As a low level feature, the CVAR management should not access game structures like actors, just to retrieve a player index. The index should be calculated by the calling code instead and passed into the function.
# Conflicts:
# src/win32/i_specialpaths.cpp
User CVARs can be stored in member variables for quicker access
Unconditional call to G_CheckDemoStatus() destructs and recreates such CVARs
Dangling pointers could be accessed during the last game tick, e.g. from event handlers
https://forum.zdoom.org/viewtopic.php?t=65150
Exit_Normal and Exit_Secret specials imply addition conditions to make a decision about proceeding to the next map
A few scriptified actors require a bypass of these conditions in order to work properly
src/posix/cocoa/i_main.mm:225:12: error: Objective-C declarations may only appear in global scope
src/posix/cocoa/i_main.mm:247:17: error: Objective-C declarations may only appear in global scope
src/posix/cocoa/i_main.mm:545:2: error: expected '}'
This shouldn't be in the hardware independent interface because the semantics on OpenGL and Vulkan are too different, so a common implementation is not possible.
Most bind calls were in the GL interface anyway, so these no longer pass through hardware independent code.
This also moves the bind calls in the shadowmap code into the GL interface - these never did anything useful in Vulkan and aren't needed there.
Last but not least, this moves the legacy buffer binding handling into FGLRenderState and performs the initial binding for the light buffer in a more suitable place so that this doesn't have to pollute the render state.
* Right/middle clicking no longer cancels out the mouse motion events
* The events for the side mouse buttons (EV_GUI_BackButtonDown/Up, EV_GUI_FwdButtonDown/Up) were never fired
* The EV_GUI_WheelRight/Left events were never fired
* The key bindings for mouse buttons above 3 (from 4-8) were broken - pressing MOUSE4 would bind to MOUSE6, for example
With JIT enabled, an implicit test for null self pointer is added to generated code
This has no effect without JIT as VM verifies a pointer before calling a native method
https://forum.zdoom.org/viewtopic.php?t=64961
IWAD wasn't detected properly, and wrong texture can be picked as a hires replacement (e.g., SLADRIP1 from Ultimate Doom instead of TNT)
https://forum.zdoom.org/viewtopic.php?t=64915
Old graphics, like mine (Intel Sandybridge Mobile, GL 3.0 Mesa) do not support this, therefore most of the time RFL_SHADER_STORAGE_BUFFER wasn't unset (I found no consequence of this on my machine, but better safe than sorry).
Found out by Valgrind.
Rendering bits got restructured and the SSE_MATTERS paths were not updated
to reflect that. A 32-bit build with clang subsequently complaints as the
files in question are not compiled with -mmmx.
During serialization, when map shutdown because of unknown class stored inside saved game may occur, level member is not assigned yet to already spawned thinkers
The message is always printed for any use of deprecated global variable
If such variable is accessed inside a deprecated function from a core script lump, the message is printed only when verbosity level is set to highest
https://forum.zdoom.org/viewtopic.php?t=64830
Script position is now initialized at the very beginning of decorate parsing process
Script position no longer contains uninitialized file name
https://forum.zdoom.org/viewtopic.php?t=64836
There was no read barrier inserted for AActor object while it's a subject of garbage collection
Pointer stored in TObjPtr<> was reinterpreted as void* because of vararg function
https://forum.zdoom.org/viewtopic.php?t=64771
The Romanian letters with comma below, the second Greek Sigma variant and a few special Cyrillic letters have replacements that should not result in rejection.
This turned out the smallest one on which a factor of 2 still works without cutting off texts but at least allows to use a larger menu display on this important laptop size.
* Fixes for Door_AnimatedClose
- Fixing that Door_AnimatedClose would be activated on an already closed door (playing the sound and the 1st frame of the animation)
- There was no check if the Door is already Animated when a tag was used with Door_AnimatedClose
* Removed doubled "door->StartClosing();".
Aside from using incorrect values this could crash the renderer if used in the very first rendering frame because the software renderer always operates on a copy.
Now it's no longer possible to manipulate TID hash from arbitrary location
For example, this prevents linking of destroyed object into the hash
TID member is still public but writing to it is limited to a few very specific cases like serialization and player traveling between levels
https://forum.zdoom.org/viewtopic.php?t=64476
The lowercase Sigma letter in Greek has two different forms (σ and ς), which changes depending on its placement in a word, but in uppercase and smallcaps contexts, it only has one look regardless of word positioning. If the character ς is missing, it should fall back to σ.
Vulkan hardware buffer for software canvas may have some padding
Software renderers should be aware of buffer's pitch in order to copy pixels properly
https://forum.zdoom.org/viewtopic.php?t=64562
src/rendering/hwrenderer/postprocessing/hw_postprocess.cpp:1007:13: warning: delete called on non-final 'PPTexture' that has virtual functions but non-virtual destructor [-Wdelete-non-virtual-dtor]
GZDoom aborts with 'Failed to submit command buffer' error on saving a game when open source Intel Vulkan driver (part of Mesa 3D package) is used on Linux
This driver generates VK_DEVICE_LOST error when vkWaitForFences() is called with zero fence count. It must be greater than zero according to Vulkan spec
Provided uniform way to handle lifetime of some of Vulkan resources
This helps to avoid issues like descriptor set that outlives its pool
https://forum.zdoom.org/viewtopic.php?t=64341
The counter variable was not only used incorrectly, it was completely redundant.
This still has a byte limit due to how the network code works so for non-Latin languages may result in shorter strings.
The episode titles required a workaround because the CQ3 episodes do not contain names in text form: If this is the case, the patch name will be used as a string table identifier to get a matching text for localization.
* - Added support for monospacing alignment modes to HUDFont / BaseStatusBar.DrawString
* - added underlying type declaration for EMonospacing
* - replaced "#include v_video.h" with a declaration of EMonospacing
If this isn't done there can be a mix of content from different sources, depending on the language. It's better to have correct English text than unfitting localized versions.
This deprecated CharAt and CharCodeAt for being unsuitable for text processing and in the case of CharCodeAt also for being buggy.
A new replacement, ByteAt has been added that reads a string byte by byte, as well as CodePointCount, which counts the amount of Unicode code points in a string and GetNextCodePoint which reads the string code point by code point.
Note that while this woll work as intended with the currently supported languages as a means to read single characters, there is no guarantee that this will remain so if Unicode support gets extended to things which break the "one code point == one character" assumption.
This is for IWADs that contain content which clashes with the engine and must be removed. The primary reason are the unimplemented characters in Raven's BigFont.
22 lines of text need to fit and to avoid packing the text too tightly onto the screen and have a little headroom, a virtual size of 400x250 is needed.
This was meant for using the VGA font in the alternative HUD but this never went beyond the Kill/Item/Secret display which isn't useful for localization.
Some reorganization to avoid code duplication plus making the log screen capable of using the generic font. This also means that the popup for the log in Strife's status bar will be disabled when in generic mode - this popup with its special font would be a bit problematic.
It's "doom.id.doom1/2" instead of "doom.doom1/2" now.
The config file's content will be renamed and for lump filtering a fallback has been added - note that you cannot combine both naming schemes! The old one has to be considered deprecated now.
This also removes the duplicated content necessitated by the old naming scheme.
src/posix/sdl/sdlglvideo.cpp:358:10: error: ‘MIN_WIDTH’ is not a member of ‘Priv’
src/posix/sdl/sdlglvideo.cpp:358:33: error: ‘MIN_HEIGHT’ is not a member of ‘Priv’
src/posix/sdl/sdlglvideo.cpp:360:7: error: ‘MIN_WIDTH’ is not a member of ‘Priv’
src/posix/sdl/sdlglvideo.cpp:361:7: error: ‘MIN_HEIGHT’ is not a member of ‘Priv’
If this is done within the intermission code, both intermission and menu will write to the same global variables and destroy their data, this became very apparent when it altered the screen scale for the conversation.
GCC 8 complains that it can't find relevant functions:
/wrkdirs/usr/ports/games/gzdoom/work/gzdoom-g3.7.2/src/m_png.cpp:669:42: error: call of overloaded 'BigLong(uint32_t)' is ambiguous
chunklen = BigLong((unsigned int)x[1]);
^
In file included from /wrkdirs/usr/ports/games/gzdoom/work/gzdoom-g3.7.2/src/m_png.cpp:44:
/wrkdirs/usr/ports/games/gzdoom/work/gzdoom-g3.7.2/src/m_swap.h:212:15: note: candidate: 'long unsigned int BigLong(long unsigned int)' <deleted>
unsigned long BigLong(unsigned long) = delete;
^~~~~~~
/wrkdirs/usr/ports/games/gzdoom/work/gzdoom-g3.7.2/src/m_swap.h:213:6: note: candidate: 'long int BigLong(long int)' <deleted>
long BigLong(long) = delete;
This is on FreeBSD/powerpc64.
The current menu system simply does not work that well with 320x200, rendering the game hard to use at that tiny screen size. This is a clear case where the work required to keep it operational stands in no relation to the benefit.
These can cause highly dangerous recursions and execute play code deep inside the renderer and since the hardware renderer does not have these, there is very little point for the software renderer to retain them.
src/rendering/vulkan/renderer/vk_renderpass.cpp:44:22: warning: comparison of integers of different signs: 'std::__1::vector<FVertexBufferAttribute, std::__1::allocator<FVertexBufferAttribute> >::size_type' (aka 'unsigned long') and 'int' [-Wsign-compare]
src/rendering/vulkan/system/vk_framebuffer.cpp:860:55: warning: format specifies type 'int' but the argument has type 'VkDeviceSize' (aka 'unsigned long long') [-Wformat]
src/rendering/vulkan/system/vk_objects.h:471:23: warning: suggest braces around initialization of subobject [-Wmissing-braces]
It should be said in no uncertain terms that OnRegister operates on an uninitialized level so it should only be used for setting up the registering process of the event handler itself and nothing else - not even the event handler's data!!!
Try to mimic OpenGL GPU selection behavior with MoltenVK when automated graphics switching is disabled
The CVAR isn't exposed in menu and its target audience is advanced users only
There are two options here - one only disables the vertical thrust and the other goes back fully to the original non-z-aware code.
Both options are settable through MAPINFO.
For the compatibility presets, the normal ones only disable the vertical thrust, the strict ones force use of the old code entirely.
This allows models to render, but a proper fix should handle mapping of static buffers for real. This workaround just allocates a non-static persistent buffer in CPU memory which is not the most efficient solution here.
In this case a map is total overkill and its poor memory locality will always incur some performance hit. A vector here can only collect a very small amount of entries so the time to search is nearly irrelevant.
It will now check all layers of a material.
Additionally it will also delete all descriptor sets of Vulkan hardware textures before precaching to make sure that nothing here can accidentally still reference a deleted texture.
Since Google can export this as well it makes a lot more sense here because it is a textual format better suited for version control and does not carry along all the formatting baggage that isn't needed for a string table.
This has to be set in the console, the default is still the regular small font. Mainly added because some mods have really hard to read fonts where it is not easy to decipher the numbers.
In fullscreen mode there is a white flash (of the default background color) on startup
To overcome this an explicit glClear() with flush buffers was used
In order to make video mode switch more generic, any backend specific functionality should be avoided here
Now, all menus will use the same scale, i.e. it only depends on the screen width and a base size of 640. This nearly universally yields better results than trying to make a 320x200 screen fit.
The only exceptions to this are the intermission screens and the level summary. These, unlike the menu need to try to make a 320x200 screen fit, but without all the hackery that was present to adjust the menu display.
Note that since this affects globally visible script variables, both the intermission and summary drawers will not use their own set but instead temporarily override the global setting as long as they run their own code.
Changing the use of variables here might cause much worse problems with menu code so it wasn't attempted
The default was changed to replace graphics if only known content is affected.
This also required blocking the feature for Hacx, Harmony and Action Doom 2 which have no localization yet and would fall back on Doom's texts if not blocked.
src/p_acs.cpp:8719:100: error: cannot pass non-trivial object of type 'FString' to variadic function; expected type from format string was 'char *' [-Wnon-pod-varargs]
* consolidated C_MidPrint and C_MidPrintBold.
* removed some unused code from the console buffer.
* handle console output of centered messages to ensure they get written to the log file and to stdout.
* replaced the non-standard bar strings with simple '-'s. These were making things needlessly complicated when redirecting console output and the new font does not have the characters anyway.
* removed some old code from a time when during console drawing new network events could come and have more text printed. This can not happen anymore with how 2D elements are being handled now so all this code was redundant.
- clamp scissors fully to avoid NVidia's awful drivers locking up the entire system if they end up out of bounds
- perform buffer clears as part of the render pass. this puts some restrictions on how FRenderState.Clear can be used
- add an offset uniform to the present shaders so the vulkan target can flip the image during presentation
The most important one is the autosave tagging. This was done because the old printout was missing the year and printed the month as a 3 character English string, sabotaging any attempt to sort the autosaves by anything meaningful.
Parts of this menu suffered badly from lack of screen space to convey the intended information due to the oversized fonts. With the new font this is a lot less problematic (unless using 320x200, of course)
Since this needs to do cursor positioning calculations it's the one spot in the entire engine where UTF-8 would simply be to messy, especially when having to deal with double wide characters.
* re-added screen blends for images from the hardware renderer.
* moved all postprocessing of the image out of the renderers.
* cleaned out a large piece of cruft for handling the palette in the frame buffer class. This was all a remnant of the old paletted backend that no longer exists. Nowadays the screen blend is just a postprocessing effect drawn over the 3D screen, there is no need to maintain any of it as global state anymore.
* since the engine doesn't produce paletted screenshots anymore there is no need to have handling for it in the generation code. This depended on otherwise obsolete information so it got removed along with that information.
Dynamic libraries placed in the directory with the main executable now can be loaded by dlopen() function
This is required in order to enable Vulkan support without additional steps like SDK installed in the system
Unlike uObjectColor2, this is more global state. uObjectColor2 is part of gradient calculation which may later be offloaded to a secondary buffer which already resulted in only conditionally setting it, resulting in broken special colormap application for the software renderer.
src/rendering/vulkan/renderer/vk_postprocess.cpp:219:54: error: cannot pass non-trivial object of type 'const FString' to variadic function; expected type from format string was 'char *' [-Wnon-pod-varargs]
This is needed to prevent compilation warnings spam with GCC and Clang
While the warning itself is useful, in this case it's rather pointless and annoying
This didn't behave like an assignment operator so it shouldn't be one, especially since the two places where it got called need different functionality.
8 bytes is the minimum header size for DMX, so for one byte of sample data it has to be 9 bytes.
This was causing access to invalid memory when trying to read the header of something too short.
For other file formats this is of no concern because none has a header this short.
In the old system it just loaded the font a second time, but with the folder based setup there is no file named CONFONT anymore so it needs to be dealt with explicitly.
This was already far too generous and caused space problems, but with localization these became a lot worse, so now it will try to allocate at least 640 virtual pixels for the menu width and only go below that for small resolution ranges where the smaller value would result in too small text.
This function was written for already validated UTF-8 but not for text that can be mixed with ISO-8859-1.
To handle that properly it needs to do a bit more validation to avoid mangling its output and instead reject invalid input.
Now a localization mod can disable the graphics patches containing text entirely so that it can properly localize the text based menu variant.
If this flag gets set in MAPINFO, it will override all user settings.
These were pixel format conversion routines used in the D3D backend. Nothing in here is needed anymore - the FBitmap class offers much of the functionality covered here in a far more concise and approachable manner.
The usedcolors array which counts the number of pixels in a given color in a font used bytes as storage, so any color that just happened to have a count that is a multiple of 256 the color was considered not present.
* added a CVAR that sets how localizable graphics need to be dealt with.
* pass the substitution string to OkForLocalization so that proper checks can be performed.
* increased item spacing on Doom's list menus to 18 from 16 pixels, because otherwise the diacritic letters would not fit. 20 would have been more ideal but 18 was the limit without compromising its visual style
* added a second text-only main menu because here the spacing cannot be changed. Doing so would render any single-patch main menu non-functional. So here the rules are that if substitution takes place, it will swap out the entire menu class.
* fixed some issues with the summary screen's "entering" and "finished" graphics.
Passing something non-constant at compile time here is extremely dangerous, especially when users can replace those strings if they like.
It now uses FString::Substitute in all cases where something needs to be inserted into a template string.
This is one of those things where the work needed to make it robust stands in no relation to the gain.
This simply isn't worth the hassle of going through the entire code and fixing every single use of the 2D texture drawing functions.
Unfortunately this means that the graphics items for the menu cannot be replaced this way because their size will most likely differ, but considering that the only candidates for this are the contents of Doom's main menu, the episode menu, the skill menu and the single player summary screen, it's simply not worth it.
In all these cases the IWAD contents can just as easily be replaced with text and user mods which want to offer localized menus will have to work within the confines of the system, e.g. making sure that all menu items are designed to have proper size for substitution to work or by requesting text based menus, which will be added as a modding feature later.
This variable is needed long after the function which sets it up will be exited. So this either needs to be dynamically allocated or static, and in this case using a static variable is simpler. However, unlike before, it is only being accessed in the one function that needs to initialize it and pass to the summary screen and nowhere else.
- Implement page string names for dialog lumps
- Create special new GZDoom name space for ZSDF
- add usdf_gzdoom spec document
- fixed: restored original behavior with negative conversation id's for the original strife dialog lumps
- reposition the binary strife fix in a more appropriate location
- add compatibility fix for negative numbers in responses in USDF/ZSDF (don't know if it's actually necessary)
For the Doom IWADs the provided font looks almost identical to the characters used on the title patches. So, for any level name that got replaced in some language, it will now check if the retrieved name comes from the default table, and if not, ignore the title patch and print the name with the specified font.
This also required removing the 'en' label from the default table, because with this present, the text would always be picked from 'en' instead of 'default'. Since 'en' and 'default' had the same contents, in any English locale the 'default' table was never hit, so this won't make any difference for the texts being chosen.
Last but not least, wminfo has been made a local variable in G_DoCompleted. There were two places where this was accessed from outside the summary screen or its setup code, and both were incorrect.
* prefer accent-less lower case over uppercase letters if an accented lower case letter cannot be found.
* added accent-less mappings for Latin Extended 1 (0x100-0x17f) and some easy to handle characters between 0x200 and 0x220. This should allow to display all Eastern European text without empty gaps for missing letters.
* the window class name was still ASCII, thanks to some totally pointless and ultimately dangerous type cast to LPCTSTR which rendered all type checks ineffective.
* use wWinMain instead of WinMain so that a Unicode argv gets created. For whatever reason, the ANSI startup leaves this variable empty.
* added a 'disablecrashlog' CCMD for Windows. It is a lot more useful with a debugger present to get the standard crash notification from the system which allows opening a debugger than the crash log and no option to open a debugger.
This still contained pieces where a multibyte string was passed through SendMessage and WM_SETTEXT. All these have been replaced with SetWindowTextW.
This commit also removes the never used crash log upload code and all associated assets because it is extremely unlikely that such a feature will ever be implemented.
This was broken by several small unicode-incompatible code fragments.
This commit also removes the input limit for the player name and the savegame description. With multibyte encoding, limiting them to a fixed length did not work right.
Currently these will just overflow the fields if the text becomes too long, this needs some additional work.
With localization for non-Latin languages on the support list the multibyte API doesn't cut it anymore. It neither can handle system text output outside the local code page nor can an ANSI window receive text input outside its own code page.
Similar problems exist for file names. With the multibyte API it is impossible to handle any file containing characters outside the active local code page.
So as of now, everything that may pass along some Unicode text will use the Unicode API with some text conversion functions. The only places where calls to the multibyte API were left are those where known string literals are passed or where the information is not used for anything but comparing it to other return values from the same API.
The Linux backend looked like it didn't handle anything non-ASCII at all, but this all needs to be tested.
Windows will be a bit more work because it requires using the Unicode API for creating the main window.
- Fix zdoom.rc to show the actual git commit tag and id for the Product Version
- Made zdoom.rc "codepage 1252" compliant as dictated by the #pragma (if this needs changed the pragma should be updated, this was messing up the version strings in the final compile)
To make things easier, DBIGFONT, SBIGFONT and HBIGFONT will now be renamed in the lump directory to make things a bit easier to handle.
Another change is to make font folders atomic units to prevent cross-pollution between incompatible fonts. The only exception to this are the def* folders because they need to piece together their fonts from both zd_extra.pk3 and the IWADs.
It was only used to avoid traversing the list if all sequences were paused which is an exceptional situation.
On the other hand, the way it counted was not correct so rather than fixing it it seemed more appropriate to remove it entirely.
This is dpne as a two-stage approach. TXT_LOGTEXTxxx will always take precedence over the log lumps, and TXT_ILOGxxx will only replace the original IWAD content.
This is so that PWADs replacing these lumps don't get overridden by the default texts.
The added table may be overkill but this way the font engine is prepared for things to come.
Currently the text placement in the menu seems a bit broken, that's a task for later.
A multi-lump font can be created by putting all characters into a subdirectory of fonts/ with the intended name. Each character needs to be named by its character index as hex number.
So far this is only active for the predefined small fonts
It now reads everything into a two-dimensional TMap and creates a list of mappings that apply to the current setting.
The constant need for reloading was the main blocker in redesigning how Dehacked strings get inserted. Currently they override everything, but IWAD-based Dehacked text shouldn't block PWAD overrides from PWADs' LANGUAGE lumps and instead be treated as coming from an [en default] block.
This also renames the main block from [enu default] to [en default], because it should be treated as the English default for all English locales and not just make it fall through to the base default as it did before.
src/p_acs.cpp:3250:75: error: cannot pass object of non-trivial type 'FString' through variadic constructor; call will abort at runtime [-Wnon-pod-varargs]
src/p_conversation.cpp:354:56: error: cannot pass object of non-trivial type 'FString' through variadic constructor; call will abort at runtime [-Wnon-pod-varargs]
src/p_conversation.cpp:438:51: error: cannot pass object of non-trivial type 'FString' through variadic constructor; call will abort at runtime [-Wnon-pod-varargs]
src/p_conversation.cpp:548:58: error: cannot pass object of non-trivial type 'FString' through variadic constructor; call will abort at runtime [-Wnon-pod-varargs]
src/p_conversation.cpp:572:59: error: cannot pass object of non-trivial type 'FString' through variadic constructor; call will abort at runtime [-Wnon-pod-varargs]
src/p_conversation.cpp:584:58: error: cannot pass object of non-trivial type 'FString' through variadic constructor; call will abort at runtime [-Wnon-pod-varargs]
Thanks to the lazy counter it used in its stat display I never noticed that the serializer was incomplete and that UnlinkFromMap did not call its super method.
After changing the counter to be actively counting on each call, all the other issues became immediately apparent.
The interpolator had been changed long ago to use proper GC tracking, so interpolations only can get collected if they had been fully orphaned.
This comment was the main reason why the design flaw in this code never got fixed until recently.
It seems there can be rare conditions where an interpolation is 'lost' and later garbage collected. If that happens after the owning map is gone, all pointers in the interpolation object will be invalid and Destroy would crash while trying to unlink it. So anything that explicitly deletes an interpolation now has to manually unlink it from the map first so that OnDestroy can be kept clean of map references.
Now all this content can be localized. However, since this is actual game content it was placed in a secondary file in zd_extra.pk3, so that it won't affect the GPL-compatible status of the main one.
src/posix/cocoa/i_input.mm:95:22: error: use of undeclared identifier 'E_CheckUiProcessors'
src/posix/cocoa/i_input.mm:200:21: error: use of undeclared identifier 'E_CheckRequireMouse'
src/posix/sdl/i_input.cpp:183:19: error: use of undeclared identifier 'E_CheckUiProcessors'
src/events.cpp:540:1: error: pasting formed '::RenderFrame', an invalid preprocessing token
src/events.cpp:541:1: error: pasting formed '::WorldLightning', an invalid preprocessing token
src/events.cpp:542:1: error: pasting formed '::WorldTick', an invalid preprocessing token
src/events.cpp:543:1: error: pasting formed '::UiTick', an invalid preprocessing token
src/events.cpp:544:1: error: pasting formed '::PostUiTick', an invalid preprocessing token
The software renderer does not have any safeguards against such a mapping and crashes on it.
This code was a quick hack from ancient times from when ZDoom did not have robust texture management and some recent changes ran afoul of this very special exception.
Having everything lumped together made this a maintenance hassle because it affected how the level has to be stored.
This hasn't been tested yet, so it may not work as intended!
The original place in I_CheckNativeMouse is unsafe because that function can get called from the system message queue which can result in a bad global state of the VM for such a call because it can be recursively invoked from code that may temporarily alter some settings.
currentUILevel is now primaryLevel.
For ZScript, currentVMLevel was added. This is also exported as 'level' and will change as needed.
This also means that no breaking deprecations will be needed in the future, because in order to sandbox a level only 4 variables need to be handled: level, players, playeringame and consoleplayer.
The remaining global variables are not relevant for the level state.
The static 'level' has been mostly removed from the code except some places that still need work.
I think these were the last two still missing it, all remaining uses of the global level variable are in code that doesn't get run through a level tick and are supposed to access the primary level.
Two files were split:
g_level.h contained both the game data definitions and some prototypes belonging to the game logic. These were split up.
decallib.cpp contained both the data and the animation thinkers. The thinkers are now in their own file.
There is one exception in ACS for a net arbitrator check.
Aside from this the bot_observer CVAR was also removed. This was never implemented properly and could stomp upon custom player settings.
The playsim really has no idea what the renderer is supposed to do here and the current system has some serious issues that eventually need addressing. So it is better to just set a flag that an actor needs to have its view interpolation reset if being used as a camera and let the render code deal with it.
This will keep the playsim clean of future changes to this feature.
src/posix/cocoa/i_video.mm:559:2: error: use of undeclared identifier 'atterm'
src/posix/sdl/hardware.cpp:85:28: error: ‘atterm’ was not declared in this scope
src/posix/sdl/i_input.cpp:261:10: error: ‘gamestate’ was not declared in this scope
src/posix/sdl/i_input.cpp:261:23: error: ‘GS_LEVEL’ was not declared in this scope
src/posix/sdl/i_input.cpp:263:48: error: ‘GS_INTERMISSION’ was not declared in this scope
src/posix/sdl/i_input.cpp:263:80: error: ‘GS_FINALE’ was not declared in this scope
src/posix/sdl/i_system.cpp:115:25: error: ‘atterm’ was not declared in this scope
src/posix/sdl/i_system.cpp:372:30: error: invalid use of incomplete type ‘const struct dirent’
src/posix/sdl/i_system.cpp:375:42: error: ‘findstate_t’ has not been declared
src/posix/sdl/i_system.cpp:391:12: error: request for member ‘current’ in ‘* fileinfo’, which is of non-class type ‘int’
src/posix/sdl/i_system.cpp:392:12: error: request for member ‘count’ in ‘* fileinfo’, which is of non-class type ‘int’
src/posix/sdl/i_system.cpp:392:56: error: request for member ‘namelist’ in ‘* fileinfo’, which is of non-class type ‘int’
src/posix/sdl/i_system.cpp:393:22: error: ‘alphasort’ was not declared in this scope
src/posix/sdl/i_system.cpp:393:31: error: ‘scandir’ was not declared in this scope
src/posix/sdl/i_system.cpp:394:16: error: request for member ‘count’ in ‘* fileinfo’, which is of non-class type ‘int’
src/posix/sdl/i_system.cpp:401:31: error: ‘findstate_t’ has not been declared
src/posix/sdl/i_system.cpp:403:2: error: ‘findstate_t’ was not declared in this scope
src/posix/sdl/i_system.cpp:403:15: error: ‘state’ was not declared in this scope
src/posix/sdl/i_system.cpp:403:37: error: expected primary-expression before ‘)’ token
src/posix/sdl/i_system.cpp:404:33: error: request for member ‘count’ in ‘* fileinfo’, which is of non-class type ‘int’
src/posix/sdl/i_system.cpp:406:39: error: request for member ‘count’ in ‘* fileinfo’, which is of non-class type ‘int’
src/posix/sdl/i_system.cpp:413:2: error: ‘findstate_t’ was not declared in this scope
src/posix/sdl/i_system.cpp:413:15: error: ‘state’ was not declared in this scope
src/posix/sdl/i_system.cpp:413:37: error: expected primary-expression before ‘)’ token
src/posix/sdl/i_system.cpp:425:16: error: ‘findstate_t’ was not declared in this scope
src/posix/sdl/i_system.cpp:425:29: error: expected primary-expression before ‘const’
src/posix/sdl/i_system.cpp:426:1: error: expected ‘,’ or ‘;’ before ‘{’ token
src/posix/sdl/i_main.cpp:160:6: error: ‘gamestate’ was not declared in this scope
src/posix/sdl/i_main.cpp:160:19: error: ‘GS_LEVEL’ was not declared in this scope
src/posix/sdl/i_main.cpp:160:44: error: ‘GS_TITLELEVEL’ was not declared in this scope
src/posix/sdl/i_main.cpp:217:18: error: ‘atterm’ was not declared in this scope
src/posix/sdl/i_main.cpp:242:11: error: ‘I_Quit’ was not declared in this scope
src/gameconfigfile.cpp:126:27: error: ‘SHARE_DIR’ was not declared in this scope
src/scripting/decorate/thingdef_parse.cpp:929:2: error: use of undeclared identifier 'strlwr'; did you mean 'strlen'?
src/scripting/decorate/thingdef_parse.cpp:937:3: error: use of undeclared identifier 'strlwr'; did you mean 'strlen'?
src/scripting/decorate/thingdef_states.cpp:564:2: error: use of undeclared identifier 'strlwr'; did you mean 'strlen'?
src/scripting/vm/jit_runtime.cpp:957:2: error: ‘unique_ptr’ is not a member of ‘std’
src/scripting/vm/jit_runtime.cpp:957:38: error: expected primary-expression before ‘>’ token
src/scripting/vm/jit_runtime.cpp:957:40: error: ‘nativeSymbols’ was not declared in this scope
src/utility/cmdlib.cpp:962:2: error: ‘DIR’ was not declared in this scope
src/utility/cmdlib.cpp:962:7: error: ‘directory’ was not declared in this scope
src/utility/cmdlib.cpp:962:34: error: ‘opendir’ was not declared in this scope
src/utility/cmdlib.cpp:967:33: error: ‘readdir’ was not declared in this scope
src/utility/cmdlib.cpp:969:10: error: invalid use of incomplete type ‘struct ScanDirectory(TArray<FFileList>&, const char*)::dirent’
src/utility/cmdlib.cpp:973:34: error: invalid use of incomplete type ‘struct ScanDirectory(TArray<FFileList>&, const char*)::dirent’
src/utility/cmdlib.cpp:985:20: error: ‘closedir’ was not declared in this scope
src\sound\mididevices\music_opl_mididevice.cpp(112): error C3861: 'I_DebugPrint': identifier not found
src\sound\mididevices\music_opl_mididevice.cpp(206): error C3861: 'I_DebugPrint': identifier not found
src\sound\mididevices\music_opl_mididevice.cpp(234): error C3861: 'I_DebugPrint': identifier not found
src\sound\mididevices\music_opl_mididevice.cpp(244): error C3861: 'I_DebugPrint': identifier not found
The menu is a very 'dirty' header, and forcing it to be pulled in with something entirely unrelated is not good - even though only two files include oalsound.h.
- Allows defining of what actor is replacing another for information.
- If multiple arachnotrons, a modder can attribute them as being a replacer of Arachnotron itself, allowing A_BossDeath and GetReplacee to work with it.
This was always used with 'consoleplayer' which really is the only thing making sense here. But this is a part of the global state which should be avoided in play code.
In particular, this makes no real sense in case of secondary maps where it should always return false.
This is supposed to be come the place where all pure play code should be placed, but for that all CVARs and CCMDs and other things that do not directly handle play data should be taken out to make code reviewing easier. These now get collected in two separate files, g_cvars.cpp and g_dumpinfo.cpp respectively.
The sole ZScript property in here has also been moved - to thingdef_properties.cpp.
This made reviewing the code for accessing the global state hard, because the doomdef.h contains mainly constants, this particular item was the only thing in there that represents actual engine state.
The Map loader may not access any global state at all - everything it can touch must be exchangable.
Furthermore, if we want to sandbox each level, there may be no direct access to any kind of global state whatsoever from the play code.
It used the current console player's camera, not the actual camera being used for rendering. Although this is the same most of the time, let's better do it right.
This also removes a few leftover references to the player array elsewhwere in the hardware renderer
All these required access to the sector's Level reference.
The remaining references to the global 'level' variable are all in deprecated functions which is ok.
LevelLocals on the left side of.a function call will now always be remapped to 'Level', which will either remap to the same-named instance variable or the global deprecated one.
In a few degenerate cases where there is a conflicting local variable named 'level' it may error out but that is unavoidable here but this is very unlikely.
- moved parts of the render setup out of the separate render functions.
Things like particle and polyobject linking were duplicated several times for rendering different things in different renderers.
These things only need to be set up once before the renderer is started so it makes a lot more sense to consolidate them into one place outside the actual rendering code.
This had two different flags that were checked totally inconsistently, and one was not even saved.
Moved everything into a few subfunctions so that these checks do not have to be scattered all over the code.
There is no need to do this deep inside the renderer where it required code duplication and made it problematic to execute on multiple levels.
This is now being done before and after the top level call into the renderer in d_main.cpp.
This also serializes the interpolator itself to avoid problems with the Serialize functions adding the interpolations into the list which can only work with a single global instance.
The shader timer may be taken from the primary level for the entire scene, because it will always be the same for all levels in a set.
The camera textures need to be prepared for all levels.
Since currently there is only one level, this will obvciously only run once on that level for the time being.
This is mainly used for CCMDs and CVARs which either print some diagnostics or change some user-settable configuration.
Depending on serialization order is not a good idea here, so now it's no longer stored as a parent in the main level script but explicitly checked for when looking for a variable.
This is what should be audible. To prevent other levels from playing sound, all entry points check whether the sound playing entity belongs to the current UI level.
src/c_dispatch.cpp:143:5: warning: delete called on 'FDelayedCommand' that is abstract but has non-virtual destructor [-Wdelete-non-virtual-dtor]
src/tarray.h:582:5: warning: delete called on 'FDelayedCommand' that is abstract but has non-virtual destructor [-Wdelete-non-virtual-dtor]
src/tarray.h:574:5: warning: delete called on 'FDelayedCommand' that is abstract but has non-virtual destructor [-Wdelete-non-virtual-dtor]
This was done to ensure that this code only runs when the thinker itself is fully set up.
With a constructor there is no control about such things, if some common initialization needs to be done it has to be in the base constructor, but that makes the entire approach chosen here to ensure proper linking into the thinker chains impossible.
ZDoom originally did it that way, which resulted in a very inflexible system and required some awful hacks to let the serializer work with it - the corresponding bSerialOverride flag is now gone.
The only thinker class still having a constructor is DFraggleThinker, because it contains non-serializable data that needs to be initialized in a piece of code that always runs, regardless of whether the object is created explicitly or from a savegame.
Doing this intermingled with the thinkers is highly unsafe because there are absolutely no guarantees about order of execution.
Effectively it ran these commands right in the middle of the playsim which could cause all sorts of synchronization issues, because CCMDs are part of the UI, not the playsim.
- pass a const string to AddCommandString.
This function manipulated the input buffer, leading to all sorts of code contortions to make sure that the passed parameter is clean for that.
This function will now create a copy of the passed parameter which it can manipulate without complicating its calling code.
# Conflicts:
# src/c_dispatch.cpp
This doesn't really write out any info for the pointer, if the level does not match it just errors out.
This is both for quick detection of badly used level data and for automatic restoring of the pointer from the serializer's working level.
This also removed the temporary workarounds in DAutomap and DLevelScript to restore these pointers when a savegame is loaded.
This was a relatively cheap change but removes a significant batch of references to the global variable, only making the entry points to the ACS interpreter relevant.
UI always runs on the primary level, so this does not need the ability to operate on multiple levels. Additionally, this can later be set to null when running play code so that scope violations result in an abort.
