Since these will live for the lifetime of the process, it is worth freezing them on creation to improve lookup performance. If the dictionaries were iterated, change to an immutable array instead to preserve the order.
As config is often meant to be loaded and then never modified, it is helpful to support collections that are read-only after creation. This allows various classes that load config from YAML and elsewhere to deserialize straight into read-only collections and enforce invariants around data mutation.
Test coverage for these classes will help prevent regressions. Major breaking changes are avoided since config files may already rely on various aspects of the behaviour, but some small breaking changes are made:
- Use `value.Split(Comma, StringSplitOptions.RemoveEmptyEntries | StringSplitOptions.TrimEntries)` consistently for all comma-separated parsing.
- Rename LoadField to LoadFieldOrProperty for clarity - since this is the one method that cares about properties in a class otherwise focused only on fields.
- Use TryParse instead of Parse and ensure to call InvalidValueAction from all parsers so we throw the intended exception when badly formatted data is encountered.
- BooleanExpression/IntegerExpression exception message updated to align with the generic one.
- Remove FromYamlKey, and update the only user SkirmishLogic to perform this logic manually instead.
- Remove unused includePrivateByDefault parameter on GetTypeLoadInfo.
- Remove unused AllowEmptyEntriesAttribute.
This rule recommends use of primary constructors. Apply the suggestions on simple POCOs, adjusting some to readonly/structs/records at the same time. For more complex classes, the use of primary constructors is more distracting than helpful, so silence the rule. IDEs can still offer fixes for using primary constructors, but it will not show up as a build issue.
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.
- 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.
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.
- Rename Bits<T> to BitSet<T>.
- Implement set based helpers for BitSet<T>.
- When representing TargetTypes of ITargetable in various traits, use a BitSet<TargetableType> instead of HashSet<string> for better performance & reduced memory usage.
- Fix FieldLoader to trim input values when generating a BitSet<T>.
- Require T in BitSet<T> and BitSetAllocator<T> to be a class since it's just a marker value. This allows the JIT to instantiate generic code for these classes once, as they don't benefit from specialized code for T. (Typically JITs will generate shared code for all reference types, and unique code for every value type encountered).
Compiling these expressions is sadly expensive, and we needed new ones for every trait on every actor each time one was generated. The expressions thankfully can be shared as they are pure functions, which removes this overhead.
