace_character#Non-space_blanks
Had to Hangul Jungseong Filler to that Wiki article as it was missing.
<!-- 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>
3191: Fix IME state softlock under fcitx and macos + Update DoEditBox logic to be the same as chat r=def- a=TsFreddie
IME state fix closes#3185 and closes#3150
DoEditBox update closes#3186, now cursor behave the same as chat where only one cursor (either the real one or the editing one) will be shown.
## 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
- [x] 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: TsFreddie <tsfreddiewang@gmail.com>
3161: Fix texture create with multi units r=def- a=Jupeyy
~Possibly `fixes` #3104~
Ravie can you test?
Does not fix the issue.
[only affects gl 3.3]
Still is required, bcs that looks wrong
Co-authored-by: Jupeyy <jupjopjap@gmail.com>
3115: Smooth Camera r=def- a=TsFreddie
closes#3096
Since nobody expressed any opinions, how about we get this in and see some reactions. it is off by default anyway.
Co-authored-by: TsFreddie <tsfreddiewang@gmail.com>
3157: Show option to disable system messages in settings r=Jupeyy a=def-
## Summary
Makes it a bit easier to find, prefer it on a busy server. Also clarified the friend message toggle is for highlighting only.
## Checklist
- [x] Tested the change ingame
- [x] Provided screenshots if it is a visual change
- [x] Tested in combination with possibly related configuration options [does not apply]
- [x] Written a unit test if it works standalone, system.c especially [does not apply]
- [x] Considered possible null pointers and out of bounds array indexing
- [x] Changed no physics that affect existing maps
- [x] 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>
3160: Make gfx_quad_as_triangle 0 default r=def- a=Jupeyy
fixes#3097
I added a TrianglesBeing/End() instead for ingame quads
Let me quickly explain why #177 failed.
In the screenshots of #177 you already see the problem
the driver can built a quad like:
or like:
or somehow it wants, but it will always do them as triangles
And thats why the ingame quads were basically flipped in the screenshots of #177
But this almost never matters.
Tiles are single colored, text is single colored, ui elements are single colored.
So it only matters for ingame quads probably, which specify the color of each edge
Also even if it would matter, we can simply change the quads to triangles now.
Pro:
- Generally less vertices upload
- Less vertices building on CPU side
Cons:
- nothing
Co-authored-by: Jupeyy <jupjopjap@gmail.com>
3122: Fix arithmentic problems r=def- a=Jupeyy
The first three commits are safe
The ones i'll add later will be inside the gamecore, so might alter behaviour.
I'll see which ones are critical, but takes me some time
i fixed most in #3081
3137: Add gfx debug r=def- a=Jupeyy
If someone wants to try, you have to execute the client with a config that contains dbg_gfx 1
If opengl debug is supported it will show "Enabled OpenGL debug mode" in the console
Co-authored-by: Jupeyy <jupjopjap@gmail.com>
/home/teeworlds/src/master/src/game/server/ddracechat.cpp:1108:65: runtime error: signed integer overflow: 99999999 * 50 cannot be represented in type 'int'
3149: Fix memory leak on team finish and read of uninitialized data r=def- a=Zwelf
Fixes#2522, fixes#3144 and fixes second leak from #3134.
Co-authored-by: Zwelf <zwelf@strct.cc>
3139: Use NUM_DUMMYS for dummy array length r=def- a=ChillerDragon
Should make code more readable by explaining that the array is used for dummy. Also makes modders life easier if there is a need to scale up dummy count.
I left out two places that do not include client.h
3be8a592e5/src/game/client/prediction/gameworld.h (L92)3be8a592e5/src/game/gamecore.h (L212)
Moving the enum to protocol.h for example is probably not nice.
Co-authored-by: ChillerDrgon <ChillerDragon@gmail.com>
3129: Editor: Fix selection of None value in auto rules (fixes#3013) r=Jupeyy a=def-
by starting the indexing at 0
3138: Fix out of bounds access in GetSolo r=Jupeyy a=def-
As discovered in #3135
3143: Ignore ddnet-info.json dl failure again r=Jupeyy a=def-
too aggressive, might just be bad internet at the moment
Co-authored-by: def <dennis@felsin9.de>
3147: Clamp font size to atleast 2 (fixes#3145) r=def- a=Jupeyy
We dont need to `close` #3146
i think setting it to 0 makes no sense anyway the config to disable entities text is `cl_text_entities` not this one
Co-authored-by: Jupeyy <jupjopjap@gmail.com>
Direct leak of 768 byte(s) in 2 object(s) allocated from:
#0 0x4e91ad in operator new(unsigned long)
#1 0x7f0c54a3bf63 in sql::mysql::MySQL_Prepared_ResultSet::getBlob(unsigned int) const (/usr/lib/x86_64-linux-gnu/libmysqlcppconn.so.7+0x77f63)
