diff --git a/internal/cbm/doclink_go.c b/internal/cbm/doclink_go.c index 97dab618d..aeb0c5858 100644 --- a/internal/cbm/doclink_go.c +++ b/internal/cbm/doclink_go.c @@ -43,10 +43,10 @@ * drops these lines and so carries no line number * at all. * G path go.mod: the module path - * w says who keeps the node. A Go method node has a FLAT qualified name - * (.), so a function, the methods of every receiver, a type and - * a value of one name all claim the node ., and within a file the - * declaration on the LAST line keeps it (cbm_gbuf_upsert_node). w is 1 for + * w says who keeps the node .. A function, a type and a value of + * one name all claim it, and within a file the declaration on the LAST line + * keeps it (cbm_gbuf_upsert_node). A method claims nothing: its node is + * receiver-qualified (.., #1913), and its w is `-`. w is 1 for * that declaration, 0 for the others of that name, x when two of them start * on that last line, and - for a value the extractor gives no node at all * (it gives one to the first name of a constant spec and of a parenthesis- @@ -1511,9 +1511,11 @@ static void go_scan_method(go_scan_t *s, TSNode m) { recv = "?"; recv_len = SKIP_ONE; } - go_push_claim(s, - cbm_arena_sprintf(s->tmp, "M\t%.*s\t%.*s", (int)recv_len, recv, (int)len, name), - name, len, m); + /* #1913: a Go method node is receiver-qualified (..), + * so a method claims no flat name: w is `-`, and the method is found + * through its receiver (the member table), never through .. */ + (void)go_push_rec( + s, cbm_arena_sprintf(s->tmp, "M\t%.*s\t%.*s\t-", (int)recv_len, recv, (int)len, name)); } /* The implied name of an embedded field: the type's own name, without diff --git a/internal/cbm/extract_calls.c b/internal/cbm/extract_calls.c index 2f512e5c7..4eb02659c 100644 --- a/internal/cbm/extract_calls.c +++ b/internal/cbm/extract_calls.c @@ -4493,6 +4493,24 @@ CBMInvocationDescriptor handle_calls(CBMExtractCtx *ctx, TSNode node, const CBML } } } + // Go receiver-aware guard (same direction as the TS/JS flag above). + // Flag a selector call x.foo(). The Go AST cannot separate a method + // call on a value from a package-qualified call — but every selector + // call the Go LSP or the import/qualified registry strategies CAN + // place never reaches the weak short-name guards, so the flag only + // bites on unresolvable receivers (`f.Close()` on an os.File, + // `sha256.New()` behind an unindexed import), where a project-wide + // short-name match fabricates an edge to an unrelated project + // symbol sharing the name. Bare calls (helper()) keep + // is_method=false and resolve same-module/import paths as before. + if (ctx->language == CBM_LANG_GO && + strcmp(ts_node_type(node), "call_expression") == 0) { + TSNode gofn = ts_node_child_by_field_name(node, TS_FIELD("function")); + if (!ts_node_is_null(gofn) && + strcmp(ts_node_type(gofn), "selector_expression") == 0) { + call.is_method = true; + } + } TSNode args = ts_node_child_by_field_name(node, TS_FIELD("arguments")); /* tree-sitter-elixir attaches NO field name to a call's arguments diff --git a/internal/cbm/extract_defs.c b/internal/cbm/extract_defs.c index 6491a677f..eb654cf33 100644 --- a/internal/cbm/extract_defs.c +++ b/internal/cbm/extract_defs.c @@ -5453,7 +5453,7 @@ static void set_def_complexity(CBMDefinition *def, TSNode body, const CBMLangSpe * Walks to the parameter_declaration's `type` field, unwrapping pointer_type * and generic_type, and returns the type_identifier text (e.g. "OrderService"). * Returns NULL if no type_identifier is found. */ -static char *go_receiver_type_name(CBMArena *a, TSNode recv, const char *source) { +char *cbm_go_receiver_type_name(CBMArena *a, TSNode recv, const char *source) { uint32_t nc = ts_node_child_count(recv); for (uint32_t i = 0; i < nc; i++) { TSNode child = ts_node_child(recv, i); @@ -6687,12 +6687,27 @@ static void extract_func_def(CBMExtractCtx *ctx, TSNode node, const CBMLangSpec * (and downstream Go IMPLEMENTS/OVERRIDE) link the method to its owning * struct/type node. The parent QN must match the type's node QN, which * is computed the same way (cbm_fqn_compute on the type name). */ - char *recv_type = go_receiver_type_name(a, recv, ctx->source); + char *recv_type = cbm_go_receiver_type_name(a, recv, ctx->source); if (recv_type && recv_type[0]) { /* Must match the Go type node QN (directory-based module) so the * DEFINES_METHOD edge links the method to its owning type. */ def.parent_class = cbm_fqn_compute_source_lang(a, ctx->project, ctx->rel_path, recv_type, ctx->language); + /* Receiver-qualify the method QN (proj.pkg.Recv.method) — same + * shape as the C++ out-of-line path below and Go interface + * members. With the flat proj.pkg.method QN every same-name + * method in a package collided in the graph upsert: one body + * survived and the twins' call edges accreted onto it. The + * call-scope side (compute_func_qn in extract_unified.c) mirrors + * this formula, and go_lsp consumers read the def QN and + * parent_class (receiver_type) verbatim, so resolution joins + * stay exact. Go only: templ's Go-derived grammar reaches this + * branch too, but its call-scope QN (compute_func_qn) stays flat, + * and a def QN the calls do not share would orphan every call + * inside a templ method. */ + if (ctx->language == CBM_LANG_GO && def.parent_class) { + def.qualified_name = cbm_arena_sprintf(a, "%s.%s", def.parent_class, name); + } } } diff --git a/internal/cbm/extract_unified.c b/internal/cbm/extract_unified.c index aa3db1e0d..fbb7e12c3 100644 --- a/internal/cbm/extract_unified.c +++ b/internal/cbm/extract_unified.c @@ -1110,6 +1110,29 @@ static const char *compute_func_qn(CBMExtractCtx *ctx, TSNode node, const CBMLan } } + /* Go method `func (s *Storage) Close() {...}`: the def extractor records + * this as Method "proj.pkg.Storage.Close" (receiver-qualified, mirroring + * the C++ out-of-line rule above). The call-scope QN must match — a bare + * "proj.pkg.Close" names a node that no longer exists, so every call + * inside the method body would fall back to File-node attribution + * (calls_find_source). Same ONE-formula contract as the def side: + * cbm_go_receiver_type_name + cbm_fqn_compute_source_lang. */ + if (ctx->language == CBM_LANG_GO && strcmp(ts_node_type(node), "method_declaration") == 0) { + TSNode recv = ts_node_child_by_field_name(node, TS_FIELD("receiver")); + if (!ts_node_is_null(recv)) { + char *recv_type = cbm_go_receiver_type_name(ctx->arena, recv, ctx->source); + if (recv_type && recv_type[0]) { + const char *type_qn = cbm_fqn_compute_source_lang( + ctx->arena, ctx->project, ctx->rel_path, recv_type, ctx->language); + /* NULL only on allocation failure: fall through to the flat + * QN, as the def side does (extract_defs.c). */ + if (type_qn) { + return cbm_arena_sprintf(ctx->arena, "%s.%s", type_qn, name); + } + } + } + } + /* Nix: a binding's own attrpath contributes scope (`a.b.fn = …`), and the def * extractor bakes it into the def QN. Compose it identically here — otherwise * an in-body call sources to a QN one or more segments short of the def, and diff --git a/internal/cbm/helpers.h b/internal/cbm/helpers.h index f7c18c808..3ffa80bc4 100644 --- a/internal/cbm/helpers.h +++ b/internal/cbm/helpers.h @@ -173,6 +173,12 @@ TSNode cbm_resolve_func_name(TSNode node, CBMLanguage lang); // def extractor — drift dropped the class qualifier from in-body calls (#554/#621). char *cbm_cpp_out_of_line_parent_class(CBMArena *a, TSNode node, const char *source); +/* Go: resolve a method_declaration's receiver parameter_list down to the bare + * receiver type_identifier (unwrapping pointer_type / generic_type). Shared by + * def extraction (extract_defs.c) and call-scope attribution + * (extract_unified.c) so the receiver-qualified method QN has ONE formula. */ +char *cbm_go_receiver_type_name(CBMArena *a, TSNode recv, const char *source); + // Find a child node by kind string. TSNode cbm_find_child_by_kind(TSNode parent, const char *kind); diff --git a/internal/cbm/lsp/go_lsp.c b/internal/cbm/lsp/go_lsp.c index 58c151a53..34fad4b9e 100644 --- a/internal/cbm/lsp/go_lsp.c +++ b/internal/cbm/lsp/go_lsp.c @@ -1987,15 +1987,24 @@ static void process_function(GoLSPContext *ctx, TSNode func_node) { if (!func_name || !func_name[0]) return; - // Enclosing-function QN must be the BARE package.Func form (no receiver - // type segment). The textual call events (extract_unified.c) source calls - // as package_qn.func_name — methods included — and the defs pass creates - // the graph Method node under the same QN, so any other form breaks the - // caller-QN join in cbm_pipeline_find_lsp_resolution and the LSP-resolved - // call silently falls back to the registry short-name resolver. The - // receiver type still reaches the registry via the def's parent_class / - // method->receiver_type; it just does not appear in the caller QN. + // Enclosing-function QN must be EXACTLY the QN the textual call events + // (extract_unified.c compute_func_qn) and the defs pass produce, or the + // caller-QN join in cbm_pipeline_find_lsp_resolution breaks and the + // LSP-resolved call silently falls back to the registry short-name + // resolver. Since #1909 receiver-qualified Go method QNs, that shared + // form is package_qn.ReceiverType.method for methods (the ONE-formula + // contract: cbm_go_receiver_type_name) and package_qn.func for functions. ctx->enclosing_func_qn = cbm_arena_sprintf(ctx->arena, "%s.%s", ctx->package_qn, func_name); + if (strcmp(ts_node_type(func_node), "method_declaration") == 0) { + TSNode recv = ts_node_child_by_field_name(func_node, "receiver", 8); + if (!ts_node_is_null(recv)) { + char *recv_type = cbm_go_receiver_type_name(ctx->arena, recv, ctx->source); + if (recv_type && recv_type[0]) { + ctx->enclosing_func_qn = cbm_arena_sprintf(ctx->arena, "%s.%s.%s", ctx->package_qn, + recv_type, func_name); + } + } + } // Push function scope CBMScope *saved_scope = ctx->current_scope; diff --git a/src/pipeline/doc_links_go.c b/src/pipeline/doc_links_go.c index 047a7654e..b43289c95 100644 --- a/src/pipeline/doc_links_go.c +++ b/src/pipeline/doc_links_go.c @@ -32,10 +32,11 @@ * fields of its own struct declaration and the methods of its own interface * declaration: nothing promoted through embedding. * - * Binding. The graph keeps one node per qualified name, Go method nodes have - * a FLAT name (.), and is the directory, not the package - * clause: functions, methods of every receiver, types and values of one name, - * in every file of the directory, share one node, and the node is ONE of those + * Binding. The graph keeps one node per qualified name, and is the + * directory, not the package clause. Go method nodes are receiver-qualified + * (.., #1913) and are found through their receiver only. + * Functions, types and values of one name, in every file of the directory, + * share one node ., and the node is ONE of those * declarations (cbm_gbuf_upsert_node keeps the smallest file path, then the * last line). So a reference binds only when the node IS the entity it names: * the node's file is a file of the named package, the declaration that keeps @@ -59,7 +60,7 @@ * not), missing (a repository package that does not declare the name; an * import path inside a module without a package there), graph_gap (declared, * but the node is another declaration's or absent: block variables, later - * names of a declaration, embedded fields, methods that lost the flat name; + * names of a declaration, embedded fields; * also an undeclared name in a package with a parse error), test_only_target * (product code naming a declaration that only test files have), ambiguous * (several modules or packages fit; a name declared as two kinds; a name two @@ -1063,7 +1064,7 @@ static void go_record_decl(go_index_t *ix, go_file_t *file, char **f, int nf) { FIELD_FIELDS = 4, IM_FIELDS = 3 }; - enum { F_OWNER = 2, T_ALIAS = 3, T_OWNER = 4, V_OWNER = 2, M_OWNER = 3, FIELD_NODE = 3 }; + enum { F_OWNER = 2, T_ALIAS = 3, T_OWNER = 4, V_OWNER = 2, FIELD_NODE = 3 }; char kind = f[0][0]; if (nf < IM_FIELDS || !go_name_ok(f[SKIP_ONE])) { return; @@ -1078,10 +1079,12 @@ static void go_record_decl(go_index_t *ix, go_file_t *file, char **f, int nf) { go_add_sym(ix, file, GO_KIND_VALUE, f[SKIP_ONE], NULL); go_add_claim(ix, file, GO_KIND_VALUE, f[SKIP_ONE], NULL, f[V_OWNER]); } else if (kind == 'M' && nf == M_FIELDS && go_name_ok(f[PAIR_LEN])) { + /* #1913: a method is found through its receiver only; it claims no + * flat . (its record says w `-`), so a type, function or + * value of the same name keeps that node. */ if (strcmp(f[SKIP_ONE], "?") != 0) { go_add_member(ix, file, GO_MEM_METHOD, f[SKIP_ONE], f[PAIR_LEN], true); } - go_add_claim(ix, file, GO_KIND_METHOD, f[PAIR_LEN], f[SKIP_ONE], f[M_OWNER]); } else if (kind == 'f' && nf == FIELD_FIELDS && go_name_ok(f[PAIR_LEN])) { go_add_member(ix, file, GO_MEM_FIELD, f[SKIP_ONE], f[PAIR_LEN], f[FIELD_NODE][0] == '1'); } else if (kind == 'm' && nf == IM_FIELDS && go_name_ok(f[PAIR_LEN])) { @@ -1342,9 +1345,9 @@ static const cbm_gbuf_node_t *go_member_node(const go_ctx_t *c, const go_pkg_t * return (fm && (fm->node_kinds & kind)) ? n : NULL; } -/* The node of the method `recv.name`: its receiver-qualified node, else the - * flat node . when the method is the one that kept it (the last - * declaration of that name in the node's file). */ +/* The node of the method `recv.name`: its receiver-qualified node. The flat + * fallback below (the method that kept .) is unreachable since + * #1913: no method record claims a flat name any more. */ static const cbm_gbuf_node_t *go_method_node(const go_ctx_t *c, const go_pkg_t *pkg, go_str_t recv, go_str_t name) { const cbm_gbuf_node_t *qualified = go_member_node(c, pkg, GO_MEM_METHOD, recv, name); diff --git a/src/pipeline/doc_links_pdf.c b/src/pipeline/doc_links_pdf.c index a667fb6c7..1eaf057ae 100644 --- a/src/pipeline/doc_links_pdf.c +++ b/src/pipeline/doc_links_pdf.c @@ -328,9 +328,10 @@ static void qn_candidate(const pdfr_index_t *x, const cbm_gbuf_node_t *n, const const char *rel = rel_qn(x, n->qualified_name); char alt[PDFR_KEY_CAP]; bool owned = owner_name(x, n, alt, sizeof(alt)); - /* A member whose QN leaves its owner out (a Go method's has no receiver) - * is named through the owner: `time.Now` is package time's function, - * never the method Now of a type that package time declares. */ + /* A member whose QN leaves its owner out (a templ method's keeps the flat + * package QN; a Go method's did before index format 4) is named through + * the owner: `time.Now` is package time's function, never the method Now + * of a type that package time declares. */ bool ownerless_qn = owned && !comp_suffix(rel, alt); bool via_rel = !ownerless_qn && comp_suffix(rel, s); bool via_alt = !via_rel && owned && strcmp(alt, rel) != 0 && comp_suffix(alt, s); diff --git a/src/pipeline/pass_calls.c b/src/pipeline/pass_calls.c index dfbb8f7fb..f06544ae5 100644 --- a/src/pipeline/pass_calls.c +++ b/src/pipeline/pass_calls.c @@ -1103,7 +1103,13 @@ static int resolve_single_call(cbm_pipeline_ctx_t *ctx, CBMCall *call, const CBM * language gated on only one resolver produces an edge on the sequential * path and not the parallel one (or vice versa), breaking MT determinism. * ArkTS belongs to the JS/TS family here (#1842); dropping it would - * reintroduce the #592/#606 false-edge class for .ets files. */ + * reintroduce the #592/#606 false-edge class for .ets files. + * + * Go (#1906) rides the same deferred-drop plumbing through its OWN + * predicate: its drop-list differs (field_type_hint is receiver-aware for + * Go, and unique_name drops only when import-unreachability-penalized), so + * it composes via cbm_go_suppress_weak_method_match instead of widening + * the shared gate. Same lockstep rule: mirror pass_parallel.c. */ bool suppress_weak_member = lang == CBM_LANG_PYTHON || lang == CBM_LANG_JAVASCRIPT || lang == CBM_LANG_TYPESCRIPT || lang == CBM_LANG_TSX || lang == CBM_LANG_ARKTS || @@ -1131,6 +1137,8 @@ static int resolve_single_call(cbm_pipeline_ctx_t *ctx, CBMCall *call, const CBM res.strategy)) || cbm_suppress_weak_local_binding_call(suppress_weak_local_binding, call->callee_is_locally_bound, res.strategy) || + cbm_go_suppress_weak_method_match(lang == CBM_LANG_GO, call->is_method, res.strategy, + res.confidence) || /* Import-binding suppression (#2127): an external `from m import f` * binds `f` for the whole module, so a weak short-name match onto a * project `X.f` outside m is fabricated. Python-only (the import diff --git a/src/pipeline/pass_parallel.c b/src/pipeline/pass_parallel.c index 298b6b8d2..22531d167 100644 --- a/src/pipeline/pass_parallel.c +++ b/src/pipeline/pass_parallel.c @@ -3364,7 +3364,15 @@ static void resolve_file_calls(resolve_ctx_t *rc, resolve_worker_state_t *ws, CB } _rc_t0 = extract_now_ns(); - try_field_type_hint(rc, &res, call->callee_name, source_node->id); + /* Not for Go (F-G9): the hint matches the capitalized receiver + * VARIABLE name as a substring of a candidate QN (`t.Run` -> any + * `...T...Run`), with no type evidence. The sequential resolver has no + * such step, so for Go it produced parallel-only edges (par/seq CALLS + * diverged), and the Go selector guard could not see them as the weak + * pick they are. SCIP precision of Go field_type_hint: 0.016. */ + if (lang != CBM_LANG_GO) { + try_field_type_hint(rc, &res, call->callee_name, source_node->id); + } /* Cross-language veto — the same predicate as pass_calls.c, applied * AFTER the field-type hint so the hint may still re-pick a compatible * candidate first, and BEFORE the empty-resolution fallbacks below so a @@ -3401,7 +3409,9 @@ static void resolve_file_calls(resolve_ctx_t *rc, resolve_worker_state_t *ws, CB * #606 direction. * * This language set MUST match the one in pass_calls.c exactly — see the - * note there. ArkTS belongs to the JS/TS family (#1842). */ + * note there. ArkTS belongs to the JS/TS family (#1842). Go (#1906) + * composes via its own predicate (different drop-list — see + * cbm_go_suppress_weak_method_match), mirrored in pass_calls.c. */ bool suppress_weak_member = lang == CBM_LANG_PYTHON || lang == CBM_LANG_JAVASCRIPT || lang == CBM_LANG_TYPESCRIPT || lang == CBM_LANG_TSX || lang == CBM_LANG_ARKTS || @@ -3419,6 +3429,8 @@ static void resolve_file_calls(resolve_ctx_t *rc, resolve_worker_state_t *ws, CB call->callee_name, res.strategy)) || cbm_suppress_weak_local_binding_call(suppress_weak_local_binding, call->callee_is_locally_bound, res.strategy) || + cbm_go_suppress_weak_method_match(lang == CBM_LANG_GO, call->is_method, res.strategy, + res.confidence) || /* Import-binding suppression (#2127) — see pass_calls.c; this gate * MUST stay identical to the one there. */ (lang == CBM_LANG_PYTHON && diff --git a/src/pipeline/pass_semantic.c b/src/pipeline/pass_semantic.c index 186b2dfdc..bc23e0d26 100644 --- a/src/pipeline/pass_semantic.c +++ b/src/pipeline/pass_semantic.c @@ -266,9 +266,9 @@ static int check_go_class_implements(cbm_pipeline_ctx_t *ctx, const cbm_gbuf_nod } /* Resolve the struct's methods two ways and use whichever finds them: * (a) the struct's DEFINES_METHOD edges, matched by method NAME — the real - * pipeline path, where Go receiver methods carry a flat QN (e.g. - * "pkg.Area" rather than "pkg.Circle.Area"), so QN-string - * reconstruction would miss; and + * pipeline path. Go method QNs are receiver-qualified since #1913 + * ("pkg.Circle.Area"), so (b) matches them as well, but the edges + * stay the primary join; and * (b) the reconstructed QN "." — used by tests and any * extractor that does emit class-qualified method QNs without * DEFINES_METHOD edges from the class. */ diff --git a/src/pipeline/pipeline.h b/src/pipeline/pipeline.h index 072674d8d..a04e0144a 100644 --- a/src/pipeline/pipeline.h +++ b/src/pipeline/pipeline.h @@ -412,6 +412,15 @@ bool cbm_weak_member_unique_name_exempt(bool is_python, bool receiver_is_self_at bool cbm_suppress_weak_local_binding_call(bool enabled, bool callee_is_locally_bound, const char *strategy); +/* Go analog of the TS/JS guard, same failure class: a selector call whose + * receiver the Go LSP could not type must not be bound by a receiver-blind + * short-name strategy. Drops suffix_match / fuzzy always, and unique_name only + * when its confidence is import-unreachability-penalized (the stdlib/vendor + * hijack shape). field_type_hint is deliberately NOT dropped for Go — struct + * fields carry declared types, so the hint is receiver-aware there. */ +bool cbm_go_suppress_weak_method_match(bool is_go, bool is_method, const char *strategy, + double confidence); + /* Import-binding counterpart (#2127). True when the callee's root identifier * is bound by an import of this file whose module chain contradicts the * resolved target (`from unittest.mock import patch; patch()` must not bind a diff --git a/src/pipeline/registry.c b/src/pipeline/registry.c index 5dda868a4..d3d035e57 100644 --- a/src/pipeline/registry.c +++ b/src/pipeline/registry.c @@ -750,6 +750,35 @@ bool cbm_suppress_weak_local_binding_call(bool enabled, bool callee_is_locally_b return weak_short_name_strategy(strategy); } +bool cbm_go_suppress_weak_method_match(bool is_go, bool is_method, const char *strategy, + double confidence) { + /* Go analog of the TS/JS guard above, same failure class: a selector call + * whose receiver the Go LSP could not type reaches the registry and a bare + * short-name strategy binds it to an arbitrary same-named project symbol + * (`f.Close()` on an os.File -> a project `Close`, suffix_match over 15 + * candidates). Built on the shared drop-list so "weak" means one thing + * for every guard; Go narrows it in one place only (unique_name below). + * field_type_hint stays weak: it is a receiver-NAME substring match, not + * a declared-type lookup (the parallel resolver no longer produces it for + * Go; a typed field call such as lrp_go_s8's `s.repo.Find()` resolves + * through the Go LSP). */ + if (!is_go || !is_method || !weak_short_name_strategy(strategy)) { + return false; + } + if (strcmp(strategy, "unique_name") != 0) { + return true; /* suffix_match, field_type_hint, fuzzy */ + } + /* unique_name is dropped only when PENALIZED: resolve_name_lookup scales + * CONF_UNIQUE_NAME by DEFAULT_CONFIDENCE exactly when the lone candidate + * is not reachable through the caller's imports — the stdlib/vendor + * hijack shape (`io.Copy` -> a project `Copy`). An unpenalized + * unique_name target sits inside the caller's import closure (or the + * file has no imports, e.g. a same-package call) and must be kept — + * dropping it kills genuinely-typed lone-candidate calls that never + * enter the field-type-hint upgrade (candidate_count == 1). */ + return confidence < CONF_UNIQUE_NAME; +} + /* Import-binding counterpart of the parameter guard above (#2127). An import * binds the identifier for the whole module; when that import is external it * never materializes, and the chain falls through to project-wide short-name diff --git a/src/store/store.h b/src/store/store.h index 803df81d8..0f291daf9 100644 --- a/src/store/store.h +++ b/src/store/store.h @@ -32,8 +32,10 @@ typedef struct cbm_store_read_scope cbm_store_read_scope_t; * Dockerfile. keeps its suffix in its QNs. * 3 Image identity invents no registry: a reference without a host is * __image__, registry "unknown" -- no longer - * __image__docker.io/[library/]. */ -#define CBM_INDEX_FORMAT_VERSION 3 + * __image__docker.io/[library/]. + * 4 #1913: Go method QNs are receiver-qualified (pkg.Recv.method, generic + * receivers unwrapped); free functions stay pkg.func. */ +#define CBM_INDEX_FORMAT_VERSION 4 #define CBM_STORE_CANCELLED (-3) #define CBM_STORE_SCAN_LIMIT (-4) #define CBM_STORE_CALLBACK_ERR (-5) diff --git a/tests/test_doc_links_pdf.c b/tests/test_doc_links_pdf.c index 1a78b7254..9c7bf90ff 100644 --- a/tests/test_doc_links_pdf.c +++ b/tests/test_doc_links_pdf.c @@ -1087,8 +1087,9 @@ TEST(pdf_links_pipeline) { ASSERT_NOT_NULL(strstr(props, "\"line\":1")); /* qualified names (pdf_qn) passed a second held-out audit (97 of 99 * correct, Wilson 95 % low 0.929) once a member whose QN leaves its owner - * out binds only through the owner. A Go method through its receiver: */ - dm_edge(db, "docs.design.page_1", "pkg.server.Serve", props, sizeof(props), &n); + * out binds only through the owner. A Go method through its receiver + * (receiver-qualified QN since index format 4): */ + dm_edge(db, "docs.design.page_1", "pkg.server.Handler.Serve", props, sizeof(props), &n); ASSERT_EQ(n, 1); ASSERT_NOT_NULL(strstr(props, "\"syntax\":\"pdf_qn\"")); ASSERT_NOT_NULL(strstr(props, "\"tier\":\"unique\"")); @@ -1142,10 +1143,12 @@ TEST(pdf_links_pipeline) { PASS(); } -/* A held-out finding of the PDF audit: a Go method's QN has no receiver, so +/* A held-out finding of the PDF audit: a Go method's QN had no receiver, so * `time.Now` (the standard library's function, in a spec) matched the method * Now of a type that the repository's own package time declares. A method is - * named through its owner: `time.Provider.Now` still binds it. */ + * named through its owner: `time.Provider.Now` still binds it. Since index + * format 4 the QN carries the receiver (pkg.time.Provider.Now), and both + * directions are checked against that node. */ TEST(pdf_qn_method_through_owner) { cbm_doclink_test_set_ships(CBM_DOCLINK_PDF_QN, true); char tmp[256]; @@ -1169,9 +1172,9 @@ TEST(pdf_qn_method_through_owner) { ASSERT_EQ(dm_index(repo, db, NULL), 0); char props[512]; int n; - dm_edge(db, "docs.spec.page_1", "pkg.time.Now", props, sizeof(props), &n); + dm_edge(db, "docs.spec.page_1", "pkg.time.Provider.Now", props, sizeof(props), &n); ASSERT_EQ(n, 0); - dm_edge(db, "docs.spec.page_2", "pkg.time.Now", props, sizeof(props), &n); + dm_edge(db, "docs.spec.page_2", "pkg.time.Provider.Now", props, sizeof(props), &n); ASSERT_EQ(n, 1); ASSERT_NOT_NULL(strstr(props, "\"syntax\":\"pdf_qn\"")); th_cleanup(tmp); diff --git a/tests/test_doc_mentions_go.c b/tests/test_doc_mentions_go.c index 78f1934e0..0d4462f76 100644 --- a/tests/test_doc_mentions_go.c +++ b/tests/test_doc_mentions_go.c @@ -228,10 +228,10 @@ TEST(doc_mentions_go_sources) { const char *raw; const char *source; } want[] = { - {"RefF", "p.pkg.F"}, {"RefS", "p.pkg.S"}, {"RefField", "p.pkg.S.Field"}, - {"RefM", "p.pkg.M"}, /* Go method nodes have a flat name */ - {"RefI", "p.pkg.I"}, {"RefDo", "p.pkg.I.Do"}, {"RefC", "p.pkg.C"}, - {"RefK", "p.pkg.K"}, {"RefV", "p.pkg.V"}, {"RefA", "p.pkg.A"}, + {"RefF", "p.pkg.F"}, {"RefS", "p.pkg.S"}, {"RefField", "p.pkg.S.Field"}, + {"RefM", "p.pkg.S.M"}, /* #1913: Go method nodes are receiver-qualified */ + {"RefI", "p.pkg.I"}, {"RefDo", "p.pkg.I.Do"}, {"RefC", "p.pkg.C"}, + {"RefK", "p.pkg.K"}, {"RefV", "p.pkg.V"}, {"RefA", "p.pkg.A"}, }; for (size_t i = 0; i < GO_COUNT(want); i++) { const CBMDocLink *t = dm_find_token(r, want[i].raw); @@ -772,14 +772,15 @@ TEST(doc_mentions_go_scope_blob) { "T\tAlias\ta\tal.Widget\t1", "T\tSame\ta\tKind\t1", "T\tList\ts\t-\t1", - /* functions and methods; two methods named Run compete for one node - * and the later one keeps it */ + /* functions and methods; a method node is receiver-qualified + * (#1913), so a method claims no flat name and its w is `-`: the two + * methods named Run no longer compete */ "F\tNew\t1", "F\tMap\t1", - "M\tWidget\tRun\t0", - "M\tKind\tRun\t1", - "M\tList\tPush\t1", - "M\tlower\tExported\t1", + "M\tWidget\tRun\t-", + "M\tKind\tRun\t-", + "M\tList\tPush\t-", + "M\tlower\tExported\t-", /* own type parameters, by the declaration's first and last line */ "g\t36\t36\tT,U", "g\t42\t42\tA,B", @@ -814,15 +815,20 @@ TEST(doc_mentions_go_scope_blob) { char *portable = cbm_doclink_portable_scope(s); ASSERT_NOT_NULL(portable); ASSERT_NULL(strstr(portable, "\ng\t")); - ASSERT_TRUE(go_has_rec(portable, "M\tKind\tRun\t1")); + ASSERT_TRUE(go_has_rec(portable, "M\tKind\tRun\t-")); ASSERT(strncmp(portable, "go1\n", 4) == 0); cbm_free(CBM_MEM_CLASS_OTHER, portable); cbm_free_result(r); - /* moving declarations changes nothing unless a node changes its owner */ + /* moving declarations changes nothing unless a node changes its owner. + * Methods own no flat name (#1913), so swapping two methods of one name + * changes nothing either; two package-level declarations of one name + * still hand . over when swapped */ const char *a = "package p\n\nfunc (a A) Run() {}\nfunc (b B) Run() {}\nfunc F() {}\n"; const char *moved = "package p\n\nfunc F() {}\n\n\nfunc (a A) Run() {}\nfunc (b B) Run() {}\n"; const char *swapped = "package p\n\nfunc (b B) Run() {}\nfunc (a A) Run() {}\nfunc F() {}\n"; + const char *pair = "package p\n\nfunc Open() {}\ntype Open struct{}\n"; + const char *pair_swapped = "package p\n\ntype Open struct{}\nfunc Open() {}\n"; CBMFileResult *ra = go_extract(a, "p/x.go"); CBMFileResult *rm = go_extract(moved, "p/x.go"); CBMFileResult *rs = go_extract(swapped, "p/x.go"); @@ -830,16 +836,25 @@ TEST(doc_mentions_go_scope_blob) { ASSERT_NOT_NULL(rm); ASSERT_NOT_NULL(rs); ASSERT_STR_EQ(ra->doc_scope, rm->doc_scope); - ASSERT(strcmp(ra->doc_scope, rs->doc_scope) != 0); - ASSERT_TRUE(go_has_rec(rs->doc_scope, "M\tA\tRun\t1")); - ASSERT_TRUE(go_has_rec(rs->doc_scope, "M\tB\tRun\t0")); + ASSERT_STR_EQ(ra->doc_scope, rs->doc_scope); + ASSERT_TRUE(go_has_rec(rs->doc_scope, "M\tA\tRun\t-")); + ASSERT_TRUE(go_has_rec(rs->doc_scope, "M\tB\tRun\t-")); cbm_free_result(ra); cbm_free_result(rm); cbm_free_result(rs); - - /* the node . is one for a function, a type, a value and the - * methods of that name: the last declaration keeps it, whatever its kind, - * and two on one line contest it */ + CBMFileResult *rp = go_extract(pair, "p/x.go"); + CBMFileResult *rq = go_extract(pair_swapped, "p/x.go"); + ASSERT_NOT_NULL(rp); + ASSERT_NOT_NULL(rq); + ASSERT(strcmp(rp->doc_scope, rq->doc_scope) != 0); + ASSERT_TRUE(go_has_rec(rp->doc_scope, "T\tOpen\ts\t-\t1")); + ASSERT_TRUE(go_has_rec(rq->doc_scope, "F\tOpen\t1")); + cbm_free_result(rp); + cbm_free_result(rq); + + /* the node . is one for a function, a type and a value of + * that name: the last declaration keeps it, whatever its kind, and two on + * one line contest it. A method of that name claims nothing (#1913) */ CBMFileResult *k1 = go_extract("package p\n\nfunc Open() {}\nfunc (t T) Open() {}\n", "p/k.go"); CBMFileResult *k2 = go_extract("package p\n\nfunc (t T) Open() {}\nfunc Open() {}\n", "p/k.go"); CBMFileResult *k3 = go_extract("package p\n\ntype Open int\nvar Open = 1\n" @@ -848,14 +863,21 @@ TEST(doc_mentions_go_scope_blob) { ASSERT_NOT_NULL(k1); ASSERT_NOT_NULL(k2); ASSERT_NOT_NULL(k3); - ASSERT_TRUE(go_has_rec(k1->doc_scope, "F\tOpen\t0")); - ASSERT_TRUE(go_has_rec(k1->doc_scope, "M\tT\tOpen\t1")); + ASSERT_TRUE(go_has_rec(k1->doc_scope, "F\tOpen\t1")); + ASSERT_TRUE(go_has_rec(k1->doc_scope, "M\tT\tOpen\t-")); ASSERT_TRUE(go_has_rec(k2->doc_scope, "F\tOpen\t1")); - ASSERT_TRUE(go_has_rec(k2->doc_scope, "M\tT\tOpen\t0")); + ASSERT_TRUE(go_has_rec(k2->doc_scope, "M\tT\tOpen\t-")); ASSERT_TRUE(go_has_rec(k3->doc_scope, "T\tOpen\to\t-\t0")); ASSERT_TRUE(go_has_rec(k3->doc_scope, "V\tOpen\t0")); - ASSERT_TRUE(go_has_rec(k3->doc_scope, "V\tOpen\tx")); - ASSERT_TRUE(go_has_rec(k3->doc_scope, "M\tT\tOpen\tx")); + ASSERT_TRUE(go_has_rec(k3->doc_scope, "V\tOpen\t1")); + ASSERT_TRUE(go_has_rec(k3->doc_scope, "M\tT\tOpen\t-")); + /* two package-level declarations on one line contest the name */ + CBMFileResult *k4 = + go_extract("package p\n\ntype Open int\nvar Open = 1; const Open = 2\n", "p/k.go"); + ASSERT_NOT_NULL(k4); + ASSERT_TRUE(go_has_rec(k4->doc_scope, "T\tOpen\to\t-\t0")); + ASSERT_TRUE(go_has_rec(k4->doc_scope, "V\tOpen\tx")); + cbm_free_result(k4); cbm_free_result(k1); cbm_free_result(k2); cbm_free_result(k3); @@ -1030,7 +1052,7 @@ static const char GO_ALPHA_C[] = "package alpha\n" "// Target is a target; see [A.Close] and [B.Close].\n" "func Target() {}\n" "\n" - "// Open is a type that loses its node name to the method below.\n" + "// Open is a type; a method of B below has the same name.\n" "type Open struct{}\n" "\n" "// Open opens: a method named like the type.\n" @@ -1295,7 +1317,7 @@ TEST(doc_mentions_go_forms) { static const go_want_edge_t edges[] = { /* [Name], [Recv.Method], an interface's method, a constant, a variable */ {"alpha.Widget", "alpha.New", "unique"}, - {"alpha.Widget", "alpha.Run", "unique"}, + {"alpha.Widget", "alpha.Widget.Run", "unique"}, {"alpha.Widget", "alpha.Runner.Run", "unique"}, {"alpha.Widget", "alpha.Max", "unique"}, {"alpha.Widget", "alpha.Default", "unique"}, @@ -1311,7 +1333,7 @@ TEST(doc_mentions_go_forms) { {"alpha.New", "sub.deep.Deep", "exact"}, {"alpha.New", "nested.pkg.Far", "exact"}, /* [pkg.Recv.Name]: a method, a field, an interface method */ - {"alpha.New", "beta.Do", "exact"}, + {"alpha.New", "beta.Thing.Do", "exact"}, {"alpha.New", "beta.Thing.Field", "exact"}, {"alpha.New", "beta.Doer.Act", "exact"}, /* an identifier with letters beyond ASCII */ @@ -1395,7 +1417,7 @@ TEST(doc_mentions_go_plain_text) { ASSERT_EQ(go_bad_rows(db, rows, GO_COUNT(rows)), 0); /* and none of them is an edge: Run's doc holds rows, plain text and one * link, to the field Widget.Size */ - ASSERT_EQ(dm_mentions_from(db, "alpha.Run"), 1); + ASSERT_EQ(dm_mentions_from(db, "alpha.Widget.Run"), 1); ASSERT_EQ(dm_mentions_from(db, "alpha.Bare"), 0); static const go_want_edge_t edges[] = { {"alpha.Sibling", "alpha.Widget", "unique"}, @@ -1424,14 +1446,14 @@ TEST(doc_mentions_go_struct_fields) { static const go_want_edge_t edges[] = { /* the same package: a field named from a method's doc and from a * sibling field's doc */ - {"alpha.Run", "alpha.Widget.Size", "exact"}, + {"alpha.Widget.Run", "alpha.Widget.Size", "exact"}, {"alpha.Widget.Name", "alpha.Widget.Size", "exact"}, /* another package, by its import */ {"alpha.New", "beta.Thing.Field", "exact"}, {"gamma.ViaAlias", "alpha.Widget.Name", "exact"}, /* Depth is a field of beta.Embedded, which Widget embeds: it is no * field of Widget */ - {"alpha.Run", "beta.Embedded.Depth", ""}, + {"alpha.Widget.Run", "beta.Embedded.Depth", ""}, }; ASSERT_EQ(go_bad_edges(db, edges, GO_COUNT(edges)), 0); /* the promoted field exists as a node: the test above is not vacuous */ @@ -1545,7 +1567,7 @@ TEST(doc_mentions_go_lookup_order) { * file that imports "fmt", the repository's own in a file that imports * example.com/demo/fmt */ ASSERT_STR_EQ(go_reason(db, "alpha/a.go", "fmt.Println"), "external"); - ASSERT_STR_EQ(go_tier(db, "alpha.Run", "fmt.Println"), ""); + ASSERT_STR_EQ(go_tier(db, "alpha.Widget.Run", "fmt.Println"), ""); ASSERT_STR_EQ(go_tier(db, "gamma.Local", "fmt.Println"), "exact"); /* 5. the longest module path decides: a directory inside the root module * that belongs to a nested module of another path is not reachable through @@ -1555,17 +1577,17 @@ TEST(doc_mentions_go_lookup_order) { ASSERT_STR_EQ(go_tier(db, "alpha.New", "sub.deep.Deep"), "exact"); /* 6. the bare name stops at the package: beta.Make is imported and still * no [Make] */ - ASSERT_STR_EQ(go_tier(db, "alpha.Run", "beta.Make"), ""); + ASSERT_STR_EQ(go_tier(db, "alpha.Widget.Run", "beta.Make"), ""); /* 7. members stop at the type: nothing promoted through embedding */ ASSERT_STR_EQ(go_reason(db, "gamma/k.go", "alpha.Widget.Promoted"), "missing"); - ASSERT_STR_EQ(go_tier(db, "gamma.ViaAlias", "beta.Promoted"), ""); + ASSERT_STR_EQ(go_tier(db, "gamma.ViaAlias", "beta.Embedded.Promoted"), ""); /* 8. an alias is followed for a member written with its package, to the * type that declares the member */ static const go_want_edge_t alias[] = { - {"gamma.ViaAlias", "beta.Do", "exact"}, + {"gamma.ViaAlias", "beta.Thing.Do", "exact"}, {"gamma.ViaAlias", "beta.Thing.Field", "exact"}, {"gamma.ViaAlias", "alpha.Alias", "exact"}, - {"gamma.ViaAlias", "alpha.End", "exact"}, /* seven aliases away */ + {"gamma.ViaAlias", "alpha.Deep10.End", "exact"}, /* seven aliases away */ {"gamma.ViaAlias", "alpha.Widget.Name", "exact"}, }; ASSERT_EQ(go_bad_edges(db, alias, GO_COUNT(alias)), 0); @@ -1638,28 +1660,42 @@ TEST(doc_mentions_go_node_owner) { char tmp[256]; char db[512]; ASSERT_EQ(go_index_lookup(tmp, sizeof(tmp), db, sizeof(db)), 0); - /* Go method nodes have a flat name: A.Close and B.Close share - * alpha.Close, and the later declaration keeps it */ - ASSERT_EQ(dm_count(db, "SELECT COUNT(*) FROM nodes WHERE qualified_name LIKE '%.alpha.Close'"), + /* #1913: Go method nodes are receiver-qualified. A.Close and B.Close are + * two nodes (no shared alpha.Close), each a target of its own */ + ASSERT_EQ(dm_count(db, "SELECT COUNT(*) FROM nodes WHERE label = 'Method' AND " + "qualified_name LIKE '%.alpha.A.Close'"), 1); - ASSERT_STR_EQ(go_tier(db, "alpha.Target", "alpha.Close"), "unique"); - ASSERT_STR_EQ(go_reason(db, "alpha/c.go", "A.Close"), "graph_gap"); + ASSERT_EQ(dm_count(db, "SELECT COUNT(*) FROM nodes WHERE label = 'Method' AND " + "qualified_name LIKE '%.alpha.B.Close'"), + 1); + ASSERT_EQ(dm_count(db, "SELECT COUNT(*) FROM nodes WHERE qualified_name LIKE '%.alpha.Close'"), + 0); + ASSERT_STR_EQ(go_tier(db, "alpha.Target", "alpha.A.Close"), "unique"); + ASSERT_STR_EQ(go_tier(db, "alpha.Target", "alpha.B.Close"), "unique"); + ASSERT_STR_EQ(go_reason(db, "alpha/c.go", "A.Close"), ""); ASSERT_STR_EQ(go_reason(db, "alpha/c.go", "B.Close"), ""); - /* the doc of the method that lost the name is not attributed to the one - * that kept it: B.Close's [Target] is the edge, A.Close's a row */ + /* each method's doc is attributed to its own node: both [Target]s are + * edges, from the line of that method's doc, and neither is a row */ char props[512]; int n = 0; - dm_edge(db, "alpha.Close", "alpha.Target", props, sizeof(props), &n); + dm_edge(db, "alpha.A.Close", "alpha.Target", props, sizeof(props), &n); + ASSERT_EQ(n, 1); + ASSERT_NOT_NULL(strstr(props, "\"line\":9,\"count\":1")); + dm_edge(db, "alpha.B.Close", "alpha.Target", props, sizeof(props), &n); ASSERT_EQ(n, 1); ASSERT_NOT_NULL(strstr(props, "\"line\":12,\"count\":1")); - ASSERT_EQ(dm_count(db, "SELECT line FROM doc_link_unresolved WHERE rel_path = 'alpha/c.go' " - "AND raw = 'Target'"), - 9); - /* a type that lost its name to a method of that name is no target, the - * method is */ - ASSERT_STR_EQ(go_reason(db, "alpha/c.go", "Open"), "graph_gap"); + ASSERT_EQ(dm_count(db, "SELECT COUNT(*) FROM doc_link_unresolved WHERE rel_path = " + "'alpha/c.go' AND raw = 'Target'"), + 0); + /* a type and a method of the same name no longer share a node: [Open] + * is the type, [B.Open] the method */ + ASSERT_STR_EQ(go_reason(db, "alpha/c.go", "Open"), ""); + ASSERT_EQ(dm_count(db, "SELECT COUNT(*) FROM nodes WHERE label <> 'Method' AND " + "qualified_name LIKE '%.alpha.Open'"), + 1); ASSERT_STR_EQ(go_tier(db, "alpha.UsesOpen", "alpha.Open"), "unique"); - ASSERT_EQ(dm_mentions_from(db, "alpha.UsesOpen"), 1); + ASSERT_STR_EQ(go_tier(db, "alpha.UsesOpen", "alpha.B.Open"), "unique"); + ASSERT_EQ(dm_mentions_from(db, "alpha.UsesOpen"), 2); /* build variants are one entity: whichever file holds the node */ static const go_want_edge_t variants[] = { @@ -1836,15 +1872,16 @@ TEST(doc_mentions_go_incremental_equals_full) { ASSERT_EQ(dm_index(repo, inc_db, NULL), 0); /* the preconditions the steps below move away from */ static const go_want_edge_t before[] = { - {"a.Use", "b.Target", "exact"}, {"a.Use", "b.Gone", "exact"}, - {"a.Use", "b.M", "exact"}, /* T.M, in the first file of the directory, keeps b.M */ - {"a.Use", "b.Name", "exact"}, /* W.Name, the later declaration, keeps b.Name */ - {"a.Use", "a.Local", "unique"}, {"a.doc.go.__file__", "a.Local", "unique"}, + {"a.Use", "b.Target", "exact"}, {"a.Use", "b.Gone", "exact"}, + {"a.Use", "b.T.M", "exact"}, /* #1913: every method has its own node */ + {"a.Use", "b.W.Name", "exact"}, {"a.Use", "b.V.Name", "exact"}, + {"b.U.M", "b.Target", "unique"}, /* U.M is a source of its own */ + {"a.Use", "a.Local", "unique"}, {"a.doc.go.__file__", "a.Local", "unique"}, {"a.Other", "b.Target", ""}, /* other.go does not import b */ }; ASSERT_EQ(go_bad_edges(inc_db, before, GO_COUNT(before)), 0); - ASSERT_STR_EQ(go_reason(inc_db, "a/a.go", "b.V.Name"), "graph_gap"); - ASSERT_STR_EQ(go_reason(inc_db, "b/c.go", "Target"), "graph_gap"); /* U.M has no node */ + ASSERT_STR_EQ(go_reason(inc_db, "a/a.go", "b.V.Name"), ""); + ASSERT_STR_EQ(go_reason(inc_db, "b/c.go", "Target"), ""); /* a doc-comment edit re-extracts the file alone */ go_w(repo, "a/a.go", GO_INC_A_DOC_EDIT); @@ -1880,23 +1917,27 @@ TEST(doc_mentions_go_incremental_equals_full) { 0); ASSERT_STR_EQ(go_reason(inc_db, "a/a.go", "b.Gone"), "missing"); - /* the method that kept a shared node name goes: the other file's method of - * that name takes the node over, and its doc becomes a source */ + /* a method goes: links to it break, and another receiver's method of + * the same name, in another file, is untouched */ go_w(repo, "b/b.go", GO_INC_B_NO_M); ASSERT_EQ( dm_step(repo, inc_db, full_db, "name keeper removed", CBM_INCREMENTAL_ROUTE_CLOSURE_REPAIR), 0); ASSERT_STR_EQ(go_reason(inc_db, "a/a.go", "b.T.M"), "missing"); - ASSERT_STR_EQ(go_tier(inc_db, "b.M", "b.Target"), "unique"); + ASSERT_STR_EQ(go_tier(inc_db, "b.U.M", "b.Target"), "unique"); ASSERT_STR_EQ(go_reason(inc_db, "b/c.go", "Target"), ""); - /* declaration order decides which declaration keeps a shared name: - * swapping two methods moves the node, for files with no edge into it too */ + /* declaration order no longer decides anything: swapping two methods of + * the same name keeps both nodes and both links */ go_w(repo, "b/d.go", GO_INC_D_SWAPPED); - ASSERT_EQ(dm_step(repo, inc_db, full_db, "methods swapped", CBM_INCREMENTAL_ROUTE_FORCED_FULL), - 0); - ASSERT_STR_EQ(go_reason(inc_db, "a/a.go", "b.W.Name"), "graph_gap"); + /* and the scope blob does not change (no method w moves), so the step + * is an ordinary closure repair, no longer a forced full rebuild */ + ASSERT_EQ( + dm_step(repo, inc_db, full_db, "methods swapped", CBM_INCREMENTAL_ROUTE_CLOSURE_REPAIR), 0); + ASSERT_STR_EQ(go_reason(inc_db, "a/a.go", "b.W.Name"), ""); ASSERT_STR_EQ(go_reason(inc_db, "a/a.go", "b.V.Name"), ""); + ASSERT_STR_EQ(go_tier(inc_db, "a.Use", "b.W.Name"), "exact"); + ASSERT_STR_EQ(go_tier(inc_db, "a.Use", "b.V.Name"), "exact"); /* renaming a target */ go_w(repo, "b/b.go", GO_INC_B_RENAME); @@ -2006,10 +2047,20 @@ TEST(doc_mentions_go_scope_delta_rules) { GO_DELTA_HEAD GO_DELTA_F GO_DELTA_G GO_DELTA_T GO_DELTA_M GO_DELTA_I GO_DELTA_C "var (\n\tV = 2\n)\n", CBM_DOCLINK_DELTA_GLOBAL, NULL}, - /* the declaration that keeps a shared node name changes */ - {"same-named declarations swapped", twins, - "package p\n\nfunc (T) Open() {}\nfunc Open() {}\n", CBM_DOCLINK_DELTA_GLOBAL, NULL}, - {"one of two same-named declarations removed", twins, "package p\n\nfunc Open() {}\n", + /* a method and a function of one name no longer share a node + * (#1913): swapping them changes no keeper */ + {"method and function swapped", twins, "package p\n\nfunc (T) Open() {}\nfunc Open() {}\n", + CBM_DOCLINK_DELTA_LOCAL, ""}, + /* the declaration that keeps a shared node name changes: two + * package-level declarations of one name still share . */ + {"same-named declarations swapped", "package p\n\nfunc Open() {}\ntype Open struct{}\n", + "package p\n\ntype Open struct{}\nfunc Open() {}\n", CBM_DOCLINK_DELTA_GLOBAL, NULL}, + /* the method of a method/function pair removed: a member left, the + * function keeps its node either way */ + {"method of a same-named pair removed", twins, "package p\n\nfunc Open() {}\n", + CBM_DOCLINK_DELTA_LOCAL, "Open"}, + {"one of two same-named declarations removed", + "package p\n\nfunc Open() {}\ntype Open struct{}\n", "package p\n\nfunc Open() {}\n", CBM_DOCLINK_DELTA_GLOBAL, NULL}, /* other files follow `type X = q.Y` through this file's imports */ {"import changed under a qualified alias", alias, @@ -2091,7 +2142,10 @@ TEST(doc_mentions_go_workers_agree) { ASSERT_STR_EQ(go_tier(par_db, "p00.T0", "p01.T0"), "exact"); ASSERT_STR_EQ(go_reason(par_db, "p00/f0.go", "p01.Gone"), "missing"); ASSERT_STR_EQ(go_reason(par_db, "p00/f0.go", "fmt.Println"), "external"); - ASSERT_STR_EQ(go_reason(par_db, "p00/f0.go", "p01.T3.Close"), "graph_gap"); + /* #1913: [p01.T3.Close] lands on its own method node now; the graph_gap + * reason stays covered by Shared, one name declared by every file */ + ASSERT_STR_EQ(go_reason(par_db, "p00/f0.go", "p01.T3.Close"), ""); + ASSERT_STR_EQ(go_tier(par_db, "p00.T0", "p01.T3.Close"), "exact"); ASSERT_STR_EQ(go_reason(par_db, "p00/f3.go", "Shared"), "graph_gap"); ASSERT_TRUE(dm_count(par_db, "SELECT COUNT(*) FROM edges WHERE type = 'MENTIONS'") > 100); dm_unlink_db(par_db); diff --git a/tests/test_extraction.c b/tests/test_extraction.c index 8c37e8c5d..4a8392205 100644 --- a/tests/test_extraction.c +++ b/tests/test_extraction.c @@ -6565,12 +6565,12 @@ TEST(extract_go_no_filename_in_module_qn) { ASSERT_NOT_NULL(conn); ASSERT_STR_EQ(conn->qualified_name, "proj.myapp.db.Conn"); - /* Go method nodes keep a FLAT QN (module + name) with a separate - * parent_class link to the receiver type — the QN must carry the + /* Go method nodes carry a receiver-qualified QN (module + receiver type + + * name) plus the parent_class link — the QN must carry the * directory-based module and NOT the `.conn.` filename segment. */ const CBMDefinition *query = find_def_by_name(r, "Query"); ASSERT_NOT_NULL(query); - ASSERT_STR_EQ(query->qualified_name, "proj.myapp.db.Query"); + ASSERT_STR_EQ(query->qualified_name, "proj.myapp.db.Conn.Query"); ASSERT_EQ(strstr(query->qualified_name, ".conn."), NULL); /* The method's parent_class must match the type node QN (for DEFINES_METHOD). */ ASSERT_NOT_NULL(query->parent_class); @@ -7235,12 +7235,17 @@ TEST(extract_rust_generic_impl_call_scope_matches_def) { PASS(); } -/* Languages OUTSIDE the is_method flag set (only Perl and TS/JS/TSX set it) must - * be unaffected: a Go method call never sets is_method. */ +/* Languages OUTSIDE the is_method flag set (Perl, Python, TS/JS/TSX and Go set + * it) must be unaffected: a Rust method call never sets is_method. */ TEST(extract_flag_exempt_method_call_not_flagged_is_method) { - CBMFileResult *r = extract("package m\n" - "func run(o Obj) { o.Commit(); helper() }\n", - CBM_LANG_GO, "t", "x.go"); + /* Rust is flag-exempt: only Perl, Python, TS/JS and Go set is_method. + * Guards the blast radius of the receiver-aware flags for every other + * language. */ + CBMFileResult *r = extract("fn run(o: Obj) {\n" + " o.commit();\n" + " helper();\n" + "}\n", + CBM_LANG_RUST, "t", "x.rs"); ASSERT_NOT_NULL(r); ASSERT_FALSE(r->has_error); for (int i = 0; i < r->calls.count; i++) { @@ -7398,6 +7403,33 @@ TEST(extract_python_bare_call_flags_locally_bound_callee) { PASS(); } +TEST(extract_go_selector_call_flags_is_method) { + /* Go selector calls are flagged so the weak-match guard can fire when the + * Go LSP cannot type the receiver; bare calls stay unflagged. */ + CBMFileResult *r = extract("package m\n" + "func run(o Obj) { o.Commit(); helper() }\n", + CBM_LANG_GO, "t", "x.go"); + ASSERT_NOT_NULL(r); + ASSERT_FALSE(r->has_error); + bool saw_selector = false; + bool saw_bare = false; + for (int i = 0; i < r->calls.count; i++) { + const CBMCall *c = &r->calls.items[i]; + if (c->callee_name && strstr(c->callee_name, "Commit") != NULL) { + ASSERT_TRUE(c->is_method); + saw_selector = true; + } + if (c->callee_name && strcmp(c->callee_name, "helper") == 0) { + ASSERT_FALSE(c->is_method); + saw_bare = true; + } + } + ASSERT_TRUE(saw_selector); + ASSERT_TRUE(saw_bare); + cbm_free_result(r); + PASS(); +} + /* The bare-call flag is DEPTH-INDEPENDENT, and the binding is UNWOUND when its * scope closes. * @@ -7503,6 +7535,91 @@ TEST(extract_python_bare_call_flag_is_depth_independent) { PASS(); } +TEST(extract_go_method_receiver_qualified_qn) { + /* A Go method QN carries the receiver type (proj.pkg.Recv.method), same + * shape as C++ out-of-line methods and Go interface members — so two + * same-name methods on different receivers no longer collide in the + * graph upsert. Free functions keep the flat package QN. */ + CBMFileResult *r = extract("package m\n" + "type Storage struct{}\n" + "type Cache struct{}\n" + "func (s *Storage) Close() {}\n" + "func (c Cache) Close() {}\n" + "func Shutdown() {}\n", + CBM_LANG_GO, "t", "x.go"); + ASSERT_NOT_NULL(r); + ASSERT_FALSE(r->has_error); + bool saw_storage = false; + bool saw_cache = false; + bool saw_free = false; + for (int i = 0; i < r->defs.count; i++) { + const CBMDefinition *d = &r->defs.items[i]; + if (!d->name || !d->qualified_name) { + continue; + } + if (strcmp(d->name, "Close") == 0 && strcmp(d->qualified_name, "t.Storage.Close") == 0) { + ASSERT_STR_EQ(d->label, "Method"); + ASSERT_STR_EQ(d->parent_class, "t.Storage"); + saw_storage = true; + } + if (strcmp(d->name, "Close") == 0 && strcmp(d->qualified_name, "t.Cache.Close") == 0) { + ASSERT_STR_EQ(d->parent_class, "t.Cache"); + saw_cache = true; + } + if (strcmp(d->name, "Shutdown") == 0) { + ASSERT_STR_EQ(d->qualified_name, "t.Shutdown"); + saw_free = true; + } + } + ASSERT_TRUE(saw_storage); + ASSERT_TRUE(saw_cache); + ASSERT_TRUE(saw_free); + cbm_free_result(r); + PASS(); +} + +/* Templ shares Go's method grammar (method_declaration with a receiver), but + * only Go method QNs are receiver-qualified (#1913); templ's call-scope QN + * stays flat. The method's def QN and the enclosing-function QN of a call + * inside it must agree, or the call has no source node and degrades to the + * File node. */ +TEST(extract_templ_method_qn_matches_call_scope) { + CBMFileResult *r = extract("package views\n" + "\n" + "type Page struct{}\n" + "\n" + "func (p Page) Title() string {\n" + "\treturn helper()\n" + "}\n" + "\n" + "func helper() string { return \"x\" }\n" + "\n" + "templ hello(name string) {\n" + "\t
{ name }
\n" + "}\n", + CBM_LANG_TEMPL, "t", "x.templ"); + ASSERT_NOT_NULL(r); + const char *title_qn = NULL; + for (int i = 0; i < r->defs.count; i++) { + const CBMDefinition *d = &r->defs.items[i]; + if (d->name && d->qualified_name && strcmp(d->name, "Title") == 0) { + title_qn = d->qualified_name; + } + } + const char *scope = NULL; + for (int i = 0; i < r->calls.count; i++) { + const CBMCall *c = &r->calls.items[i]; + if (c->callee_name && strcmp(c->callee_name, "helper") == 0) { + scope = c->enclosing_func_qn; + } + } + ASSERT_NOT_NULL(title_qn); + ASSERT_NOT_NULL(scope); + ASSERT_STR_EQ(scope, title_qn); + cbm_free_result(r); + PASS(); +} + /* TS/JS/TSX receiver-aware flag (#592/#606; same intent as the Perl flag above). * A member call x.foo() with a non-this/super receiver is flagged is_method so * the resolver can suppress a weak short-name match (`re.test()` must not bind a @@ -12574,6 +12691,9 @@ SUITE(extraction) { RUN_TEST(extract_python_member_call_flags_is_method); RUN_TEST(extract_python_bare_call_flags_locally_bound_callee); RUN_TEST(extract_python_bare_call_flag_is_depth_independent); + RUN_TEST(extract_go_selector_call_flags_is_method); + RUN_TEST(extract_go_method_receiver_qualified_qn); + RUN_TEST(extract_templ_method_qn_matches_call_scope); RUN_TEST(extract_ts_member_call_flags_is_method); RUN_TEST(extract_ts_this_super_receiver_not_flagged); RUN_TEST(extract_js_member_call_flags_is_method); diff --git a/tests/test_parallel.c b/tests/test_parallel.c index fc97332e6..f5952baf2 100644 --- a/tests/test_parallel.c +++ b/tests/test_parallel.c @@ -4080,8 +4080,9 @@ TEST(parallel_go_cross_package_field_chain_resolves) { cbm_gbuf_t *gbuf = run_go_field_chain_sequential("go_field_fold", tmpdir, files, 2); ASSERT_NOT_NULL(gbuf); - const cbm_gbuf_edge_t *edge = - find_call_edge_to_target_fragment(gbuf, "handler.PlaceOrder", ".service.PlaceOrder"); + /* Since #1909, Go method QNs are receiver-qualified on both endpoints. */ + const cbm_gbuf_edge_t *edge = find_call_edge_to_target_fragment( + gbuf, "OrderHandler.PlaceOrder", ".service.OrderService.PlaceOrder"); const bool found = edge != NULL; const bool dispatch = edge && edge->properties_json && strstr(edge->properties_json, "\"strategy\":\"lsp_type_dispatch\""); diff --git a/tests/test_pipeline.c b/tests/test_pipeline.c index 9a756f688..4ebf40693 100644 --- a/tests/test_pipeline.c +++ b/tests/test_pipeline.c @@ -8761,6 +8761,283 @@ TEST(pipeline_html_embedded_member_call_stays_unbound) { PASS(); } +/* CALLS edges whose source and target QNs end with the given suffixes and + * whose serialized properties contain `props_fragment` (NULL: any edge). QN + * suffixes pin the exact node, not just a same-named one. */ +static int go_calls_between(cbm_store_t *s, const char *project, const char *caller_suffix, + const char *callee_suffix, const char *props_fragment) { + cbm_edge_t *edges = NULL; + int edge_count = 0; + cbm_store_find_edges_by_type(s, project, "CALLS", &edges, &edge_count); + size_t cs = strlen(caller_suffix); + size_t ts = strlen(callee_suffix); + int hits = 0; + for (int i = 0; i < edge_count; i++) { + cbm_node_t src = {0}; + cbm_node_t tgt = {0}; + if (cbm_store_find_node_by_id(s, edges[i].source_id, &src) == CBM_STORE_OK && + cbm_store_find_node_by_id(s, edges[i].target_id, &tgt) == CBM_STORE_OK && + src.qualified_name && tgt.qualified_name) { + size_t sl = strlen(src.qualified_name); + size_t tl = strlen(tgt.qualified_name); + if (sl >= cs && tl >= ts && strcmp(src.qualified_name + sl - cs, caller_suffix) == 0 && + strcmp(tgt.qualified_name + tl - ts, callee_suffix) == 0 && + (!props_fragment || + (edges[i].properties_json && strstr(edges[i].properties_json, props_fragment)))) { + hits++; + } + } + cbm_node_free_fields(&src); + cbm_node_free_fields(&tgt); + } + if (edges) { + cbm_store_free_edges(edges, edge_count); + } + return hits; +} + +TEST(pipeline_go_method_caller_keeps_lsp_join) { + /* #1909: receiver-qualifying method QNs moves the def/textual-call QN to + * package.Type.method — go_lsp's enclosing-function QN must move with it + * (the ONE-formula contract), or every LSP resolution sourced from inside + * a method body loses its caller join and the edge dies. The callee name + * collides across receivers so no short-name fallback can mask the loss. */ + char tmp[256]; + snprintf(tmp, sizeof(tmp), "/tmp/cbm_go_mcall_XXXXXX"); + if (!cbm_mkdtemp(tmp)) { + FAIL("tmpdir"); + } + write_temp_file(tmp, "go.mod", "module example.com/fxmcall\n\ngo 1.22\n"); + write_temp_file(tmp, "svc/repo.go", + "package svc\n" + "\n" + "type Repo struct {\n" + "\tn int\n" + "}\n" + "\n" + "func (r *Repo) Install() int {\n" + "\treturn r.helper()\n" + "}\n"); + write_temp_file(tmp, "svc/helper.go", + "package svc\n" + "\n" + "func (r *Repo) helper() int {\n" + "\treturn r.n\n" + "}\n"); + write_temp_file(tmp, "other/other.go", + "package other\n" + "\n" + "type Decoy struct{}\n" + "\n" + "func (d *Decoy) helper() int { return 2 }\n"); + + char db_path[512]; + snprintf(db_path, sizeof(db_path), "%s/go_mcall.db", tmp); + cbm_pipeline_t *p = cbm_pipeline_new(tmp, db_path, CBM_MODE_FULL); + ASSERT_NOT_NULL(p); + ASSERT_EQ(cbm_pipeline_run(p), 0); + const char *project = cbm_pipeline_project_name(p); + + cbm_store_t *s = cbm_store_open_path(db_path); + ASSERT_NOT_NULL(s); + ASSERT_TRUE(cross_file_call_exists(s, project, "Install", "helper")); + /* Pinned to the receiver-qualified target in svc/helper.go and to an LSP + * strategy: a name-only binding, or one onto other.Decoy.helper, would + * still satisfy the by-name check above. */ + ASSERT_GTE(go_calls_between(s, project, ".svc.Repo.Install", ".svc.Repo.helper", + "\"strategy\":\"lsp_"), + 1); + ASSERT_EQ(go_calls_between(s, project, ".svc.Repo.Install", ".other.Decoy.helper", NULL), 0); + cbm_store_close(s); + cbm_pipeline_free(p); + th_rmtree(tmp); + PASS(); +} + +/* Fixture for the #1906 Go selector-guard probes (sequential and parallel + * twins). pad_files > 0 adds filler Go files to push the index over + * MIN_FILES_FOR_PARALLEL, because the guard is wired into both resolvers + * (pass_calls.c and pass_parallel.c) and each needs its own probe. */ +static void write_go_receiver_fixture(const char *tmp, int pad_files) { + /* go.mod makes project imports resolvable — real Go repos always have one, + * and import reachability (the unique_name penalty) depends on it. */ + write_temp_file(tmp, "go.mod", "module example.com/myapp\n\ngo 1.22\n"); + /* The lone project symbol named "Close" — a real method. */ + write_temp_file(tmp, "storage/storage.go", + "package storage\n" + "\n" + "type Storage struct{ open bool }\n" + "\n" + "func NewStorage() *Storage { return &Storage{open: true} }\n" + "\n" + "func (s *Storage) Close() {\n" + "\ts.open = false\n" + "}\n" + "\n" + "func Boot() {\n" + "\ts := NewStorage()\n" + "\ts.Close()\n" + "}\n"); + /* Cross-package control target: imported by hash.go, so the caller file + * has a non-empty import map (like any real Go file) and unreachable + * unique_name candidates get the import penalty. */ + write_temp_file(tmp, "util/util.go", + "package util\n" + "\n" + "func Tag() string { return \"t\" }\n"); + /* Stdlib receiver: `f.Close()` closes an *os.File, NOT the project method. + * The Go LSP cannot bind it to a project symbol → the registry would guess + * Close by short name (weak). This is the false edge to suppress — + * the exact shape that attached every file/rows/gzip Close in a real Go + * repo to one unrelated project method. */ + write_temp_file(tmp, "hash/hash.go", + "package hash\n" + "\n" + "import (\n" + "\t\"os\"\n" + "\n" + "\t\"example.com/myapp/util\"\n" + ")\n" + "\n" + "func FileLen(path string) int64 {\n" + "\tf, err := os.Open(path)\n" + "\tif err != nil {\n" + "\t\treturn 0\n" + "\t}\n" + "\tdefer f.Close()\n" + "\tst, err := f.Stat()\n" + "\tif err != nil {\n" + "\t\treturn 0\n" + "\t}\n" + "\treturn st.Size()\n" + "}\n" + "\n" + "func localHelper() int { return 1 }\n" + "\n" + "func CallsLocal() int { return localHelper() }\n" + "\n" + "func UsesUtil() string { return util.Tag() }\n"); + /* F-G9: two project methods named Run, one under a directory whose QN + * carries an uppercase T. `t.Run` on a *testing.T must bind neither. The + * parallel resolver field-type hint matched the capitalized receiver + * variable ("T") as a QN substring and produced a par-only edge. */ + write_temp_file(tmp, "TBench/bench.go", + "package tbench\n" + "\n" + "type Bench struct{}\n" + "\n" + "func (b *Bench) Run(name string) bool { return name != \"\" }\n"); + write_temp_file(tmp, "worker/worker.go", + "package worker\n" + "\n" + "type Worker struct{}\n" + "\n" + "func (w *Worker) Run(name string) bool { return false }\n"); + write_temp_file(tmp, "hash/hash_test.go", + "package hash\n" + "\n" + "import \"testing\"\n" + "\n" + "func TestSubRun(t *testing.T) {\n" + "\tt.Run(\"a\", func(t *testing.T) {})\n" + "}\n"); + for (int i = 0; i < pad_files; i++) { + char name[64]; + char body[128]; + snprintf(name, sizeof(name), "filler/filler%d.go", i); + snprintf(body, sizeof(body), "package filler\n\nfunc Filler%d() int { return %d }\n", i, i); + write_temp_file(tmp, name, body); + } +} + +TEST(pipeline_go_receiver_suppresses_weak_method_edge) { + char tmp[256]; + snprintf(tmp, sizeof(tmp), "/tmp/cbm_go_recv_XXXXXX"); + if (!cbm_mkdtemp(tmp)) { + FAIL("tmpdir"); + } + write_go_receiver_fixture(tmp, 0); + + char db_path[512]; + snprintf(db_path, sizeof(db_path), "%s/go_recv.db", tmp); + cbm_pipeline_t *p = cbm_pipeline_new(tmp, db_path, CBM_MODE_FULL); + ASSERT_NOT_NULL(p); + ASSERT_EQ(cbm_pipeline_run(p), 0); + const char *project = cbm_pipeline_project_name(p); + + cbm_store_t *s = cbm_store_open_path(db_path); + ASSERT_NOT_NULL(s); + + /* (1) The false edge is suppressed (reproduce-first: RED before the fix). */ + ASSERT_FALSE(cross_file_call_exists(s, project, "FileLen", "Close")); + /* (2) The same-package typed-receiver call survives (LSP / same_module — + * both outside the weak drop-list). */ + ASSERT_TRUE(cross_file_call_exists(s, project, "Boot", "Close")); + /* (3) The bare local call survives (is_method stays false for bare calls). */ + ASSERT_TRUE(cross_file_call_exists(s, project, "CallsLocal", "localHelper")); + /* (4) The import-qualified cross-package call survives (import-aware + * strategies are outside the drop-list). */ + ASSERT_TRUE(cross_file_call_exists(s, project, "UsesUtil", "Tag")); + /* (5) F-G9: the stdlib *testing.T Run binds no project Run. */ + ASSERT_FALSE(cross_file_call_exists(s, project, "TestSubRun", "Run")); + + cbm_store_close(s); + cbm_pipeline_free(p); + th_rmtree(tmp); + PASS(); +} + +/* Parallel twin of the test above. The guard is wired into pass_parallel.c as + * well as pass_calls.c, but the four-file fixture only ever runs the + * sequential resolver, so the parallel hunk was bound by no test. Real + * repositories take the parallel path: >= MIN_FILES_FOR_PARALLEL (50) files + * plus CBM_WORKERS=4 force it, the shape of + * pipeline_python_receiver_parallel_suppresses_weak_method_edges. Same + * both-directions pin as the sequential test. */ +TEST(pipeline_go_receiver_parallel_suppresses_weak_method_edge) { + char tmp[256]; + snprintf(tmp, sizeof(tmp), "/tmp/cbm_go_recv_par_XXXXXX"); + if (!cbm_mkdtemp(tmp)) { + FAIL("tmpdir"); + } + write_go_receiver_fixture(tmp, 52); + + char *old_workers = getenv("CBM_WORKERS"); + char *saved = old_workers ? strdup(old_workers) : NULL; + cbm_setenv("CBM_WORKERS", "4", 1); + + char db_path[512]; + snprintf(db_path, sizeof(db_path), "%s/go_recv_par.db", tmp); + cbm_pipeline_t *p = cbm_pipeline_new(tmp, db_path, CBM_MODE_FULL); + ASSERT_NOT_NULL(p); + ASSERT_EQ(cbm_pipeline_run(p), 0); + const char *project = cbm_pipeline_project_name(p); + + cbm_store_t *s = cbm_store_open_path(db_path); + ASSERT_NOT_NULL(s); + + /* NEGATIVE: the stdlib *os.File Close must not bind the project method. */ + ASSERT_FALSE(cross_file_call_exists(s, project, "FileLen", "Close")); + /* POSITIVE: typed same-package, bare local and import-qualified calls + * survive the parallel resolver too. */ + ASSERT_TRUE(cross_file_call_exists(s, project, "Boot", "Close")); + ASSERT_TRUE(cross_file_call_exists(s, project, "CallsLocal", "localHelper")); + ASSERT_TRUE(cross_file_call_exists(s, project, "UsesUtil", "Tag")); + /* F-G9: the parallel-only field-type hint must not bind `t.Run` either. */ + ASSERT_FALSE(cross_file_call_exists(s, project, "TestSubRun", "Run")); + + cbm_store_close(s); + cbm_pipeline_free(p); + if (saved) { + cbm_setenv("CBM_WORKERS", saved, 1); + free(saved); + } else { + cbm_unsetenv("CBM_WORKERS"); + } + th_rmtree(tmp); + PASS(); +} + /* Fixture for the #1928 cross-language reference-guard probes (sequential and * parallel twins). pad_files > 0 adds filler files to push the index over the * parallel-pipeline threshold, since USAGE/WRITES/READS have one resolver per @@ -24830,6 +25107,9 @@ SUITE(pipeline) { RUN_TEST(pipeline_python_receiver_suppresses_weak_method_edge); RUN_TEST(pipeline_python_receiver_keeps_specific_unique_name_member_call); RUN_TEST(pipeline_html_embedded_member_call_stays_unbound); + RUN_TEST(pipeline_go_method_caller_keeps_lsp_join); + RUN_TEST(pipeline_go_receiver_suppresses_weak_method_edge); + RUN_TEST(pipeline_go_receiver_parallel_suppresses_weak_method_edge); RUN_TEST(pipeline_go_rw_usage_never_cross_into_c); RUN_TEST(pipeline_go_rw_usage_never_cross_into_c_parallel); RUN_TEST(pipeline_rust_call_never_binds_another_language); diff --git a/tests/test_registry.c b/tests/test_registry.c index 0f3e805f2..391fd3b88 100644 --- a/tests/test_registry.c +++ b/tests/test_registry.c @@ -1673,10 +1673,61 @@ TEST(weak_call_guards_share_one_drop_list) { strategies[i], member, binding); } ASSERT_EQ(member, binding); + /* The Go selector guard builds on the same list. At a penalized + * confidence (0.75 * 0.5) it agrees with it exactly; its one narrowing + * (an unpenalized unique_name) does not apply here. Any drift from + * the shared list fails here. */ + bool go = cbm_go_suppress_weak_method_match(true, true, strategies[i], 0.375); + bool go_expected = member; + if (go != go_expected) { + printf(" Go drop-list divergence on strategy \"%s\": go=%d expected=%d\n", + strategies[i], go, go_expected); + } + ASSERT_EQ(go, go_expected); } PASS(); } +TEST(go_suppress_drops_weak_selector_matches) { + /* Go selector call with an untyped receiver, landed via a receiver-blind + * short-name strategy → drop (same failure class as #592/#606). + * suffix_match/fuzzy drop at any confidence; unique_name drops only when + * import-unreachability-penalized (CONF_UNIQUE_NAME 0.75 * 0.5 = 0.375 — + * the `io.Copy` -> project `Copy` stdlib-hijack shape). */ + ASSERT_TRUE(cbm_go_suppress_weak_method_match(true, true, "suffix_match", 0.9)); + ASSERT_TRUE(cbm_go_suppress_weak_method_match(true, true, "suffix_match", 0.11)); + ASSERT_TRUE(cbm_go_suppress_weak_method_match(true, true, "fuzzy", 0.9)); + ASSERT_TRUE(cbm_go_suppress_weak_method_match(true, true, "unique_name", 0.375)); + /* field_type_hint is a receiver-NAME substring match, not type evidence + * (F-G9): weak for Go like the shared list. */ + ASSERT_TRUE(cbm_go_suppress_weak_method_match(true, true, "field_type_hint", 0.85)); + PASS(); +} + +TEST(go_suppress_keeps_typed_and_import_aware_matches) { + /* Unpenalized unique_name = lone candidate inside the caller's import + * closure (or an import-free file, e.g. same-package) — a genuinely-typed + * lone-candidate call never enters the field-type-hint upgrade, so it must + * survive (lrp_go_s8_field_type_hint). */ + ASSERT_FALSE(cbm_go_suppress_weak_method_match(true, true, "unique_name", 0.75)); + ASSERT_FALSE(cbm_go_suppress_weak_method_match(true, true, "same_module", 0.9)); + ASSERT_FALSE(cbm_go_suppress_weak_method_match(true, true, "import_map", 0.95)); + ASSERT_FALSE(cbm_go_suppress_weak_method_match(true, true, "import_map_suffix", 0.9)); + ASSERT_FALSE(cbm_go_suppress_weak_method_match(true, true, "qualified_suffix", 0.9)); + ASSERT_FALSE(cbm_go_suppress_weak_method_match(true, true, "callee_suffix", 0.5)); + ASSERT_FALSE(cbm_go_suppress_weak_method_match(true, true, "service_pattern", 0.5)); + ASSERT_FALSE(cbm_go_suppress_weak_method_match(true, true, "lsp_strategy_cross_file", 0.92)); + ASSERT_FALSE(cbm_go_suppress_weak_method_match(true, true, "lsp_direct", 0.95)); + /* A bare call (is_method=false) is a free-function call → never suppressed. */ + ASSERT_FALSE(cbm_go_suppress_weak_method_match(true, false, "suffix_match", 0.11)); + /* Non-Go languages are never affected by this gate. */ + ASSERT_FALSE(cbm_go_suppress_weak_method_match(false, true, "suffix_match", 0.11)); + /* No match (NULL/empty strategy) → nothing to suppress. */ + ASSERT_FALSE(cbm_go_suppress_weak_method_match(true, true, NULL, 0.5)); + ASSERT_FALSE(cbm_go_suppress_weak_method_match(true, true, "", 0.5)); + PASS(); +} + /* ── Suite ─────────────────────────────────────────────────────── */ /* Method call THROUGH an imported symbol that is itself an indexed node @@ -2235,6 +2286,8 @@ SUITE(registry) { RUN_TEST(python_import_binding_contradicts_only_foreign_chains); RUN_TEST(python_external_base_contradicts_django_shapes); RUN_TEST(weak_call_guards_share_one_drop_list); + RUN_TEST(go_suppress_drops_weak_selector_matches); + RUN_TEST(go_suppress_keeps_typed_and_import_aware_matches); RUN_TEST(registry_python_function_local_is_exact_only_issue1548); RUN_TEST(registry_resolve_lexical_walks_enclosing_defs_issue1548); RUN_TEST(registry_python_local_bound_at_own_file_issue1548);