fix(extract): descend into Go struct field_declaration_list by ilyabrykau-orca · Pull Request #1940 · DeusData/codebase-memory-mcp

ilyabrykau-orca · 2026-08-30T12:41:41Z

Fixes #1935. Stacked on #1937 (the USAGE/WRITES cross-language guard) — the issue's own measurement shows why this must not land alone: ~4600 recovered Field nodes named err/ctx/path/_ would otherwise mint ~5700 WRITES / ~22000 USAGE name-collision edges through the unguarded reference resolver. The two tip commits + the test are this PR's own change; the base commit is #1937's.

What

Go struct fields were never extracted — 0 Field nodes for ~5055 field declarations on the measured repo, while the 850 Field nodes in the graph were all C/C++. find_class_body() returns type_spec's type child:

interface: interface_type holds its method specs directly → always worked;
struct: struct_type's only named child is a field_declaration_list, one level above the field_declaration nodes the member loop looks for → every field silently skipped. go_field_types[] = {"field_declaration"} was already registered in lang_specs.c; the extractor just never reached them.

How

go_normalize_struct_body() descends struct_type → field_declaration_list before extract_class_fields() iterates, the same normalization step the Java enum_body_declarations path already does. A second commit skips the Go blank identifier _ (struct padding in generated code — not a referenceable field; 55 nodes that collected 241 collision edges).

Measured effect (from the issue)

0 → 4588 Go Field nodes, each carrying its declared type in return_type. Embedded fields (~208) stay unrepresented by construction (no name child) — separate follow-up; lsp_embed_dispatch is unaffected. With #1937 under it, CALLS onto Go fields stay guarded (2466 → 417 by #1907's gate in the issue's stacked measurement) and the WRITES/USAGE collision classes are dropped by the reference guard. Also unblocks func-typed-field call dispatch (r.fn(4) → Router.fn), which needs the field node to exist.

Tests

Reproduce-first extract_go_struct_fields_have_nodes: RED with the descend fix reverted — count_defs_with_label(r, "Field") == 0, expected 3 — GREEN on the branch. Asserts the three named fields with declared types (string, int, *Config), the blank identifier's absence, and that interface members keep extracting exactly as before.
Full scripts/test.sh venue leg (ASan+UBSan, all suites + contract steps): green on the stack. git clang-format --diff: clean.

#1932 tracks the family; ordering constraint (#1928 first) is from the issue's own two-index measurement.

USAGE, WRITES and READS edges resolve through the same short-name registry as CALLS but never consulted the DeusData#725 cross-language guard. On a Go tree with eBPF C probes every Go identifier spelled like a C one produced a reference edge into the C file: 31.5% of all WRITES on the originally measured repo crossed the Go->C boundary, led by dozens of Go test locals named event writing a C probe's automatic variable. Add cbm_suppress_cross_language_ref() - the reference-edge analog of cbm_suppress_cross_language_suffix_match - and consult it on BOTH resolvers of each edge type: the sequential pass (pass_usages.c resolve_usage_edges registry-fallback branch, resolve_rw_edges) and their parallel twins (pass_parallel.c resolve_file_usages, resolve_file_rw). The sequential-only version of this change left 344 Go->C WRITES alive on a ~1150-file repo because large repos resolve through pass_parallel.c - the field census caught it, and the parallel-twin test now pins it. Unlike the CALLS guard the predicate takes no strategy parameter: a reference edge carries no import-closure evidence, so every registry strategy is a bare-name guess across a boundary. LSP-backed semantic references resolve before the fallback and are unaffected. JS/TS stay one family, and C/C++ count as one family too (.h maps to CBM_LANG_CPP, so a .c file referencing its own header is not a boundary). Field-validated on the ~1150-file Go+C repo: Go->C/C++ WRITES 835->0, USAGE 1545->0; C->Go 15/90->0; 6432 reference edges dropped in total, every one cross-language (Go->.json 3034, Go->.hpp/.h 1612, Go->.c 746, Go->.sh/.yaml/.yml 443, ...) and none same-language. CALLS and IMPORTS totals are byte-identical to main. Fixes DeusData#1928 Signed-off-by: Ilya Brykau <ilya.brykau@orca.security>

Signed-off-by: Ilya Brykau <ilya.brykau@orca.security>

Reproduce-first probe for the two fixes on this branch: RED without the descend fix (count_defs_with_label(r, "Field") == 0, expected 3), GREEN with it. Asserts the three named fields with their declared types in return_type, the absence of the blank identifier, and that interface members keep extracting exactly as before. Signed-off-by: Ilya Brykau <ilya.brykau@orca.security>

github-actions · 2026-08-30T13:09:30Z

Thanks for opening this — it has been seen, and it is queued.

This note is automated, but it is not a brush-off: it exists so you know where your PR stands instead of having to guess from silence.

Current review status: working through a backlog. 0.9.1-rc.1 is out, so the release freeze that held reviews is over — but it left a large queue of open pull requests behind it, and we are reading through them oldest-first. The background is in discussion #1144.

What that means for this PR, concretely:

It will not be closed for inactivity. No stale bot touches pull requests here.
It may still sit a while before a human reads it. That is on us, not on you.
Older PRs are read first, so a recent one is not being skipped — it is behind a queue.

Things that will genuinely speed it up whenever review does happen:

Keep it rebased on main — the tree is moving quickly right now, and a conflicting branch cannot be reviewed as the diff you intended.
Get CI green, or say which failures you believe are pre-existing.
Keep the change to one claim. Bundled features and refactors get split before they get merged, which costs you a round trip.
Every commit needs a sign-off (git commit -s) — CI enforces DCO.

If this fixes a bug, a reproduction we can run is worth more than a description of the symptom.

Thanks for contributing, and sorry in advance for the wait.

ilyabrykau-orca · 2026-08-30T13:22:11Z

Rebased onto the amended #1937 (its guard now covers the parallel resolvers). No semantic changes to this PR's commits; extraction+pipeline suites green, full venue leg green on the stack.

Heads-up from field validation: with #1937 under it, this fix still surfaces ~26k same-language bare-name reference edges onto the new Field nodes (a local err := … binding whichever field is named err). That is its own bounded defect — filed as #1942 with counts, fix PR incoming stacked on this branch. Suggested merge order: #1937 → this → #1942's PR (or the last two together).

The READS/WRITES resolvers and the USAGE registry fallback hand bare reference text to the short-name registry, which contains Field nodes - so once Go struct fields exist (DeusData#1935), every Go local err := ... binds whichever struct field is named err, project-wide: 21308 USAGE and 5191 WRITES onto Go fields on the measured repo, top target a test struct's field T collecting 3013 edges. In Go that binding is impossible by construction: a field is only reachable through a selector expression (x.f), and selector references resolve on the LSP path - every Field-targeted reference edge in the census carried dot-less text. Add cbm_go_suppress_bare_field_ref() next to the DeusData#1928 predicate and consult it at the same four sites (both READS/WRITES resolvers, both USAGE registry fallbacks): drop the bind when the file is Go, the target label is Field, and the reference text has no '.'. Go-gated because C#/Java/C++/Python method bodies legitimately reference their own members bare (cp_reads_writes_cs_static_field pins that shape). Field-validated on the DeusData#1940 stack: USAGE onto Go fields 21308 -> 0, WRITES 5191 -> 0; the only remaining field-targeted edges are 2466 CALLS, which are DeusData#1906/DeusData#1907's selector-guard territory. Reproduce- first pipeline probes (sequential + parallel twins) were RED on the stack without this commit. Fixes DeusData#1942 Signed-off-by: Ilya Brykau <ilya.brykau@orca.security>

ilyabrykau-orca requested a review from DeusData as a code owner August 30, 2026 12:41

This was referenced Aug 30, 2026

fix(extract): Go struct fields are never extracted — 0 Field nodes for ~5000 field declarations (structs miss one level of descent) #1935

Open

Meta: Go / cgo / Go+C extraction — tracking issue #1932

Open

ilyabrykau-orca added 4 commits August 30, 2026 14:53

fix(extract): descend into Go struct field_declaration_list

47116b8

Signed-off-by: Ilya Brykau <ilya.brykau@orca.security>

fix(extract): skip Go blank-identifier struct fields

cb7cb44

Signed-off-by: Ilya Brykau <ilya.brykau@orca.security>

ilyabrykau-orca mentioned this pull request Aug 30, 2026

fix(pipeline): a bare Go reference can bind a struct Field — READS/WRITES/USAGE attach every local err to whichever Field is named err #1942

Open

ilyabrykau-orca force-pushed the fix/go-struct-fields branch from 914aac0 to e07a12b Compare August 30, 2026 13:22

This was referenced Aug 30, 2026

fix(pipeline): a bare Go reference never binds a struct Field #1944

Open

fix(pipeline): USAGE and WRITES edges have no cross-language guard — 31% of WRITES on a Go+C repo cross the Go/C boundary #1928

Open

Navigation Menu

fix(extract): descend into Go struct field_declaration_list - #1940

fix(extract): descend into Go struct field_declaration_list#1940
ilyabrykau-orca wants to merge 4 commits into
DeusData:mainfrom
ilyabrykau-orca:fix/go-struct-fields

ilyabrykau-orca commented Aug 30, 2026

Uh oh!

github-actions Bot commented Aug 30, 2026

Uh oh!

ilyabrykau-orca commented Aug 30, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Conversation

ilyabrykau-orca commented Aug 30, 2026

What

How

Measured effect (from the issue)

Tests

Uh oh!

github-actions Bot commented Aug 30, 2026

Uh oh!

ilyabrykau-orca commented Aug 30, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant