Instead of returning the number of bytes written, which are platform specific, return `true` on success and `false` on failure, so no platform specific code is required when checking the result.
Use `bytes_be_to_uint` and `uint_to_bytes_be` instead.
As casting between `int` and `unsigned` preserves the bit representation of the value, it's not necessary to apply additional tricks to convert between `char` arrays and `int`.
6313: Use `str_copy` instead of `str_format`, minor fixes and refactoring r=Chairn a=Robyt3
## 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 (especially base/) or added coverage to integration test
- [ ] 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 <robytemueller@gmail.com>
Using `str_format` without format arguments is equivalent to `str_copy`, but using the latter is more efficient and readable.
As `str_format` also returns the result `str_utf8_fix_truncation`, i.e. the potentially truncated string length, the return value is also added to `str_copy` so existing invocations don't need to be adjusted.
6299: Show error message when downloaded map cannot be saved r=def- a=Robyt3
Check if deleting the old map file or renaming the temporary downloaded map fails. If so, show an error message which indicates that the user should delete the map file manually.
Sometimes downloaded map files seem to end up with wrong permissions, ownership or with read-only flag set, which makes the client unable to delete them.
Closes#5825.
## 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 (especially base/) or added coverage to integration test
- [ ] 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 <robytemueller@gmail.com>
6289: Fix `sv_vanilla_antispoof` for vanilla 0.6.5 clients r=def- a=Robyt3
This fixes issues that prevented vanilla 0.6.5 clients from joining DDNet servers with `sv_vanilla_antispoof 1` (closes#2074).
The dummy map was not valid. The size of the tiles and game layers was calculated incorrectly. The last member variable included in `CMapItemLayerTilemap` version 2 should be `m_Data`. Previously only the size of the member `m_aName` was subtracted from the total size, which was resulting in an incorrect item size, as the size of the following 5 members also needs to be subtracted.
The handshake messages were packed incorrectly. The message ID and the system flag were not added to the six packers that are passed to `SendMsgs`. The lines that are removed in this function seemed to assume that the message ID was already added but not packed, which was no longer the case. Presumable at some point `CMsgPacker` was changed without adapting the vanilla antispoof feature. Now, the message ID and system flag are properly packed when initially creating the message packers.
The `dummy_map` tool is improved to also print the generated map file's hashes and data as a C style array, so the data can immediately be copied to `network_server.cpp`.
## 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 (especially base/) or added coverage to integration test
- [ ] 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: Robert Müller <robytemueller@gmail.com>
As described in #5844, sometimes the color kept changing when activating a color picker.
By rounding the color components before packing them into an unsigned, the color only changes very little and only at most once when activating a color picker.
6217: Register application separately to specify its displayed name r=def- a=Robyt3
Follow-up from #6199.
Adding version information to the executable (#6203) doesn't result in the name being shown in the Windows settings. There is a separate registry key where applications can register a readable name.
See: https://learn.microsoft.com/en-us/windows/win32/shell/app-registration
## 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 (especially base/) or added coverage to integration test
- [ ] 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 <robytemueller@gmail.com>
6199: Register protocol and file extensions on client launch on Windows r=def- a=Robyt3
When launching the client on Windows, associate the protocol `ddnet` and the file extensions `.map` and `.demo` with the client executable.
See #6072.
## 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 (especially base/) or added coverage to integration test
- [ ] 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 <robytemueller@gmail.com>
This function removes the registry keys that are created with `shell_register_protocol` and `shell_register_extension`.
According to the Microsoft documentation, only the keys for the program IDs should be deleted.
The keys that associate the file extensions with the program IDs should be kept, as Windows will automatically ignore the value if the program ID does not exist.
See: https://learn.microsoft.com/en-us/windows/win32/shell/fa-file-types#deleting-registry-information-during-uninstallation
The `shell_update` function notifies the system when the shell needs to be updated due to a changed protocol or file association.
An output parameter is added to the `shell_register_protocol` and `shell_register_extension` functions, to determine whether the shell needs to be updated after calling the functions.
We only check whether the application path and program association were changed, instead of checking whether any key or value was changed, as this reduces the amount of necessary checks and we assume that the other values are not externally changed.
Because updating the shell is a potentially expensive operation, this should only be done when necessary and only once after registering all protocols and extensions.
This function register a file extension on Windows by creating the necessary registry keys and values.
The file extension is only registered for the current user in the registry key `HKEY_CURRENT_USER\SOFTWARE\Classes`, so admin privileges are not required.
Windows does not allow changing the default file extension handler programatically. When the user opens a file for which a new file extension handler exists, a dialog is shown that allows the user to change the default handler.
This function registers a protocol handler on Windows by creating the necessary registry keys and values.
The handler is only registered for the current user in the registry key `HKEY_CURRENT_USER\SOFTWARE\Classes`, so admin privileges are not required.
6191: Use own defines of format specifiers for MinGW (fixes#6187) r=Jupeyy a=Chairn
Counter proposition to #6188. Redefines our own specifiers
## 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 (especially base/) or added coverage to integration test
- [ ] 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: Chairn <chairn.nq@hotmail.fr>
6035: Fix various issues reported by cppcheck static analyser r=def- a=Robyt3
After generating `compile_commands.json` with cmake, I ran [cppcheck](https://cppcheck.sourceforge.io/) like this:
```
cppcheck --project=compile_commands.json -DWIN64 --suppressions-list=cppcheck.supp --enable=all 2>cppcheck.log
```
With these suppressions in `cppcheck.supp`:
```
cstyleCast
useStlAlgorithm
unusedFunction
variableScope
noExplicitConstructor
useInitializationList
noConstructor
uninitMemberVar
uninitMemberVarPrivate
uninitDerivedMemberVar
uninitStructMember
uninitvar
shadowFunction
memleakOnRealloc
internalAstError
virtualCallInConstructor
unknownMacro
noOperatorEq
noCopyConstructor
```
Many of these occur too often or are false positives.
Here is a list of all remaining non-suppressed issues reported by cppcheck: [cppcheck.log](https://github.com/ddnet/ddnet/files/9997663/cppcheck.log)
And here is a list of all remaining issues including the suppressed ones: [cppcheck_all.log](https://github.com/ddnet/ddnet/files/9997662/cppcheck_all.log)
I couldn't get cppcheck's command line argument to ignore the external folders to work correctly, so I manually removed those entries from the 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 (especially base/) or added coverage to integration test
- [ ] 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 <robytemueller@gmail.com>
According to cppchecker's `constParameter` error:
```
src\engine\gfx\image_manipulation.cpp:7:58: style: Parameter 'pSrc' can be declared as pointer to const [constParameter]
static void Dilate(int w, int h, int BPP, unsigned char *pSrc, unsigned char *pDest, unsigned char AlphaThreshold = TW_DILATE_ALPHA_THRESHOLD)
^
src\engine\gfx\image_manipulation.cpp:58:67: style: Parameter 'pSrc' can be declared as pointer to const [constParameter]
static void CopyColorValues(int w, int h, int BPP, unsigned char *pSrc, unsigned char *pDest)
^
src\engine\shared\network_conn.cpp:241:42: style: Parameter 'Addr' can be declared as reference to const [constParameter]
void CNetConnection::DirectInit(NETADDR &Addr, SECURITY_TOKEN SecurityToken, SECURITY_TOKEN Token, bool Sixup)
^
src\base\system.cpp:4060:71: style: Parameter 'random' can be declared as pointer to const [constParameter]
void generate_password(char *buffer, unsigned length, unsigned short *random, unsigned random_length)
^
src\engine\client\backend\vulkan\backend_vulkan.cpp:263:38: style: Parameter 'AllocatedMemory' can be declared as reference to const [constParameter]
void Free(SMemoryHeapQueueElement &AllocatedMemory)
^
src\engine\client\backend\vulkan\backend_vulkan.cpp:1708:47: style: Parameter 'ImgExtent' can be declared as reference to const [constParameter]
static size_t ImageMipLevelCount(VkExtent3D &ImgExtent)
^
src\engine\client\backend\vulkan\backend_vulkan.cpp:2801:29: style: Parameter 'Image' can be declared as reference to const [constParameter]
void ImageBarrier(VkImage &Image, size_t MipMapBase, size_t MipMapCount, size_t LayerBase, size_t LayerCount, VkFormat Format, VkImageLayout OldLayout, VkImageLayout NewLayout)
^
src\engine\client\backend\vulkan\backend_vulkan.cpp:6495:46: style: Parameter 'ExecBuffer' can be declared as reference to const [constParameter]
void Cmd_Clear(SRenderCommandExecuteBuffer &ExecBuffer, const CCommandBuffer::SCommand_Clear *pCommand)
^
src\game\client\components\skins.cpp:83:72: style: Parameter 'pImg' can be declared as pointer to const [constParameter]
static void CheckMetrics(CSkin::SSkinMetricVariable &Metrics, uint8_t *pImg, int ImgWidth, int ImgX, int ImgY, int CheckWidth, int CheckHeight)
^
src\game\client\prediction\entities\character.h:106:37: style: Parameter 'pNewInput' can be declared as pointer to const [constParameter]
void SetInput(CNetObj_PlayerInput *pNewInput)
^
src\game\client\prediction\gameworld.cpp:245:106: style: Parameter 'pNotThis' can be declared as pointer to const [constParameter]
CCharacter *CGameWorld::IntersectCharacter(vec2 Pos0, vec2 Pos1, float Radius, vec2 &NewPos, CCharacter *pNotThis, int CollideWith, class CCharacter *pThisOnly)
^
src\game\client\prediction\gameworld.cpp:245:151: style: Parameter 'pThisOnly' can be declared as pointer to const [constParameter]
CCharacter *CGameWorld::IntersectCharacter(vec2 Pos0, vec2 Pos1, float Radius, vec2 &NewPos, CCharacter *pNotThis, int CollideWith, class CCharacter *pThisOnly)
^
src\game\client\prediction\gameworld.cpp:283:116: style: Parameter 'pNotThis' can be declared as pointer to const [constParameter]
std::list<class CCharacter *> CGameWorld::IntersectedCharacters(vec2 Pos0, vec2 Pos1, float Radius, class CEntity *pNotThis)
^
src\game\client\ui.cpp:522:180: style: Parameter 'pReadCursor' can be declared as pointer to const [constParameter]
void CUI::DoLabel(CUIElement::SUIElementRect &RectEl, const CUIRect *pRect, const char *pText, float Size, int Align, const SLabelProperties &LabelProps, int StrLen, CTextCursor *pReadCursor)
^
src\game\client\ui_scrollregion.cpp:23:86: style: Parameter 'pParams' can be declared as pointer to const [constParameter]
void CScrollRegion::Begin(CUIRect *pClipRect, vec2 *pOutOffset, CScrollRegionParams *pParams)
^
src\game\server\scoreworker.h:239:29: style: Parameter 'aTimeCp' can be declared as const array [constParameter]
void Set(float Time, float aTimeCp[NUM_CHECKPOINTS])
^
src\game\server\score.cpp:135:80: style: Parameter 'aTimeCp' can be declared as const array [constParameter]
void CScore::SaveScore(int ClientID, float Time, const char *pTimestamp, float aTimeCp[NUM_CHECKPOINTS], bool NotEligible)
^
src\game\server\teeinfo.cpp:40:57: style: Parameter 'pUseCustomColors' can be declared as pointer to const [constParameter]
CTeeInfo::CTeeInfo(const char *apSkinPartNames[6], int *pUseCustomColors, int *pSkinPartColors)
^
src\game\server\teeinfo.cpp:40:80: style: Parameter 'pSkinPartColors' can be declared as pointer to const [constParameter]
CTeeInfo::CTeeInfo(const char *apSkinPartNames[6], int *pUseCustomColors, int *pSkinPartColors)
^
```
The functions `str_utf8_skeleton_begin` and `str_utf8_skeleton_next` and the `struct SKELETON` are only used internally, so they don't need to be exported.
5599: Add support for Rust code in DDNet r=def- a=heinrich5991
The glue is done using the [cxx crate](https://cxx.rs/) on the Rust side.
As a proof-of-concept, only a small console command (`rust_version`) printing the currently used Rust version was added.
You can generate and open the Rust documentation using `DDNET_TEST_NO_LINK=1 cargo doc --open`.
You can run the Rust tests using `cmake --build <build dir> --target run_rust_tests`, they're automatically included in the `run_tests` target as well.
Rust tests don't work on Windows in debug mode on Windows because Rust cannot currently link with the debug version of the C stdlib on Windows: https://github.com/rust-lang/rust/issues/39016.
---
The stuff in `src/rust-bridge` is generated using
```
cxxbridge src/engine/shared/rust_version.rs --output src/rust-bridge/engine/shared/rust_version.cpp --output src/rust-bridge/engine/shared/rust_version.h
cxxbridge src/engine/console.rs --output src/rust-bridge/cpp/console.cpp --output src/rust-bridge/cpp/console.h
```
Co-authored-by: heinrich5991 <heinrich5991@gmail.com>
Converting wide-char (UTF-16) to multi-byte (UTF-8) takes 1-2 wide-chars from the input and transforms them to 1-4 bytes, so having `char` and `WCHAR` buffers with equal static lengths means that this function can fail due to insufficient buffer sizes when the user has folder or files with very long names (many unicode codepoints).
Therefore checks are added that allow only the error `ERROR_INSUFFICIENT_BUFFER` when `WideCharToMultiByte` fails, which is expected on very long paths, as these would also lead to further errors later in the code. The respective functions will now fail with a return and ignore files that have too long names.
This could only completely be fixed using dynamically sized buffers for all paths, which seems like too much work and overhead.
Other errors are not expected and hence caught by the assertions, as those would indicate programming errors like wrong arguments being passed.
Reference: https://learn.microsoft.com/en-us/windows/win32/api/stringapiset/nf-stringapiset-widechartomultibyte
The second buffer was previously used when calling `fs_is_dir` explicity, but it has become from obsolete from #4657 by using the flag `FILE_ATTRIBUTE_DIRECTORY` instead.
Extract `windows_format_system_message` and allocate appropriate buffer for the formatted system error messages, so messages of any length can be displayed.
The flag `FORMAT_MESSAGE_ALLOCATE_BUFFER` is used so `FormatMessageW` allocates the buffer which must later be freed with `LocalFree`.
The flag `FORMAT_MESSAGE_MAX_WIDTH_MASK` is also added so the formatted message will not contain any line breaks at the end, which would make log messages less readable.
Reference: https://learn.microsoft.com/en-us/windows/win32/api/winbase/nf-winbase-formatmessagew
Converting multi-byte (UTF-8) to wide-char (UTF-16) take 1-4 bytes from the input and transforms them to 1-2 wide-chars, so having `char` and `WCHAR` buffers with equal lengths should mean that this function can't fail due to insufficient buffer sizes, unless a path that's already too long for Windows is being used internally.
Other errors are also not expected, as those would indicate programming errors like wrong arguments being passed.
The maximum length for paths in the Win32 API is 260 characters. This limitation can be lifted in Windows 10, but it must be done explicitly (opt-in) both in the Windows Registry by the user and in the application's manifest by us.
References:
- https://learn.microsoft.com/en-us/windows/win32/api/stringapiset/nf-stringapiset-multibytetowidechar
- https://learn.microsoft.com/en-us/windows/win32/fileio/maximum-file-path-limitation?tabs=registry
Make sure the Windows debugger logger does not fail when log messages are too long.
The first call to `MultiByteToWideChar` gets the required size of the wide-char buffer, which should be at least one for the null termination. The second call is expected to produce exactly the same number of characters.
The function `MultiByteToWideChar` can't fail unless wrong arguments are passed, when the buffer size is set correctly by calling the function twice. The returned buffer size on the second call should always match the buffer size determined with the first call.
The glue is done using the [cxx crate](https://cxx.rs/) on the Rust
side.
As a proof-of-concept, only a small console command (`rust_version`)
printing the currently used Rust version was added.
You can generate and open the Rust documentation using
`DDNET_TEST_NO_LINK=1 cargo doc --open`.
You can run the Rust tests using `cmake --build <build dir> --target
run_rust_tests`, they're automatically included in the `run_tests`
target as well.
Rust tests don't work on Windows in debug mode on Windows because Rust
cannot currently link with the debug version of the C stdlib on Windows:
https://github.com/rust-lang/rust/issues/39016.
---
The stuff in `src/rust-bridge` is generated using
```
cxxbridge src/engine/shared/rust_version.rs --output src/rust-bridge/engine/shared/rust_version.cpp --output src/rust-bridge/engine/shared/rust_version.h
cxxbridge src/engine/console.rs --output src/rust-bridge/cpp/console.cpp --output src/rust-bridge/cpp/console.h
```
By moving the calling convention `APIENTRY`, which expands to `__stdcall`, inside the parenthesis.
```
Severity Code Description Project File Line Suppression State
Error C2143 syntax error: missing ')' before '(' engine-shared src\base\system.cpp 4242
Warning C4229 anachronism used: modifiers on data are ignored engine-shared src\base\system.cpp 4242
Error C2059 syntax error: ')' engine-shared src\base\system.cpp 4242
Error C2059 syntax error: ')' engine-shared src\base\system.cpp 4242
Error C3536 'exception_log_file_path_func': cannot be used before it is initialized engine-shared src\base\system.cpp 4246
Error C2446 '==': no conversion from 'nullptr' to 'int' engine-shared src\base\system.cpp 4246
Error C2064 term does not evaluate to a function taking 1 arguments engine-shared src\base\system.cpp 4249
```
5942: Support unicode with ExcHndl, use upstream module offsets, handle errors r=def- a=Robyt3
Update Dr. Mingw (ExcHndl) to 0.9.8.
Use the new `ExcHndlSetLogFileNameW` function to set the exception log file name using wide characters, to support paths containing unicode.
It's not necessary to call `ExcHndlInit` explicitly after loading `exchndl.dll`, as the `DllMain` will already initialize the exception handler when the DLL is loaded. Module offsets are supported by upstream ExcHndl now, so we don't need to provide our own version that supplies the module offset to `ExcHndlInit` anymore. Upstream ExcHndl will also resolve the source code lines for addresses automatically, when the executable is build with debug information.
Handle the cases that the exception handling module cannot be loaded and that the `ExcHndlSetLogFileNameW` function cannot be found in the module.
Update `scripts/parse_drmingw.sh`:
- Parse both old and new module offsets.
- Use tabs instead of spaces consistently.
- Reset the ANSI color after printing colored messages.
Closes#5877.
Needs https://github.com/ddnet/ddnet-libs/pull/34.
Example crash logs:
- [crash_debug_old.RTP.txt](https://github.com/ddnet/ddnet/files/9768008/crash_debug_old.RTP.txt)
- [crash_debug_new.RTP.txt](https://github.com/ddnet/ddnet/files/9768011/crash_debug_new.RTP.txt)
- [crash_release_old.RTP.txt](https://github.com/ddnet/ddnet/files/9768010/crash_release_old.RTP.txt)
- [crash_release_new.RTP.txt](https://github.com/ddnet/ddnet/files/9768009/crash_release_new.RTP.txt)
## 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 (especially base/) or added coverage to integration test
- [ ] 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 <robytemueller@gmail.com>
Update Dr. Mingw (ExcHndl) to 0.9.8.
Use the new `ExcHndlSetLogFileNameW` function to set the exception log file name using wide characters, to support paths containing unicode.
It's not necessary to call `ExcHndlInit` explicitly after loading `exchndl.dll`, as the `DllMain` will already initialize the exception handler when the DLL is loaded.
Module offsets are supported by upstream ExcHndl now, so we don't need to provide our own version that supplies the module offset to `ExcHndlInit` anymore.
Upstream ExcHndl will also resolve the source code lines for addresses automatically, when the executable is build with debug information.
Handle the cases that the exception handling module cannot be loaded and that the `ExcHndlSetLogFileNameW` function cannot be found in the module.
Update `scripts/parse_drmingw.sh`:
- Parse both old and new module offsets.
- Use tabs instead of spaces consistently.
- Reset the ANSI color after printing colored messages.
- Retrieve version information of `kernel32.dll` without providing a fixed path to the Windows directory, to handle the case that Windows is not installed in the default location `C:\Windows`. This works because `GetFileVersionInfoSizeW` and `GetFileVersionInfoW` use the search sequence defined by `LoadLibrary`, so they will find the DLL without the path explicitly being specified. The version is retrieved from `kernel32.dll` instead of `user32.dll`, as the former is more commonly used for this purpose.
- Use the unicode (wide character) functions consistently instead of the ANSI functions.
- Use the correct format specifier `%hu` for `unsigned short` (i.e. `WORD` in the Windows API).
References:
- https://learn.microsoft.com/en-us/windows/win32/api/winver/nf-winver-getfileversioninfosizew
- https://learn.microsoft.com/en-us/windows/win32/api/winver/nf-winver-getfileversioninfow
- https://stackoverflow.com/a/44672263/1708371
Update the `str_skip_to_whitespace(_const)` functions according to their documentation, which already stated that `\r` was also considered as whitespace.
This changes `str_comp_filenames` so it sorts filenames case insensitive while also comparing digits characters as numbers.
This makes filename sorting consistent with the behavior in Windows Explorer.
- Change argument and return value to `const char *`, as the string is not modified by this function.
- Use our own `str_isspace` instead of standard library `isspace`.
- Always trim leading whitespace to correctly handle inputs with leading whitespace.
For more consistent usage of the Windows API. `_wgetcwd` seems to be a wrapper around `GetCurrentDirectoryW` anyway, at least in the [Wine API](https://source.winehq.org/ident?_i=_wgetcwd).
Pass the correct size of the wide char buffer instead of passing the size of the output parameter buffer.
Remove a nullptr check. The argument should never be null and we don't check for null arguments anywhere else.
For more consistent usage of the Windows API. `_wchdir` seems to be a wrapper around `SetCurrentDirectoryW` anyway, at least in the [Wine API](https://source.winehq.org/ident?_i=_wchdir).
For more consistent usage of the Windows API. `_wmkdir` seems to be a wrapper around `CreateDirectoryW` anyway, at least in the [Wine API](https://source.winehq.org/ident?_i=_wmkdir).
Previously, the socket addresses were truncated as the `msg_namelen`
field is both input **and** output: After receiving an IPv4 packet, the
socket address field would be too short for an IPv6 address.
Remove square brackets to reduce the amount of space used.
Add log level indicator. The position between timestamp and system was
chosen because it is at a fixed position (unlike after the system) but
the log still remains naively sortable (which wouldn't happen if we were
to place it in front of the timestamp.
Before:
```
[2022-04-29 15:25:37][engine]: running on unix-linux-amd64
[2022-04-29 15:25:37][engine]: arch is little endian
[2022-04-29 15:25:37][storage]: added path '$USERDIR' ('/path/to/home/.teeworlds')
[2022-04-29 15:25:37][storage]: added path '$DATADIR' ('data')
[2022-04-29 15:25:37][storage]: added path '$CURRENTDIR' ('/path/to/ddnet')
[2022-04-29 15:25:37][host_lookup]: host='localhost' port=0 1
[2022-04-29 15:25:37][host_lookup]: host='localhost' port=0 2
[2022-04-29 15:25:37][console]: executing 'autoexec_server.cfg'
```
After:
```
2022-04-29 15:25:37 I engine: running on unix-linux-amd64
2022-04-29 15:25:37 I engine: arch is little endian
2022-04-29 15:25:37 I storage: added path '$USERDIR' ('/path/to/home/.teeworlds')
2022-04-29 15:25:37 I storage: added path '$DATADIR' ('data')
2022-04-29 15:25:37 I storage: added path '$CURRENTDIR' ('/path/to/ddnet')
2022-04-29 15:25:37 I host_lookup: host='localhost' port=0 1
2022-04-29 15:25:37 I host_lookup: host='localhost' port=0 2
2022-04-29 15:25:37 I console: executing 'autoexec_server.cfg'
```
5760: Fix variable shadow warnings with MinGW r=def- a=Robyt3
```
ddnet/src/base/system.cpp: In function 'int byteval(const char*, unsigned char*)':
ddnet/src/base/system.cpp:3044:32: warning: declaration of 'byte' shadows a global declaration [-Wshadow]
3044 | static int byteval(const char *byte, unsigned char *dst)
| ~~~~~~~~~~~~^~~~
In file included from msys64/mingw64/include/objbase.h:8,
from ddnet/src/base/system.cpp:68:
msys64/mingw64/include/rpcndr.h:63:25: note: shadowed declaration is here
63 | typedef unsigned char byte;
| ^~~~
ddnet/src/base/system.cpp: In function 'unsigned char str_byte_next(const char**)':
ddnet/src/base/system.cpp:3557:23: warning: declaration of 'byte' shadows a global declaration [-Wshadow]
3557 | unsigned char byte = **ptr;
| ^~~~
msys64/mingw64/include/rpcndr.h:63:25: note: shadowed declaration is here
63 | typedef unsigned char byte;
| ^~~~
ddnet/src/base/system.cpp: In function 'int str_utf8_decode(const char**)':
ddnet/src/base/system.cpp:3577:31: warning: declaration of 'byte' shadows a global declaration [-Wshadow]
3577 | unsigned char byte = str_byte_next(ptr);
| ^~~~
msys64/mingw64/include/rpcndr.h:63:25: note: shadowed declaration is here
63 | typedef unsigned char byte;
| ^~~~
```
<!-- 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 (especially base/) or added coverage to integration test
- [ ] 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 <robytemueller@gmail.com>
5754: Auto refresh skins when changing related settings r=def- a=Jupeyy
motivation: downloaded skins aren't resettet, e.g. if they failed before. config change -> resets everything
## 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 (especially base/) or added coverage to integration test
- [ ] 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)
5757: Make file link absolute, add `fs_is_relative_path` r=def- a=Robyt3
This fixes links not opening for relative paths, as links like `file://temp/skins` cannot be resolved by the shell.
Closes#5746.
## Checklist
- [X] Tested the change ingame (only on Windows)
- [ ] Provided screenshots if it is a visual change
- [ ] Tested in combination with possibly related configuration options
- [ ] Written a unit test (especially base/) or added coverage to integration test
- [ ] 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>
Co-authored-by: Robert Müller <robytemueller@gmail.com>
On Windows, the check was incorrectly using logical or instead of bitwise or. Checking for the flag `FILE_ATTRIBUTE_REPARSE_POINT` is not necessary, as the `FILE_ATTRIBUTE_DIRECTORY` will correctly be set when the target of a symbolic link is a directory. This otherwise causes symbolic links to files to be incorrectly handled as directories.
On Linux, the minor performance optimization of using `entry->d_type` is reverted and `fs_is_dir` is used instead. This internally uses `stat`, which correctly returns the attributes for the symbolic link targets.
5205: Allow multiple addresses per server in the serverbrowser r=def- a=heinrich5991
Support is incomplete for `leak_ip_address_to_all_servers` (will only
ping the first address of each server) and for the `leak_ip` setting
(which will also only ping the first address of each server).
Fixes#5158.
## 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>
Support is incomplete for `leak_ip_address_to_all_servers` (will only
ping the first address of each server) and for the `leak_ip` setting
(which will also only ping the first address of each server).
5504: Remove useless typedef in c++ r=Jupeyy a=Chairn
## 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: Chairn <chairn.nq@hotmail.fr>
- `io_read_all` reads all bytes from a file handle into a new buffer. It should only need one allocation per file in cases where the actual file size matches the expected file size. Otherwise it falls back to doubling the buffer size if the actual file size is larger than expected to avoid TOCTOU problems.
- `io_read_all_str` reads all bytes from a file handle into a new buffer and also ensures that the buffer is null-terminated and contains no other null-characters.
- `mem_has_null` is a utility used by `io_read_all_str` to ensure that no null-characters exist in the bytes read from the file.
5406: Remove namespace tw r=Jupeyy a=heinrich5991
It didn't have a clear role, it just acted as a distinguisher between
two functions with the same name.
Rename `tw::time_get` to `time_get_nanoseconds` and delete the old
`time_get_nanoseconds`. Move `CCmdlineFix` and the typed
`net_socket_read_wait` function to the global namespace.
## 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: heinrich5991 <heinrich5991@gmail.com>
It didn't have a clear role, it just acted as a distinguisher between
two functions with the same name.
Rename `tw::time_get` to `time_get_nanoseconds` and delete the old
`time_get_nanoseconds`. Move `CCmdlineFix` and the typed
`net_socket_read_wait` function to the global namespace.
5351: Refactoring: Use MultiByteToWideChar instead of str_utf16le_encode r=heinrich5991 a=Robyt3
## 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 <robytemueller@gmail.com>
5332: Pass output parameter to WriteFile that is required on Windows 7 r=heinrich5991 a=Robyt3
Maybe closes#5322. I can't reproduce the crash. Can you try if this fixes it, `@Chairn?`
Looks like Windows 7 needs this parameter.
Alternative to #5331.
## 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: Robert Müller <robytemueller@gmail.com>
5327: Add CLockScope r=heinrich5991 a=def-
Remove unused lock and scope in threading.h
Not sure if this is preferred
We could also try switching to std::mutex and lock_guard
<!-- 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: Dennis Felsing <dennis@felsin9.de>
5260: Pr thread safety negative r=heinrich5991 a=def-
WorkerThread is hard because `REQUIRES(!((CJobPool *)pUser)->m_Lock)` would require alias analysis or the function using that everywhere.
https://clang.llvm.org/docs/ThreadSafetyAnalysis.html#negative
## 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>
5291: Windows: Use current console background color, restore original color on exit, refactoring r=heinrich5991 a=Robyt3
- Remember initial console foreground and background color.
- Use the original background color with the message specific foreground color, so it looks better when the console background color isn't black.
- Restore the initial console foreground and background color on exit instead of leaving the color be whatever was last used.
- Add locks to ensure `Log`-calls from different threads don't overlap their output/color.
- Refactoring: Replace numbers with color constants defined in `wincon.h` (4312d20923/include/wincon.h (L51-L67)).
- Refactoring: Slightly reorder the `if`s and remove redundant checks which are ensured by the `else-if` structure.
## 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 <robytemueller@gmail.com>
5179: Add loglevel config (fixes#5178, fixes#5174) 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
- [ ] 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>
5272: Handle failed socket creation (fixes#5267) r=heinrich5991 a=def-
by returning nullpointer, as the calling functions expect
Broken in 471bb441a1
<!-- 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>
5264: Prevent users from setting port 1-1023 r=Jupeyy a=def-
Following user report that they set cl_port 1 and it didn't work
<!-- 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)
5266: Use std::make_unique (fixes#5166) r=Jupeyy a=def-
No idea why clang-tidy's modernize-... didn't work
<!-- 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>
5252: Use std::vector instead of array, remove base/tl/array.h, algorithm.h, allocator.h, range.h r=def- a=Robyt3
Replaces all usages of `array` with `std::vector`.
I adjusted variable names of variables I changed to use the `v` prefix. Not so in the editor however, as there are already many changes due to clang-tidy enforcing the use of for-each loops.
This allows us to remove all remaining `base/tl` headers except `threading.h`.
Clang-tidy now finds `clang-analyzer-cplusplus.NewDelete` (Use of memory after it is freed), which is also fixed here, though it appears to be a false-positive.
This last remaining usages of `goto` are also removed.
## 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 <robytemueller@gmail.com>
ChillerDragon
