track(2.0): breaking-change roadmap for the 2.x major bump

Summary

Tracking issue for the 2.x major bump. A major bump is the only SemVer
window in which the breaking changes below may land. This collects:

the loose ends already documented in
STABILITY.md under "On the 2.0 horizon", and
additional candidates surfaced by a pre-2.0 project review (linked under
Review findings below).

2.0 is not scheduled; this is the staging ground so nothing is forgotten
when it is cut. All issues here are filed against the 2.0 milestone.

Known loose ends (from STABILITY.md)

Mark the per-metric Stats structs #[non_exhaustive] so field
additions stop being a strict-SemVer shape break.
Remove the deprecated metrics / metrics_with_options shims (in
favour of analyze).
Decide the fate of the remaining #[doc(hidden)] extension traits
(ParserTrait, the per-metric compute traits, Parser<T>, Filter):
delete or formally expose. The hidden state is a holding pattern.
Land the clean metric-value re-baseline note in the 2.0
CHANGELOG entry, folding in every value drift accumulated across 1.x
so consumers have one diff to reason about across the major boundary.
FilesData / ConcurrentRunner reshape into a terminal file-set
processor (refactor(cli): single anchored walk seam — library globsets are dead and files are double-stat'd after #489 #495 — already closed; confirm it is reflected at 2.0).

Review findings (filed separately)

Core library API:

fix(tools): guess_language returns unbound-lifetime &'a str (latent unsoundness) #506 — guess_language returns an unbound-lifetime &'a str (latent
unsoundness; tighten to &'static str or drop the tuple element) (bug) — ✅ landed on main
refactor(api): normalize &LANG-by-value and drop Java-style get_ getter prefixes #507 — normalize &LANG-by-value and drop Java-style get_ getter
prefixes across LANG / LanguageInfo / ParserTrait
feat(lang): implement Display/FromStr/Hash for the LANG enum #508 — implement Display / FromStr / Hash for LANG — ✅ landed on main
refactor(lib): get_ops/metrics_from_tree still key identity off a lossy path #509 — get_ops / metrics_from_tree still key identity off a lossy
path; add a Source-based seam and deprecate the path-positional form
refactor(vcs): decide vcs::Options exhaustiveness (non_exhaustive/builder) before 2.0 #584 — decide vcs::Options construction contract
(#[non_exhaustive] / builder) so future additive VCS fields are not
external breaks; migrate CLI / web / Python construction sites

Core architecture (2.0-gated):

refactor(checker): reshape Checker trait (default predicates, is_primitive(&Node), narrow bounds) #520 — reshape the Checker trait: default bodies for trivial
predicates (kills per-language stub boilerplate), is_primitive(&Node)
instead of is_primitive(u16), narrow count_specific_ancestors's
bound from ParserTrait to Checker

Metrics & output:

refactor(output): normalize metric JSON/YAML key naming across all Stats #510 — normalize metric JSON/YAML key naming across all Stats
(sum/average/min/max scheme; misleading nom.functions /
nargs.total_functions; npa/npm average-as-ratio; abc magnitude
asymmetry; serialize_struct names)
fix(halstead): n1/N1/n2/N2 keys are case-only-distinct (CSV/header collisions) #511 — Halstead n1/N1/n2/N2 case-only-distinct keys collide for
case-insensitive consumers (CSV/env) (bug)
refactor(metrics): unify average-divisor guarding and denominator convention #512 — unify average-divisor guarding and per-space-vs-per-function
denominator convention (re-baseline)

CLI:

refactor(cli): unify output-format selection across subcommands #513 — unify output-format selection across subcommands (flag vs
positional, -O vs --format, enum consolidation)
fix(cli): reconcile metric-name spelling between diff --metric and check --threshold #514 — reconcile metric-name spelling between diff --metric and
check --threshold — ✅ landed on main
feat(cli): give --ls/--le descriptive long names; reconsider global scope #518 — give --ls/--le descriptive long names; reconsider global

Web (not strictly 2.0-gated — schema/HTTP, not library SemVer):

feat(web): version REST routes (/v1) before 2.0; revisit Option-in-200 errors #517 — version REST routes (/v1) before 2.0; revisit
Option-in-200-body error idiom

Filed without the 2.0 milestone (fix anytime — not breaking):

✅ fix(web): content-type guards reject 'application/json; charset=utf-8' (silent 404) #515 — web content-type guards reject application/json; charset=utf-8
(silent 404) (bug) — closed, landed on main
refactor(web): replace eprintln! error logging with tracing/log #516 — replace eprintln! web error logging with tracing/log
fix(groovy): closure bodies don't count toward cognitive lambda nesting #519 — (bug) Groovy closure bodies don't count toward cognitive lambda
nesting (cross-language parity violation; metric fix, lands pre-2.0)
✅ perf(node): has_sibling allocates a TreeCursor per call on the JS/TS hot path #521 — perf: Node::has_sibling allocates a TreeCursor per call on the
JS/TS hot path — closed, landed on main
✅ perf(spaces): Elixir per-node source scan runs regardless of metric selection #522 — perf: Elixir per-node source scan runs regardless of metric
selection — closed, landed on main
✅ refactor(node): children() iterator desyncs length (child_count) from data (cursor) #523 — Node::children() iterator desyncs length from data (latent
dup-yield) — closed, landed on main
✅ build(deps): exact-pin tree-sitter-kotlin-ng and CI-guard enums/ grammar-pin lockstep #524 — exact-pin tree-sitter-kotlin-ng; CI-guard enums/ grammar-pin
lockstep (dependencies) — closed, landed on main
✅ build(deps): drop num meta-crate; hoist csv/tempfile into [workspace.dependencies] #525 — drop the num meta-crate; hoist csv/tempfile to workspace deps
(dependencies) — closed, landed on main
✅ docs: stale supported-languages lists (phantom C/Mozcpp; deb/rpm omit 5 langs) #526 — stale supported-languages lists (phantom C/Mozcpp; deb/rpm omit
5 langs) (documentation) — closed, landed on main

Second-pass review findings (2026-06-05)

A fresh pre-2.0 review across the public Rust API, serialized output schema,
CLI grammar, REST surface, Python bindings, and cross-language metrics surfaced
the following, none overlapping the items above. All filed against the 2.0
milestone except where noted additive.

Bug (fix before the 2.0 re-baseline so values fold in cleanly):

fix(cognitive): PHP double-counts two-word else if form #529 — (bug) PHP cognitive double-counts the two-word else if form
(5.0 vs the correct 3.0; isolated to cognitive)

Serialized output schema:

refactor(output): serialize integer-valued metrics as integers, not f64 #530 — integer-valued metrics are typed f64 and serialize as floats
("sloc":5.0); four formats disagree; CBOR bloat
refactor(output): uniform non-finite & precision policy at the serialize boundary #531 — no uniform non-finite (NaN/Inf→silent JSON null) or precision
policy at the serialize boundary
feat(lib): implement Deserialize on the serialization-root types #532 — implement Deserialize on the serialization-root types (the
library can't read its own output)
refactor(output): normalize AstNode dump key casing; name the Span tuple #535 — normalize AstNode dump key casing (Type/TextValue/…); name
the Span 4-tuple
test(output): add round-trip coverage for TOML/YAML/CBOR formats #543 — add round-trip test coverage for TOML/YAML/CBOR (additive;
de-risks refactor(output): serialize integer-valued metrics as integers, not f64 #530/refactor(output): uniform non-finite & precision policy at the serialize boundary #531)

Core library API:

refactor(lib): hide builder-type internal fields (Source/MetricsOptions/MetricsCfg) #533 — hide builder-type internal pub fields
(Source/MetricsOptions/MetricsCfg)
refactor(lib): tidy internal-plumbing visibility (Cursor/Callback/LanguageInfo/Node) #534 — tidy internal-plumbing visibility (Cursor is dead public surface;
Callback/LanguageInfo require a doc-hidden ParserTrait)
refactor(lib): 2.0 shape & error-type ergonomics cleanups #536 — 2.0 shape & error ergonomics: parse-error accessors, reserved
MetricsError variants, exit/nexits triple-spelling, FunctionSpan
name: Option<String>
refactor(lib): CountCfg/FindCfg leak Arc<Mutex>/Arc<[String]> representation #537 — CountCfg/FindCfg leak Arc<Mutex<…>>/Arc<[String]>
(companion to the refactor(cli): single anchored walk seam — library globsets are dead and files are double-stat'd after #489 #495 reshape)

CLI:

refactor(cli): cross-subcommand parity (only-filter naming, --output/--strip-prefix on diff family) #538 — cross-subcommand parity (only-filter naming; --output/
--strip-prefix on the diff family)
docs(cli): unify/clarify list-merge semantics (--paths-from union vs manifest replace) #539 — clarify/unify list-merge semantics (--paths-from union vs
manifest replace)

Web & Python (published surfaces, newly represented here):

fix(web): /metrics language field emits non-lookup display names (c/c++, c#) #540 — /metrics language emits non-lookup display names
(c/c++, c#) that disagree with the Python surface
refactor(web): consistent error schema, response envelope, and introspection endpoints #541 — consistent REST error schema, response envelope, and introspection
endpoints
docs(stability): bring Python bindings & REST schema under the 2.0 stability contract #542 — bring the Python bindings & REST schema under the written 2.0
stability contract (name-lock before first PyPI publish; analyze_batch
default polarity)

Third-pass review findings (2026-06-06)

A third pre-2.0 review (focused on the SemVer-frozen surfaces that cannot be
changed after the major bump) filed #551–#562.

Core library API (2.0-gated / breaking):

refactor(api)!: mark remaining public enums #[non_exhaustive] (suppression + Severity + SpaceKind) #551 — mark remaining public enums #[non_exhaustive] (MetricKind,
Severity, SuppressionScope, SpaceKind; reason about the closed
ones) — generalizes the Stats non_exhaustive item; breaking-to-add-later
feat(lib): impl Display + std::error::Error + #[non_exhaustive] on ConcurrentErrors #553 — impl Display + std::error::Error + #[non_exhaustive] on
ConcurrentErrors (separate error type refactor(lib): 2.0 shape & error-type ergonomics cleanups #536 missed)
refactor(suppression)!: give MetricKind::FromStr a real error type (Err=()) #554 — MetricKind::FromStr has Err = (); give it a real error type
refactor(api)!: reconcile the two parallel metric enums (Metric vs MetricKind) #555 — reconcile the two parallel metric enums (Metric vs MetricKind:
Nexits/Exit, NArgs/Nargs, missing Tokens)
refactor(node)!: hide Node's tree_sitter::Node pub field behind an accessor #556 — hide Node's pub tree_sitter::Node field behind an accessor
(welds the value-not-stable dep into a stable layout)

Core library API (additive, bundle into the 2.0 sweep):

feat(lib): derive PartialEq/Hash/Ord on public value types (Stats, Severity, SpaceKind) #552 — derive PartialEq on the compute Stats/CodeMetrics/FuncSpace/
Metrics (wire mirrors already have it); Hash/Ord on
Severity/SpaceKind/Direction

Metrics (re-baseline):

fix(abc): wire Fitzpatrick unary-condition walker for Kotlin, Ruby, Elixir #557 — (bug) ABC Fitzpatrick unary-condition walker missing for
Kotlin/Ruby/Elixir (cross-language parity; reproduced) — lands with the
2.0 re-baseline; also fixes a stale abc.rs policy doc-comment

Output / stability contract:

docs(stability): lock output-format contracts (CSV/SARIF/code-climate/AST) for 2.0 #559 — lock the output-format contracts (CSV header + integer-column
nullability, SARIF schema version + rule shape, code-climate field set +
fingerprint, AST JSON shape + one-way status, wire::CyclomaticModified)
fix(output): use underscore keys for dump purity_ratio/estimated_program_length #562 — use underscore keys for dump purity_ratio/estimated_program_length

Web (2.0-gated schema):

fix(web): /comment empty result returns 200 {code:null} on JSON but 204 on octet-stream #558 — (bug) /comment empty result returns 200 {code:null} on JSON
but 204 on octet-stream (content-type-dependent envelope)

Filed without the 2.0 milestone (additive, fix anytime):

feat(web): bca-web --num-jobs should accept 'auto' and be cgroup-aware (CLI parity) #560 — bca-web --num-jobs should accept auto + be cgroup-aware (CLI parity)
docs(cli): document the workspace-wide exit-code convention #561 — document the workspace-wide exit-code convention

Grammars (2.0-gated default flip, tracked epic):

track(lang): default C/C++ to upstream grammars, add C language, mozcpp opt-in #718 — track(lang): default C/C++ to upstream tree-sitter-cpp/tree-sitter-c,
add a real LANG::C, demote the vendored mozcpp fork to an opt-in feature
(children chore(grammar): measure mozcpp vs upstream cpp/c grammar divergence #719-feat(lang): add C language backed by tree-sitter-c #721, chore(grammar): re-evaluate mozcpp-era skips, workarounds, and docs post-flip #723; mirrors the refactor(api): normalize &LANG-by-value and drop Java-style get_ getter prefixes #507 JS/mozjs shape; (breaking):
.c reported language changes cpp -> c, cpp feature contents change)

Release-prep follow-ups (no separate issue — handle at cut)

STABILITY.md is internally inconsistent on version: the header frames the
crate as 1.x while the body/horizon describe already-staged 2.0 breaks in
past tense. At cut, flip the header to 2.0.0, move landed-break descriptions
out of the "1.x" framing, and prune the "On the 2.0 horizon" list of
implemented items.
CHANGELOG has no ## [2.0.0] section yet, though [Unreleased] carries the
(breaking) entries and several say "the release-prep commit moves this into
the 2.0.0 section." Stage the section and author the refactor(output): normalize metric JSON/YAML key naming across all Stats #510/fix(halstead): n1/N1/n2/N2 keys are case-only-distinct (CSV/header collisions) #511 key map that
STABILITY.md (:629) promises the 2.0.0 entry will carry.

Acceptance

Every checkbox above resolved or explicitly deferred with rationale.
STABILITY.md "On the 2.0 horizon" section emptied/updated.
CHANGELOG 2.0.0 section lists every (breaking) change and the
value re-baseline.

Filed the pre-2.0 review findings as individual issues against the 2.0
milestone and linked them in the body above.

Bugs worth attention regardless of the 2.0 timeline:

fix(tools): guess_language returns unbound-lifetime &'a str (latent unsoundness) #506 — guess_language's unbound output lifetime (only sound today because
every arm happens to be 'static).
fix(halstead): n1/N1/n2/N2 keys are case-only-distinct (CSV/header collisions) #511 — Halstead n1/N1/n2/N2 case-only keys silently collide in
case-insensitive CSV/env consumers.
fix(web): content-type guards reject 'application/json; charset=utf-8' (silent 404) #515 — web rejects application/json; charset=utf-8 with a bodyless 404
(not 2.0-gated; can ship now).

The largest single win is #510 (output key-naming normalization): the
serialized JSON is the documented-stable surface, so the inconsistent
sum/average/min/max spellings can only be unified at the major bump.

Review scope covered the core library API, all per-metric Stats + output
serializers, and the CLI / web / Python crates. The Python bindings were in
good shape — no 2.0-blocking findings there.

Second review wave (core AST architecture, per-language metric mirroring,
docs/build hygiene) added the following:

2.0-gated: #520 (Checker trait reshape — the biggest "add-a-language"
ergonomics win).

Bugs / fixes that should land pre-2.0 (not breaking):

fix(groovy): closure bodies don't count toward cognitive lambda nesting #519 — Groovy closures invisible to cognitive lambda-nesting (verified:
Java/Kotlin/JS/etc. all have the lambda += 1 arm Groovy lacks). This is
a cross-language parity violation per lesson build(deps): update tree-sitter-python requirement from =0.23.6 to =0.25.0 in /enums #11.
build(deps): exact-pin tree-sitter-kotlin-ng and CI-guard enums/ grammar-pin lockstep #524 — tree-sitter-kotlin-ng is the only grammar not exact-pinned; a
dependabot minor bump can silently shift Kotlin metric values.

Perf (whole-repo hot path): #521 (has_sibling cursor alloc, the same
#217 pattern, missed), #522 (Elixir double source-scan per node).

Robustness/hygiene: #523 (children() iterator length/data desync),
#525 (drop num meta-crate), #526 (stale supported-languages lists).

The per-language modules are otherwise unusually well-mirrored — a
dedicated cross-language parity suite already guards ternary/safe-nav/
boolean-sequence/receiver-exclusion handling. The only genuine divergence
found was the Groovy closure gap (#519).

Batch progress update — the following findings landed on main (fast-forward merge of fix/batch-2026-06-03, rebased atop the #510/#511 key-normalization; fmt + clippy --all-features + cargo test --workspace --all-features all green before merge) and are now closed:

Issue	Commit
#506 — `guess_language` → `&'static str`	`fbdcec78`
#508 — `LANG` `Display`/`FromStr`/`Hash`	`e0ba5fe1`
#514 — diff/check metric-name spelling	`498d92da`
#515 — web content-type essence match	`ad152f42`
#521 — `has_sibling` TreeCursor alloc	`0d8564a7`
#522 — Elixir lazy space-kind scan	`6c74dfd6`
#523 — `children()` cursor termination	`f6823957`
#524 — kotlin-ng exact-pin + lockstep guard	`830eae90`
#525 — drop `num`; hoist `csv`/`tempfile`	`e48e79ab`
#526 — stale supported-languages docs	`4bc19af0`

(#510 / #511 — the metric-key normalization — landed earlier in 1d0606e9 and were already closed.)

Still open: #507, #509, #512, #513, #516, #517, #518, #519, #520.

refactor(api): drop get_ getter prefixes, take LANG by value

refactor(metrics): unify average-divisor guard, cyclomatic per-function

Second-pass pre-2.0 review complete (2026-06-05). Fanned out across six
surfaces — public Rust API, serialized output schema, CLI grammar, REST API,
Python bindings, and cross-language metrics — explicitly excluding the
already-filed #506–#526 findings. Filed 15 new issues (#529–#543), all linked
in the body's Second-pass review findings section.

Headline items:

fix(cognitive): PHP double-counts two-word else if form #529 (bug, verified by reproduction): PHP cognitive double-counts the
two-word else if. Should land before the 2.0 re-baseline.
refactor(output): serialize integer-valued metrics as integers, not f64 #530 / refactor(output): uniform non-finite & precision policy at the serialize boundary #531: the output schema emits integers as floats and has no
uniform non-finite/precision policy — the two biggest schema decisions to
settle alongside the refactor(output): normalize metric JSON/YAML key naming across all Stats #510/fix(halstead): n1/N1/n2/N2 keys are case-only-distinct (CSV/header collisions) #511 key normalization.
feat(lib): implement Deserialize on the serialization-root types #532: the library cannot deserialize its own output.
docs(stability): bring Python bindings & REST schema under the 2.0 stability contract #542: the Python bindings and REST schema were entirely unrepresented in
this roadmap; fix(web): /metrics language field emits non-lookup display names (c/c++, c#) #540–docs(stability): bring Python bindings & REST schema under the 2.0 stability contract #542 close that gap. Python distribution/import names must
be locked in STABILITY.md before the first stable PyPI publish.

Each issue carries file:line evidence; the public-API and integer-as-float
claims were spot-checked against main, and #529 was reproduced with bca metrics. The Python surface and the web /v1 routing came out otherwise clean
(no bugs); their findings are mostly contract-documentation and one breaking
default.

Third-pass pre-2.0 review (2026-06-06). Ran a fresh review focused on the
SemVer-frozen surfaces that cannot be fixed after the major bump. Filed
#551–#562; body updated with a "Third-pass review findings" section.

Highest-leverage (breaking-to-add-later — must land at 2.0):

refactor(api)!: mark remaining public enums #[non_exhaustive] (suppression + Severity + SpaceKind) #551 — #[non_exhaustive] on the remaining public enums. This is the one
change that is genuinely free only before 2.0; the track(2.0): breaking-change roadmap for the 2.x major bump #505 horizon had it for
Stats but missed MetricKind/Severity/SpaceKind/SuppressionScope.
refactor(node)!: hide Node's tree_sitter::Node pub field behind an accessor #556 — Node's pub tree_sitter::Node field welds the value-not-stable
dependency into a stable struct's layout; a tree-sitter minor bump silently
breaks our API.
refactor(suppression)!: give MetricKind::FromStr a real error type (Err=()) #554/refactor(api)!: reconcile the two parallel metric enums (Metric vs MetricKind) #555 — MetricKind::FromStr returns Err = (); Metric and
MetricKind are two divergent vocabularies for the same domain.
feat(lib): impl Display + std::error::Error + #[non_exhaustive] on ConcurrentErrors #553 — ConcurrentErrors (a public returned error) has no Display/
Error impl and isn't #[non_exhaustive].

Bugs found (reproduced):

fix(abc): wire Fitzpatrick unary-condition walker for Kotlin, Ruby, Elixir #557 — ABC's Fitzpatrick unary-condition walker is missing for
Kotlin/Ruby/Elixir; verified via CLI (if (a && b || c) scores 0 conditions
vs Java's 3). Internally inconsistent with the same languages' cyclomatic.
Lands with the planned 2.0 re-baseline.
fix(web): /comment empty result returns 200 {code:null} on JSON but 204 on octet-stream #558 — /comment returns a content-type-dependent envelope for the empty
result (200 {code:null} vs 204).

Plus a consolidated stability-contract issue (#559) covering the output formats
(CSV/SARIF/code-climate/AST) that are public + serialized but not in
STABILITY.md, and additive parity/doc items (#552, #560, #561, #562).

Two release-prep follow-ups (no separate issue, noted in the body): the
STABILITY.md 1.x-vs-2.0 tense inconsistency and the missing ## [2.0.0]
CHANGELOG section + the #510/#511 key map it promises.

Scope note: deliberately did not file marginal additive nice-to-haves
(/v1/metrics endpoint, BCA_LOG env var, analyze(language=) kwarg,
MetricSet bitwise ops) or items the closed issues already settled (the Python
str-vs-StrEnum surface is intentional per #542's recorded decision; the
--format value-vocabulary divergence is documented per #513).

Progress check — 2026-06-06

All filed review-finding sub-issues are resolved. #506–#562 are
closed and landed on main; #505 is now the only open issue in the
2.0 milestone. Ticked every - [ ] #NNN review-finding checkbox
in the body accordingly.

The 35 (breaking) entries each sit in CHANGELOG [Unreleased],
flagged "deferred to 2.0", with the release-prep commit slated to
move them into the 2.0.0 section.

Deliberately left unticked — this is the at-cut release-prep

The implementation work is done, but #505's own acceptance criteria
require the release cut itself, which has not happened (2.0 is not
scheduled):

Known loose ends — staged, not yet executed. Stats/enum
#[non_exhaustive] (refactor(api)!: mark remaining public enums #[non_exhaustive] (suppression + Severity + SpaceKind) #551), builder-field narrowing (refactor(lib): hide builder-type internal fields (Source/MetricsOptions/MetricsCfg) #533), the
#[doc(hidden)] trait fate (refactor(lib): tidy internal-plumbing visibility (Cursor/Callback/LanguageInfo/Node) #534), the FilesData/ConcurrentRunner
reshape (refactor(cli): single anchored walk seam — library globsets are dead and files are double-stat'd after #489 #495), and the re-baseline note are all decided and staged
in [Unreleased], but the breaking edits physically land at 2.0.
STABILITY.md "On the 2.0 horizon" is still fully populated and
the header still frames the crate as 1.x (1.1.0).
No ## [2.0.0] CHANGELOG section exists yet, and the refactor(output): normalize metric JSON/YAML key naming across all Stats #510/fix(halstead): n1/N1/n2/N2 keys are case-only-distinct (CSV/header collisions) #511
key map + value re-baseline note promised for that entry are not yet
authored.

Recommendation

Keep open. #505 is a release-gate checklist, not a work-tracker —
the remaining boxes are the cut (flip STABILITY.md to 2.0.0, stage
the 2.0.0 CHANGELOG section with the key map + re-baseline, move the
(breaking) entries across). Closing now would orphan the one
checklist that fires when 2.0 is actually cut. Close this as part of
the 2.0 release-prep commit.

Trait/shim checkboxes — scoped into an execution epic (#566)

The two open #505 loose ends — "Remove the deprecated shims" and "Decide
the fate of the remaining #[doc(hidden)] extension traits" — turned out
to be one coupled body of work, now planned in #566 and broken into
four landable waves:

feat(lib): complete the Ast seam — strip_comments/functions/dump methods #567 — complete the Ast seam (strip_comments/functions/dump); additive
refactor(cli)!: migrate off path-positional shims to analyze/Ast; remove the shims #568 — migrate bca-cli off the path-positional shims; delete them
refactor(web): migrate the four handlers off action/Callback onto Ast #569 — migrate bca-web's four handlers off action/Callback onto Ast
refactor(lib)!: retire parser-generic surface; demote extension traits to pub(crate) #570 — delete the parser-generic surface; demote ParserTrait/Callback/
Getter/Checker/compute traits to pub(crate)

Resolution of "delete or formally expose": delete from the public
surface — the traits survive as private AstInner machinery and Ast
becomes the single documented analysis seam. (Filter in the original
list is a struct, not a trait — no action.) #256 already closed having
parked these as the doc-hidden holding pattern; #566 is the deliberate
exit from it. No metric re-baseline — output is byte-identical.

Update: the Ast-seam unification epic #566 is complete and closed (all of #567–#572 landed on main, verified green). That resolves the two open trait/shim loose ends tracked here (deprecated shims removed; #[doc(hidden)] extension traits resolved as demote to pub(crate), with Ast the single public analysis seam).

#505 still cannot close: the remaining work is the 2.0 release cut itself — flip STABILITY.md to 2.0.0, author the ## [2.0.0] CHANGELOG section (moving the (breaking, deferred to 2.0) entries, now including #570's), bump the crate version, and the value re-baseline. None of this series is pushed yet either. These are the at-cut steps; #505 closes as part of that release-prep commit.

Partial progress on the STABILITY.md release-prep item: the API-reference reconciliation is now done on main (the deleted action/Callback/parser-generic fns/path-positional shims and the demoted ParserTrait/Parser/<Lang>*/compute traits are described in past-tense "removed / demoted to pub(crate) at 2.0" throughout — commit 4e03f135). What remains for the at-cut commit is narrower: flip the header 1.x/1.1.0 → 2.0.0, bump the crate version, and the value re-baseline + ## [2.0.0] CHANGELOG section.

docs(stability): finish reconciling STABILITY.md with the retired sur…

Added #584 (decide vcs::Options exhaustiveness / builder before 2.0) to the Core library API checklist. Surfaced during the #576 file-type-scoping review: Options keeps gaining public fields, and once the VCS surface ships under SemVer each additive field is a breaking change for external constructors. Not a defect today — forward-looking for the 2.x bump.

refactor(cli)!: make -I/-X single-value and repeatable

refactor(cli)!: unify --top semantics (0 = all)

Status: all milestone-2.0 issues implemented on local `feat/2.0` (pending review/merge)

The full milestone-2.0 issue set has been implemented in one batched effort
on a local feat/2.0 integration branch (off main, not yet pushed),
in seven dependency-ordered waves. Each wave was followed by a four-skill
quality pass (simplify / optimize / audit-tests / code-review) with findings
remediated before commit.

Wave	Issues
1 — lib accessor vocabulary	#588
2 — wire / output shape	#589 #590 #674 #684 #635
3 — CLI surface	#597 #651 #659 #661 #663 #666 #669 #683 #688 #690 #691 #692 #662 #603
4 — web REST schema	#631 #633 #636 #654 #638 #637
5 — Python bindings	#612 #614 #664 #667 #670 #682 #658
6 — report presentation	#677 #687 #685 #678 #679 #675 #668 #689 #686 #680 #592
7 — docs / gates	#656 #660 #673 (+ a new `make py-stubtest` gate)

The branch passes the full local gate: the cargo trio, cargo doc -D warnings, self-scan at both tiers, snapshot-anchors, man-page drift, the
markdown/TOML/Makefile lints, and the Python pytest / mypy --strict /
pyright / stubtest / ruff stages. The big-code-analysis-output
submodule carries one matching local snapshot commit.

Remaining (release-cut, intentionally deferred)

Per the "release-prep follow-ups" in this issue, the following are left for
the deliberate 2.0 cut and are not done on this branch:

Bump Cargo.toml 1.x → 2.0.0 across the crates.
Move CHANGELOG ## [Unreleased] → ## [2.0.0] and author the refactor(output): normalize metric JSON/YAML key naming across all Stats #510/fix(halstead): n1/N1/n2/N2 keys are case-only-distinct (CSV/header collisions) #511
key map + the cumulative value re-baseline note.
Flip the STABILITY.md header to 2.0.0 and empty the "On the 2.0 horizon"
list.

Close this tracker (and the child issues) once feat/2.0 is reviewed,
merged, and the release is cut.

refactor(lib)!: align Stats accessor names with wire vocabulary (#588)

refactor(output)!: align wire/dump/VCS shapes for 2.0 (#589 #590 #674 #…

refactor(cli)!: format/metric/diff CLI surface for 2.0 (part 1)

refactor(cli)!: scope flags to subcommands and accept positional paths (

refactor(cli)!: fail on explicit unknown-language files; unify --outp…

refactor(cli)!: tiered thresholds, manifest/flag name alignment, CI o…

refactor(web)!: align 2.0 REST schema, errors, and defaults (#631 #633 …

refactor(python)!: namespace VCS, single-source shapes, walker entry …

refactor(cli)!: 2.0 AST-report presentation overhaul (#677 #687 #685 #…

docs(2.0): rest.md, python docs, stubtest gate, umbrella doc cleanup (#…

Added a new 2.0-gated child: #718 — default C/C++ parsing to upstream
tree-sitter-cpp (=0.23.4, the exact base mozcpp already inherits) and a
new upstream-tree-sitter-c-backed LANG::C, with the vendored mozcpp
fork demoted to an opt-in feature — the same shape #507 gave
JavaScript/Mozjs. Children: #719 (empirical go/no-go measurement), #720
(Cpp flip + mozcpp opt-in), #721 (C language), #723 (post-flip cleanup).
Related but not 2.0-gated: #724 (ObjC .m/.mm misrouting).

Unlike the zero-drift JS flip, this one expects metric drift and an
integration-submodule re-baseline (mozcpp's primitive_type/
class_specifier overrides; .c re-routing) — #719 scopes it before any
code moves.

Navigation Menu

Repository navigation

track(2.0): breaking-change roadmap for the 2.x major bump #505

Summary

Known loose ends (from STABILITY.md)

Review findings (filed separately)

Second-pass review findings (2026-06-05)

Third-pass review findings (2026-06-06)

Release-prep follow-ups (no separate issue — handle at cut)

Acceptance

Progress check — 2026-06-06

Deliberately left unticked — this is the at-cut release-prep

Recommendation

Trait/shim checkboxes — scoped into an execution epic (#566)

27 remaining items

Status: all milestone-2.0 issues implemented on local `feat/2.0` (pending review/merge)

Remaining (release-cut, intentionally deferred)

Metadata

Assignees

Labels

Projects

Milestone

Relationships

Development

track(2.0): breaking-change roadmap for the 2.x major bump #505

Description

Summary

Known loose ends (from STABILITY.md)

Review findings (filed separately)

Second-pass review findings (2026-06-05)

Third-pass review findings (2026-06-06)

Release-prep follow-ups (no separate issue — handle at cut)

Acceptance

Activity

dekobon commented on Jun 4, 2026

dekobon commented on Jun 4, 2026

dekobon commented on Jun 4, 2026

dekobon commented on Jun 5, 2026

dekobon commented on Jun 7, 2026

dekobon commented on Jun 7, 2026

Progress check — 2026-06-06

Deliberately left unticked — this is the at-cut release-prep

Recommendation

dekobon commented on Jun 7, 2026

Trait/shim checkboxes — scoped into an execution epic (#566)

dekobon commented on Jun 7, 2026

dekobon commented on Jun 7, 2026

dekobon commented on Jun 9, 2026

27 remaining items

dekobon commented on Jun 11, 2026

Status: all milestone-2.0 issues implemented on local feat/2.0 (pending review/merge)

Remaining (release-cut, intentionally deferred)

dekobon commented on Jun 12, 2026

Metadata

Metadata

Assignees

Labels

Projects

Milestone

Relationships

Development

Issue actions

Status: all milestone-2.0 issues implemented on local `feat/2.0` (pending review/merge)