diff --git a/CHANGELOG.md b/CHANGELOG.md index 9eef754..1c4652d 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -14,6 +14,7 @@ and adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html). - C++ forward declarations no longer crowd out the real class definition. A `class Foo;` forward declaration — common in large C++ and Unreal Engine codebases, where a heavily used class is forward-declared across dozens of headers — was indexed as its own class node every time it appeared. So exploring that class returned mostly forward-declaration sites, and could even pick one of them as the representative for blast-radius, burying the actual definition and its members and callers. Bodiless forward declarations are now skipped for C and C++, exactly as forward-declared structs and enums already were, so only the real definition is indexed. Languages where a class with no body is a complete definition — such as Kotlin's `class Empty` and Scala — are unaffected. Thanks @luoyxy for the report and root-cause analysis. (#1093) - C++ methods that return a reference, and user-defined conversion operators, are now indexed under their correct names. An inline getter like `const FGameplayTagContainer& GetActiveTags() const` — everywhere in Unreal Engine headers — was indexed as `& GetActiveTags() const` instead of `GetActiveTags`, and a conversion operator like `operator EALSMovementState() const` kept its trailing `() const` instead of reading `operator EALSMovementState`. In both cases the garbled name meant you couldn't find the symbol by name and its callers weren't linked. Both now read cleanly, matching how pointer-returning and value-returning methods already worked. (#1096) - C++ functions written with an inline-specifier macro before the return type are now indexed correctly. In Unreal Engine, inline helpers are commonly written `FORCEINLINE FString GetEnumerationToString(...)`; the `FORCEINLINE` macro made the parser read the return type as part of the function's name (`FString GetEnumerationToString` instead of `GetEnumerationToString`) and lose the real return type, so the function couldn't be found by name and its callers weren't linked. CodeGraph now recognizes the standard Unreal inline macros (`FORCEINLINE`, `FORCENOINLINE`, `FORCEINLINE_DEBUGGABLE`), so both the name and the return type are captured. (#1100) +- The same function-name recovery now covers inline macros from common third-party C++ libraries, not just Unreal Engine — including pugixml (`PUGI__FN`, `PUGIXML_FUNCTION`), Godot (`_FORCE_INLINE_`), Boost (`BOOST_FORCEINLINE`), and generic `ALWAYS_INLINE` / `FORCE_INLINE`. Functions decorated with these are now indexed under their real names. On a large Unreal project vendoring these libraries this cleaned up the large majority of remaining function-name garbling. (#1101) ## [1.1.6] - 2026-06-30 diff --git a/__tests__/extraction.test.ts b/__tests__/extraction.test.ts index 9f76acb..e2e7f03 100644 --- a/__tests__/extraction.test.ts +++ b/__tests__/extraction.test.ts @@ -2963,6 +2963,19 @@ class APXCharacter { // the one real definition expect(infoOf('static FORCEINLINE const FString& GetRef(int V) { return H(V); }').map((x) => x.name)).toContain('GetRef'); }); + it('handles common third-party inline macros (pugixml, Godot, Boost, generic)', () => { + // pugixml: PUGI__FN before the return type; PUGIXML_FUNCTION (linkage) + // between the return type and the name — both recovered. + expect(infoOf('PUGI__FN void* default_allocate(size_t n) { return H(n); }').map((x) => x.name)).toContain('default_allocate'); + expect(infoOf('PUGI__FN_NO_INLINE bool strequal(const char_t* a) { return H(a); }').map((x) => x.name)).toContain('strequal'); + expect(infoOf('std::string PUGIXML_FUNCTION as_utf8(const wchar_t* s) { return H(s); }').map((x) => x.name)).toContain('as_utf8'); + // Godot / Boost / generic inline hints + expect(infoOf('_FORCE_INLINE_ String get_name() const { return H(); }').map((x) => x.name)).toContain('get_name'); + expect(infoOf('_ALWAYS_INLINE_ Vector2 get_pos() { return H(); }').map((x) => x.name)).toContain('get_pos'); + expect(infoOf('BOOST_FORCEINLINE result_type call() { return H(); }').map((x) => x.name)).toContain('call'); + expect(infoOf('ALWAYS_INLINE MyType compute() { return H(); }').map((x) => x.name)).toContain('compute'); + }); + it('leaves ordinary functions and real all-caps return types untouched (controls)', () => { expect(infoOf('FString GetName(int V) { return H(V); }')).toEqual([{ name: 'GetName', ret: 'FString' }]); // A real all-caps type that is NOT a listed inline macro stays the return type. diff --git a/src/extraction/languages/c-cpp.ts b/src/extraction/languages/c-cpp.ts index 35db1c0..8aae0de 100644 --- a/src/extraction/languages/c-cpp.ts +++ b/src/extraction/languages/c-cpp.ts @@ -247,27 +247,45 @@ export function blankCppExportMacros(source: string): string { * becomes the return type and, for a non-primitive return, the return type gets * glued onto the name (`GetName` → `"FString GetName"`), so the function can't * be found by name and its callers don't link. This is pervasive in Unreal - * Engine, where inline helpers are written `FORCEINLINE (…)`. - * Replacing the macro with equal-length spaces preserves every byte offset (so - * line/column stay exact) and the declaration then parses as an ordinary - * function — recovering the real name AND the return type — mirroring how - * `blankCppExportMacros` recovers macro-annotated classes (#946/#1061). + * Engine (`FORCEINLINE (…)`) and in vendored third-party libraries + * that define their own inline macro (pugixml's `PUGI__FN`, Godot's + * `_FORCE_INLINE_`, Boost's `BOOST_FORCEINLINE`, …). Replacing the macro with + * equal-length spaces preserves every byte offset (so line/column stay exact) + * and the declaration then parses as an ordinary function — recovering the real + * name AND the return type — mirroring how `blankCppExportMacros` recovers + * macro-annotated classes (#946/#1061). * * Matched tightly so it can't touch an ordinary identifier: only the exact, - * well-known UE inline specifiers, and only in specifier position — immediately - * followed by whitespace and the identifier that starts the return type or name. - * That lookahead leaves value/expression uses (`x = FORCEINLINE ? …`), string - * literals, and `FORCEINLINE_SOMETHINGELSE` (word-boundary) alone. To cover a - * new codebase's inline macro, add its exact token here. + * curated inline-specifier tokens below (never an arbitrary all-caps token, so a + * real return type like `HRESULT DoIt()` is untouched), and only in specifier + * position — immediately followed by whitespace and the identifier that starts + * the return type or name. That lookahead leaves value/expression uses + * (`x = FORCEINLINE ? …`), string literals, and longer words + * (`FORCEINLINE_SOMETHINGELSE`, word-boundary) alone. To cover a new codebase's + * inline macro, add its exact token to the list. */ -const CPP_INLINE_MACROS = ['FORCEINLINE_DEBUGGABLE', 'FORCENOINLINE', 'FORCEINLINE'] as const; +const CPP_INLINE_MACROS = [ + // Unreal Engine + 'FORCEINLINE_DEBUGGABLE', 'FORCENOINLINE', 'FORCEINLINE', + // pugixml (ubiquitous vendored XML parser): `#define PUGI__FN inline` before + // the return type, plus `PUGIXML_FUNCTION` (linkage macro) between the return + // type and the name — the blank mechanism handles both positions. + 'PUGI__FN_NO_INLINE', 'PUGI__FN', 'PUGIXML_FUNCTION', + // Godot + '_ALWAYS_INLINE_', '_FORCE_INLINE_', + // Boost + 'BOOST_FORCEINLINE', 'BOOST_NOINLINE', + // Common cross-ecosystem inline/attribute hints + 'ALWAYS_INLINE', 'FORCE_INLINE', 'NOINLINE', +] as const; +// One alternation, longest token first so a longer macro wins over a prefix. +const CPP_INLINE_MACRO_RE = new RegExp( + `\\b(${[...CPP_INLINE_MACROS].sort((a, b) => b.length - a.length).join('|')})\\b(?=\\s+[A-Za-z_])`, + 'g' +); export function blankCppInlineMacros(source: string): string { if (!CPP_INLINE_MACROS.some((m) => source.indexOf(m) !== -1)) return source; - return source.replace( - // `FORCEINLINE_DEBUGGABLE` before `FORCEINLINE` so the longer token wins. - /\b(FORCEINLINE_DEBUGGABLE|FORCENOINLINE|FORCEINLINE)\b(?=\s+[A-Za-z_])/g, - (_m, macro) => ' '.repeat(macro.length) - ); + return source.replace(CPP_INLINE_MACRO_RE, (m) => ' '.repeat(m.length)); } /** C/C++ source pre-processing before tree-sitter: recover both macro-annotated