For indexed PNGs, we only need to allocate a palette large enough to accommodate the number of indexed colours in the image. For images that don't use all 256 colours, this slightly reduces the memory usage for these images.
When LoadFromManifest is called, trim the various backing collections. These backing collections tend to live a long time without further modifications.
As the SpriteCache is used as a one-shot operation in practise, holding on to the capacity of backing collections is not required. Memory usage can be reduced by allowing the capacity to be reset after the SpriteCache has resolved items.
- Once LoadReservations is called, reset the reservation dictionaries so their backing collections can be reclaimed.
- When ResolveSprites is called, shrink the resolved dictionary as resolutions take place.
These implementations are often backed by a Dictionary, and tend to live a long time after being loaded. Ensure TrimExcess is called on the backing dictionaries to reduce the long term memory usage. In some cases, we can also preallocate the dictionary size for efficiency.
If you edit an actor name, then delete the actor - it fails to be removed from the map in the editor. This is because the actor previews are keyed by ID. Editing their name edits their ID and breaks the stability of their hash code. This unstable hash code means the preview will now fail to be removed from collections, even though it's the "same" object.
Fix this by making the ID immutable to ensure hash stability - this means that a preview can be added and removed from collections successfully. Now when we edit the ID in the UI, we can't update the ID in place on the preview. Instead we must generate a new preview with the correct ID and swap it with the preview currently in use.
The intended check was "has any trait", but TraitOrDefault throws if there is more than one. Adjust this check so it doesn't throw in the face of multiple trait instances.
Resolves a regression introduced in 63de527d9e.
In a3d0a50f4d, SpriteCache is updated to sort sprites by height before adding them onto the sheet. This improves packing by reducing wasted space as the sprites are packed onto the sheet. D2kSpriteSequence does not fully benefit from this change, as it creates additional sprites afterwards in the ResolveSprites method. These are not sorted, so they often waste space due to height changes between adjacent sprites and cause an inefficient packing. Sorting them in place is insufficient, as each sequence performs the operation independently. So sets of sprites across different sequences end up with poor packing overall. We need all the sprites to be collected together and sorted in one place for best effect.
We restructure SpriteCache to allow a frame mutation function to be provided when reserving sprites. This removes the need for the ReserveFrames and ResolveFrames methods in SpriteCache. D2kSpriteSequence can use this new function to pass in the required modification, and no longer has to add frames to the sheet builder itself. Now the SpriteCache can apply the desired frame mutations, it can batch together these mutated frames with the other frames and sort them all as a single batch. With all frames sorted together the maximum benefit of this packing approach is realised.
This reduces the number of BGRA sheets required for the d2k mod from 3 to 2.
Several parts of bot module logic, often through the AIUtils helper class, will query or count over all actors in the world. This is not a fast operation and the AI tends to repeat it often.
Introduce some ActorIndex classes that can maintain an index of actors in the world that match a query based on a mix of actor name, owner or trait. These indexes introduce some overhead to maintain, but allow the queries or counts that bot modules needs to perform to be greatly sped up, as the index means there is a much smaller starting set of actors to consider. This is beneficial to the bot logic as the TraitDictionary index maintained by the world works only in terms of traits and doesn't allow the bot logic to perform a sufficiently selective lookup. This is because the bot logic is usually defined in terms of actor names rather than traits.
When sheet builders are adding sprites to a sheet, they work left to right along each row. They reserve height for the highest sprite seen along that row, resetting the height reservation when the row runs out of space and it moves down to the next row.
As the SpriteCache adds the sprites in a giant batch, it can optimise this operation by ordering the sprites by their height. This reduces wastage where shorter sprites don't use the the full height reserved within the row. The reduced wastage can help the sheet builder allocate fewer sheets, improving load times and improving GPU memory usage as less texture memory is required.
The repeated small stream reads of ReadUInt16 generate a lot of overhead. Instead, consume the data in a single ReadBytes call and then unpack within the same buffer.
The assets for the Tiberian Dawn HD mod are much larger than assets for the default mods, causing a lot of load time to be spent in Util.FastCopyIntoChannel.
We can provide a special case for the SpriteFrameType.Bgra32 format, which is the same format as the destination buffer. In this scenario we can just perform memory copies between the source and destination. Additionally, whilst the default mods require all their assets to get their alpha premultiplied, many of the Tiberian Dawn assets are already premultiplied. Being able to skip this step for these assets saves us having to interpret the bytes into colors and back again.
For the default mods, there isn't a noticeable timing difference. For Tiberian Dawn HD or other mods with modern assets sizes, a large speedup is achieved.
This is a more natural representation than int that allows removal of casts in many places that require uint. Additionally, we can change the internal representation from long to uint, making the Color struct smaller. Since arrays of colors are common, this can save on memory.
Handle the client being null. Previously, a service could be created with a null client. This would leads to NREs when invoking the static Update methods. Now we guard against a null client.
- Rename the filename parameter to name and make it mandatory. Review all callers and ensure a useful string is provided as input, to ensure sufficient context is included for logging and debugging. This can be a filename, url, or any arbitrary text so include whatever context seems reasonable.
- When several MiniYamls are created that have similar content, provide a shared string pool. This allows strings that are common between all the yaml to be shared, reducing long term memory usage. We also change the pool from a dictionary to a set. Originally a Dictionary had to be used so we could call TryGetValue to get a reference to the pooled string. Now that more recent versions of dotnet provide a TryGetValue on HashSet, we can use a set directly without the memory wasted by having to store both keys and values in a dictionary.
Some existing widget are too small to accommodate their text. Adjust their sizes to fit. Text can be rendered outside the widget bounds so visually this often has no impact, but adjusting this now will help in the future for checking translation text for other languages fit in their widgets.
The Text element of these widgets was changed from display text to a translation key as part of adding translation support. Functions interested in the display text need to invoke GetText instead. Lots of functions have not been updated, resulting in symptoms such as measuring the font size of the translation key rather than the display text and resizing a widget to the wrong size.
Update all callers to use GetText when getting or setting display text. This ensure their existing functionality that was intended to work in terms of the display text and not the translation key works as expected.
An empty MiniYaml comment that was indented was previously not recognized, and instead parsed as a key named '#'. Now, indented comments are recognized as comments, which matches the behaviour for unindented lines.
This provides a single utility command for interacting with maps, that takes an arg for the map operation. The filename filter allows all maps in the mod to be operated on by default, or a regex can be passed to limit the operation to certain maps.
Reduces size used for png files from 13,366,660 bytes to 13,055,285 bytes in total. Changes size used for oramap files from 2,601,027 bytes to 2,605,779 bytes in total (contained PNGs are smaller, but the oramap zip wrapper didn't compress as well). This slight filesize improvement doesn't noticeably impact loading times.
zopfilpng is used for compression with the following command line:
'zopflipng.exe -y -m image.png image.png'
This follows on from 78bef8a98f and bc5e7d1497. Except now that the PNG decoder supports bit depths of 1, 2 or 4 we don't have to preserve the original bit depth of the image, allowing for more compression.
The oramap files were updated by:
- Running utility command "<mod> --unpack-map unpack" for each mod.
- Compressing the png files using the command above.
- Running utility command "<mod> --unpack-map repack" for each mod, except in Map.Save the line `if (!LockPreview) { var previewData = ...` is replaced with `if (false) { var previewData = ...` to save the existing optimized image on disk rather than generating a fresh preview.
This command allows either unpacking oramap files into folders, or packing folders into oramap files.
Example invocations:
"d2k --unpack-map unpack" to unpack maps of the d2k mod into folders.
"cnc --unpack-map repack" to repack maps of the cnc mod into oramap files (but will only pack folders that were unpacked previously).
The PNG decoder, when dealing when indexed images with a palette, could only decode a bit depth of 8. Teach it to decode depths of 1, 2 and 4 as well. As the palette data is exposed to consumers of the PNG class, unpack the data into a 8 bit depth so consumers don't need to also handle the new bit depths.
This allows the LINQ spelling to be used, but benefits from the performance improvement of the specific methods for these classes that provide the same result.
This rule no longer appears to be buggy, so enforce it. Some of the automated fixes are adjusted in order to improve the result. #pragma directives have no option to control indentation, so remove them where possible.
Calculate a rolling average of FPS over the last second. This allows the FPS counter to be updated every frame - and in particular means it can display a rough figure immediately rather than needing to wait one second to collect information at the start of a game.
When the widget is created, use the current frame as reference rather than always using zero. That avoids the first FPS reading from a new widget calculating as if all frames rendered since the game started occurred in the first second.
Previously the StartGameNotification and MusicPlaylist traits used the IWorldLoaded interface to play an audio notification and begin music when the game started. However this interface is used by many traits to perform initial loading whilst the load screen was visible, and this loading can take time. Since the traits could run in any order, then audio notification might fire before another trait with a long loading time. This is not ideal as we want the time between the audio notification occurring and the player being able to interact to be as short and reliable as possible.
Now, we introduce a new IPostWorldLoaded which runs after all other loading activity, and we switch StartGameNotification and MusicPlaylist to use it. This allows timing sensitive traits that want to run right at the end of loading to fire reliably and with minimal delay. The player perception of hearing the notification and being able to interact is now much snappier.
Ensure stale perf history data, to ensure the data is useful and the perf graph widget displays useful information.
- Remove stale data from the previous game when starting a new game. This avoids the graph showing values from the previous game when a new game starts.
- Remove data that was collected during loading. This avoids displaying data points that were collected whilst the loading screen was visible. Data collected whilst loading is not relevant to the in-game performance graph.
The performance graph when starting a new game will now display accurate information from the first tick of the game, whereas previously it displayed some stale information as well.
Also avoid ReadBytes calls that allocate a buffer by either updating the stream position (if not interested in the bytes), by reusing an input buffer (if interested in the bytes), or using a stackalloc buffer to avoid the allocation (for small reads).
This takes a list of map UIDs which may be locally installed or hosted
on the resource center. If any maps aren't found, startup will be
delayed by up to 10 seconds while it attempts to query the resource
center.
Disabling the shroud is sufficient to allow seeing the map. This fixes a game with the "Explored Map" option enabled. Previously using the `/all` command twice to toggle it on and off again would also reset the shroud, causing the map to no longer be explored. Now, using it twice will cause the map to remain explored, as intended when the "Explored Map" option is enabled.
As the `touched` cell layer uses Boolean values, Array.IndexOf is able to use a fast vectorised search. Most values in the array are false, so the search is able to significantly improve the performance of finding the next true value in the array.
The domains in HierarchicalPathFinder can be compared to find disjoint areas. For example islands on a water map will belong to different domains. Use these domains in path searches to allow us to bail out early if a path is impossible, e.g. trying to path between different islands. Keeping the domains updated via the RebuildDomains method adds some cost to the average path search, but that savings from path searches that can bail early pays for this many times over.
The ignoreSelf flag is intended to allow the current actor to be ignored when checking for blocking actors. This check worked correctly for cells occupied by a single actor. When a cell was occupied by multiple actors, the check was only working if the current actor happened to be the first actor. This is incorrect, if the current actor is anywhere in the cell then this flag should apply.
This flag failing to be as effective as intended meant that checks in methods such as PathFinder.FindPathToTargetCells would consider the source cell inaccessible, when it should have considered the cell accessible. This is a disaster for performance as an inaccessible cell requires a slow fallback path that performs a local path search. This means pathfinding was unexpectedly slow when this occurred. One scenario is force attacking with a group of infantry sharing the same cell. They should benefit from this check to do a fast path search, but failed to benefit from this check and the search would be slow instead.
Applying the flag correctly resolves the performance impact.
A recent update in the Lua extension makes it consider all fields that are defined as table entries annotated with @type to be readonly (providing a somewhat misleading warning saying that they don't exist). Defining them as @field annotations on the class makes it tread them normally.
This affects ScriptActorProperties and ScriptPlayerProperties.
The Lua extension would report missing/uninitialized fields on actor creation because it thought they were required. This makes them all optional, except for OwnerInit, which is special.
When handling the Nodes collection in MiniYaml, individual nodes are located via one of two methods:
// Lookup a single key with linear search.
var node = yaml.Nodes.FirstOrDefault(n => n.Key == "SomeKey");
// Convert to dictionary, expecting many key lookups.
var dict = nodes.ToDictionary();
// Lookup a single key in the dictionary.
var node = dict["SomeKey"];
To simplify lookup of individual keys via linear search, provide helper methods NodeWithKeyOrDefault and NodeWithKey. These helpers do the equivalent of Single{OrDefault} searches. Whilst this requires checking the whole list, it provides a useful correctness check. Two duplicated keys in TS yaml are fixed as a result. We can also optimize the helpers to not use LINQ, avoiding allocation of the delegate to search for a key.
Adjust existing code to use either lnear searches or dictionary lookups based on whether it will be resolving many keys. Resolving few keys can be done with linear searches to avoid building a dictionary. Resolving many keys should be done with a dictionary to avoid quaradtic runtime from repeated linear searches.
added unit test
update ToHex(byte[]) to support mono
added punctuations to unit test summary and parameter description
Replaced with Convert.ToHexString(), public ToHex() + use from Color.ToString()
Adjusted back to a simpler mono compatible version only, with lowered allocation
Ensure the target location can be pathed to by all units in the squad, so the squad won't get stuck if some units can't make it. Improve the choice of leader for the squad. We attempt to a choose a leader whose locomotor is the most restrictive in terms of passable terrain. This maximises the chance that the squad will be able to follow the leader along the path to the target. We also keep this choice of leader as the squad advances, this avoids the squad constantly switching leaders and regrouping backwards in some cases.
Previously, the ClosestTo and PositionClosestTo existed to perform a simple distance based check to choose the closest location from a choice of locations to a single other location. For some functions this is sufficient, but for many functions we want to then move between the locations. If the location selected is in fact unreachable (e.g. on another island) then we would not want to consider it.
We now introduce ClosestToIgnoringPath for checks where we don't care about a path existing, e.g. weapons hitting nearby targets. When we do care about paths, we introduce ClosestToWithPathFrom and ClosestToWithPathTo which will check that a path exists. The PathFrom check will make sure one of the actors from the list can make it to the single target location. The PathTo check will make sure the single actor can make it to one of the target locations. This difference allows us to specify which actor will be doing the moving. This is important as a path might exists for one actor, but not another. Consider two islands with a hovercraft on one and a tank on the other. The hovercraft can path to the tank, but the tank cannot path to the hovercraft.
We also introduce WithPathFrom and WithPathTo. These will perform filtering by checking for valid paths, but won't select the closest location.
By employing the new methods that filter for paths, we fix various behaviour that would cause actors to get confused. Imagine an islands map, by checking for paths we ensure logic will locate reachable locations on the island, rather than considering a location on a nearby island that is physically closer but unreachable. This fixes AI squad automation, and other automated behaviours such as rearming.
I noticed even on a naval only map, the naval AI doesn't necessarily beat a Normal AI. This makes it much more likely that it will.
- Drop number of ore refineries and ore trucks. As Naval AI is mostly suited for islands I haven't found a map that really needs as many as the other AIs.
- Reduce number of ground based base defenses - and delay Tesla coil a lot.
- Reduce number of migs as yaks more useful if they just get blown up.
- Add Flak trucks and v2s for base defense for Soviet
- Add Jeep and Arty for base defense for Allied
- Add delay for building ore truck so now chance of building one first from War Factory
- A service depot is not useful for this AI except for building an MCV so delay it a lot.
Tested with Ukraine and Germany and can consistently beat normal on island map.
1. Only allow new item being queued when cash above a certain number
2. Only tick one kind of queues at one tick, reduce the pressure on the actived tick
3. 'BaseBuilderBotModule' will check all buildings in producing, avoid queue mutiple same buildings.
If nested calls to RunUnsynced are running, then using a bool would cause the flag to be reset once the inner function completes, but an outer function may still be running and not yet ready for the flag to be reset. To correctly handle nested calls, we track a count and only reset the flag once all functions have completed.
As the solution currently targets .NET 6, a variety of style rules only introduced in .NET 7 are not suitable for enforcing as warnings (which are treated as errors in the CI pipeline). Anybody compiling locally with a .NET 6 SDK won't be able to trigger these rules locally, but the Linux CI agent comes with the .NET 7 SDK and will trigger these rules. This provides a poor dev experience as the CI run will report errors that don't reproduce locally.
To remove this developer friction, reduce the severity of these rules to avoid CI runs failing.
This changeset is motivated by a simple concept - get rid of the MiniYaml.Clone and MiniYamlNode.Clone methods to avoid deep copying yaml trees during merging. MiniYaml becoming immutable allows the merge function to reuse existing yaml trees rather than cloning them, saving on memory and improving merge performance. On initial loading the YAML for all maps is processed, so this provides a small reduction in initial loading time.
The rest of the changeset is dealing with the change in the exposed API surface. Some With* helper methods are introduced to allow creating new YAML from existing YAML. Areas of code that generated small amounts of YAML are able to transition directly to the immutable model without too much ceremony. Some use cases are far less ergonomic even with these helper methods and so a MiniYamlBuilder is introduced to retain mutable creation functionality. This allows those areas to continue to use the old mutable structures. The main users are the update rules and linting capabilities.
When laying mine with PreLayDelay > 0, end activity's tick immediately. That means don't try to immediately move to next cell.
This change unifies the behavior with scenario when a mine is laid without any PreLayDelay.
Use CollectionsMarshal to hold a ref to the dictionary entry. When the value needs to be added this allows us to set the value directly into it without having to locate the entry a second time.
- Seal the classes, and make SourceLocation a readonly struct.
- In ToDictionary, use TryAdd to avoid a try-catch.
- In Merge, use ToList to ensure sources is only enumerated once.
Moving the key duplication check allows a redundant check on top-level nodes to be avoided. Add tests to ensure key checks are functioning as expected.
- In FieldLoader, cache boxed bools and some boxed ints.
- In FieldLoader, presize collections when parsing a List, HashSet or Dictionary.
- In FieldLoader, don't allocate a list of missing items until required.
- In FieldLoader, when a string value is passed, avoid wrapping this in a MiniYaml object by allowing both strings and yaml to be passed in the GetValue overload that does the real work.
- In Animation, avoid allocating no-op actions.
- In VxlReader, use EnsureCapcity to better size the Dictionary.
- In VxlReader change VxlElement to a struct.
- In Locomotor, presize TerrainSpeeds dictionary.
- In MixFile, the Distinct call doesn't presize the HashSet it uses internally. As we know we will enumerate all results, create the HashSet ourselves so that is it presized correctly.
- In ObjectCreator, stream the assembly when hashing rather than reading all bytes into memory.
These changes avoid some allocations on the large object heap, in turn this means the GC can avoid performing unnecessary Gen 2 collections just to clear down the LOH.
- In RLEZerosCompression use dedicated Array.Clear method instead of open-coded loop.
- In VoxelLoader.GenerateSlicePlanes.Get use TryGetValue to avoid repeated array and dictionary lookups.
- In TmpTSLoader.UnpackTileData use ReadBytes to populate array with less overhead compared to repeated one byte reads.
- Resolve TODO in VqaVideo.
Provides two callbacks using INotifyMineLaying interface: MineLaying (just before laying a mine), MineLaid after mine actor has been created (in FrameEndTask)
Apparently building/publishing/packaging for Windows on Linux using .NET 6 sets some of the assembly / Portable Executable information for the generated DLL, but nothing for the generated EXE (which isn't the case when building on Windows).
Reduces size used for oramap files from 2,774,405 bytes to 2,614,332 bytes in total. The smaller files also improve loading times as the zlib decoder has less bytes to schlep through.
The PNG decoder shipped with OpenRA only supports a bit depth of 8 for images with a palette, so must must ensure that other depths (1,2 or 4) are not used.
zopfilpng is used for compression with the following command line:
'zopflipng.exe --keepcolortype -y -m image.png image.png'
The keepcolortype flag ensures the bit depth is not changed.
The oramap files were updated by:
- Saving as a folder, unpacking the data in loose files.
- Compressing the png files using the command above.
- Saving back as an oramap, except instead of regenerating the map preview on save we save the existing version on disk.
Reduces size used for png files from 19,034,446 bytes to 13,305,798 bytes in total. The smaller files also improve loading times as the zlib decoder has less bytes to schlep through.
The PNG decoder shipped with OpenRA only supports a bit depth of 8 for images with a palette, so must must ensure that other depths (1,2 or 4) are not used.
zopfilpng is used for compression with the following command line:
'zopflipng.exe --keepcolortype -y -m image.png image.png'
The keepcolortype flag ensures the bit depth is not changed.
Most cells are accessible, so presizing to the full size of the grid is sensible to avoid allocations to resize as it is filled up. The set can also be reused across all layers to avoid allocating it many times.
This avoids loading, parsing and merging YAML rules for the mod during loading of each individual map. This saves significant time resolving custom rules on each map loaded.
When a search is initiated from an inaccessible source location, a path is still allowed if there is an adjacent, accessible location the unit can move into. The local pathfinder and HierarchicalPathFinder already account for this logic, but HPF has some bugs.
Firstly, when the HierarchicalPathFinder heuristic is being evaluated, it assumes all cells being explored are accessible - this is important for performance as it avoids constantly rechecking the accessibility of cells. Although this fact holds true for cells explored by the path search, it does not hold true for cells being added as the initial starting points of the search.
Secondly, when checking for adjacent locations to an inaccessible source cell, we checked only if these were still on the map. This is insufficient - we need to know if movement between the source cell and the adjacent cell is possible.
The fixes resolve this by:
- Teaching the heuristic an extra parameter to know if the location is known to be accessible. This allow an accessibility check to be performed for starting locations which stops HPF mistakenly assuming the abstract node for that location is the one we need to consider, and to correctly check the adjacent locations and their abstract nodes instead. The parameter means will can still skip the accessibility check in the typical case where the path search is being expanded, preserving performance.
- When adjacent cells are considered we now check if movement to them is possible. This stops HPF from allowing jumps over height discontinuities (i.e. no magically jumping up or down cliffs) and thinking a path is possible when it is not.
- Consuming methods cared only about the count and not the actual actors, so only counts the actors rather that creating lists.
- ProvidesPrerequisites implementations return cached objects rather then allocating new enumerables on each call.
# Example Scenario
MockString:
CollectionOfStrings:
StringA: A
MockString:
CollectionOfStrings:
StringB: B
MockString:
-CollectionOfStrings:
MockString:
CollectionOfStrings:
StringC: C
MockString:
CollectionOfStrings:
StringD: D
# Previous MergePartial result
# The CollectionOfStrings is merged into a single unit, so the C and D items are dragged upwards and jump ahead of the Removal
# When this is processed, the final result removes CollectionOfStrings entirely
MockString:
CollectionOfStrings:
StringA: A
StringB: B
StringC: C
StringD: D
-CollectionOfStrings:
# New MergePartial result
# When merging nodes, we no longer allow merges to jump an intervening removal node.
# This means we can have multiple of a certain key (CollectionOfStrings in this example) which was not the case previously.
# When this is processed, the final result includes C/D but not A/B.
MockString:
CollectionOfStrings:
StringA: A
StringB: B
-CollectionOfStrings:
CollectionOfStrings:
StringC: C
StringD: D
MockString:
CollectionOfStrings:
StringA: A
MockString:
CollectionOfStrings:
StringB: B
MockString:
-CollectionOfStrings:
MockString:
CollectionOfStrings:
StringC: C
MockString:
CollectionOfStrings:
StringD: D
MockString:
CollectionOfStrings:
StringA: A
StringB: B
StringC: C
StringD: D
-CollectionOfStrings:
MockString:
CollectionOfStrings:
StringA: A
StringB: B
-CollectionOfStrings:
CollectionOfStrings:
StringC: C
StringD: D
Split out ImportGen2MapCommand from ImportTSMapCommand, to also serve as a base for a RA2 map importer.
Also renamed TS importer to ImportTiberianSunMapCommand.
Enforces a variety of CAxxxx rules that do not have existing violations.
For the benefit of dotnet_code_quality.CA2241.try_determine_additional_string_formatting_methods_automatically = true, rename parameters of methods that forward to string.Format so format issues will get detected automatically.
The Harvester trait and MoveAdjacentTo activity called the pathfinder but had a single source and multiple targets. The pathfinder interface only allows for the opposite: multiple sources and a single target. To work around this they would swap the inputs. This works in most cases but not all cases. One aspect of asymmetry is that an actor may move out of an inaccessible source cell, but not onto an inaccessible target cell.
Searches that involved an inaccessible source cell and that applied this swapping method would therefore fail to return a path, when a valid path was possible. Although a rare case, once good way to reproduce is to use a production building that spawns actors on inaccessible cells around it, such as the RA naval yard. A move order uses the pathfinder correctly and the unit will move out. Using a force attack causes the unit to use the broken "swapped" mechanism in MoveAdjacentTo and it will be stuck.
This asymmetry has been longstanding but the pathfinding infrastructure only sporadically accounted for it. It is now documented and applied consistently. Create a new overload on the pathfinder trait that allows a single source and multiple targets, so callers have an overload that does what they need and won't be tempted to swap the positions and run into this issue.
Internally, this requires us to teach Locomotor to ignore the self actor when performing movement cost checks for these "in reverse" searches so the unit doesn't consider the cell blocked by itself.
Issues:
- The generated API file caused a ton of diagnostics warnings.
- Perhaps due to EmmyLua moving forward since this was created, we were missing out on some features.
Changes:
- Disabled diagnostics about missing return values for functions.
- Added add operator definitions because by default CPos+CVec would be assumed to return a number and assignments would cause warnings about type mismatches.
- Added explicit @enum annotations. This also fixed warnings in the generated API file about missing types.
- Changed default type of properties from an empty table to nil. This fixed warnings in the generated API file about type mismatch.
- Renamed local variable "required" -> "requiredTraits" for readability.
- Disabled duplicate function/property name warnings
- Updated trait docs URL.
Since # 19174 bounds are calculated from WDist instead of pixels. This causes many of the older maps to crash games, And this lint test makes it easier for map makers to detect these issues.
As calculating the bounds requires MapGrid, we need to write a separate lint test for Interactable instead of defaulting to RulesetLoaded checks.
- Providing the comparer as a type argument that is a struct allows the calls to be devirtualised, leading to approx a 3x performance improvement.
- Use a single backing array, rather than a list of arrays.
- Enforce SA1604 ElementDocumentationShouldHaveSummary.
- Enforce SA1629 DocumentationTextShouldEndWithAPeriod.
- Turn off some rules covered by IDExxxx rules.
- Remaining rules are treated as part of OpenRA style.
- Group style rules with their associated options in a way that matches the documentation. This makes it easier to pair rules and their options.
- Remove OpenRA.ruleset and move all rules into .editorconfig file.
- Centralise IDE0005 workaround in Directory.Build.props file.
ContentPackages are defined in mod.yaml and list Installers that support them, but then the Installers and their SourceActions knew nothing about ContentPackages.
Also added BeforeInstall and AfterInstall sections for SourceActions in the Installers.
Multiple layers of Lazy<T>ness are replaced with
an explicit two-part loading scheme.
Sequences are parsed immediately, without the need
for the sprite assets, and tell the SpriteCache
which frames they need. Use-cases that want the
actual sprites can then tell the SpriteCache to
load the frames and the sequences to resolve the
sprites.
PlayerColorRemap expected colors in linear space yet we provided them in gamma. We fix this by instead expecting gamma space colors and then converting them into linear space ourselves.
Previous commits removed the async loading of videos, which can be a problem for videos played in the radar widget mid-game because it can cause a lag spike. This loads the video on a separate thread and runs it on the main thread whenever it is loaded, not blocking the main thread meanwhile and allowing the game to continue while the video loads.
Also add back cancelling of already playing video and add a check to not try to run onComplete if it is null.
Somehow this would divide by 0 on the next line, not crashing but resulting in an integer overflow on `overlayHeight`, which would crash when trying to render.
- Improve consistency
- Remove dead code
- Removed AsyncAction!
- Delegate.BeginInvoke is unsupported since .NET Core and throws an "Operation is not supported on this platform." exception.
This moves initialization of Initialize getters for left/right arrow image from constructor Initialize method making it possible to override fields Decorations, DecorationScrollLeft and DecorationScrollRight after widget creation.
All magic behaviour for constructing sprite filenames
has been removed in favour of an explicit Filename
(and TilesetFilenames for tileset-specific sequences)
property.
The previous MiniYaml.Merge implementation interacted
poorly with yaml inheritance, making it complicated
(or impossible) to override certain keys from Defaults.
The new implementation is simpler: If a key is defined
it will be used. If it isn't, the default (if defined)
will be used. Defaults can be masked by making sure
the same key is defined (even with an empty value)
in the sequence.
This also fixes naming within the sequence code to
distinguish between images (a group of sequences),
sequences (defining a specific sprite/animation),
and filenames for a specific sprite/animation.
The cache in Locomotor that is populated via the UpdateCellBlocking method disagreed with the non-cached logic of IsBlockedBy when dealing with Mobile actors. The cache determined an actor to be moving if it was both movable and had horizontal movement types. IsBlockedBy determined an actor to be moving if it had horizontal movement types, but did not check if it was movable. This difference in checks could allow a mobile trait that was disabled or paused and which had horizontal movement to be treated differently be the two methods. UpdateCellBlocking would consider it not moving due to the disabled/paused trait. IsBlockedBy would consider it moving as it didn't care about the disabled/paused state of the trait.
Now, we unify the two methods to consider a mobile trait that is disabled/paused as not moving. This prevents HierarchicalPathFinder from crashing on the inconsistent state, i.e. when asked to path search through a cell of a mobile unit which has disabled or paused movement, but which has horizontal movement types from prior movement.
* Minimum macOS version is raised to 10.11.
* App bundles ship 3 versions of the runtime and engine binaries,
and a fat launcher that selects the appropriate runtime/apphost.
* Mono is used for macOS 10.11 - 10.14, or if OPENRA_PREFER_MONO
environment variable has been set.
Set an actor moving along several waypoints whilst the /path-debug command is active, then deselect the actor. Each waypoint will add more information to the debugging overlay until it crashes with "Maximum two records permitted." This resolves the crash by no longer adding new debugging information once the actor is deselected.
An event is added to Map to indicate when the cell projection is changed. This is important as this can mean Map.Contains(CPos) could now return different results for the cell. The HierarchicalPathFinder is made aware of these changes so it can rebuild any out-of-date information. This fixes prevent a crash if a cell that was previously outside the map changes height and becomes inside the map. The local path search will explore the cell as it is inside the map - but if the HPF was unaware if had been updated, it will still consider the cell to be outside the map and unreachable, resulting in a crash.
If a path search is performed by the HierarchicalPathFinder when the source cell is unreachable location, a path is still allowed and starts from one of the cells adjacent to the location. If moving into one of these cells results in the actor moving into an isolated area that cannot reach the target this would previously crash as no abstract path could be found. Now we handle such locations by giving them a unreachable cost so the path search will not attempt to explore them.
Imagine a map split into two by a one tile wide line of impassable cliffs. If an actor is on top of these cliffs it is allowed to path because it can "jump off" the cliff and move into the cell immediately either side of it. However it is important which side it chooses to jump into, as one it has moved off the cliff it loses access to the other side. The previous error came about because the path might try and search on the side that couldn't reach the target location. Now we avoid that being considered.
- Add prefixes to all message keys to provide context
- Use messages with attributes for some UI elements (dropdowns, dialogs, checkboxes, menus)
- Rename some class fields for consistency with translation keys
During a game notification duration should be the same regardless of
game speed. Switch to using wall-clock time defined in milliseconds
instead of game ticks. Also use the opportunity to rename the field
to "Duration" because "RemoveTime" is not so clear.
DeployForGrantedCondition is wrapped around the Move activity, so the AttackMoveActivity thinks that DeployForGrantedCondition is the Move activity.
All it means is that we need to forward the target line request to the Move activity
The Locomotor IsMoving check was allowing us to consider another actor that moving as not a blocker. However for some reason it also considered the actor trying to path being mobile as sufficient for this check to pass which did not make sense. We remove that extra check and inline the method.
This was a regression from 4a609bbee8 which changed the method from IsMovingInMyDirection (which required the lookup of the mobile trait) to just IsMoving. It should have removed the lookup as not required.
This fixes a crash in HPF which was considered the location as blocked when Locomotor considered it unblocked because the logic was not aligned. Removing this check aligns the logic and resolves the crash.
Using the local pathfinder, you could not find a path to an unreachable destination cell, but it was possible to find a path from an unreachable source cell if there was a reachable cells adjacent to it.
The hierarchical pathfinder did not have this behaviour and considering an unreachable source cell to block attempts to find a path.
Now, we unify the pathfinders to use a consistent behaviour, allowing paths from unreachable source cells to be found.
Occupied cells was defined by height yet we didn't update actor map on changing height. This in some scenarios could have caused the aircraft to forget to remove its influence from actor map
The variable `${MOD}` was not enclosed in quotes, but the value contained a space. This caused the argument to be split into two parts.
Using **find** isn't necessary due to globbing.
- Fix an instance where "silo-usage" translation was used without
arguments
- Use the same translation reference for the "Power usage"
- Make the ResourceBarWidget accept a cached transform with the tooltip
text
so it won't have to build the string itself
- Display an infinity symbol when the infinite power cheat is used
- Removes a magic number that is no longer used (>1000000 to check for
unlimited power)
- Use LongBitSetAllocator and not BitSetAllocator. Using the wrong allocator means all string based checks and displays would provide incorrect results.
- Remove LongBitSetAllocator.Mask which wasn't being calculated or Reset correctly. We can use world.AllPlayersMask to provide the same effect at use sites.
When crushables and crates change their Location/TopLeft, their crushability is cached, but when their CenterPosition is changed, their cached crushability is not refreshed. Since their CrushableBy functions depends on IsAtGroundLevel, which depends on the CenterPosition, this means that when the crushability is cached it will depend on the current height of the object. If the height of the object changes, the cache is not refreshed and now contains out of date information.
The Locomotor cache and the HPF both cache this same information, but at different times. HPF caches immediately, but Locomotor caches on demand which means there can be a delay. This means they can have inconsistent, differing views of the crushability information. This eventually surfaces in a "The abstract path should never be searched for an unreachable point." error from HPF when it detects the inconsistency.
The bug is that Locomotor was caching information without refreshing it when required. Fixing this to refresh the cache when the CenterPosition changes is likely to have negative performance impacts. As would removing crushability from the cache. These would both be fixes that address the underlying bug.
The high impacts of a proper fix lead us to a workaround instead. If we set the CenterPosition before setting the Location, then when the Location is set and the caches are refreshed, the new CenterPosition is available when caching the crushability information. This means logic depending on IsAtGroundLevel will get the new information and cache a more up-to-date view of things. This means when changing both the CenterPosition and Location together we now cache correct information. However calls that set only the CenterPosition and not the Location can still result in a bad cache state. Although this is imperfect it is an improvement over current affairs, and has less impact.
Minor adjustment into the D2k
- Add rally point into the Palace
- removed harvester MustBeDestroyed in campaing
- Players can see they carryalls and ornothopers under the fog.
- Increased CameraRemoveDelay on Superweapons so player can see superweapon impact.
When the Land activity is run, the aircraft adds influence to the cell so it cannot be used by other actors. When the TakeOff activity runs, it removes the influence so the cell can be used by other actors.
However, when a Carryall picks up a unit, it is told to Land with a vertical offset - it never reaches ground level. When the TakeOff activity runs, it saw the aircraft was above ground level and bailed out. The means the influence is never removed. The cell is now unusable despite the fact the Carryall has left.
To fix this, TakeOff now checks if influence was applied instead of checking if the aircraft is above ground level. If so, we know the Land activity had decided that influence was required, even if the aircraft has not made it to ground level. When TakeOff runs, it will treat it as a proper take off event even though the aircraft is already above ground level. This means influence will be removed and the cell will become accessible as intended.
In ActorMap, we also fix a design flaw where disposed actors where excluded from queries. This caused cache inconsistencies with clients using ActorMap.CellUpdated event to rely on updates. This event will not get called when the actor was disposed, so the downsteam client may have cached the actors at that location, only for them to "change" when the actor is later disposed. This could cause the Locomotor and HierarchicalPathFInder to have inconsistent views of the actors on the map, causing crashes if the inconsistent state broken some internal invariants. The only reason to exclude disposed actors would be to cover up for the actors not being removed properly from the map, which is fixed now aircraft are handled correctly. If ever an actor isn't removed from the actor map, then the caller needs fixing rather than having the actor map exclude it.
If a path search is attempted from a location outside the map, then PathSearch will filter these out to prevent any crashes. The path search will result in no path. However if the location is within the map but on a custom movement layer that the locomotor cannot use, this currently crashes. To fix this we apply a similar filtering logic to ignore any source locations that cannot be used, and so the path search will result in no path for these as well.
Disallow join button without IP address Input
Closes#20234
Trying to join a server with an empty address crashes the game. This fix disallows pressing join button without ip address field input. Updated to reuse search for joinButton vs 2 separate calls to search.
The BlockingCollection would have `IsAddingCompleted` to true, but `IsComplete` to false, slipping through the cracks and causing an InvalidOperationException ("The collection has been marked as complete with regards to additions.") when trying to add to it.
We now add a check on `(Try)SendData` to only try to add if we can. The collection is still viable for reading until empty/`IsComplete`.
When using the internal AbstractCellForLocalCell method to check if a local cell is reachable, this should return null when the cell is unreachable. If multiple abstract cells were required for that grid, this worked as intended. Only reachable cells are stored in the localCellToAbstractCell mapping. For a grid that required only a single abstract cell, which is the common case, we optimize this to store only the single abstract cell rather than the whole mapping for potentially 100 cells in that grid. However this makes no distinction between the reachable and unreachable cells, so when we check later we get incorrect results. If a cell is unreachable but belongs to the same grid as a single group of reachable cells then we incorrectly report it as reachable. The easiest way to see this incorrect behaviour is when the PathExists is called and can sometimes indicate a path exists when it does not.
To fix this, we now ensure we perform a check to see if the cell is reachable in this single layer case, this allows us to retain the optimization where we don't need to store the whole mapping, but allows us to correctly indicate when cells are unreachable.
Switched the Utility's ExtractTraitDocsCommand output to JSON.
Updated documentation generation to use that and the new Python script to generate the Markdown file, same as the Weapon documentation.
Modifying the list potentially several thousand times is really slow, so notify the child elements that they are being removed and then clear the list in one go.
By tracking updates on the ActorMap the HierarchicalPathFinder can be aware of actors moving around the map. We track a subset of immovable actors that always block. These actors can be treated as impassable obstacles just like terrain. When a path needs to be found the abstract path will guide the search around this subset of immovable actors just like it can guide the search around impassable terrain. For path searches that were previously imperformant because some immovable actors created a bottleneck that needed to be routed around, these will now be performant instead. Path searches with bottlenecks created by items such as trees, walls and buildings should see a performance improvement. Bottlenecks created by other units will not benefit.
We now maintain two sets of HPFs. One is aware of immovable actors and will be used for path searches that request BlockedByActor.Immovable, BlockedByActor.Stationary and BlockedByActor.All to guide that around the immovable obstacles. The other is aware of terrain only and will be used for searches that request BlockedByActor.None, or if an ignoreActor is provided. A new UI dropdown when using the `/hpf` command will allow switching between the visuals of the two sets.
When the UpdateCellBlocking encountered a transit-only cell (the bibs around a building) it would bail from the loop. This would leave the cellCrushablePlayers set to all players. It would update the cell cache and mark that cell as a crushable location.
When CanMoveFreelyInto would later evaluate a cell, it would consider it passable because the crushable check would pass (cellCache.Crushable.Overlaps(actor.Owner.PlayerMask)) rather than because the transit check (otherActor.OccupiesSpace is Building building && building.TransitOnlyCells().Contains(cell)) would pass.
Although this meant the cell was treated as passable in either scenario, it means the cache contained incorrect data. The cell does not contain any crushable actors but the cache would indicate it did. Correcting this means we can rely on the crushability information stored in the cache to be accurate.
When this cheat is used by notifying of shroud changes we invoke the usual logic that would occur if the visibility had been granted by units. Without this change any cached information about the visibility is not refreshed. Without this refresh actors with different visibility may not act correctly.
One aspect this improves is frozen actors. Using the visibility cheat will show up all actors on the map. If the cheat is then disabled than frozen actors will appear in their place. Prior to this change a frozen actor would fail to appear if the cheat had caused it to be revealed. Healthbars and selection boxes are also made consistent for similar reasons.
Since bbf5970bc1 we update frozen actors only when required.
In 8339c6843e a regression was fixed where actors created in line of sight would be invisible.
Here, we fix a related regression where cloaked units that are revealed, and then frozen when you move out of line of sight would lack tooltips.
The fix centers around the setting of the Hidden flag. In the old code this used CanBeViewedByPlayer which checks for visibility modifiers and then uses the default visibility. The bug with this code is that when a visibility modifier was not hiding the actor, then we would report the default visibility state instead. However the frozen visibility state applies here which means if the frozen actor is visible, then we consider the actor to be hidden and therefore tooltips will not appear. In the fixed version we only consider the modifiers. This means a visibility modifier such as Cloak can hide the frozen actor tooltips. But otherwise we do not consider the frozen actor to be hidden. This prevents a frozen actor from hiding its own tooltips in some unintended circular logic. Hidden now becomes just a flag to indicate if the visibility modifiers are overriding things, as intended.
Since bbf5970bc1 we only update frozen actor state on demand rather than every tick. However when the actor was initially created we were failing to set the initial visibility state if the frozen actor was invisible.
With this fix, we now set the visibility states on creation correctly. This fixes an issue where enemy actors created within line of sight would not appear.
During the refactoring to introduce HierarchicalPathFinder, custom costs were no longer applied to source locations. The logic being that if we are already in the source location, then there should be no cost added to it to get there - we are already there!
Path searches support the ability to go from multiple sources to a single target, but the reverse is not supported. Some callers that require a search from a single source to one of multiple targets perform their search in reverse, swapping the source and targets so they can run the search, then reversing the path they are given so things are the correct way around again. For callers that also use a custom cost like the harvester code that do this in order to find free refineries, they might want the custom cost to be applied to the source location. The harvester code uses this to filter out overloaded refineries. It wants to search from a harvester to multiple refineries, and thus reverses this in order to perform the path search. Without the custom cost being applied to the "source" locations, this filtering logic never gets applied.
To fix this, we now apply the custom cost to source locations. If the custom cost provides an invalid path, then the source location is excluded entirely. Although this seems unintuitive on its own, this allows searches done "in reverse" to work again.
When asked to find a path from multiple source locations, the abstract search is used to determine which source locations are viable. Source locations that cannot be reached on the abstract graph are excluded from the local path search. As we know the locations are unreachable, this prevents the local path search from expanding over the entire search space in an attempt to find these unreachable locations, preventing wasted effort.
In order to determine these reachable locations, the abstract search is expanded successively trying to reach each source location each time. However, this failed to account for a property of the ExpandToTarget for which a comment is now added. If the location was found previously, then expanding to try and find it again will fail. If the source locations were close together, it was likely that the initial expansions of the search space would have included them, and thus they would not be found on a later expansion. This would mean these locations would incorrectly be thought unreachable.
To fix this, we check if the location has already been explored (has CellStatus.Closed in the graph). If so we can check the cost to determine if it is reachable.
Previously, actors that were visible would refresh their frozen actor state every tick in preparation for the actor becoming hidden, and the frozen actor appearing as a placeholder instead.
By using ICreatesFrozenActors.OnVisibilityChanged when can avoid refreshing the state constantly, and instead just refresh it the moment the frozen actor needs to appear. This provides a nice performance improvement on the cost on managing frozen actors.
Activated with the '/path-debug' chat command, this displays the explored search space and costs when searching for paths. It supports custom movement layers, bi-directional searches as well as visualizing searches over the abstract graph of the HierarchicalPathFinder. The most recent search among selected units is shown.
Teach HierarchicalPathFinder to keep a cache of domain indices, refreshing them only on demand and when invalidated by terrain changes. This provides an accurate and quick determination for checking if paths exist between given locations.
By exposing PathExistsForLocomotor on the IPathFinder interface, we can remove the DomainIndex trait entirely.
Replaces the existing bi-directional search between points used by the pathfinder with a guided hierarchical search. The old search was a standard A* search with a heuristic of advancing in straight line towards the target. This heuristic performs well if a mostly direct path to the target exists, it performs poorly it the path has to navigate around blockages in the terrain. The hierarchical path finder maintains a simplified, abstract graph. When a path search is performed it uses this abstract graph to inform the heuristic. Instead of moving blindly towards the target, it will instead steer around major obstacles, almost as if it had been provided a map which ensures it can move in roughly the right direction. This allows it to explore less of the area overall, improving performance.
When a path needs to steer around terrain on the map, the hierarchical path finder is able to greatly improve on the previous performance. When a path is able to proceed in a straight line, no performance benefit will be seen. If the path needs to steer around actors on the map instead of terrain (e.g. trees, buildings, units) then the same poor pathfinding performance as before will be observed.
Commands are registered in WorldLoaded event handlers, and IngameChatLogic takes all registered commands and provides tab completion. However IngameChatLogic is also created during WorldLoaded via LoadWidgetAtGameStart. No initialization order is enforced between commands and LoadWidgetAtGameStart, so they can appear in any order.
If a command gets registered before LoadWidgetAtGameStart runs, then it will get tab completion. If it gets registered after then no tab completion is available, even though the command can still be used and appears when using '/help'.
To fix this, we allow the tab completion to check for available commands lazily, meaning it will check for available commands every time the tab key is pressed. This means it will always have the full list of commands available regardless of the initialization order.
Error messages are displayed using the following methods:
* **zenity** parses pango markup and replaces escaped characters
* **kdialog** replaces (some) escaped characters
* **gtk-dialog.py** replaces `\n`
* **printf** interprets format strings and replaces escaped characters
* **echo** just displays the text
The error messages themself contain escaped characters and paths from variables.
This PR unifies the behavior by:
* Use **printf** to format error messages and replace escaped characters
* Setting `--no-markup` for **zenity** to disable pango markup and escaped characters
* Remove `\n` replacement from **gtk-dialog.py**.
* Use plain **echo** instead of **printf**
Use **xargs** to pass results of **find** instead of word splitting. Word splitting fails when filenames contain white spaces (or if no files are found).
To prepare them for documentation generation.
Also added descriptions to SpriteSequence implementations and their properties.
Also made a few code style fixes.
Two different issues were causing a path search to not explore cells in order of the cheapest estimated route first. This meant the search could sometimes miss a cheaper route and return a suboptimal path.
- PriorityQueue had a bug which would cause it to not check some elements when restoring the heap property of its internal data structure. Failing to do this would invalidate the heap property, meaning it would not longer return the items in correct priority order. Additional tests ensure this is covered.
- When a path search encountered the same cell again with a lower cost, it would not update the priority queue with the new cost. This meant the cell was not explored early enough as it was in the queue with its original, higher cost. Exploring other paths might close off surrounding cells, preventing the cell with the lower cost from progressing. Instead we now add a duplicate with the lower cost to ensure it gets explored at the right time. We remove the duplicate with the higher cost in CanExpand by checking for already Closed cells.
Prior to ef44c31a72, Locomotor would be earlier in the trait initialization sequence than SpawnStartingUnits. After this commit, the initialization sequence was perturbed and SpawnStartingUnits would initialize first. When SpawnStartingUnits would query CanEnterCell this would generate a null reference as Locomotor had not yet initialized.
SpawnStartingUnitsInfo is made to initialize NotBefore LocomotorInfo to enforce the required trait ordering.
The following error occurs on a map where the radar activation sound plays:
Exception has occurred: CLR/System.ArgumentException
An exception of type 'System.ArgumentException' occurred in System.Private.CoreLib.dll but was not handled in user code: 'Property set method not found.'
at System.Reflection.RuntimePropertyInfo.SetValue(Object obj, Object value, BindingFlags invokeAttr, Binder binder, Object[] index, CultureInfo culture)
at System.Reflection.RuntimePropertyInfo.SetValue(Object obj, Object value, Object[] index)
at OpenRA.FieldLoader.LoadField(Object target, String key, String value) in /home/jason/git/OpenRA/OpenRA.Game/FieldLoader.cs:line 609
at OpenRA.WidgetLoader.LoadWidget(WidgetArgs args, Widget parent, MiniYamlNode node) in /home/jason/git/OpenRA/OpenRA.Game/Widgets/WidgetLoader.cs:line 60
at OpenRA.WidgetLoader.LoadWidget(WidgetArgs args, Widget parent, MiniYamlNode node) in /home/jason/git/OpenRA/OpenRA.Game/Widgets/WidgetLoader.cs:line 67
at OpenRA.WidgetLoader.LoadWidget(WidgetArgs args, Widget parent, MiniYamlNode node) in /home/jason/git/OpenRA/OpenRA.Game/Widgets/WidgetLoader.cs:line 67
at OpenRA.WidgetLoader.LoadWidget(WidgetArgs args, Widget parent, MiniYamlNode node) in /home/jason/git/OpenRA/OpenRA.Game/Widgets/WidgetLoader.cs:line 67
at OpenRA.WidgetLoader.LoadWidget(WidgetArgs args, Widget parent, String w) in /home/jason/git/OpenRA/OpenRA.Game/Widgets/WidgetLoader.cs:line 43
at OpenRA.Game.LoadWidget(World world, String id, Widget parent, WidgetArgs args) in /home/jason/git/OpenRA/OpenRA.Game/Game.cs:line 160
at OpenRA.Mods.Common.Widgets.Logic.LoadIngamePlayerOrObserverUILogic..ctor(Widget widget, World world) in /home/jason/git/OpenRA/OpenRA.Mods.Common/Widgets/Logic/Ingame/LoadIngamePlayerOrObserverUILogic.cs:line 34
Move the domain logic involved into a base class named DensePathGraph. The base class contains all the domain logic necessary to traverse a graph including concepts such as custom movement layer.
PathGraph becomes responsible for proving a backing array for the pathfinding information, and is where the pooling logic lives instead, helping split the two concepts out.
The restores the previous behaviour before FindUnitPathToTargetCell was introduced. This prevents callers such as the harvester code crashing when a harvester tries to route home to a refinery, but there are no refineries.
Prior to ef44c31a72, Locomotor would be earlier in the trait initialization sequence than SpawnMapActors. Locomotor would assume no actors on the map, and register to update blocked cells when new ones were added. When SpawnMapActors created actors, Locomotor was made aware and kept up-to-date.
After this commit, the initialization sequence was perturbed and SpawnMapActors would initialize first. Locomotor would assume no actors on the map and thus be unaware of these starting units, meaning those starting units would not cause blocking, allowing units to pass through them.
There are two possible fixes. SpawnMapActorsInfo can initialize NotBefore<LocomotorInfo>, enforcing that actors are spawned after locomotor is ready. Or we can remove the assumption in Locomotor that the map starts empty, and have it update blocked cells on startup. The latter seems cleaner, so any other traits that may want to spawn actors don't have to be aware sequencing their initialization with the Locomotor trait, instead things would "just work".
The types for Int32 and Boolean are currently replaced with friendly names of int and bool for the docs. Ensure we apply the same handling when these are nullable types, changing the output from Int32? and Boolean? to int? and bool?
Some path searches, using PathSearch, were created directly at the callsite rather than using the pathfinder trait. This means some searches did not not benefit from the performance checks done in the pathfinder trait. It also means the pathfinder trait was not responsible for all pathing done in the game. Fix this with the following changes:
- Create a sensible shape for the IPathFinder interface and promote it to a trait interface, allowing theoretical replacements of the implementation. Ensure none of the concrete classes in OpenRA.Mods.Common.Pathfinder are exposed in the interface to ensure this is possible.
- Update the PathFinder class to implement the interface, and update several callsites manually running pathfinding code to instead call the IPathFinder interface.
- Overall, this allows any implementation of the IPathFinder interface to intercept and control all path searching performed by the game. Previously some searches would not have used it, and no alternate implementations were possible as the existing implementation was hardcoded into the interface shape.
Additionally:
- Move the responsibility of finding paths on completed path searches from pathfinder to path search, which is a more sensible location.
- Clean up the pathfinder pre-search optimizations.
Requires<T> means that trait of type T will be initialized first, and asserts that at least one exists. The new NotBefore<T> means that trait of type T will be initialized first, but allows no traits.
This allows traits to control initialization order for optional dependencies. They want to be initialized second so they can rely on the dependencies having been initialized. But if the dependencies are optional then to not throw if none are present.
We apply this to Locomotor which was previously using AddFrameEndTask to work around trait order initialization. This improves the user experience as the initialization is applied whilst the loading screen is still visible, rather than the game starting and creating jank by performing initialization on the first tick.
This helps improve the safety of code the uses reflection when methods may get renamed, and helps navigating code as the nameof will show up when searching for references to members.
1. If it follow the refinery placing logic, then we can use Facings in PlaceBuildingVariants to help BaseBuilderBotModule "rotates" it to minefield.
2. If it is a normal building, BaseBuilderBotModule will place a random variant actor.
PlayerReference colors in D2k missions only affect chat text and minimap colors because actors use specific palette colors.
So using the colors from the original game's minimap.
- Add a new widget type for input and extend it from other input widgets
- Add a new label type that can be linked to an input widget
- Change the label color when the input's disabled state changes
The `Refinery` trait has a hardcoded usage of `SpriteHarvesterDockSequence`, which requires the harvester to have `WithDockingAnimation`, making it inconvenient-at-best to NOT have a docking/unloading animation.
Actor previously cached targetable locations for static actors as an optimization. As we can no longer reference the IPositionable interface, move this optimization to HitShape instead. Although we lose some of the efficiency of caching the final result on the actor, we gain some by allowing HitShape to cache the results as long as they have not changed. So instead of being limited to static actors, we can extend the caching to currently stationary actor.
Aligns the naming conventions defined in editorconfig (dotnet_naming_style, dotnet_naming_symbols, dotnet_naming_rule) which are reported under the IDE1006 rule with the existing StyleCop rules from the SA13XX range.
This ensures the two rulesets agree when rejecting and accepting naming conventions within the IDE, with a few edges cases where only one ruleset can enforce the convention. IDE1006 allows use to specify a naming convention for type parameters, const locals and protected readonly fields which SA13XX cannot enforce. Some StyleCop SA13XX rules such as SA1309 'Field names should not begin with underscore' are not possible to enforce with the naming rules of IDE1006.
Therefore we enable the IDE1006 as a build time warning to enforce conventions and extend them. We disable SA13XX rules that can now be covered by IDE1006 to avoid double-reporting but leave the remaining SA13XX rules that cover additional cases enabled.
We also re-enable the SA1311 rule convention but enforce it via IDE1006, requiring some violations to be fixed or duplication of existing suppressions. Most violations fixes are trivial renames with the following exception. In ActorInitializer.cs, we prefer to make the fields private instead. ValueActorInit provides a publicly accessible property for access and OwnerInit provides a publicly accessible method. Health.cs is adjusted to access the property base instead when overriding. The reflection calls must be adjusted to target the base class specifically, as searching for a private field from the derived class will fail to locate it on the base class.
Unused suppressions were removed.
- When a path search is being performed the path search will not attempt route to inaccessible cells, so domain index checks to avoid inaccessible cells in the search predicate are redundant and can be removed.
- DomainIndex is a required world trait, so we don't need to use TraitOrDefault and therefore can avoid dealing with the null case.
The existing APIs surfaces for pathfinding are in a wonky shape. We rearrange various responsibilities to better locations and simplify some abstractions that aren't providing value.
- IPathSearch, BasePathSearch and PathSearch are combined into only PathSearch. Its role is now to run a search space over a graph, maintaining the open queue and evaluating the provided heuristic function. The builder-like methods (WithHeuristic, Reverse, FromPoint, etc) are removed in favour of optional parameters in static creation methods. This removes confusion between the builder-aspect and the search function itself. It also becomes responsible for applying the heuristic weight to the heuristic. This fixes an issue where an externally provided heuristic ignored the weighting adjustment, as previously the weight was baked into the default heuristic only.
- Reduce the IGraph interface to the concepts of nodes and edges. Make it non-generic as it is specifically for pathfinding, and rename to IPathGraph accordingly. This is sufficient for a PathSearch to perform a search over any given IGraph. The various customization options are concrete properties of PathGraph only.
- PathFinder does not need to deal with disposal of the search/graph, that is the caller's responsibility.
- Remove CustomBlock from PathGraph as it was unused.
- Remove FindUnitPathToRange as it was unused.
- Use PathFinder.NoPath as the single helper to represent no/empty paths.
- Split control groups management to its own interface
- Add hotkeys for selecting, creating, adding to and combining with control groups
- Add a ControlGroups widget to manage the player interaction
Also added a rule to silence StyleCop complaining about StaticReadonlyFieldsMustBeginWithUpperCaseLetter to match what we already have configured for the IDE.
Having this set to "none" disabled the IDE's option to add braces, whereas "silent" lets it do it on the user's request while still not suggesting it on its own.
Added optional padding to video frames because that's what VideoPlayerWidget expects.
Keeping the option to not use padding for other use-cases like converting frames to PNG.
Removed property backing fields where applicable, introduced C#7 syntax for properties.
Renamed a bunch of interface properties and class private members with more descriptive names.
Did some inconsequential reordering.
- The DEFAULT_PRIVATE_MODIFIER behaviour is now handled by the .editorconfig file via `dotnet_style_require_accessibility_modifiers = omit_if_default:warning`.
Also added `dotnet_diagnostic.IDE0040.severity = warning` there to raise compile-time errors in the CI.
- The field naming conventions seem to already be covered by (some) analyzer rules (checked in both VS and VSCode) - IDE1006/SA1306 and SA1307.
- Fixed a rounding issue in `WavReader.WaveLength()`.
- Fixed `AudReader.SoundLength()` not resetting the stream position.
- Fixed crashes caused by disposed streams because `LengthInSeconds` would try and calculate the length on the fly. It is now precalculated and cached (making it consistent across all 5 current `ISoundFormat` implementations).
- Fixed a crash in `AudReader.LoadSound()`'s `out Func<Stream> result` because that func would try and access the disposed stream's `Length` property. That works for `SegmentStream`, but not for `FileStream`.
- Fixed frameCount/soundLength label positioning in the AssetBrowser window to avoid text clipping .
Splitting them from one array into separate allows us to then reliably pick how each asset should be presented. Also lets us unhardcode some checks like "if file is .vxl ... else is sprite".
Each successive value of BlockedByActor is a superset of the previous value. Having a mixed up order of values in PathSearchOrder is not useful.
In the previous ordering, if a search for Immovable failed to find a path, it would then attempt Stationary. However Stationary is *more* restrictive then Immovable. If Immovable failed, there is no way Stationary could succeed. This means the search for Stationary is wasted effort.
In the fixed ordering, we try Stationary first. In the fixed ordering there are no pointless searches. Every search might succeed where the previous one failed and is therefore useful to try.
The path cache was originally a moderate benefit, but over time a couple of things have conspired against it:
- Only paths with BlockedByActor.None are cached. Originally all paths regardless of blocking were cached but this was deemed unacceptable due to potentially returning outdated paths as actors move about. Paths with BlockedByActor.None are only invalidated if terrain conditions change, which are rarer.
- Move will try and find a path 4 times, trying with a different BlockedByActor check each time. BlockedByActor.None is the last check and only reached if the other searches fail. This is a rare scenario.
Overall, this means the hit rate for the cache is almost non-existent. Given the constraints on path validity it seems unlikely that the hit rate could be improved significantly, therefore it seems reasonable to remove the cache entirely to remove the overhead of cache management.
Fixed code based on feedback
Replaced try/catch block with a null check and exception throw
Fixed code based on feedback
Fixed code based on feedback
Simplified Launch.Map parameter to use map name directly
- Remove functionality for tracking cells considered during the search, as nothing relies on this.
- Rename various parameters in the expand function to closer match naming of fields used in CellInfo, intended to improve clarity.
- Make Status the first field.
- Rename EstimatedTotal to EstimatedTotalCost to make it clearer it has the same unit as the CostSoFar field.
- Rename PreviousPos to PreviousNode as node terminology is a better match for usage.
Firstly, when dealing with maps with height discontinuities, the neighbouring cells we need to search are more that the set we need to search on flat maps. We ensure that as we traverse a map with varying height, we now consider cells "behind" us that may have become accessible due to a height change.
Secondly, when considering connections available via Custom Movement Layers, make sure the target cell on the new layer is actually enterable. Previously this cell would be reported as a valid connection, even if it wasn't actually possible to enter the cell as it was blocked. We also apply the same optimization of ignoring already closed cells.
- Stream lines in as memory rather than needing to realise a string for each line, via a new method in StreamExts.
- Use span to avoid string allocations during parsing until we want to realise the node itself, in MiniYaml.FromLines.
- Change several callsites to use the streaming extension method rather than string method where possible.
This is a follow-up to PR 19379, which aimed to "provide an easy debug option for VSCode developers", but only did so for non-Windows users. VSCode has been building in Release mode for ever and continued to do so on Windows after that PR.
Previously, some paths used a separator and some did not. This broke some de-duplication logic in ExternalMods which tried to enumerate distinct paths but would end up running logic on the same directory more than one as it was provided both with and without a trailing directory separator. By normalizing the path this logic now works.
As there are few custom movement layers, using an array is good for improving lookup speed. Additionally, we can simplify some code by reserving index 0 of the array for the ground layer. Code that needs to maintain a state for the ground layer and every custom movement layer can now maintain a flat array of state using index 0 for the ground layer, and the the ICustomMovementLayer.Index for the custom movement layer. This removes a lot of ternary statements checking for the ground layer special case.
This also removes a workaround that allowed the current mod to be
registered even if it defined a bogus path. Uses of Game.ExternalMods
should therefore always use TryGetValue and correctly handle it
returning false.
- Extract chat line templates and logic so they can be reused across widgets
- Make text notification styling entirely template driven (by removing chat color configuration and making color optional for `TextNotification`)
- Add a new TextNotificationsDisplay widget (based on and replacing ChatDisplayWidget)
- Add timestamp support to text notifications
In CellInfoLayerPool, instead of having to store a layer with the default values, we know we can just clear the pooled layer in order to reset it. This saves on memory, and also makes resetting marginally faster.
In PathSearch, we need to amend a check to ensure a cell info is not Unvisited before we check on its cost.
Unfortunately due to bugs in the analyzers or something else, the IDE0005 doesn't work as expected. The "officially suggested" workaround is to enable XML documentation generation to trigger IDE0005 during compiling. Then we need to add three more rules to silence the warnings that come from the XML documentation generation. We also need to enable code style enforcing on build for all of this to work.
Known issue is that all of this produces a bunch (tens to hundreds) of obscure analyzer warnings on older versions of Visual Studio, but those seem to not be causing issues.
- Rename CostForInvalidCell to PathCostForInvalidPath
- Add MovementCostForUnreachableCell
- Update usages of int.MaxValue and short.Maxvalue to use named constants where relevant.
- Update costs on ICustomMovementLayer to return short, for consistency with costs from Locomotor.
- Rename some methods to distinguish between path/movement cost.
- _http_ `->` _https_ where appropriate
- _wiki.openra.net_ `->` _https://github.com/OpenRA/OpenRA/wiki_
- _bugs.openra.net_ `->` _https://github.com/OpenRA/OpenRA/issues_
- the IRC channel is now on Libera.Chat + make it into a URI
This is in order to be consistent (one link to GitHub wiki already
exists), reflect reality, as well as avoid unnecessary redirects.
We observe that most cells within a map lie within a region where no matter their height, their projection would still remain in map bounds. We can utilise this to perform a fast check for such cells and skipping the expensive checks on their actual height. We only need to check the actual height of a cell if this could cause the projection to go out of bounds.
By making the constructor take non-optional parameters, this highlights some calls sites which were forgetting to set these values. These are now fixed.
Set the path debug to have a marker size of 2 for better visibility.
The ping/pong orders are replaced with a dedicated
(and much smaller) Ping packet that is handled
directly in the client and server Connection wrappers.
This allows clients to respond when the orders are
processed, instead of queuing the pong order to be
sent in the next frame (which added an extra 120ms
of unwanted latency).
The ping frequency has been raised to 1Hz, and pings
are now routed through the server events queue in
preparation for the future dynamic latency system.
The raw ping numbers are no longer sent to clients,
the server instead evaluates a single ConnectionQuality
value that in the future may be based on more than
just the ping times.
* EchoConnection is now a trivial buffer that stores
and repeats orders directly without serialization.
* NetworkConnection no longer subclasses EchoConnection,
and now also caches local orders without serialization.
* Replay recording was moved to NetworkConnection
(it is never used on EchoConnection).
Well, 05eb really, but until we implement campaign progression,
calling it 'c' is the easiest way to differentiate it without breaking
our mission title length/alignment conventions.
* TSVeinsRenderer now shows border cells on the radar
* BuildableTerrainLayer now uses the radar colors defined on the individual tiles
* CliffBackImpassabilityLayer no longer overrides the underlying terrain color.
This is less realistic, but better matches the original
game and is the only practical way to reduce visual issues
caused by long shadows being cast over multiple cells.
The trait documentation specified that the speed
and offset values are px/tick, but they have actually
always been treated as px/render.
Fix the update logic and rescale the map definitions
to account for the fixed behaviour.
Bibs and other effects that should be drawn at ground level
can now simply define ZRamp: 1, Offset: <X>,<Y>,1, avoiding the
need to account for the Y offset or internal sprite offsets.
This explicitly tells KDE to associate the OpenRA window with
the integrated appimage desktop file, allowing it to use the
correct resolution icon for the task switcher.
Forces the chat and performance panels to be re-initialized when a
player transitions to spectators. This ensures that spectators don't
get to see faction themed widgets.
In path finding GetConnections considers connections to already closed cells and calculates the cost for them. The caller afterwards ignores them. These are 15% of all connections.
If progress == Distance, we must not move again on the same tick,
but still 'return true' to avoid losing a tick in the case
when this is the last Move tick followed by a different activity
(or a new queued Move, for example via waypoints).
This was previously decreased to support legacy GPUs
that only supported 8 texture image units and we need
to reserve one of these for the palette texture.
OpenGL 3.X mandates a minimum of 16 (and most most GL2
cards also supported it) so we can now safely increase
this limit.
The previous asynchronous approach did not work particularly well,
leading to large janks when switching to custom maps or opening the
mission browser.
This commit introduces two key changes:
* Rule loading for WorldActorInfo and PlayerActorInfo is made
synchronous, in preparation for the next commit which will
significantly optimize this path.
* The full ruleset loading, which is required for map validation,
is moved to the server-side and managed by a new ServerMapStatusCache.
The previous syntax check is expanded to include the ability to run
lint tests.
As far as I could tell, this was the last place that still
used 'seconds' instead of ticks, apart from
some sound notification intervals (which are better
converted to real [milli]seconds).
Also renamed ScaredyCat.PanicLength to PanicDuration for
consistency and easier finding.
There were 2 issues at work here, both when progress
would overshoot Distance (which is the usual case,
rather than the exception):
- the overshot progress was passed on by MoveFirstHalf, however
OnComplete would make the next MovePart start ticking the
same tick on which the old MovePart reached Distance,
but move by carryoverProgress +(!!!) terrain speed instead of moving
by just the left-over carryoverProgress.
- MoveSecondHalf would not pass any overshot progress to the
next MoveFirstHalf queued by parent Move, leading to
the next MoveFirstHalf performing a full-speed move the same tick
MoveSecondHalf finished its last move.
InnerActivity and UpdateCenterLocation made this
overly complex and hard to read & debug.
This also fixes a bug that would make an outdated
facing being passed during OnComplete (because
InnerActivity was cached before UpdateCenterLocation
could set the correct final facing).
Both writing to perf.log frequently as well as GetTimestamp
aren't free and hurt performance particularly on slower systems
(which can have notably higher output to perf.log, further
amplifying the problem).
Therefore we make simulation perf logging opt-in.
Additionally, logging of the current tick and tick type
(local/net) is removed from debug.log, and some
remnant debug logging for kills and pips is removed
to keep performance-sensitive logging limited to
perf.log.
*Remove internal GameSpeed defaults
Enforce setting values explicitly all the time
Require definition of a DefaultSpeed
*Remove Global.Timestep default
*Remove the hacky Timestep/OrderLatency setting via LobbyInfo
*Fix shellmaps ignoring mod-defined gamespeeds
*Make DateTimeGlobal use the MapOptions gamespeed
A mere int comparison is obviously cheaper than
a comparison between two multiplications,
so pulling this above the checks of other damage states
allows us to bail early for undamaged actors.
Profiling has shown that filtering them out early is cheaper
than applying those percentage modifiers anyway.
Additionally, profiling shows foreach loops to be cheaper
than LINQ here as well.
* ResourceType trait has been removed.
* Simulation-related data is now defined on the
ResourceLayer (which mods can subclass/replace).
* Support non-money resources by moving the resource
values to the PlayerResources trait.
* Allow mods to disable the neighbour density override
and instead always use the map-defined densities.
* Allow mods to define their own resource placement
logic (e.g. allow resources on slopes) by subclassing
(Editor)ResourceLayer.
* Improve ability to subclass/override ResourceRenderer
by exposing more virtual methods.
While they may be only 'visual' in terms of influence/cell grid,
they all do update CenterPosition, which is essentially the
actual world position of the actor.
'Visual' would imply that it only affects the position where the
actor is drawn, which is inaccurate.
Furthermore, using the term 'Visual' here would make
naming future methods/properties related to visual interpolation
unnecessarily complicated, because that's where we might
need a real 'Visual(Only)Position'.
Since it's music it's optional, so the players still won't know anything happened, but if they do open the content manager they will have a chance for an automatic installation.
Alpha can specify a single value for the sequence
or values for each frame in the sequence.
AlphaFade: True can be specified to linearly fade
to transparent over the length of the animation.
Unlike its snow counterpart, the transition between building
and its shadow was jarring, making it look as if there was a
1-pixel gap between them. Fixed that.
Also applied some very subtle polish by replacing some non-
remapable red pixels (nothing worth porting to snow variant, though).
A few small improvements:
* The type `desktop` was renamed to `desktop-appliaction` (a long time ago)
* Add `launchable` to tell how the application can be launched
* Use `https` for homepage link
* Add link to the bugtracker
* Update `oars-1.0` to `oars-1.1`
* Remove all unnecessary `content_attribute` entries with value `none`
The `VERSION` variable doesn't work with a release tarball, because it requires git to be set correctly.
Instead this reads the version from the `VERSION` file.
ClientWithIndex may rarely be null, causing a crash.
In any case we do want to report these changes to the first client, as
somebody else may have changed the settings and left.
Removes it from destroyed fields, trees and rocks in RA and C&C.
This trait isn't free in terms of performance due to all the
INotify* calls, and the mentioned actors should rarely
- if ever - need it. Besides, if a mission ever really needs
this on those actors, the trait can be added back via
map rules.
Fix the Building's SoundOnDamageTransition DamagedSounds with the correct one and add a second sound file to DestroyedSounds for more variation like in the original Tiberian Dawn.
Our SpriteFrameType names refer to the byte channel order rather than
the bit order, meaning that SpriteFrameType.BGRA corresponds to the
standard Color.ToArgb() etc byte order when the (little-endian) integer
is read as 4 individual bytes.
The previous code did not account for the fact that non-indexed Png
uses big-endian storage for its RGBA colours, and that SheetBuilder
had the color channels incorrectly swapped to match and cancel this out.
New SpriteFrameType enums are introduced to distinguish between BGRA
(little-endian) and RGBA (big-endian) formats, and also for 24bit data
without alpha. The channel swizzling / alpha creation is now handled
when copying into the texture atlas, removing the need for non-png
ISpriteLoader implementations to allocate an additional temporary array
and reorder the channels during load.
Do not call SkipDoneActivities method recursively via the
NextActivity property. Rather use the nextActivity member.
Avoiding additional function calls and a recursively
growing stack.
Do not call ChildActivity and NextActivity properties
twice in a row. Once to test for null and after to access
it's value. It will cause the complete list of activities
to be traversed twice looking for non done activities.
Replace Queue method with a version that does not the
NextActivity property causing an extra call to
SkipDoneActivities. Avoid calling Queue recursively.
Similar replace QueueChild with a version that does
not call additional methods.
Note that ActivitiesImplementing returns only non
done activities. The method name does not suggest this.
Please consider making NextActivity a method to cleary indicate it
involves the logic of skipping Done activities. To let
the called know it is 'expensive'.
Please consider renaming the protected property ChildActivity to
FirstChildActivityNotDone to avoid it being used as childActivity.
Please consider maintaining a pointer to the first
non done activity. This avoids the need the each time find it.
@@ -13,4 +13,4 @@ You can help speed up the review process by following a few steps:
* Respond to review comments as soon as you reasonably can. Reviewers will usually prioritize Pull Requests that are still fresh in their minds. Make sure to leave a comment when you push new changes, otherwise GitHub does not automatically notify reviewers!
* Leave a polite comment asking for reviews if a week or more has passed without feedback.
If you need any help you can ask in the #openra IRC channel on freenode (most active during European evenings).
If you need any help you can ask on Discord (https://discord.openra.net) or in the #openra IRC channel on Libera (not as active as Discord).
Compiling OpenRA requires the following dependencies:
* [Windows PowerShell >= 4.0](http://microsoft.com/powershell) (included by default in recent Windows 10 versions)
* [.NET Framework 4.7.2 (Developer Pack)](https://dotnet.microsoft.com/download/dotnet-framework/net472) (or via Visual Studio 2017)
* [.NET Core 2.2 SDK](https://dotnet.microsoft.com/download/dotnet-core/2.2) (or via Visual Studio 2017)
* [.NET 6 SDK](https://dotnet.microsoft.com/download/dotnet/6.0) (or via Visual Studio)
To compile OpenRA, open the `OpenRA.sln` solution in the main folder, build it from the command-line with MSBuild or use the Makefile analogue command `make all` scripted in PowerShell syntax.
To compile OpenRA, open the `OpenRA.sln` solution in the main folder, build it from the command-line with `dotnet` or use the Makefile analogue command `make all` scripted in PowerShell syntax.
Run the game with `launch-game.cmd`. It can be handed arguments that specify the exact mod one wishes to run, for example, run `launch-game.cmd Game.Mod=ra` to launch Red Alert, `launch-game.cmd Game.Mod=cnc` to start Tiberian dawn or `launch-game.cmd Game.Mod=d2k` to launch Dune 2000.
Linux
=====
Mono, version 5.18 or later, is required to compile OpenRA. You can add the [upstream mono repository](https://www.mono-project.com/download/stable/#download-lin) for your distro to obtain the latest version if your system packages are not sufficient.
.NET 6 or Mono (version 6.12 or later) is required to compile OpenRA. We recommend using .NET 6 when possible, as Mono is poorly packaged by most Linux distributions (e.g. missing the required `msbuild` toolchain), and has been deprecated as a standalone project.
To compile OpenRA, run `make` from the command line. After this one can run the game with `./launch-game.sh`. It is also possible to specify the mod you wish to run from the command line, e.g. with `./launch-game.sh Game.Mod=ts` if you wish to try the experimental Tiberian Sun mod.
The [.NET 6 download page](https://dotnet.microsoft.com/download/dotnet/6.0) provides repositories for various package managers and binary releases for several architectures. If you prefer to use Mono, we suggest adding the [upstream repository](https://www.mono-project.com/download/stable/#download-lin) for your distro to obtain the latest version and the `msbuild` toolchain.
To compile OpenRA, run `make` from the command line (or `make RUNTIME=mono` if using Mono). After this one can run the game with `./launch-game.sh`. It is also possible to specify the mod you wish to run from the command line, e.g. with `./launch-game.sh Game.Mod=ts` if you wish to try the experimental Tiberian Sun mod.
The default behaviour on the x86_64 architecture is to download several pre-compiled native libraries using the Nuget packaging manager. If you prefer to use system libraries, compile instead using `make TARGETPLATFORM=unix-generic`.
If you choose to use system libraries, or your system is not x86_64, you will need to install the following using your system package manager:
If you choose to use system libraries, or your system is not x86_64, you will need to install [SDL 2](https://www.libsdl.org/download-2.0.php), [FreeType](http://gnuwin32.sourceforge.net/packages/freetype.htm), [OpenAL](https://openal-soft.org/), and [liblua 5.1](http://luabinaries.sourceforge.net/download.html) before compiling OpenRA.
Type `sudo make install` for system-wide installation. Run `sudo make install-linux-shortcuts` to get startup scripts, icons and desktop files. You can then run the Red Alert by executing the `openra-ra` command, the Dune 2000 mod by running the `openra-d2k` command and Tiberian Dawn by the `openra-cnc` command. Alternatively, you can also run these mods by clicking on their desktop shortcuts if you ran `sudo make install-linux-shortcuts`.
These can be installed using your package manager on various distros:
Arch Linux
----------
It is important to note there is an unofficial [`openra-git`](https://aur.archlinux.org/packages/openra-git) package in the Arch User Repository (AUR) of Arch Linux. If manually compiling is the way you wish to go the build and runtime dependencies can be installed with:
Type `sudo make install` for system-wide installation. Run `sudo make install-linux-shortcuts` to get startup scripts, icons and desktop files. You can then run the Red Alert by executing the `openra-ra` command, the Dune 2000 mod by running the `openra-d2k` command and Tiberian Dawn by the `openra-cnc` command. Alternatively, you can also run these mods by clicking on their desktop shortcuts if you ran `sudo make install-linux-shortcuts`.
macOS
=====
Before compiling OpenRA you must install the following dependencies:
To compile OpenRA, run `make` from the command line. Run with `./launch-game.sh`.
The default behaviour is to download several pre-compiled native libraries using the Nuget packaging manager. If you prefer to use system libraries, compile instead using `make TARGETPLATFORM=unix-generic`. If you choose to use system libraries you will need to install:
[.NET 6](https://dotnet.microsoft.com/download/dotnet/6.0) or [Mono](https://www.mono-project.com/download/stable/#download-mac) (version 6.12 or later) is required to compile OpenRA. We recommend using .NET 6 unless you are running a very old version of macOS (10.9 through 10.14).
To compile OpenRA, run `make` from the command line (or `make RUNTIME=mono` if using Mono). Run with `./launch-game.sh`.
/// Called every tick to run activity logic. Returns false if the activity should
/// remain active, or true if it is complete. Cancelled activities must ensure they
/// return the actor to a consistent state before returning true.
///
/// </para>
/// <para>
/// Child activities can be queued using QueueChild, and these will be ticked
/// instead of the parent while they are active. Activities that need to run logic
/// in parallel with child activities should set ChildHasPriority to false and
/// manually call TickChildren.
///
/// </para>
/// <para>
/// Queuing one or more child activities and returning true is valid, and causes
/// the activity to be completed immediately (without ticking again) once the
/// children have completed.
/// </para>
/// </summary>
publicvirtualboolTick(Actorself)
{
@@ -206,25 +211,26 @@ namespace OpenRA.Activities
publicvoidQueue(Activityactivity)
{
if(NextActivity!=null)
NextActivity.Queue(activity);
else
NextActivity=activity;
var it=this;
while(it.nextActivity!=null)
it=it.nextActivity;
it.nextActivity=activity;
}
publicvoidQueueChild(Activityactivity)
{
if(ChildActivity!=null)
ChildActivity.Queue(activity);
if(childActivity!=null)
childActivity.Queue(activity);
else
ChildActivity=activity;
childActivity=activity;
}
/// <summary>
/// Prints the activity tree, starting from the top or optionally from a given origin.
///
/// <para>Prints the activity tree, starting from the top or optionally from a given origin.</para>
/// <para>
/// Call this method from any place that's called during a tick, such as the Tick() method itself or
/// the Before(First|Last)Run() methods. The origin activity will be marked in the output.
/// </para>
/// </summary>
/// <param name="self">The actor performing this activity.</param>
/// <param name="origin">Activity from which to start traversing, and which to mark. If null, mark the calling activity, and start traversal from the top.</param>
thrownewLuaException("Attempted to call CPos.Add(CPos, CVec) with invalid arguments ({0}, {1})".F(left.WrappedClrType().Name,right.WrappedClrType().Name));
thrownewLuaException($"Attempted to call CPos.Add(CPos, CVec) with invalid arguments ({left.WrappedClrType().Name}, {right.WrappedClrType().Name})");
returnnewLuaCustomClrObject(a+b);
}
@@ -99,7 +105,7 @@ namespace OpenRA
{
varrightType=right.WrappedClrType();
if(!left.TryGetClrValue(outCPosa))
thrownewLuaException("Attempted to call CPos.Subtract(CPos, (CPos|CVec)) with invalid arguments ({0}, {1})".F(left.WrappedClrType().Name, rightType.Name));
thrownewLuaException($"Attempted to call CPos.Subtract(CPos, (CPos|CVec)) with invalid arguments ({left.WrappedClrType().Name}, {rightType.Name})");
if(rightType==typeof(CPos))
{
@@ -112,7 +118,7 @@ namespace OpenRA
returnnewLuaCustomClrObject(a-b);
}
thrownewLuaException("Attempted to call CPos.Subtract(CPos, (CPos|CVec)) with invalid arguments ({0}, {1})".F(left.WrappedClrType().Name, rightType.Name));
thrownewLuaException($"Attempted to call CPos.Subtract(CPos, (CPos|CVec)) with invalid arguments ({left.WrappedClrType().Name}, {rightType.Name})");
thrownewLuaException("Attempted to call CVec.Add(CVec, CVec) with invalid arguments ({0}, {1})".F(left.WrappedClrType().Name,right.WrappedClrType().Name));
thrownewLuaException($"Attempted to call CVec.Add(CVec, CVec) with invalid arguments ({left.WrappedClrType().Name}, {right.WrappedClrType().Name})");
thrownewLuaException("Attempted to call CVec.Subtract(CVec, CVec) with invalid arguments ({0}, {1})".F(left.WrappedClrType().Name,right.WrappedClrType().Name));
thrownewLuaException($"Attempted to call CVec.Subtract(CVec, CVec) with invalid arguments ({left.WrappedClrType().Name}, {right.WrappedClrType().Name})");
returnnewLuaCustomClrObject(a-b);
}
@@ -110,14 +110,11 @@ namespace OpenRA
{
case"X":returnX;
case"Y":returnY;
default:thrownewLuaException("CVec does not define a member '{0}'".F(key));
default:thrownewLuaException($"CVec does not define a member '{key}'");
}
}
set
{
thrownewLuaException("CVec is read-only. Use CVec.New to create a new value");
}
set=>thrownewLuaException("CVec is read-only. Use CVec.New to create a new value");
Log.Write("debug","Failed to generate keypair with exception:");
Log.Write("debug",e);
Console.WriteLine("Key generation failed:");
Console.WriteLine(e);
innerState=LinkState.Uninitialized;
}
@@ -154,12 +158,14 @@ namespace OpenRA
}
catch(Exceptione)
{
Log.Write("debug","Failed to delete keypair with exception: {1}",e);
Console.WriteLine("Key deletion failed: {0}",e);
Log.Write("debug","Failed to delete keypair with exception:");
Log.Write("debug",e);
Console.WriteLine("Key deletion failed:");
Console.WriteLine(e);
}
innerState=LinkState.Uninitialized;
parameters=default(RSAParameters);
parameters=default;
innerFingerprint=null;
innerData=null;
}
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.