Skip to content

Fix 2.0.1 release blockers: engine bugs, typings, packaging/CI and docs - #282

Draft
ffMathy wants to merge 150 commits into
masterfrom
ccr-f7030c4e-rhdc1v
Draft

ffMathy wants to merge 150 commits into
masterfrom
ccr-f7030c4e-rhdc1v

Conversation

@ffMathy

@ffMathy ffMathy commented Nov 16, 2026 •

Copy link
Copy Markdown
Owner

This PR fixes the problems found in an audit of the unreleased 2.0.1 on master, before it is published. It covers the runtime engine, the typings, packaging/CI and the docs. CHANGELOG.md lists every user-visible change, and MIGRATION.md explains how to upgrade from 1.x and from the betas.

The work went through several rounds. Each round fixed the open findings, then re-ran every earlier repro and started fresh adversarial reviews on Node.js 22, 24 and 26. The last review round is still running. Any further fix will come as a new commit here, so the PR stays a draft until then.

What changed

Engine (src/internals/, rewritten around per-substitute state)

  • Setups and precedence
    • Configuring a call whose arguments overlap an existing setup works again; before, it ran the old setup. The most specific setup wins (literal arguments, then matchers, then Arg.all()), and among setups of the same kind the most recent one wins.
    • Checks never run setups.
    • A call with an argument matcher is always a setup or a check, never a received call.
  • Which calls are recorded
    • Every call through a stored or destructured method reference, a callback, .call/.apply/.bind or Reflect.apply is recorded.
    • The reads and calls that lead up to a setup in the same expression are not recorded. This is decided by call site, and the README states the exact rule.
  • Auto-substitutes are cached, so nested setups and checks (app.logger.received().info('hi')) work. Root setups no longer leak into them.
  • New API: sub[configure](), sub[mimick](instance), sub[clearReceivedCalls](), matchers nested inside objects and arrays, Arg.all() after leading arguments, and an exported SubstituteException.
  • Failure messages
    • Failed checks throw synchronously whenever the outcome is known.
    • The rare later report is thrown from a microtask, so jest attributes it to the right test instead of crashing.
    • Messages say what kind of call was expected, show the substitute path, cap the call list at 50 calls, and keep source-mapped "called at" lines.
  • Performance
    • 10k calls take well under a second for every argument shape we tested: nested objects, cycles, errors, 1 MB strings, 150k-item arrays, wide objects, Maps and Sets, and shared tables.
    • The per-call hot path is back to the speed it had before the round-4 fixes slowed it down.
  • Robustness: works with Error.stackTraceLimit = 0, with a read-only or frozen Error, with custom Error.prepareStackTrace formatters (stack-chain, source-map-support), and with fake timers that fake queueMicrotask and nextTick.
  • Protocol members: Symbol.iterator, Symbol.hasInstance, then, util.promisify.custom, util.inspect.custom and jest's asymmetric-matcher check behave sensibly instead of answering with a substitute.

Typings

  • Fixed:
    • optional members;
    • index signatures, also through a generic constraint;
    • overloads;
    • unions, including members whose signatures differ between variants and null-only members;
    • function-type and callable-interface substitutes;
    • constrained generic code (<T extends Calc>(sub: SubstituteOf<T>));
    • Arg.any('undefined');
    • .mimicks() on union and boolean returns;
    • editor docs on .not.
  • Type-level tests run with npm run test:types, plus a TypeScript 3.6 minimum check in CI. The remaining limitations on old TypeScript versions are documented with workarounds in the README's "Typing limitations".

Packaging and CI

  • An exports map and an ES module entry that shares one copy of the library with CommonJS, so the default import works under native ESM. It is checked with Node and with Rollup/Vite/webpack/esbuild/Rolldown bundles.
  • A prepack build, engines verified down to Node 10, and the CI matrix on Node 22/24/26 plus minimum-Node and TypeScript 3.6 jobs.
  • Compatibility checks for ava, jest 29/30, mocha, node:test and vitest against the packed package.
  • The publish workflow sets the latest/beta dist-tag and provenance, and runs after CI. The GitHub Packages job, which could not publish the @fluffy-spoon scope, is removed.

Docs

  • README rewrite: every snippet is type-checked and executed in CI (npm run test:docs, 81 blocks).
  • New MIGRATION.md, CHANGELOG.md (closes Update documentation #234) and CONTRIBUTING.md.

Verification at 74b8588

  • npm test passes 277 tests on Node 22, 24 and 26.
  • test:types, the TypeScript 3.6 check and test:docs pass.
  • The compatibility checks pass 29/29 on Node 22, 24 and 26.
  • The minimum-Node smoke tests pass on Node 10.0, 12.17, 13.0 and 13.7.

Worth a look before merging

  • Debugger stops. A debugger set to pause on caught exceptions stops inside the library on chained steps wherever Error.prepareStackTrace is not a function: Node < 20.20 without a formatter, jest 30, and vitest vm* pools. The README gives the skipFiles workaround.
  • Bundler caveat. CommonJS code bundled by @rollup/plugin-commonjs with the package external, and bundles that emit CommonJS, need the named import or a documented setting. The README and MIGRATION spell this out.
  • Mutated arguments. An argument that is mutated after it was used in a setup is matched by identity only.

Before publishing

  • NPM_PUBLISH_TOKEN must be a valid granular npm token, or trusted publishing must be configured for the package.
  • Publishing the v2.0.1 GitHub release makes 2.x the npm latest. Until now, latest is 1.208.0.

🤖 Generated with Claude Code

https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt

- run the spec/**/*.test.ts regression files again (lost in #280)
- drop inline 'type' import modifiers so the typings parse on TypeScript < 4.5
- import 'util' instead of 'node:util' (Node < 14.18 compatibility)
- use registered symbols (Symbol.for) so duplicate package copies interoperate
- add a 'configure' context symbol (runtime support follows)
- remove the unused src/api/Constants.ts

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
Replace the node/recorder model with an explicit SubstituteState per
substitute (root or cached auto-substitute): append-only received-call
records indexed by member, configured substitutions kept apart from
them, an optional mimicked instance, and a cache of auto-substitutes
keyed by (kind, member, arguments). Thin proxies translate operations
onto it: real use, setup chains, assertion chains and a configure view.

Existing specs updated where a fix-spec decision changes behaviour:
- RecordedArguments.spec: argument matching is one-directional (D3)
- received.spec (#18): a bare add(Arg.all()) is a setup, not a call (D3)
- issues/178: received(3).subSection fails both ways, so it throws
  synchronously; an ambiguous case keeps the deferred report (D9)
- clearReceivedCalls.spec: asserts behaviour instead of the removed
  recorder internals
Recorder.spec and RecordsSet.spec are removed with the classes.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
Add one regression test per audit finding (named with its id) under
spec/regression/fixes/, plus tests for the decisions that add or
change behaviour (D2 auto-substitute caching, D5 configure, D9
deferral, D11 string assertion forms, D12 mimick, D13, D14, D16-D19)
and for two copies of the package interoperating (G:RG-8).

Engine changes found while testing:
- every proxy target carries the printer, because util.inspect
  inspects the target of a proxy instead of calling its traps
- only the call-site frame is captured per call, which halves the
  cost of a call (D19)

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
After sub[mimick](instance), reading a member that the instance does
not have now returns undefined, as on the instance itself. The
substitute keeps an auto-substitute only where it needs one: the
string received/didNotReceive forms, members with configured calls,
and members configured through nested setups.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
… spec output

- Add an "exports" map with a generated native ES module entry
  (dist/index.mjs + dist/index.d.mts, written by packaging/build-esm.js),
  so `import Substitute from '@fluffy-spoon/substitute'` is the Substitute
  class under native ESM and TypeScript node16/nodenext/bundler. CommonJS,
  "main"/"types", package.json and deep imports of dist/ files keep working.
- Build before packing (prepack), so npm pack / npm publish always ship a
  fresh dist/ (G:PK-3).
- Compile the specs into dist-spec/ (npm run build:spec), so npm test never
  leaves dist/ in an unpublishable state (G:PK-3).
- Sync package-lock.json to 2.0.1 (G:PK-4).
- engines.node stays >=10: the packed build was smoke-tested on Node
  10.0.0 to 26 (CommonJS) and 12.17.0+ (ESM); Node 8 lacks
  util.isDeepStrictEqual (G:PK-2).
- publishConfig pins the npm registry; packaging/dist-tag.js computes the
  npm dist-tag for a release; better keywords; volta pin moves to Node 24.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
…:test check

- The pretest script (npm run install:local) packs the repository root like
  npm publish does and installs the tarball with --no-save; the runners
  import '@fluffy-spoon/substitute' instead of ../../../src (G:CI-2).
- Strip NODE_TEST_* from the environment of the nested runners. A nested
  `node --test` inherited NODE_TEST_CONTEXT and exited 0 although its test
  failed, so the node:test check failed on Node 20.11+ (G:CI-2).
- Accept the runners' reports for a synchronous failure as well as for an
  assertion thrown on process.nextTick, and match the exception name and
  message separately (ava and node:test print them apart).
- Update the runners (ava 6.4, jest 30, mocha 12, vitest 5, tsx, ts-jest,
  TypeScript 5.9) so the checks run on Node 22, 24 and 26; mocha 10 failed
  to start on Node 26. This also removes the critical/high audit findings.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
…rs; fix publishing

- Node CI: run on push and pull_request (and as a reusable workflow) on
  Node 22, 24 and 26 instead of 18/20/21, and run
  `npm run test:types --if-present`.
- New compatibility job: the compatibility checks against the packed
  package (G:CI-2).
- New minimum-node job: packs the package and smoke-tests it on
  MINIMUM_NODE_VERSION (10.0.0, = engines.node) and on 12.17.0 for the ES
  module entry (G:PK-2).
- Publish: trigger on `published` releases, run the full CI workflow
  first, check that the release tag matches package.json, publish stable
  versions as `latest` and prereleases as `beta`, with provenance.
- Remove the GitHub Packages job. GitHub Packages only accepts packages
  whose scope is the repository owner (@ffMathy), so it could never host
  @fluffy-spoon/substitute and failed every release (G:CI-1).
- CodeQL: action v4, drop the obsolete HEAD^2 checkout, explicit
  permissions. Dependabot also watches compatibility-checks/. The bug
  report template asks for the module system and the test runner.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
packaging/package.test.js runs with `npm test`. It packs the repository like
npm publish does (starting from a stale dist/), installs the tarball into a
fresh consumer project and checks:
- G:PK-3: the tarball has the entry points and no spec/stale output; the
  spec build goes to dist-spec/.
- DOCS:ESM: the default import works under native ESM and in TypeScript
  (node10, node16, nodenext, bundler); require, deep imports and
  package.json still work.
- G:PK-4: package-lock.json version = package.json version.
- G:PK-2: engines.node = the version the minimum-node CI job tests.
- G:CI-1 / CI:dist-tag: publishing only targets npm, tag logic.
- G:CI-2: nested node --test runs report failures; CI runs the
  compatibility checks against the packed build.
- CI:matrix: supported LTS lines and the type tests.

CONTRIBUTING.md (linked from the README, previously a 404) documents
install, build, test, test:types, the compatibility checks, CI and the
release process.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
- One rule for the setup methods (T:TY-2, G:RG-4): `.returns` / `.throws` /
  `.mimicks` / `.resolves` / `.rejects` are offered unless the value they are
  accessed on (property type or return type) has a member with that name; the
  symbol forms are always offered. Checks are non-distributive for the method
  types, so union return types keep a single `mimicks` (T:TY-1).
- D11: `sub.received` / `sub.didNotReceive` are only offered when T has no
  public or private member named received, didNotReceive, returns, throws,
  resolves, rejects or mimicks; the symbol forms are always available,
  including on SubstituteOf<T> for a generic T (T:TY-3, G:RG-3, T:TY-14).
- D12: `mimick` is symbol-only (`sub[mimick](instance)`) (T:TY-5).
- D5: `sub[configure]()` returns a view whose members only expose the setup API.
- Overloads: up to 8 overloads in one bivariant inference step (T:TY-11),
  zero-parameter overloads anywhere (T:TY-10), `(...args: unknown[])`
  (T:TY-13), positional signatures before the Arg.all() ones.
- Optional members (T:TY-7), index signatures next to named members (T:TY-8),
  members returning `never` (T:TY-12), `then` typed `never` and `toJSON`
  without setup API (T:TY-16, D20).
- Export SubstituteException from the index through a public interface, so the
  typings do not reach the internals or @types/node (G:RG-9).
- overlap.spec.ts: use the string received/didNotReceive forms through `any`,
  since D11 no longer offers them in the typings for these types.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
- spec-types/: fixtures type-checked by `npm run test:types` (strict, and
  spec-types/loose without strict), one file per finding; must-fail lines use
  `@ts-expect-error`. `npm test` runs them first.
- spec-types/min-ts: `npm run test:types:min-ts` builds dist/ and type-checks
  a consumer of the built typings (resolved through package.json, without
  @types/node) with an old TypeScript installed by npx (default 3.6, the
  measured minimum; other versions as arguments).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
With the D9 engine change, the old scenario fails synchronously. Three
calls followed by a property-style received(2) can only be reported on
process.nextTick, so the runners are still checked for that path.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
Editors then show the documentation of received, didNotReceive,
configure, mimick and the setup methods on hover.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
sub[returns](1) (and [throws], [resolves], [rejects], [mimicks]) on a root
substitute, or on its configure() view, used to read a member keyed by the
symbol and return an auto-substitute, so nothing was configured. It now
throws a SubstituteException usage error that points to member/call setups
and to sub[configure]() for a substitute returned by a configured call.
The string names stay ordinary members (D11).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
A failing assertion on a member with hundreds of received calls produced a
message with one or two lines per call. The listing now stops after 50
entries with "› … and N more", using the listing's own marker. The counts in
the expectation line still cover every call.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
The colour decision now follows Node's own policy more closely:
FORCE_COLOR still wins ('0'/'false' off, anything else on); otherwise a
non-empty NO_COLOR or NODE_DISABLE_COLORS, or TERM=dumb, turn colours off;
otherwise stdout must be a TTY. The decision is a pure function,
shouldUseColors(env, stdoutIsTTY), so it is tested without a real TTY.

The existing A:RT-14 test now also clears NODE_DISABLE_COLORS and TERM, so
its TTY case does not depend on the environment the tests run in.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
Two hints, after the listing of received calls:

- When a method assertion finds too few calls and the member was also read
  without being invoked: "@Substitute.m was read N time(s) without being
  called; reading a method does not count as calling it." The same hint is
  given when a member that was read and invoked fails a property-style check
  with a count above its calls.
- When didNotReceive().m(...) fails on a member that was only read, for
  example passed as a callback (the A:RT-3 residual forced by D9): the hint
  explains that the assertion is decided when the member is read and that
  sub[clearReceivedCalls]() after the arrange step resets the reads.

SubstituteException's mismatch description gains optional `explanation`
(continues the expectation line) and `hints` (lines after the listing).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
… method

Amended D9: a property-style assertion such as `sub.received(2).m` on a
member that has only ever been invoked is reported on process.nextTick when
it is not invoked. Its message said "Expected to receive 2 property reads
matching m, but received none." and nothing else. It now continues with
"m was never read as a property; it was called 1 time as a method. To assert
calls, invoke it: sub.received(2).m(...)", with the path of nested
substitutes (sub.logger.received(2).info(...)).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
…utated arguments

Resolving a call compared it against every literal substitution of the
member, and configuring one compared it against all earlier ones, so 8000
literal setups plus 8000 calls took about 4.6 s. All-literal substitutions
are now indexed per member by a stable fingerprint of their arguments:
primitives by value, functions, substitutes and symbols by identity, and an
object only as "an object", so the key never goes stale when an argument
object is mutated after the setup (literal arguments are compared as they
are when the call is made, D15). The deep-equal confirmation and the D4
precedence (most specific class, then most recent) are unchanged. The same
workload now takes about 0.17 s. Literal setups with object arguments still
compare against the setups with the same primitives in the same places.

The auto-substitute cache looked entries up by a structural fingerprint
only, which goes stale when an argument object is mutated: after
api.get(request).body.returns('ok') and request.headers = {...},
api.get(request).body was a new auto-substitute (D2). Entries are now also
found by the identity of the argument objects.

The ratio test 'the cost of a call does not grow' now takes the best of
three runs and compares 2000 with 32000 calls against a bound of 48x
(linear is about 16x, quadratic about 256x), so CPU load does not fail it.
New: 8000 literal setups plus 8000 matching calls in under 1 s (best of
three, directly and through configure()) and a growth check for them.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
Every example now compiles under strict TypeScript and does what the text
says (the old snippets mostly failed to compile or crashed at runtime).

- Fix the malformed badge HTML and the requirements: Node.js 10 for
  CommonJS, 12.17 for the ES module entry point, TypeScript 3.6.
- Document importing (CommonJS, ES module default and named imports),
  sequences that repeat their last value, promises on properties,
  mimicks, the Arg.any type strings, Arg.all() for the rest of the
  arguments, the .not matchers, match order and precedence, configure(),
  partial mocks with [mimick](), auto-substitutes and nested members,
  received(count) validation, property checks, what counts as a call,
  clearReceivedCalls, SubstituteException, the assertions reported on
  process.nextTick, the name-conflict rules for TypeScript and
  JavaScript, the symbol API, special members, typing limitations and
  test runners.
- Replace the Substitute.disableFor section, which described an API that
  no longer exists.
- Link the migration guide, the changelog and CONTRIBUTING.md.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
Covers npm latest moving from 1.208.0 to 2.x, the Node.js and TypeScript
minimums, the removed APIs (Substitute.disableFor, clearSubstitute and
ClearType, the string mimick(instance)) with their replacements, the
symbol API and the name-conflict rules, and the behaviour changes:
sequences repeating their last value, cached auto-substitutes and nested
checks that really check, matcher calls being setups, match order,
configure(), what counts as a call, checks never running setups,
received(count) validation, matchers and messages. Also the ES module
default import, the exports map and the moved deep import paths. Each
section says which old versions it applies to; the "after" examples are
checked like the README's.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
Keep a Changelog format. The 2.0.1 entry summarises the fixes, changes,
additions and removals since 2.0.0-beta.4 in user terms; the beta entries
are reconstructed from the git history.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
`npm run test:docs` builds the package, then spec-docs/check.js extracts
every typescript and js code block from README.md, MIGRATION.md and
CHANGELOG.md, type-checks the TypeScript ones under strict against the
built package (resolved through package.json and its exports map, like an
installed copy) and runs each block in its own Node.js process.

Examples state what they do in comments that the check enforces:
`// prints <output>` must print exactly that, and `// throws <Name>`
must throw. Any other output, exception, uncaught exception or unhandled
rejection fails the check. HTML comment directives share setup code
between blocks (name=/uses=), mark ES modules, compile-only and skipped
blocks, an expected uncaught exception, and an error message text that
must match. Relative links and #anchors are checked too.

spec-docs/self-test.js runs the check on fixtures/broken.md, whose blocks
and links are broken on purpose, and fails unless each problem is
reported. The whole run takes about 10 seconds. CI runs it in the test
job.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
…tion (D10, D11)

A call of the string `received` / `didNotReceive` with an Arg matcher among
its arguments was always taken as a setup (D3) of a member with that name,
so `sub.received(Arg.any()).add(1)` silently configured something and the
"assertion" passed, and `sub.didNotReceive(Arg.any()).value = 1` threw
"Property assignments cannot be configured". By D11 such a call followed by
a member access or an assignment is the assertion API: received() now
reports the invalid count (D10) and didNotReceive() asserts, ignoring its
arguments as it does without matchers. Followed by a substitution method,
and inside a configure() view, it is still a setup of the user's member.

Found by porting engine A's D11 test.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
…tion

A failed `sub.received(1).isEnabled = false` listed the assignments under
"All property or method calls to @Substitute.isEnabled received so far:".
The listing of an assignment assertion only contains assignments, so it is
now headed "All property assignments to ... received so far:". The other
headings are unchanged.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
… setups

Code under test that reads a member called returns (or throws, resolves,
rejects, mimicks) on an unconfigured call result gets the setup method,
so calling it reconfigures the substitute. Configuring the value first
avoids that.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
T42 (R5TD-1): a concrete union whose variants declare a same-named
method with different parameter lists (an extra optional parameter,
other parameter types or required-argument counts, overloads, a generic
method in one variant) has the full setup API on calls of that method
again. Setups accept the return types of the variants whose overloads
the call can reach. The residual limits are pinned in
spec-types/limitations.ts and in the TypeScript 3.6 consumer.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
…fixes

- Performance (R5V-1, R5TD-2, E13, E14): wide objects, Maps and Sets hash
  every key and value (records with a budget of 4 objects), a sample of
  the values above 65,536 entries, families of 1,025 keys for objects
  with the keys of an earlier one, and a hash remembered across calls.
  Every number re-measured on Node.js 22 against the build (best of 3):
  hot path, shared and distinct wide shapes, records, the 200 literal
  setups with copies (0.2 s of 0.25 s is isDeepStrictEqual), colliding
  strings and symbols, unreadable values (about 50 to 70 ms for 4,000),
  matcher setups (0.8 s / 7.5 s) and memory (1,350 bytes per chain call).
- How arguments are compared (E13, E14): unreadable values are equal only
  to themselves; a wide container keeps the hash of its first use, so a
  copy does not match a setup made after it changed, and a matcher put
  into it later is not recognised. Checked examples.
- Matchers inside objects (E13): containers with more than 256 items or
  keys without a matcher are searched only the first time.
- Error.stackTraceLimit (R5V-3, E15): read-only, accessor, missing and
  frozen limits work and are left as they were; accessors give no
  "called at" lines. Checked example with Object.freeze(Error).
- "called at" text (E16): the formatter in place when the message is
  made, or the last one seen for reads in a chain.
- Debuggers (R5-ADV-2, E16): no stops when Error.prepareStackTrace is a
  function (Node.js 20.20+, source-map-support, jest 29, vitest forks and
  threads); 40 stops for a 6-line test otherwise (Node.js 18, deleted
  formatter, vm contexts: jest 30, vitest vm pools), with the skipFiles
  hint.
- Typing limitations, MIGRATION "TypeScript" and CHANGELOG (R5TD-1, T42):
  methods whose parameter lists differ between union variants have the
  setup API; exact rule, the two pinned limitations with casts, and
  unnamed parameters in editor hints. Checked example.
- CHANGELOG: entries framed against 2.0.0-beta.4 (union methods with
  differing parameter lists, debugger stops as a change, tables passed
  with every call, stackTraceLimit shapes).
- CONTRIBUTING: what a test run leaves in the temporary directory.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
…() values (E27, E28)

R6FE-6: an unconfigured read of __esModule is undefined and not recorded, so a
substitute used as a whole module is its own default export under the interop
helpers. [configure]() and [mimick] can still configure it.

R6FE-5: Arg.any(type) and Arg.any.not(type) throw a TypeError for a type name
outside 'any', 'string', 'number', 'boolean', 'symbol', 'undefined', 'object',
'function', 'array' and 'bigint'. mimicks() without functions throws a
TypeError at the setup that names the member or call.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
…E22)

hashArguments, coarseHashArguments and identityHashArguments leave trailing
undefined arguments out, so f(1) and f(1, undefined), which compare as equal
argument lists from now on, land in the same buckets of the literal-setup
index and the auto-substitute cache.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
R6FE-2: when a call is compared with a setup or a check, a position one side
does not have reads as undefined. It matches only a literal undefined or a
matcher that accepts undefined (Arg.any(), Arg.any('undefined'), ...), so
f(1) and f(1, undefined) match each other and set('a', 1, Arg.any()) matches
set('a', 1), while f(1) and f(1, 2) stay different. Re-configuring f(1) as
f(1, undefined) replaces the setup. Arg.all() is unchanged.

Existing tests updated where this deliberately changes the outcome: an
omitted argument now matches Arg.any() (decisions D3/D4, amendments D4/D15,
ported-from-a D14), f() is f(undefined) (amendments D15), and multiply(1)
after multiply(1, Arg.any()) is re-configured through the configure view
(returns.spec).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
…ing or not a number, as V8 does (R6-ADV-6, E21)

A missing, non-number or below-1 limit gives no frames for any error in V8. The library
used to define the limit as 1 and delete it again around every capture, which changed
Error's shape on every operation (1.4-1.9x slower hot path). Now such an operation
shares one capture without a stack: no "called at" line, and D25 falls back to
adjacency, as with Error.stackTraceLimit = 0. The property is never touched then.

The round-6 E15 test expected frames with a deleted limit; E21 changes that.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
R6FE-3: the construct trap records `new sub(...)` and `new ns.Client(...)`
as a call (shown as `new Client(args)` / `new @substitute(args)` in messages
and String()), so setups and checks written with call syntax apply to it.
It answers with the configured value when that is an object or a function,
and with the cached auto-substitute of the call otherwise. A mimicked class,
or a class or ordinary function given to mimicks(), is constructed; an arrow
function is called. new on a bound substitute and on the deferred-read,
configure and assertion views behaves like invoking them.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
R6FE-1: String()/util.inspect of an auto-substitute print its path
(@Substitute.logger, @Substitute.getUser(1)) and, when it has any, its
received calls and setups, like a substitute; no dangling ': '. A substitute
also shows the calls and setups of its auto-substitutes under their paths
relative to it (logger.info('started'), logger.level => returns(3)), calls
in the order they were made. Inside another printout (an argument in a
failure message, a path) an auto-substitute shows only its path.

Existing tests that compared a root's printout to '@substitute {\n}' or to
the old 'property<name>: ' label are updated to the new output.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
…tes (E25, E26)

R6FE-4: Object.freeze, Object.seal, Object.preventExtensions and
Object.defineProperty no longer make later reads crash with a Proxy-invariant
TypeError. The proxy hides the own length/name of its bound-function target
(ownKeys and getOwnPropertyDescriptor report no own properties), and the first
structural change removes them, so the target's own properties are only the
ones the code under test defined. A defined property is a real property that
reads and assignments use and that is not recorded; a frozen substitute keeps
answering setups and recording calls. The views the library hands out
(assertions, configure views, bound functions, members configured both as a
property and as a method) refuse these operations with a SubstituteException
that names the operation.

R6FE-8: a has trap makes `key in sub` and Reflect.has true for keys with a
setup (property, method or nested setup) and for keys the mimicked instance
has, and false otherwise, including protocol keys and the library's own
members unless configured. Nothing is recorded.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
…tes in jest and vitest (E24-E27)

The jest-matchers suite checks that a substitute in a failure message shows
its path ([Function @Substitute.logger.name] from pretty-format), that a
frozen substitute still compares by identity, and that the has trap leaves
toHaveProperty, toHaveBeenCalledWith, isMockFunction and clearAllMocks as
they were. A new passing vitest-matchers suite checks the same with vitest's
expect, util.inspect of an auto-substitute, and the __esModule probe.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
… family anchors cost small records nothing (R6-ADV-4, R6V-1, R6-ADV-1, R6V-2, R6-ADV-5, R6V-3, R6-ADV-3; E18-E20)

E18: every hash belongs to an operation (newOperation). The memo of a wide
container written by an earlier operation may be stale; a hash that used one
says so (lastHashUsedMemo). Literal setups are filed under fresh hashes, and a
call whose memo-based hash finds no setup looks again with its fresh hash when
the coarse hash allows a candidate. The auto-substitute cache does the same, and
files the one entry per coarse hash that was filed under a memo-based hash
without such a look under its fresh hash before it looks again. An unreadable
value's identity verdict lives for its operation only, so a value that became
readable matches an equal setup again. hashSettings.memos (internal) turns all
memos off, for the property test that compares random scenarios with and
without them.

E19: families stay (100 calls with distinct 70,000-key objects need them), but a
family is probed by the key with the lowest key hash and the one with the
highest, not by its first key, and an object's family is no longer cached, so
small records that share a field name with a wide object pay nothing.

E20: a wide object found inside an entry, or being hashed, is marked in its
memo for the operation, so shared wide objects and back-references are listed
once per operation. The matcher search counts a cut only for a plain container,
and runs its unbounded re-scan only for a container seen a second time.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
…ash has a candidate (E18)

The coarse hash alone sends every call with a shared wide table plus another
object argument (render(props, theme), save(entity, config)) through a fresh
hash, which lists the table every time: 10,000 such calls took about as long as
at 45facfb (13 s) again. The narrow hash is the structural hash with every
argument that holds a wide container counted as one tag. It is exact without
memos, provided every container whose memo the hash used is still wide, which
is checked without listing keys (Maps and Sets by size, objects by 257 of the
keys they had). Literal setups and auto-substitute entries are grouped by it.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
…p only entries that met a wide container (E18, E19)

E19: array items and the values inside an entry of a wide container are mostly
records, which a wide object rarely is: they are not probed against the families
before their keys are listed (by the hash or by the matcher search). A wide one
among them is still found to be wide, or in its family, once listed, with the
same hash. With 16 families, batches of small records cost 0.9-1.05x of the
same library copy without families (74b8588: 3.2x).

E18: the auto-substitute cache groups by narrow hash only the entries whose hash
met a wide container, as only those can be equal to a call whose hash used
memos; ordinary object arguments pay nothing for it. A call whose fresh hash is
known already (a shrunk container, or the setup lookup computed it) looks with it.

The E19 test now loads the library twice in each child process and measures the
two copies by turns, in both load orders.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
…6V-4, T43)

A member type such as Record<K, () => number>[K] in generic code chose its
signatures through a chain of two indexed accesses on the same object type.
When TypeScript resolves such a member through its constraint, the chain is
deep enough that it stops computing the inner access (it shares the outer
one's recursion identity), and the member lost its substitute signatures on
TypeScript 4.7 and newer. The box is now unwrapped with a conditional type.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
…shape-only memo uses (E18, E19, E20)

- A memo of an earlier operation that the hash only takes as "this container is
  wide" (inside an entry) is exact as long as the container is still wide
  (checked once per operation without listing): sessions sharing a config object
  or tables of records no longer count as possibly stale.
- The narrow hash is a separate pass in which every wide container (not every
  argument that holds one) is one tag, so calls that share a wide container but
  differ elsewhere are told apart without a fresh hash.
- The auto-substitute cache first asks the coarse hash whether any entry has
  objects at the same places (R5TD-2: a distinct first argument per call needs
  no further work), then the narrow hash.
- The search for nested matchers hands the families it found to the hash of the
  same operation, as it does the keys it listed.
- Memo uses are deduplicated per hash; the E19 and R6-ADV-5 ratio tests measure
  their two cases by turns.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
The R6-ADV-5 ratio test takes the best of six interleaved runs and allows 2x for
flat records (74b8588: 3.7x-5x for nested records, which is what it pins); the
long tests get a timeout of their own.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
…s (R6V-5, R6TD-2, T44)

A concrete union member whose variants all return the same type (the
addEventListener of every lib.dom element, fetch(url) | fetch(url, options?))
now gets one signature per overload position, whose rest parameter is the
union of the variants' parameter lists, computed from the whole member type.
The per-variant boxes are only built when the return types differ;
intersecting them compared every variant with every other one, which made the
union of all element types check 1.6-1.8 times as slowly as before T42 on
TypeScript 3.6-3.8. A callback argument without parameter types gets no
contextual type there when the variants' callback types differ in more than
their this type (pinned in spec-types/limitations.ts).

spec-types/R6V-5.R6TD-2.T44.check-cost.js (run by test:types:min-ts) checks a
deterministic instantiation budget for the element-union fixture.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
…-1, T45)

For concrete unions whose variants return different types, every overload
now gives one signature per number of arguments it accepts (its first n
parameters), whose result also takes what the variants that accept those
arguments return: get('k') of get(key, options?: V1): number and
get(key, options?: V2): string takes number | string, and a call of a
variant with a rest parameter reaches the other variant too. A single
Arg.all() takes what any overload of any variant returns. Signatures without
parameters come last, after the Arg.all() signatures, so that TypeScript 5.3
and older do not pair other signatures with them. Arg.any() and untyped
Arg.is() predicates on parameters whose types differ per variant stay
limited to the first variant's overload (pinned in limitations.ts).

The same-return fast path (T44) now writes each slot's parameters position by
position: TypeScript compared two rest parameters typed as unions of
parameter lists only by their first element and dropped one of the
signatures (postMessage of MessagePort | Worker | BroadcastChannel). It is
taken only for lists of up to 6 parameters without a rest parameter, and in
generic code the boxes are used.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
- An overload whose parameter list is a union of lists (a rest parameter
  computed from a type parameter, erased to its constraint) stays one
  overload. It was split into one overload per list, and the member's
  substitute signatures then became a union that TypeScript paired with the
  plain signatures of T: a typed emitter's emit() had no setup API, and the
  configure view was not callable.
- A member with a single generic signature also gets a generic signature
  after all others that takes the arguments the member itself accepts
  (TypeScript checks them against the generic signature). Node's
  EventEmitter<Events>.emit, whose erased parameters are never, is then
  callable, and calls with Arg.any() arguments, which TypeScript's subtype
  pass used to resolve to the plain generic signature, keep the setup API.

Arguments and values of generic methods are checked against the type
parameters' constraints (pinned in spec-types/limitations.ts).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
new on a substitute or a member is recorded and answered like a call (E23),
so a type with construct signatures but no call signatures now also gets
call signatures with the constructor's parameters on the substitute, the
configure view and the assertion views: Client('u').returns(client),
Client.received(1)('u'), ns.Client('u').returns(x),
ns.received(1).Client('u'). new keeps the constructor's own typing. On
TypeScript 5.0 and older only the last construct signature of an overloaded
constructor may be matched this way.

The instantiation budget of the T44 cost check is set to 470 000 (the
element-union fixture needs about 442 000 now, 525 000 at 74b8588).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
…ints (R6FE-5, R6TD-3, R6TD-4, T48)

- Arg.any('bigint') and Arg.any.not('bigint') are typed (the runtime
  accepts 'bigint' officially, E28).
- A concrete union of types with a string index signature is mapped as one
  type: its named members with the union of the variants' member types,
  next to its string index signature. The homomorphic mapping mapped each
  variant on its own, so speak() of IA | IB had the setup API of each
  variant's speak and .returns() took never. Generic code keeps T.
- Generic code with a union constraint whose variants declare a method with
  different parameter lists has no setup API on TypeScript 5.4 and newer;
  pinned as a documented limitation with the configure-view and cast
  workarounds.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
…l() signatures across variants (T45)

The T45 prefix signatures left a variant's own method in the union signatures when one of its parameters has a
literal type (TypeScript tries such signatures first): add each overload's whole parameter list to the boxes.
The prefix and Arg.all() signatures of variants with the same parameter list are the same types again (call results
by the parameter list without names, Arg.all() signatures through an alias), and ReturnsAlike tells return types
apart by identity: a union of 60 variants with a `this`-returning method checked 3.3 s instead of 8.4 s on TS 4.5.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
…rameter, DOM unions (T45)

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
…ipt versions have

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
…rs (T45)

The signature with an overload's whole parameter list is the same as its last prefix signature when the list has no
optional parameters; having it twice made TypeScript 3.7 and 3.8 check a union of 60 variants with a
`this`-returning method in 24 s instead of 3 s. Empty lists keep it, so that calls without arguments still resolve
to the signatures of the variants without parameters first.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
…p API (T46)

The parameter lists of signatures are no longer split per tuple (T46), which also gives
`log(...args: [string] | [string, number])` the setup API, on every supported TypeScript version.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
…E18-E21)

E18 (R6-ADV-4, R6V-1, R6-ADV-1): memos never change results. A lookup
through a possibly stale memoized hash that finds no setup or cached
entry looks again with a fresh hash, first the coarse hash and then an
exact narrow hash, only where a candidate can exist. An unreadable value
is matched by identity for the current operation only, so a value that
becomes readable matches an equal setup again.
E19 (R6V-3, R6-ADV-3): families stay (a fix-spec perf test needs them),
without the membership cache; a family is probed by its lowest and
highest key hash, and array items and values inside wide containers are
never probed, so small records cost the same with or without families.
E20 (R6V-2, R6-ADV-5): a wide object met inside entries or being hashed
is listed at most once per operation, and in later operations only its
width is checked; the unbounded matcher re-scan runs only for a
container seen a second time.
E21 (R6-ADV-6, R6-ADV-2): a missing, non-number or below-1
Error.stackTraceLimit captures no frames, as V8 does, and the library
never defines or deletes the property.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
E22 (R6FE-2): an omitted trailing argument compares as undefined, so
f(1) and f(1, undefined) match each other and set('a', 1, Arg.any())
matches set('a', 1), while f(1) and f(1, 2) still differ. Trailing
undefined arguments do not change the argument hashes.
E23 (R6FE-3): new on a substitute or a member is a recorded call, shown
as `new Client(args)`; setups and checks written with call syntax apply.
E24 (R6FE-1): auto-substitutes print their path and, when they have
any, their setups and received calls; the root lists nested ones under
their paths.
E25 (R6FE-4): Object.freeze/seal/preventExtensions/defineProperty work
on substitutes and auto-substitutes; short-lived views refuse them with
a SubstituteException.
E26 (R6FE-8): `key in sub` is true for configured keys and keys of a
mimicked instance, false otherwise; it records nothing.
E27 (R6FE-6): '__esModule' is a protocol key.
E28 (R6FE-5): Arg.any(type) and Arg.any.not(type) throw a TypeError for
an unknown type name ('bigint' accepted); mimicks() without a function
throws a TypeError at the setup.

Conflicts: ArgumentHash.ts hashList takes engine B's argumentCount with
engine A's hashValue signature, so the narrow and refined hashes also
ignore trailing undefined arguments; the SubstituteState.ts import is
the union of both branches.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
T43 (R6V-4): members reached through a generic key keep the setup API
on TypeScript 4.7 and newer.
T44 (R6V-5, R6TD-2): concrete unions whose variants all return the same
type take a fast path, so large unions such as all HTML element types
check within about 1.2x of 45facfb on TS 3.6-3.8; a cost check pins it.
T45 (R6TD-1): union calls resolve by the number of arguments they pass,
and take the return types of the variants that accept them; a single
Arg.all() takes every variant's return types.
T46 (R6FE-7): generic methods (EventEmitter<Events>, typed emitters,
find<K>, get<K>) get a usable signature instead of never.
T47 (R6FE-3): construct-signature substitutes and constructor members
type-check with the call syntax of E23; `new` keeps its typing.
T48 (R6FE-5, R6TD-3, R6TD-4): Arg.any('bigint') is typed; unions with a
string index signature take the union of return types; the union
constraint limitation on TS 5.4+ is pinned.

src/api/Arg.ts merged cleanly: engine B's runtime check of the type
names and this branch's types and JSDoc both list 'bigint'.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
README:
- `new` on substitutes and members is a call (E23, T47), with an example;
  mimicks() validation and construction of classes (E28, E23).
- Matching by type: 'bigint', unknown type names throw (E28); Arg.any('undefined')
  matches an omitted argument.
- How arguments are compared: omitted trailing arguments (E22) with an
  example; unreadable values are decided per call (E18, R6V-1, R6-ADV-1);
  the wide-object hash memo never changes a result, with the
  arrange-before-act example (E18, R6-ADV-4); the snippet now prints true.
- Matcher search memo: second-pass re-scan wording (E20, R6-ADV-5).
- Re-configuring a call: the E22 consequence for D5, with an example.
- Stack frames: frozen, sealed and non-extensible Error (R6-ADV-2) and
  missing, non-number or below-1 stackTraceLimit (E21, R6-ADV-6), with the
  openConnection example under a frozen Error.
- Special members: __esModule (E27), printing of auto-substitutes and
  pretty-format (E24), freeze/seal/defineProperty and own keys (E25), the
  `in` operator (E26).
- Typing limitations: generic methods (T46), union constraints with
  different parameter lists on TS 5.4+ (R6TD-4), call resolution by
  argument count and the Arg.any() case (T45), untyped callbacks in large
  same-return unions (T44), unions with index signatures (R6TD-3); the
  union-of-tuples limitation is gone.
- Debuggers: throwing getters of the code under test; Performance numbers
  re-measured on Node.js 22 (best of 3).

MIGRATION: omitted arguments, `new`, printing, freezing and `in`,
__esModule, Arg.any/mimicks validation, the wide-object wording, and the
TypeScript changes. CHANGELOG: the same changes against 2.0.0-beta.4.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KD7sVgSiP9KbiEpaM8wCyt
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Update documentation

2 participants