tree-sitter-go puts the comment above `type Foo struct{}` before the
`type_declaration`, but the node is made from the spec inside it, so both
extractors found no docstring for an ungrouped struct, interface, defined
type or alias. A type inside a `type ( ... )` group kept its own comment,
which sits beside it in the parentheses.
Go's docstring is now read from the declaration when the declaration
holds one spec and nothing precedes that spec inside it. The leading
comment of a group with several members stays the group's.
Reading from the declaration exposed comments that trail the line above:
`const _ = proto.GoGoProtoPackageIsVersion3 // please upgrade the proto
package` would have become `type MetricType int32`'s doc. For every Go
declaration, a comment that begins after code on its line, and one that
begins on the line such a comment ends, is now left out, as go/parser
reads them. Seven function docstrings on etcd and prometheus lose one.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A Dart member's annotations are siblings that stand between it and the
dartdoc written above them:
/// Builds the widget.
@override
Widget build(BuildContext context) { ... }
Both extractors read a docstring from the comments directly before the
declaration (or its `declaration` wrapper) and stopped at the first node
that was not a comment, so the annotation ended the run before the `///`
was reached. Members behind `@override`, `@protected`, `@mustCallSuper`,
`@pragma(...)`, `@internal` and the like lost their documentation, and
searching for words in it didn't find them.
Extraction (TS + kernel, at parity):
- getPrecedingDocstring takes an optional list of sibling types to step
over, and Dart sets `docstringStepOverTypes: ['annotation']`. Stacked,
multi-line and inline annotations are passed, and comments on either
side of one still join (`/// a` `@x` `// b` gives "a" and "b" on two
lines), as adjacent comments already did. Any other node still ends
the walk: the previous member's body, a field, a top-level variable,
an import or another declaration.
- The kernel's shared docstring.rs gains preceding_docstring_stepping_over.
preceding_docstring passes an empty list, so every other language is
unchanged; only Dart's docstring_of passes ["annotation"].
- Class-like declarations needed nothing: a class, mixin, extension,
extension type, enum or typedef node starts at its first annotation,
so the dartdoc above it is already the node's previous sibling.
Validated on felangel/bloc @b9be1e2 and rrousselGit/riverpod @4ba1be2 at
main 249d9a86 (and at 1b752de2) with the kernel loaded: docstrings bloc
+26, riverpod +1,072 with 9 changed and none removed. Nodes otherwise,
edges, refs and files are byte-identical, full-index kernel and wasm
dumps are byte-identical, and the kernel parity sweeps show 0 diffs.
Not changed: a comment between an annotation and the declaration still
hides the annotation from the decorator scan. Recording those would only
add wrong edges until Dart `decorates` resolution stops preferring
methods: riverpod's `@riverpod` edges land on an analyzer getter rather
than `const riverpod`.
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
On CRLF checkouts (every Windows autocrlf clone) the JS reference's
block-continuation strip /^\s*\*\s?/gm finds a line start after the \r and
its greedy \s* consumes the \n, leaving a bare \r in the docstring; the
kernel's (?m)^ pass matched after \n only and kept \r\n. Caught by the O2
Windows VM leg (6 kernel-tsjs-parity failures), reproduced on macOS by
CRLF-converting the fixtures.
js_multiline_strip now replicates the JS anchor set (\n, \r, U+2028, U+2029)
for all five line-marker passes; CRLF variants of every torture fixture are
pinned in kernel-tsjs-parity, derived in-memory so nothing can normalize
them away.
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
Java joins the native kernel (codegraph-kernel/src/java.rs), mirroring
the wasm extractor's Java paths bug-for-bug: package namespaces,
imports, javadoc, annotations→decorates, type_list inheritance,
static-final constants, enum constants, anonymous classes (including
the TS side's 0-based-line quirk on the extends ref), method_invocation
calls with the this.field unwrap and the Foo.getInstance().bar() chain
encoding (#645/#608), static-member value reads, method-reference
fn-refs (#756), value-reference edges, and the full Lombok member
synthesizer (#912: Getter/Setter/Data/Value/Builder/ToString/
EqualsAndHashCode/Slf4j-family with taken-member dedup). The shared
docstring/textutil modules moved to crate level. Grammar:
tree-sitter-java 0.23.5, with the wasm grammar vendored from the same
tag (parser.c sha-matched) replacing tree-sitter-wasms' 2023-era build.
Gate (plan §4c): extraction sweeps 100% — gson 262/262, retrofit
341/341, dubbo 4,048/4,048 — plus a Java torture fixture in npm test;
full-init dump-diffs byte-identical on gson (49,766 rows), retrofit
(62,735), and dubbo (441,266 rows); all R2/R3 repos re-verified; Linux
container runs all 23 kernel tests green under CODEGRAPH_KERNEL_EXPECT=1.
The gate caught a real cross-language bug: fn-ref dedupe and value-ref
self-target checks must compare node ID STRINGS, not node-table rows —
ids collide for same-(kind, name, line) nodes, which minified one-line
bundles hit routinely (retrofit's website JS exposed it; latent in the
TS/JS walker since R2, never released). Fixed in both walkers.
Benchmark honesty: dubbo fresh-init on an 11-core Mac is ~flat
(parse-loop wall 5,020→4,394ms; total ~11.3s both arms) because that
wall is main-thread-bound (reads + store), not worker-CPU-bound — the
§6 expectation assumed otherwise. Where worker CPU binds the kernel
delivers: dubbo on a 2-CPU/6GB container drops 27.8-28.6s → 22.3-22.8s
(~1.25×). The identified lever for the many-core headline is decoding
kernel buffers directly into store rows (skipping per-node JS object
materialization); the buffer contract already carries everything.
DEFAULT_ROUTED now includes java. Full suite: 2,467 tests pass.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
