Skip to content

API-Reference.md

Latest commit

 

History

History
64 lines (44 loc) · 3.43 KB

API-Reference.md

File metadata and controls

64 lines (44 loc) · 3.43 KB

API Reference

A complete, auto-generated reference to every symbol, file, and folder in the SparkEngine codebase — no third-party tool required. Everything on this page is produced by pure-shell scripts under docs/, consumes only source files in the repository, and renders natively on GitHub.

How to regenerate: docs/update-all-docs.sh (or individually, e.g. docs/generate-symbol-index.sh generate).

Master indexes

Page Contents
Symbol Index Every class, struct, enum, function, method, macro, and type alias — alphabetical, with file-and-line source links.
Function Index Every free function and out-of-line method definition (headers + .cpp).
Class & Struct Index Every declared class and struct in the codebase.
Enum Index Every enum with its source location.
Macro & Alias Index Every #define (excluding header guards) plus typedef/using aliases.
File Tree Hierarchical listing of every source file with LOC and @brief. Includes a Mermaid module-dependency graph.
Class Hierarchy Mermaid classDiagram per top-level module showing public inheritance.

Quick-reference indexes (scoped)

Page Contents
Entity Component System ECS components and systems overview.
Architecture Overview High-level engine architecture.
Codebase Statistics LOC, file counts, subsystem metrics.
Engine Architecture Flowchart ASCII/Mermaid flowcharts of subsystems.

Per-header detail pages

Running docs/generate-api-docs.sh generate also produces ~750 per-header Markdown pages under docs/api/ (one per .h/.hpp). These are not committed — regenerate locally whenever you want the deeper view. Each page lists classes, enums, free functions, macros, and type aliases in that header, each with a clickable source-line anchor.

What gets scanned

The generator scans every .h, .hpp, and .cpp under:

  • SparkEngine/Source/
  • SparkEditor/Source/
  • SparkConsole/src/
  • SparkShaderCompiler/src/
  • SparkSDK/
  • Tests/
  • GameModules/*/Source/ (all game modules auto-discovered)

ThirdParty/ and build directories are excluded.

What gets extracted

For every source file:

  • Classes & structs — name, base class (when : public X), surrounding @brief comment, source line.
  • Enums — name, source line.
  • Functions & methods — free functions, static functions, and out-of-line method definitions (Class::Method) in .cpp. Each gets a signature and a clickable file:line anchor.
  • Macros — every #define, excluding header guards.
  • Type aliases — every using X = … and typedef … X;.

All symbols are also emitted to docs/api/.symbols.tsv for programmatic use.

Why not Doxygen?

The SparkEngine documentation pipeline is intentionally dependency-free: the generator scripts use only find, grep, awk, sed, sort, and wc — tools present on every developer machine and every CI runner. The legacy docs/Doxyfile.txt and docs/generate-docs.sh remain available for anyone who wants full HTML with GraphViz, but nothing on this page needs them.