mirror of
https://github.com/DeusData/codebase-memory-mcp.git
synced 2026-11-04 10:48:20 +00:00
Martin Vogel
4d97d7819b
The Go 1.22 pattern split ran for every route-verb callee, so other Go
calls whose name ends in Get/Post/Put/Delete/Patch became route
registrations: st.Get("users/123") on a project method minted
__route__GET__/123 and lost its CALLS edge to Store.Get, and
c.Get("https://api.example.com/v1/x") split "https:" off as a host and
minted __route__GET__//api.example.com/v1/x, in both passes.
cbm_service_pattern_mux_route_path reads the pattern only for a callee
ending ".Handle" or ".HandleFunc", and never for an argument containing
"://". The extractor's handler capture and both passes (route_reg_path,
find_route_path_in_args, cbm_pipeline_route_reg_method) go through it,
so the two passes still mint identical Routes. Two negative cases
extend pipeline_go_servemux_method_patterns_issue2537 in both passes,
and the infrascan unit covers the gate.
Refs #2537
Signed-off-by: Martin Vogel <martin.vogel.tech@gmail.com>
124 lines
6.7 KiB
C
124 lines
6.7 KiB
C
/*
|
||
* service_patterns.h — Allowlists for HTTP clients, async dispatch, and config accessors.
|
||
*
|
||
* Used during call resolution to classify CALLS edges as:
|
||
* HTTP_CALLS — synchronous HTTP client calls
|
||
* ASYNC_CALLS — async message/task dispatch
|
||
* CONFIGURES — config/env access
|
||
*
|
||
* Lookup is O(1) via hash table initialized once at startup.
|
||
*/
|
||
#ifndef CBM_SERVICE_PATTERNS_H
|
||
#define CBM_SERVICE_PATTERNS_H
|
||
|
||
#include <stdbool.h>
|
||
#include <stddef.h>
|
||
|
||
/* Edge type returned by pattern match. */
|
||
typedef enum {
|
||
CBM_SVC_NONE = 0, /* Not a service pattern — use normal CALLS */
|
||
CBM_SVC_HTTP = 1, /* Synchronous HTTP client call */
|
||
CBM_SVC_ASYNC = 2, /* Async dispatch (message broker, task queue) */
|
||
CBM_SVC_CONFIG = 3, /* Config/env accessor */
|
||
CBM_SVC_ROUTE_REG = 4, /* Route registration (router.GET, app.get, Route::post) */
|
||
CBM_SVC_GRPC = 5, /* gRPC client call (protobuf stub invocation) */
|
||
CBM_SVC_GRAPHQL = 6, /* GraphQL client query/mutation */
|
||
CBM_SVC_TRPC = 7, /* tRPC client procedure call */
|
||
} cbm_svc_kind_t;
|
||
|
||
/* Initialize the pattern lookup tables. Call once at startup. Thread-safe after init. */
|
||
void cbm_service_patterns_init(void);
|
||
|
||
/* Check if a resolved QN contains a known service library identifier.
|
||
* Returns the pattern kind, or CBM_SVC_NONE if no match.
|
||
* Matches on library name substrings in the QN (e.g., "requests" in
|
||
* "project.venv.requests.api.get"). Import-alias transparent. */
|
||
cbm_svc_kind_t cbm_service_pattern_match(const char *resolved_qn);
|
||
|
||
/* True for a bare, unqualified call to the native `fetch()` API. Deliberately
|
||
* NOT part of the substring tables above: those are matched unconditionally
|
||
* against the raw callee name before/regardless of registry resolution
|
||
* (the #523 external-library bypass), and "fetch" — unlike "axios" or
|
||
* "requests" — collides with a plausible local identifier. Callers must
|
||
* only consult this after registry resolution has come back empty, so a
|
||
* locally resolvable `function fetch(){}` / `const fetch = () => {}` is
|
||
* classified via its real resolved QN instead and never reaches this check. */
|
||
bool cbm_service_pattern_is_global_fetch(const char *callee_name);
|
||
|
||
/* True when a string literal is a plausible HTTP route for the given callee.
|
||
* Rejects filesystem paths and non-HTTP string consumers. */
|
||
bool cbm_service_pattern_is_http_route_literal(const char *literal, const char *callee_name);
|
||
|
||
/* True when a JS/TS-family call is client-side navigation rather than a
|
||
* network request (#1250): Angular Router.navigate/navigateByUrl, Ionic navigation, Nuxt
|
||
* navigateTo, react-router navigate, router/history push|replace|prefetch,
|
||
* History pushState/replaceState, Location go|assign|replace, bare goto /
|
||
* redirect / permanentRedirect. Its slash-prefixed argument is a client route,
|
||
* so it must not mint an arg_url HTTP_CALLS edge. js_family is the caller's
|
||
* FILE-language gate: false always returns false, so no other language's
|
||
* recall is traded (suppressors stay per-language). */
|
||
bool cbm_service_pattern_is_client_navigation(bool js_family, const char *callee_name);
|
||
|
||
/* True when text that reached a path slot is a source comment rather than a
|
||
* path: a block comment (slash-star … star-slash), a line comment (slash-slash
|
||
* followed by a space), or any text with a line break. The wildcard route
|
||
* (slash-star alone) is a path. */
|
||
bool cbm_service_pattern_is_comment_text(const char *text);
|
||
|
||
/* Per-worker TLS cache for cbm_service_pattern_match results. The
|
||
* pattern matcher runs once per resolved CALL edge in emit_service_
|
||
* edge — that's 6 pattern lists × ~30 patterns × strstr per call ≈
|
||
* ~180 strstrs per call. The same resolved QN repeats across most of
|
||
* the call edges in a project (e.g. "fmt.Errorf"), so caching turns
|
||
* a linear pattern-list scan into one hash lookup. Call _begin once
|
||
* per worker thread before the resolve loop and _end at the end. */
|
||
void cbm_service_pattern_cache_begin(void);
|
||
void cbm_service_pattern_cache_end(void);
|
||
|
||
/* Get the HTTP method from the callee name suffix (e.g., ".get" → "GET").
|
||
* Returns NULL if method cannot be inferred. */
|
||
const char *cbm_service_pattern_http_method(const char *callee_name);
|
||
|
||
/* Get the HTTP method from a route registration callee name suffix
|
||
* (e.g., "router.GET" → "GET", "app.post" → "POST").
|
||
* Returns NULL if not a known route registration method. */
|
||
const char *cbm_service_pattern_route_method(const char *callee_name);
|
||
|
||
/* Go 1.22+ net/http ServeMux pattern "[METHOD ][HOST]/[PATH]" (#2537):
|
||
* "POST /v1/x", "GET api.example.com/v1/x", "api.example.com/v1/x". Returns a
|
||
* pointer to PATH's leading slash inside `pattern`, or NULL when the pattern
|
||
* names neither a method nor a host before a slash (a plain "/v1/x" is NULL
|
||
* too: there is nothing to split). METHOD is a run of uppercase letters ended
|
||
* by spaces or tabs; it is copied to `method` (NUL-terminated, "" when the
|
||
* pattern names none) when `method` is non-NULL; a method that does not fit
|
||
* `method_sz` then makes the pattern NULL. The grammar is Go's: callers gate
|
||
* on the file language. */
|
||
const char *cbm_service_pattern_mux_split(const char *pattern, char *method, size_t method_sz);
|
||
|
||
/* The path of a ServeMux registration's pattern (#2537): only a callee ending
|
||
* ".Handle" or ".HandleFunc" registers one, and never with a URL ("://").
|
||
* Another route-verb call names no pattern: `store.Get("users/123")` is no
|
||
* host "users" + path "/123", `client.Get("https://...")` no host "https:".
|
||
* The split of cbm_service_pattern_mux_split for such a registration, else
|
||
* NULL. The grammar is Go's: callers gate on the file language. */
|
||
const char *cbm_service_pattern_mux_route_path(const char *callee_name, const char *pattern);
|
||
|
||
/* Get the broker name for an async QN (e.g., "pubsub" from a Pub/Sub QN).
|
||
* Returns NULL if not an async pattern. */
|
||
const char *cbm_service_pattern_broker(const char *resolved_qn);
|
||
|
||
/* GraphQL operation identity of a client call argument (#598).
|
||
* Scans `doc` as a GraphQL document (skipping comments, strings, `${...}`
|
||
* template interpolations and leading fragment definitions) and reports the
|
||
* first operation: *op_type is "query", "mutation" or "subscription" ("query"
|
||
* for the `{ ... }` shorthand). Returns true and writes the operation name to
|
||
* name_buf when the operation is named. Returns false for an anonymous
|
||
* operation, or with *op_type = "operation" when `doc` is not a GraphQL
|
||
* document (NULL, a URL, a bare word such as a dict key ...). The operation
|
||
* text itself is never copied, so callers can key Routes on a bounded
|
||
* identity. */
|
||
bool cbm_service_pattern_graphql_operation(const char *doc, const char **op_type, char *name_buf,
|
||
size_t name_sz);
|
||
|
||
#endif /* CBM_SERVICE_PATTERNS_H */
|