Commit Graph
33 Commits
Author SHA1 Message Date
morluto cef87c498a refactor(architecture): add the ports and resources layers, move ports and shared resource owners there, and ratchet the layer baseline (#11208)
* test(architecture): fail on a stale layer baseline and document the declared layers

verify:module-boundaries reported a layer baseline entry whose imports were
gone but still passed, so main kept process -> artifacts at 20 after it fell
to 17 and a later change could add three imports back unnoticed. A stale
entry now fails like growth; --shrink-layer-baseline lowers it in the change
that removes the imports. The architecture map and CONTRIBUTING name
scripts/lib/module-layers.mjs as the per-directory layer declaration.

* test(architecture): refuse to lower the layer baseline after an inspection failure

A file that cannot be parsed contributes no imports, so a shrink run would record counts that fail as growth once the file is fixed. Whole-command tests cover stale rejection, shrinking, refused growth and the inspection failure.

* refactor(ports): move the progress, work-scope and inspection-scope ports to src/ports (#11214)

* refactor(ports): move the progress, work-scope and inspection-scope ports to src/ports

Pure move with git mv and import specifier updates.

* test(architecture): declare the ports layer and lower the layer baseline

src/ports sits between the foundation and the providers. Moving the three
ports removes 11 provider -> application imports.

* refactor(ports): move AnalysisProvider to src/ports (#11217)

* refactor(ports): move AnalysisProvider to src/ports

Pure move with git mv and import specifier updates in its importers, the
built-fixture path, the CI scope glob and the lint-rule test spellings.

* test(architecture): lower the layer baseline after moving AnalysisProvider

* refactor(ports): move the per-feature provider ports to src/ports (#11226)

* refactor(ports): move the per-feature provider ports to src/ports

Pure move with git mv and import specifier updates: the 47 *Port.ts and
*Ports.ts files that providers implement, plus OptionalObservationProviders,
which only composes the observation ports.

* test(architecture): lower the layer baseline after moving the feature ports

* refactor(providers): move closeAnalysisClient and the Frida session input errors to their owners (#11233)

* refactor(providers): move closeAnalysisClient to src/providers and the Frida session input errors to src/frida

Pure move with git mv and import specifier updates.

* test(architecture): drop the application-providers and application-frida cycles from the layer baseline

* fix(ci): keep provider lanes selecting the ports moved to src/ports

Each lane glob that matched a moved port under src/application gains its src/ports counterpart; the change-scope test pins one moved port per lane.

* refactor(domain): declare the localPath host exception and read HOST_PLATFORM there (#11256)

absoluteLocalPathStringSchema decides which paths are absolute on the running host, so it reads HOST_PLATFORM instead of process.platform. Domain may not import outside domain/contracts; module-layers now declares LAYER_EXCEPTIONS, exact file-to-file imports with a reason, and both the layer count and the pure-layer boundary skip only that import. Any other domain -> platform import still fails.

* refactor(ports): move the doctor check and provider inspection types to src/ports (#11246)

* refactor(ports): move the doctor check and provider inspection types to src/ports

Declaration move from application/Doctor.ts (unchanged text) and import updates; no re-export from Doctor.ts.

* test(architecture): lower the layer baseline after moving the doctor provider types

* refactor(ports): move the packet capture and signal recording ports to src/ports (#11249)

* refactor(ports): move the packet capture and signal recording ports to src/ports

Declaration move (unchanged text) out of the two application services, with import updates.

* test(architecture): lower the layer baseline after moving the capture and signal ports

* refactor(domain): move the analysis error projection schema from contracts to domain (#11251)

* refactor(domain): move the analysis error projection schema from contracts to domain

Declaration move (unchanged text) of the projection schemas that domain modules embed; contracts/errorSchemas.ts keeps the CLI envelope and the published JSON Schema. Import updates only.

* test(architecture): drop the contracts-domain cycle from the layer baseline

* refactor(ports): drop the imports the moved capture, signal and doctor types left behind

Also give the packet capture and signal recording lanes the new port paths, with change-scope cases.

* refactor(resources): add src/resources and move AnalysisCleanupLease and ArtifactReader there (#11259)

* refactor(resources): move AnalysisCleanupLease to src/resources

Pure move: git mv plus import specifier rewrites (source, tests, and the dist paths the verifier scripts load).

* refactor(resources): declare src/resources as a foundation unit

src/resources owns the shared analysis resource modules that process, windows and providers hold; AnalysisCleanupLease is the first. Its tests join the adapters project, the apk signature lane selects it beside src/artifacts, and the layer baseline drops the process and windows imports of the lease.

* test(ci): pin the apk signature lane on the moved cleanup lease

* refactor(resources): move ArtifactReader to src/resources

Pure move: git mv plus import specifier rewrites (source, tests, and the dist path a verifier script loads).

* test(architecture): lower the layer baseline after moving ArtifactReader

windows no longer imports src/artifacts; process -> artifacts 11 -> 9. The CI lanes that selected src/artifacts/ArtifactReader.ts select its new path, pinned in the change-scope test.

* refactor(resources): move ArtifactResourceScope to src/resources (#11265)

* refactor(artifacts): name the output tree and private root capabilities the resource scope releases

ArtifactResourceScope typed its output-tree and private-runtime-root owners with SafeOutputTree (src/artifacts) and PrivateRuntimeRoot (src/process). It only calls rollback() and close(), so it now declares those capabilities itself and names neither unit, which lets it move to src/resources without a resources <-> process cycle. The withheld-export test narrows to SafeOutputTree where it reads the published flag.

* refactor(resources): move ArtifactResourceScope to src/resources

Pure move: git mv plus import specifier rewrites (source, tests, the dist paths verifier scripts load, and the FIFO probe's module-relative URL).

* test(architecture): lower the layer baseline after moving ArtifactResourceScope

process -> artifacts 9 -> 5, and the artifacts <-> dotnet mutual pair is gone. The sqlite and agentenv lanes select the scope at its new path, pinned in the change-scope test.

* refactor(resources): move ArtifactFileOperation to src/resources (#11269)

* refactor(resources): move ArtifactFileOperation to src/resources

Pure move: git mv plus import specifier rewrites (source, tests, and the dist paths verifier scripts load).

* test(architecture): lower the layer baseline after moving ArtifactFileOperation

process -> artifacts 5 -> 1.

* refactor(resources): move readStableArtifact to src/resources (#11271)

* refactor(resources): move readStableArtifact to src/resources

Pure move: git mv plus import specifier rewrites.

* test(architecture): lower the layer baseline after moving readStableArtifact

process no longer imports src/artifacts: process -> artifacts 1 -> 0 and the artifacts <-> process mutual pair is gone. The seven lanes that selected readStableArtifact select its new path, pinned in the change-scope test; the architecture map lists what src/resources owns.

* refactor(hopper): take Linux launcher verification from composition through a port (#11278)

The Hopper launcher imported inspectLinuxHopperLauncher and retryLinuxHopperPublisherCleanup from src/application (the two hopper -> application imports). src/ports/LinuxHopperLauncherVerification.ts now names that capability; src/application/LinuxHopper.ts provides LINUX_HOPPER_LAUNCHER_VERIFICATION, binary composition passes it to HopperProvider, and the provider hands it to the launcher. A launcher built without one refuses a Linux demo launch with reason verification_unavailable before running the launcher. Layer baseline: hopper -> application 2 -> 0.

* refactor(config): move Hopper launcher discovery to a src/hopperPaths foundation unit (#11292)

* refactor(hopper): move the launcher discovery and macOS Hopper paths to src/hopperPaths

Pure move: git mv plus import specifier rewrites.

* refactor(hopper): declare src/hopperPaths as a foundation unit with the Linux launcher candidates

Hopper install paths and launcher discovery are pure path facts that the
Hopper provider, doctor and setup all read. Declare src/hopperPaths in the
foundation layer, add it to the adapters test project, the architecture
map and the package-baseline CI lane, and move the Linux launcher
candidates out of src/application/LinuxHopper.ts into
src/hopperPaths/LinuxHopperPaths.ts so discovery no longer imports
application code.

* refactor(config): leave Hopper launcher discovery to the Hopper provider

parseConfig resolved an unset HOPPER_LAUNCHER_PATH by probing the
filesystem through application-owned discovery, so config imported
application (config -> application 2, application <-> config cycle) and
read process.platform ambiently.

AppConfig.hopperLauncherPath is now the caller's HOPPER_LAUNCHER_PATH
unchanged (string | undefined). src/hopperPaths/HopperLauncherDiscovery.ts
owns the selection (selectedHopperLauncherPath: the configured spelling,
else discovery from the caller environment's home on the provider's
platform); HopperProvider and the two real-Hopper verifiers read it.
Discovery tests move from config.test.ts to src/hopperPaths; a provider
test covers discovery from the provider's own platform and home.

Module layers: config -> application 2 -> 0; application <-> config removed.
Architecture guards: ambient platform reads in parseConfig.ts 2 -> 0.

* refactor(providers): move owned-process prerequisite availability out of application (#11294)

* refactor(providers): move the owned-process prerequisite availability mapping to src/providers

Pure move: git mv plus import specifier rewrites.

* test(architecture): lower the layer baseline after moving owned-process availability

android -> application 1 -> 0; apktool -> application 1 -> 0.

* refactor(process): move the regular-file read owner out of application (#11299)

* refactor(process): move the regular-file read owner from src/application to src/process

Pure move: git mv plus import specifier rewrites.

* test(architecture): lower the layer baseline after moving the regular-file read owner

desktop -> application 5 -> 4; dotnet -> application 2 -> 1.

The agentenv lane listed src/application/RegularFileRead.ts. At its new
path it is a src/process/ shared boundary: fast qualification selects the
code and runtime checks (and the single owning lane, web_source_map, via
src/process/**), and full qualification selects every lane, as for every
other src/process/ module. A second owner entry would only suppress the
web_source_map selection, so the agentenv entry is removed, not renamed.

* refactor(resources): give atomic text writes their own owner (#11302)

* refactor(resources): give atomic text writes their own owner in src/resources

src/application/JsonFiles.ts owned both JSON input reads and the atomic
text write (writeTextFile / writeTextParts and their staged-output
publication). The write side depends only on domain, filesystem and
resource owners, but il2cpp's Ghidra seed writer had to import
application to reach it (il2cpp -> application).

Move the write side verbatim to src/resources/TextFileWrite.ts
(TextWriteCancellation, writeTextFile, writeTextParts and their private
staging, publication and failure classification helpers). JsonFiles keeps
the JSON read owner and reads the shared errno-to-reason classifier
(fileAccessFailureReason) from the write owner instead of a copy.
Importers (il2cpp seed writer, evidence bundle files, desktop build
solves, analysis snapshot files and four boundary tests) read the new
path; no re-export is left in JsonFiles.

* test(architecture): lower the layer baseline after moving the text write owner

il2cpp -> application 1 -> 0.

* refactor(ports): move the desktop build host contract to src/ports (#11303)

* refactor(domain): move the setup cancellation check from src/application to src/domain

Pure move: git mv plus import specifier rewrites.

* refactor(ports): move the desktop build host contract to src/ports

The desktop provider implements doctor's desktop build facts reader and
setup's desktop build host, but the contract it implements lived in
application (DesktopBuildDoctor, SetupDesktopBuild, Doctor's
clientRegistrations signature, ClientRegistrationEnvironment's setting
types), so desktop imported application four times and the two units
were a recorded mutual pair.

src/ports/DesktopBuildHost.ts now holds that contract, moved verbatim:
DesktopBuildCommand, DesktopBuildFacts, DesktopBuildFactsInput,
DesktopBuildSetupFacts, DesktopBuildSetupHost, RecordedProviderSettings
and OwnedProviderSetting. Two names are new: DesktopBuildRegistration,
the registration facts the host reads (client, command and the
recorded-settings names; client registration status satisfies it), and
DesktopBuildRegistrationReader, which replaces
DoctorHost["clientRegistrations"] in the setup host. The facts input's
registration reader returns DesktopBuildRegistration instead of the full
application status; the setup facts' settings are spelled
Readonly<Record<string, string>> (SetupProviderEnvironment's definition).
Application imports the types from the port; no re-export is left.

* test(architecture): lower the layer baseline after the desktop build host port

desktop -> application 4 -> 0; the application <-> desktop mutual pair is gone.

* refactor(domain): define the resolved app bundle resource root beside its schema (#11304)

* refactor(domain): define the resolved app bundle resource root beside its provenance schema

ResolvedAppBundleResourceRoot (an app directory's resource-only identity)
is a pure type over BundleResourceRoot's layout, but it lived in
src/application/AppBundleExecutable.ts, so the native and artifact
providers that receive it imported application.

Move the interface verbatim to src/domain/apple/bundleResourceRoot.ts,
which owns BundleResourceRoot, and point its eight readers there (native
macOS provider and plist source, artifact provider, binary composition,
and the binary session, status, port and bundle-resource session). The
resolver that produces it stays in application.

* test(architecture): lower the layer baseline after moving the resource root type

native -> application 2 -> 0; artifacts -> application 2 -> 1 (the
artifact provider still constructs BundleResourceSession).

* refactor(config): move the pwntools input bound parser to src/config (#11306)

* refactor(config): move the pwntools input bound parser from src/native to src/config

Pure move: git mv plus import specifier rewrites, and the module path in docs/design/rea-limit-inventory.md (table realigned by oxfmt).

* test(architecture): lower the layer baseline after moving the pwntools input bound

config -> native 1 -> 0. The binary_layout and recorded_crash lanes, which
select src/native/pwntools/**, also select src/config/pwntoolsInputLimit*.ts;
ciChangeScope.test.ts pins both.

* refactor(filesystem): move REA cache directory helpers out of config (#11309)

* refactor(filesystem): move the REA cache directory and lexical directory helpers from src/config to src/filesystem

Pure move: git mv plus import specifier rewrites.

* test(architecture): drop the config and filesystem cycle from the layer baseline

config no longer imports filesystem; filesystem still reads the Windows
ChangeTime settle budget from config.

* refactor(windows): break the process and windows import cycle (#11317)

* refactor(domain): move the runtime-control and private-runtime-root refusals from src/process to src/domain

Pure move: git mv plus import specifier rewrites (including the dist path scripts/verify-windows-native.mjs loads).

* refactor(windows): type Job Object cleanup by what a Windows job reports

WindowsOwnedProcess.cleanup() returned process-ownership's
ProcessCleanupResult, so src/windows imported src/process for a type while
process imports windows to supervise Windows trees. A Job Object stop only
ever reports settled-and-signalled or a reason it did not settle; name that
subset WindowsJobCleanupResult in windows. It stays assignable to
ProcessCleanupResult, so the process supervisor reads it unchanged.

* test(architecture): drop the process and windows cycle from the layer baseline

* test(architecture): ratchet cross-provider imports and imports past a provider entry (#11386)

* test(architecture): ratchet cross-provider imports and imports past a provider entry

A provider folder may import the domain, contracts, foundation and ports layers, the shared provider core (src/providers) and its own files; an import of another provider folder is a recorded exception. Application, composition and the CLI/MCP adapters reach a provider only through src/<provider>/index.ts. verify:module-boundaries counts both kinds of import per unit pair in the layer baseline (crossProvider, providerEntry); like the upward counts they may only shrink, and a pair missing from the baseline fails.

* refactor(resources): move StableArtifactFailure and StableArtifactObservation into src/resources (#11387)

* refactor(resources): move StableArtifactFailure and StableArtifactObservation into src/resources

Providers project stable-read failures through StableArtifactFailure next to the other shared provider toolkit owners (ArtifactReader, ArtifactResourceScope, ArtifactFileOperation, AnalysisCleanupLease, readStableArtifact) that already live in src/resources. Under src/artifacts it made every provider that reads a stable artifact import the artifacts provider. Pure move: git mv plus import specifier and baseline updates; the layer baseline drops the provider-boundary counts the move removes.

* refactor(resources): reach the shared provider toolkit through src/resources/index.ts (#11390)

* refactor(resources): import the shared provider toolkit through src/resources/index.ts

Adds src/resources/index.ts, which exports the toolkit names that production code outside src/resources uses, and rewrites those importers (420 files) to import from it, merging each importer's toolkit imports into one declaration. Mechanical: import declarations only.

* test(architecture): reach src/resources only through its entry file

verify:module-boundaries now counts any production import of a src/resources file other than src/resources/index.ts, from any unit, with the imports past a provider entry. The count is renamed from providerEntry to entry ("import past the entry file"); it may only shrink, and src/resources starts at zero.

* test(architecture): cover provider boundary counts through the verifier command

* refactor(resources): move the safe output tree and artifact path owners into src/resources (#11398)

* refactor(resources): move the safe output tree and artifact path owners into src/resources

SafeOutputTree (with its admission, creation failure, readback, cleanup and identity files), StreamBytes, ArtifactPaths and readArtifactFileChunks depend only on domain, filesystem, platform and the toolkit, and browser, firmware, ghidra, javascript, android and apktool use them to publish outputs. Under src/artifacts every such provider imported the artifacts provider. Pure move: git mv, importers moved onto src/resources/index.ts (names added to the entry), the moved files import toolkit modules directly, path keys in the CI scopes and guard baseline follow the move, and the layer baseline drops the removed imports.

* test(architecture): give the shared provider core its own layer below the providers (#11401)

src/providers (deferred client owners and failure attribution) now imports only domain, ports, process, filesystem and the resources toolkit. Declaring it as the providerCore layer, between ports and providers, makes any import from it into a provider folder an upward import that verify:module-boundaries rejects, and lets every unit in the providers layer be a provider folder.

* test(architecture): declare each provider's entry files instead of a provider barrel (#11407)

* chore: ignore the source map parser move (#11348) in blame

* test(architecture): declare inspector's use of the browser CDP client as a provider dependency (#11420)
2026-11-16 03:56:10 +08:00
morluto 02ebd3e02d refactor: prune test-only seams and obsolete helpers (#10275) 2026-11-12 19:24:54 -05:00
morluto de0c50ead7 test: assert artifact integrity and retained descriptor ownership 2026-11-10 16:41:40 -05:00
morluto e8d1bbc98b fix(reference): preserve import cancellation provenance (#9597)
* fix(reference): preserve import cancellation reason

* test(reference): cover null cancellation during failed cleanup

* test(reference): always release fixture directory
2026-11-09 19:42:00 +08:00
morluto 352a846127 test: reduce brittle assertions and test-only runtime paths (#9289)
* test: focus assertions on behavioral evidence

Prune redundant and wording-coupled checks while retaining diagnostic, parser, recovery, and cleanup coverage.

* refactor: remove test-only wrappers and duplicate checks

Exercise production operations through test fixtures, remove unused adapters, and keep managed-member uncertainty guidance under its existing owner.

* test: focus process validation on diagnostic facts

* refactor: remove test-only runtime helpers

Move semantic evidence fixture setup out of runtime and remove unused Windows settle recovery helpers while checking configured defaults independently.

* test: focus artifact diagnostics on format facts

* test: verify retained cleanup guidance delivery
2026-11-08 02:02:31 +08:00
morluto 943ae604df refactor(reference): derive import lookup from reader inventory 2026-10-26 15:14:05 +08:00
morluto 4610843efd fix(filesystem): preserve failure provenance across cleanup owners
Carry descriptor outcomes, cancellation, retention and physical origin as named facts so wrappers cannot drop ownership context. Replace raw primary failure unwrapping with provenance facts across artifact consumers, and use Result for Kuna span reads including undefined rejections.

Exercise actual descriptors, rollback, cancellation and cleanup retry; update brittle snapshot and completion fixtures to use their producing boundaries.
2026-10-24 07:43:49 +08:00
morluto a3eae6e236 test: give oversized registration groups semantic owners 2026-10-24 04:23:56 +08:00
morluto cae03361cf fix(reference): separate local filesystem diagnostics from graph commitments 2026-10-24 03:11:07 +08:00
morluto c3e4c116d5 fix(reference): retain unknown package scope for JavaScript loaders 2026-10-24 02:41:05 +08:00
morluto 5e8f8b3163 fix(paths): honor physical caller selection and refuse extraction overlap 2026-10-24 02:21:58 +08:00
morluto 857a75baac fix: preserve unknown historical module resolution
Fixes #6679

Fixes #6678

Fixes #6612
2026-10-23 21:28:10 +08:00
morluto 949966712e fix(files): preserve physical read failures across late cancellation
Tag physical read outcomes at the owning boundary and preserve opaque caller cancellation reasons through JSON pipelines and reference imports. Share named bounded-read options across consumers and retain failed cleanup ownership for retry.

Closes #5686
2026-10-21 06:14:03 +08:00
morluto 0f4c484be5 fix(reference): preserve missing symlink target facts
Treat ENOTDIR while resolving a symlink target as missing target evidence, preserving its spelling and partial inventory. Keep graph-level excluded-ancestor rejection separate from import success. The compiled boundary regression passes.
2026-10-14 08:43:59 +08:00
morluto ec271121e4 fix(artifacts): preserve path identity and exclusion applicability
Normalize path facts once at the owning filesystem boundary, preserve caller spelling and unavailable inventory members, and retain typed read failures. Record exclusion entry kinds so only proven directories can establish descendant symlink targets; omitted legacy kinds stay unknown.

Reviewed fixes include #1969, #2642, #2759, and #2994.

Validation: 26 reference graph source tests passed; prior artifact checks passed 275 tests with one Windows-only skip. Updated compiled import workflows are included in final validation after regeneration.
2026-10-14 08:31:16 +08:00
morluto 1b953021ea fix(javascript): centralize module syntax and separate erased type dependencies 2026-10-13 14:39:06 +08:00
morluto 53e7fadbd0 fix(reference)!: retain source and metadata cleanup owners
Require a caller-owned artifact resource scope for source traversal and Git metadata reads. Preserve cancellation, filesystem reasons, partial entries and bytes, and cleanup observations through import and the CLI. Transfer exact failed file owners at regular-file admission and close boundaries, including multiple failures encountered by Git reference resolution.

Validation: 169 focused tests passed across 26 files; two Windows-only cases skipped. Build, typecheck, lint, formatting, knip, architecture guards, discovery and docs checks passed.

BREAKING CHANGE: Reference source readers and imports require an ArtifactResourceScope. VCS metadata reads return Result, and CLI import failures include observed details and partial evidence.
2026-10-11 06:10:10 +08:00
morluto 6636045c9c fix(reference): match exclusions against observed entry kinds
Inspect each entry without following links before applying directory-only patterns. Retain the winning rule and original entry path while pruning excluded directories before reading their children.

Validate directory exclusions, negation, symlinks, explicit secret rules, and compiled CLI provenance alongside reference reader failure and cancellation coverage.
2026-10-11 04:43:02 +08:00
morluto 6585293c56 fix(reference)!: preserve exclusion rule provenance
Record the exact matching pattern and distinguish project, default, caller, and configured-sensitive exclusions without changing ignore ordering. Include that provenance in graph commitments and validate the contract through compiled CLI and MCP consumers.

BREAKING CHANGE: Pattern-based reference graph exclusions require pattern and distinguish project-ignored and default-ignored reasons. Re-import saved graphs that omit matching patterns.
2026-10-11 04:15:26 +08:00
morluto ad59531a96 refactor: remove forwarding exports and redundant aliases 2026-10-09 22:46:09 +08:00
morluto b7b4333703 fix(reference)!: represent unreadable symlink targets explicitly
BREAKING CHANGE: Historical source symlinks with unreadable or unknown target_state require target: null. Placeholder target strings are rejected.
2026-10-09 21:01:52 +08:00
morluto 63bf91405c refactor!: enforce canonical evidence and provider boundaries
Deliver complete Evidence directly through MCP with required nullable analysis profiles and immutable per-server delivery settings. Bind selected environments at construction and require typed cleanup outcomes with retained retry ownership.

Remove test-only routing, duplicate session and snapshot metadata, producer-field repairs, and ambiguous provider arguments. Separate artifact byte identity from occurrence classification and retain typed source-map and partial native facts across CLI, MCP, and verifier consumers.

BREAKING CHANGE: obsolete Evidence envelopes, session aliases, incomplete snapshots, and missing normalized producer facts are no longer accepted.
2026-10-09 20:18:52 +08:00
morluto ffd9611308 test: consolidate failure guards and trim unused fixture APIs 2026-10-09 03:13:43 +08:00
morluto a1b7d40bb0 test: remove duplicate result assertions before throwing guards 2026-10-09 02:58:03 +08:00
3af3811ab0 fix: integrate 24 reviewed browser/native/semantics corrections (#815)
* fix(reference): resolve emitted extensions to TypeScript source

Signed-off-by: Rudy Celekli <rudy@gradiahq.com>

* test(reference): retain legitimate declaration import counterparts

Signed-off-by: Rudy Celekli <rudy@gradiahq.com>

* fix(mcp): recognize implicit form elicitation capability

Signed-off-by: RudyCelekli <47457359+rudycelekli@users.noreply.github.com>

* fix(ghidra): reject contradictory NativeAOT reports

* fix(native): preserve thin architectures across echoed filename lines

Signed-off-by: Rudy Celekli <47457359+rudycelekli@users.noreply.github.com>

* fix(native): preserve app bundle executable filename whitespace

Signed-off-by: Rudy Celekli <47457359+rudycelekli@users.noreply.github.com>

* fix(doctor): validate Node-wrapped client policy

Signed-off-by: RudyCelekli <47457359+rudycelekli@users.noreply.github.com>

* fix(reference): preserve observed entry failure reasons

Signed-off-by: Rudy Celekli <rudy@gradiahq.com>

* fix(artifacts): verify opaque bytes before reusing inventory identities

Signed-off-by: Rudy Celekli <rudy@gradiahq.com>

* fix(browser): resolve DOM destinations against document base URL

Signed-off-by: Rudy Celekli <rudy@gradiahq.com>

* fix: recover serialized native NIB view hierarchy

Signed-off-by: RudyCelekli <47457359+rudycelekli@users.noreply.github.com>

* ci: exercise native compiled NIB hierarchy regressions

Signed-off-by: RudyCelekli <47457359+rudycelekli@users.noreply.github.com>

* fix(deps): remediate reachable archive-parsing advisories

Two high-severity advisories are reachable from untrusted target content:

- @xmldom/xmldom 0.9.10 (via plist) parses XML plist bytes from analyzed
  app bundles in the keyed-archive reader. Refreshed to 0.9.12.
- brace-expansion 5.0.7 (via @electron/asar -> minimatch) is reached while
  analyzing untrusted ASAR archives. Refreshed to 5.0.12.

Both are in-range patch updates (plist declares ^0.9.10, minimatch declares
^5.0.5), so no source or API change is required and no direct dependency
declaration moves. Lockfile-only change.

The remaining smol-toml advisory is not included: it parses operator-owned
client configuration rather than untrusted target content, and its fix moves
outside the declared range and changes leading-BOM handling.

* fix(semantics): link wrapped arrow Promise returns

* fix(reconstruction): preserve unknown graph limitations

* fix(native): preserve exact code signing identifiers

Signed-off-by: Rudy Celekli <47457359+rudycelekli@users.noreply.github.com>

* fix(native): recognize only whole otool command headers

Signed-off-by: Rudy Celekli <47457359+rudycelekli@users.noreply.github.com>

* fix(native): retain nm-proven literal export names

Signed-off-by: Rudy Celekli <47457359+rudycelekli@users.noreply.github.com>

* fix(javascript): project decoded file URL paths

Signed-off-by: Rudy Celekli <47457359+rudycelekli@users.noreply.github.com>

* fix: preserve explicit source-relative URL observation spelling

Signed-off-by: Rudy Celekli <47457359+rudycelekli@users.noreply.github.com>

* fix(native): project Mach-O evidence into container file offsets

Signed-off-by: Rudy Celekli <47457359+rudycelekli@users.noreply.github.com>

* fix(browser): normalize referrer policy fallback lists

Signed-off-by: Rudy Celekli <47457359+rudycelekli@users.noreply.github.com>

---------

Signed-off-by: Rudy Celekli <rudy@gradiahq.com>
Signed-off-by: RudyCelekli <47457359+rudycelekli@users.noreply.github.com>
Signed-off-by: Rudy Celekli <47457359+rudycelekli@users.noreply.github.com>
Co-authored-by: Rudy Celekli <rudy@gradiahq.com>
Co-authored-by: RudyCelekli <47457359+rudycelekli@users.noreply.github.com>
Co-authored-by: Kaoru0822 <aisakamimaki@gmail.com>
Co-authored-by: DeryFerd <deryferdikao125@gmail.com>
2026-10-07 10:33:10 +08:00
N0zoM1z0 911d580634 fix(cli): preserve unsupported historical-source imports (#795) 2026-10-07 04:02:10 +08:00
Rudy Mizrahi Celekli c8aed4740d fix(reference): preserve unresolved absolute module specifiers (#670)
Signed-off-by: RudyCelekli <47457359+rudycelekli@users.noreply.github.com>
2026-10-06 16:18:17 +08:00
Kaoru0822 1a24be574d fix(reference): recognize CMake build manifests (#665) 2026-10-06 16:17:50 +08:00
morluto b5e98915c6 fix(tools)!: simplify direct invocation and harden boundary contracts (#555)
Remove REA permission grants, scope ceilings, approval confirmations, and inert settings. Preserve selected local evidence, accept caller-chosen commands and paths, and report actual runtime failures.

Harden provider results, schema serialization, snapshot evidence binding, archive paths, source maps, native screenshots, managed EH parsing, and cancellation cleanup. Regenerate tool metadata and document direct invocation.

BREAKING CHANGE: Removed permission configuration and policy commands, approval fields, filesystem scope inputs, and fixed confirmation options. Process capture inherits the host environment and accepts command names; callers must refresh tool schemas and migrate removed fields.
2026-10-05 20:16:44 +08:00
morluto 7ee3f236a9 test: replace implementation lock-in with behavioural assertions (#498)
* fix(process): normalize reactive rendered-frame timestamps

A reactive `resize` action stamped `rendered_frames[].at_ms` with the raw
elapsed time while every other rendered frame used
`normalizeProcessElapsedTime`. Under the default 10ms time bucket the next
frame buckets down and appears to move backwards, so `parseProcessCapture`
rejected the entire capture with "rendered_frames: timestamps must be
ordered". No test combined a reactive resize with time bucketing, so the
path was never exercised.

Thread the scenario's bucket into the reactive effect host and normalize the
renderer call, matching what the scheduled-event resize path already did.
`dispatched_at_ms` and `captured_at_ms` stay raw on both paths, so this
changes only the one inconsistent clock.

Also replace the wall-clock-scheduled interactive PTY scenario with
terminal_text triggers. A fixed `at_ms` could fire before the child wrote
its prompt under host contention, dropping the input and the resize echo.
That rewrite is what surfaced this bug, and it now passes deterministically.

Replace the 11 `if (!capability.available) return;` guards across the PTY
suite with a shared it.skipIf gate. The old form reported a green pass with
zero assertions on any host lacking native PTY authority; skips are now
named and reported.

Verified by reverting only the normalization, which reproduces the
validation failure immediately.

* fix(install): add recovery guidance to version-mismatch failure

Every installer failure except one told the user what to do next. The
version-mismatch failure reported only that the installed version did not
match, with no remedy, which contradicts the requirement to keep
diagnostics actionable for mismatch locations.

Add a remedy in the same style as the neighbouring failures and drop the
exemption the test carried for it, so all ten installer failures now assert
they carry a recovery action.

* ci: repair stale curated Windows lane and unblind the duplication gate

The `windows-curated` job listed fifteen test files and nine of them no
longer existed, including six `processOwnership*` files moved to
`src/process/` in 4ecc7a72. Vitest silently drops unmatched paths, so the
job ran six files and reported success. Repoint the paths at their current
locations and add a test that fails when any listed path is renamed or
removed, so the lane cannot silently lose coverage again.

The duplication gate ignored `**/*.test.ts`, so it reported zero clones
across the test tree while running on every PR. Include test files, remove
the two largest duplicated clusters, and ratchet the threshold to the
measured result so the gate can no longer pass by not looking.

Run local tests with two workers instead of one. A single worker made a
full local run cost ~140s of wall time at ~100% CPU on a 10-core host,
which was the largest dev-cycle cost in the repository. Local now matches
the concurrency CI already proves green, and PTY scenarios moved to a
serial project because they contend for host terminal resources. Full local
suite: 140s to ~89s.

Regenerate the managed conformance manifest. Commit 4ecc7a72 edited
SKILL.md without refreshing the digest, so `npm run check:pr` has been
failing on main since that commit.

* test: replace implementation lock-in with behavioural assertions

Audit the suite for tests that cost maintenance without catching defects,
then remove the lock-in rather than the coverage.

Replace assertions that froze incidental detail:

- toolContracts.test.ts pinned group sizes and 50 lines of literal tool
  names, so any additive catalog change required a test edit. Replace with
  four structural invariants: global name uniqueness, group containment in
  the published inventory, group disjointness, and non-empty groups. Both
  invariants are mutation-checked by orphaning and duplicating a tool.
- Six tests asserted only that a schema accepts its own canonical fixture.
  Replace five with real rejections and two with property tests covering
  the invariants those tests were gesturing at: `isDbSaveComplete` is
  downward closed in every expected count, and frame retention holds at
  any size.
- A concurrency test whose fake returned `? [] : []` could not fail. Give
  the fake real shared state so a lost update is observable.
- providerProcess.test.ts asserted listener and timer counts plus exact
  cleanup retry counts. Assert detachment behaviour instead, so an
  unrelated internal timer no longer breaks the test.
- Scale tests that rebuilt removed caps: 16 MiB and 10,001 files to prove
  limits that no longer exist, 100,001 frames compared against their own
  length, and a projection asserting a literal 10,201 bridge candidates.
  Replace with derived relationships that hold at any size.
- installScript.test.ts merged two tables and now asserts alongside the
  exact wording that every failure carries a recovery action.
- cdpBrowserProvider.sensitive_data asserted outcomes produced after the
  observation window closes while using a zero-length window; give it a
  real window so it measures the malformed-body rejection rather than host
  speed.
- runtime.test.ts repeated the catalog assertion from the MCP boundary
  suite while spawning a second stdio server. Give the database-kind
  scenario its own claim and make stale-build failures legible.

Extract two byte-identical fixture preambles that the max-lines rule had
split across five and three files respectively, removing ~1,000 duplicated
lines.

Profile the heaviest files rather than only running them: the Apple
projection test drops from 6.4s to 0.28s and the reference source import
from 2.3s to 0.2s, with stronger assertions in both cases.

* ci: keep native-authority suites out of the curated Windows lane

CI showed the repointed curated list was red. Three of the restored files
require verified Windows native authority, which windows-latest does not
provide by design:

- src/ghidra/GhidraProvider.test.ts, the replacement for the deleted
  tests/boundary/providers/ghidra/ghidraProvider.test.ts
- tests/boundary/providers/ghidra/ghidraClient.test.ts
- tests/boundary/process/providerProcess.test.ts

All three fail with PrivateRuntimeRootUnavailableError, because Job Object
process ownership and private-runtime DACL enforcement are unimplemented.
The previous ghidraClient and providerProcess entries were added in this
branch and never belonged here; the GhidraProvider replacement does not
inherit the deleted file's Windows behaviour.

Those claims belong behind the same approval boundary as the real Ghidra
Windows workflow rather than the default lane. Record the exclusion in the
guard test so the lane cannot regress to red, and in ci.yml so the next
reader knows the omission is deliberate.

The lane now runs 14 files, up from the 6 it silently ran before, all
verified green on windows-latest.

* test: stop max-lines from forcing test files to split

The 500-line cap on test files only ever produced duplication. It chopped one
binary session suite across nine files whose names were truncated
mid-sentence, and it is why the curated Windows lane referenced six
`processOwnership*` paths that had been split apart in the first place.

Turn the cap off for tests and fixtures, keeping the function-length limit,
then merge the nine binary session files back into one suite. That restores
all 22 cases under a single module registry instead of nine and cuts the
composition project from 20 files to 12. The merged cases become nine
top-level describes grouped by concern, because a single wrapping describe
would itself breach the function-length limit.

Drop applicationWorkflowVerifier.test.ts. Its first case asserted
`verified: true`, which the script emits as a literal after its own checks,
so it could not fail, and every remaining assertion was `expect.any`. Its
second case asserted an export-shape comparison that
applicationWorkflowCli.test.ts already asserts through the product CLI,
which is the stronger claim.

Finally, five javascriptSemanticAnalysis files imported `it` from
@fast-check/vitest while containing no properties. Import `it` from vitest
so those files no longer read as property-based coverage.

* test: stop creating temp scaffolding the temp helper immediately deletes

`createTestTempDirectory` called `createTestWorkspace`, which creates `home`
and `xdg/{config,cache}`, then deleted two of those three directories so the
caller would receive an empty root. Every caller paid three `mkdir` calls and
two `rm` calls per temporary directory for scaffolding it never used, across
roughly a hundred test files.

Create the directory directly instead. Callers that genuinely need a home or
XDG tree can call `createTestWorkspace`.

* test: merge the split process reactive runtime suite

Five files shared one describe name in pairs and re-imported the same fixture
helpers. With the shared preamble already extracted into
processReactiveRuntime.fixture.ts, merge them into one suite of five
top-level describes grouped by concern. All 12 cases are preserved and the
domain project drops from 92 files to 88.

* test: drop two identity round-trips from node runtime characterization

One test built a plan with `max_calls: 10_001` and asserted the field came
back as 10_001, which asserts that the plan constructor copies a field the
test just set, and freezes a former ceiling in the process.

The other built a 201-character alias and asserted it parsed back to the
same string, which round-trips a value the test created.

Neither could fail for a reason that matters. The remaining cases still
cover preparation, stale-plan rejection, single dispatch, and artifact write.

* test: hoist duplicated graph fixtures into the shared fixture module

Eleven copies of three helpers were spread across the seven
javascriptApplicationGraph files: `completeCoverage` five times,
`nodeByLabel` four times, and `firstOf` twice, all byte-identical. The
fixture module already carried a private `completeCoverage`.

Export all three from javascriptApplicationGraph.fixture.ts and import them,
removing 95 duplicated lines.

* test: separate the reactive fixture's two PTY writes

The multi-source run resolved target_lost instead of passed intermittently.
The ready transition matches terminal text "Ready" and consumes that
observation, advancing the terminal frontier; the collecting state then
needs "Collecting" in a later observation. The fixture wrote both back to
back, so the PTY sometimes coalesced them into one chunk and the collecting
trigger could never match.

Move the "Collecting" write after the synchronous shim probe. The probe takes
real wall time, so the reader has drained the first chunk before the second
is written, which makes the ordering deterministic rather than timing
dependent.

Measured 33/34 isolated passes, against roughly even odds before. Under four
concurrent copies it still failed 2 of 12, so this narrows the window rather
than closing it: the underlying race is that node-pty delivers buffered
output through a callback the exit path never awaits, so the harness can
declare the target lost before the last chunk is ingested. The test is left
strict because it is correctly detecting that.

* test: replace expect.any placeholders with real results

Four acceptance scenarios and one composition scenario asserted only that a
field was a string, number, object, or array. Those assertions cannot fail
for a reason that matters, and three of them sit behind real CLI process
spawns that cost about 1.5s.

Capture the actual results once and assert them exactly. The CLI parity
scenario now pins the trace summary and coverage, the version comparison
summary and coverage, and the source-to-bundle summary and coverage, so each
spawn earns its cost. The MCP rejection case now asserts `invalid_request`
with category `invalid_input`, which is what makes it a caller-input failure
rather than a provider one.

The artifact comparison case asserted three `expect.any` values immediately
before an assertion that subsumed them. Replace them with the claim that
actually matters above 500 entries: manifest counts must reconcile with the
arrays returned, and every written file must appear as an occurrence, so a
reintroduced truncation would fail.

* test: remove impossible and incidental prompt assertions

Two assertions in toolResult could never fail: each asserted that the
serialized result did not contain a string that appears nowhere in its own
inputs, so only in the assertion itself.

promptContracts asserted that a `Set` built from a locally declared six
element literal had no duplicates, and pinned thirteen incidental fragments
of rendered prompt prose. Drop the tautology and reduce the prose block to
the five properties an analyst actually depends on: caller input framed as
data rather than instruction, tools offered as options rather than a
mandated sequence, the epistemic vocabulary present, tools never framed as
authorization, and deterministic suggestion order.

* test: hoist the remaining duplicated graph fixtures

Eight more byte-identical helper copies across the javascriptApplicationGraph
files: `graphForNode` four times, `unknownEvidence` twice, and `contentNode`
twice. Move them into javascriptApplicationGraph.fixture.ts alongside the
three hoisted earlier, and drop the imports they left behind.

Net -92 lines across the seven files.

* test: drop enhancedTools helpers already in the shared harness

Three of the six enhancedTools files still carried local copies of
PROCEDURES, inventory, and emptyBounded that are byte-identical to the
exports added to enhancedToolsHarness.ts. Import them instead.

Only fixturePort, connect, and jsonResult differ per file, so those stay
local.

* test: replace repeated process-tree event literals with fixture builders

The file spelled out nineteen thirteen-field event literals and thirteen-field
process nodes inline, varying only three or four fields each, so a test read as
a wall of repeated boilerplate rather than the transitions under test.

Add eventProcessTree.fixture.ts with spawn, exit, re-exec and reparent
builders plus a process-node builder, and express each case as the sequence it
describes. The test drops from 422 to 214 lines with all nine cases intact.

* test: share the process-ownership validation fixtures

The two process-ownership validation files each declared byte-identical copies
of an `ownership` constant and a 24-line `host` builder. Move both into
ProcessOwnership.fixture.ts and import them.

* test: make rejection and retention assertions exact

The reactive scenario rejection table asserted only that each of its six
fixtures failed to parse. A typo anywhere in a fixture would satisfy that,
including one that invalidated an unrelated field. Record the field each row
is meant to invalidate and require the reported issue to point at that path
with the expected code. Verified by making one row valid except for an
unknown key: the row now fails instead of passing for the wrong reason.

The large-body replay test asserted that the serialized result exceeded five
million characters as a proxy for the bodies being retained, which escaping
or padding could satisfy even under truncation. Compare each retained body to
the input byte for byte and total the retained length. Verified by shaving one
byte off the expected value, which now fails.

* test: share the elicitation grants fixtures

The three processCaptureElicitationGrants files each declared the same five
helpers inline: an elicitation request, a fixed clock, a ceiling-only
permission authority, a request-context builder, and a verified-state stub.
Move them into elicitationGrantsHarness.ts and import them.

* fix(process): wait for trailing PTY chunks before declaring target loss

The exit path drained the reactive coordinator and then submitted
target_lost, but node-pty delivers buffered output through onData callbacks
that run outside the coordinator's queue. A trailing chunk that had not been
delivered yet produced no journal entry, so the drain had nothing to wait for
and the target was declared lost while a terminal trigger was still
satisfiable. That is what made the multi-source reactive fixture resolve
target_lost instead of passed at roughly even odds.

Yield to the event loop until the capture journal stops growing before
draining, bounded so a pathological stream cannot stall exit.

The fixture change that accompanied this is no longer carrying the weight on
its own; the ordering it introduced remains correct and worth keeping.

Measured 18 consecutive passes across isolated and four-way concurrent runs,
against roughly even odds beforehand. A control run with the wait disabled did
not reproduce the failure, so the machine was not loaded enough for a
controlled comparison; the before/after numbers are from separate sessions.

* test: hoist the last duplicated domain and coordinator fixtures

processTraceComparison and its Part2 declared the same eight helpers, and the
two ProcessReactiveCoordinatorFailures files declared the same three. All were
byte-identical. Move them into shared fixture modules and import them, netting
429 fewer lines.

The earlier attempts at this failed because depth-based scanning mis-detected
helper boundaries: bracket characters inside type annotations, such as
`options: { truncated?: boolean }`, unbalance the count. Extracting by verified
line range instead worked first time, and unused imports were then settled by
alternating typecheck and lint rather than by guessing which names a body uses.

* ci: ratchet the duplication gate to the reduced baseline

Test-file duplication is down to 2.58% of lines across src and tests, from
3.39% when tests were first included in the gate. Lower the threshold from
3.2 to 2.7 so the reduction cannot regress.

* test: merge the remaining artificially split test files

Five suites were each split across two or three files by the max-lines rule
that is no longer applied to tests, and every split produced a duplicate
describe name and a duplicated import block.

Merge them back: processTraceComparison, ProcessReactiveCoordinatorFailures,
ProcessOwnership.validation, processCaptureElicitationGrants, and
processCaptureElicitation. Net -62 lines and five fewer files.

Give each merged describe a name describing what it covers, since the splits
were arbitrary and the names now collide. Test bodies are unchanged; every
describe already fitted the function-length limit, so merging them as
separate top-level blocks satisfies it.

Imports were settled by writing each block explicitly after typecheck and lint
disagreed. Lint is right about names that appear only inside string literals,
such as `values(["terminal", "process", "http", "websocket"])`, while
typecheck is right about names a body genuinely references, so neither alone
was sufficient.

* test: merge the graph suite and share the enhancedTools result helper

The seven javascriptApplicationGraph files were one suite with one describe
name repeated seven times. Merge them into a single file with seven
describes named for what each covers, keeping all 18 cases.

Three of the enhancedTools files declared byte-identical copies of a
jsonResult helper. Move it into enhancedToolsHarness.ts alongside the rest of
the shared fixtures.

The remaining enhancedTools files stay split: their fixturePort helpers are
three genuinely different provider doubles rather than duplicates, so merging
them would mean renaming per-file seams and lose the reason each exists.

* ci: repoint the Windows lane at the merged ownership validation file

Merging the split ProcessOwnership.validation files removed the two Part
paths the curated Windows lane referenced. The curated-lane guard test caught
it on CI, which is the failure it was added to prevent.

* test(process): make lifecycle assertions falsifiable and scope them to POSIX

Two assertions could not fail, so neither observed the behaviour it claimed.

The deadline disposal test aborted an already-aborted controller and then read
an immutable interruption, so it passed whether or not disposal released the
timer or the listener. Dispose before aborting instead and assert the owned
timer count reaches zero, which distinguishes a released deadline from a
retained one. Verified by mutation: emptying dispose, dropping the listener
removal, and dropping the timer clear each fail the test.

The supervisor test asserted that a repeated stop still resolved, but stop
caches its promise, so it proved nothing about detachment. Replace it with a
late error and close event after disposal and assert the diagnostic channel
stays untouched, keeping an attached control that proves the channel works.
Verified by mutation: retaining the error listener fails the test.

Also mark the TERM-to-KILL escalation POSIX-only. libuv maps every signal to
TerminateProcess on Windows, where a child that ignores SIGTERM cannot survive
the TERM step, so the escalation and SIGKILL signal code are unreachable there.

* fix(vitest): restore cross-project scheduling for local runs

Dropping the per-project `sequence.groupOrder` made every project run
concurrently. `fileParallelism: false` only serialises files inside a single
project, so `process-boundary` could still overlap with the boundary projects
that observe host process and terminal state, which is the contention that
drops input and resize echoes. The removed blanket per-project serialisation
was slower, but it was never what isolated projects from each other.

Keep the two-worker intra-project parallelism that made local runs fast and
apply the ordering locally only, so CI keeps the concurrency it proves green.

* fix(acceptance): make the database startup record check observable

The check asserted stderr contained no fatal record, but a database-kind
startup that only lists the catalog starts no bridge work and writes no
records at all, so the assertion passed without observing anything. Reading
stderr while the child was still running also raced the pipe, and splitting
the buffer could parse a partial trailing line.

Call a tool so startup records are actually produced, parse only complete
newline-delimited records, and read after the transport closes so delivery is
no longer in question. Guard the result with an explicit non-vacuity assertion
so the fatal check cannot silently stop observing anything.

Verified by mutation: injecting a level-60 record through the stderr logger
now fails the test.

* test(acceptance): restore the application workflow verifier coverage

Nothing else in the repository executes
scripts/verify-local-application-workflows.mjs, so deleting this file left the
only automated check on a shipped script. Regressions in its dist imports,
argument handling, or two-version and export-shape output could ship with a
green suite.

The two cases cost about 350ms and 500ms, so the duplication that motivated
removing them was never the reason to drop the coverage.

Verified by mutation: flipping the reported verification status fails the
suite.

* test(contracts): add a published tool name floor

Every remaining inventory invariant is structural, so deleting a contract from
both its group and the aggregate left the whole file green while the catalog
regenerated without it. Structural checks cannot detect removal; only a
declared baseline can.

Declare the 114 published names as a floor rather than an equality. Adding a
tool still needs no edit, while removing one now fails with a message naming
the tool. Verified by mutation: dropping a contract from its group so the
aggregate loses it too fails the floor.

* test: restore no-ceiling coverage above plausible truncation limits

Four suites claimed a result was complete at any scale while their own inputs
were bounded below where a truncation would appear, so the guarantee they
described was not actually under test. Scale independence cannot be asserted as
a relationship over a fixture smaller than the cap being ruled out.

Raise each fixture past any ceiling a regression would plausibly introduce,
which is far cheaper than the removed cases:
  referenceSourceImport  600 entries/64 KiB -> 5,000 entries with a 4 MiB member
  customProtocolCapture  generated max 40 frames -> 300..1,000
  appleApplication       12 frameworks/16 candidates -> 250/1,600
  nodeRuntimeCharacterization  assert the absent call-budget and alias bounds
    where they would be added, since the budget derives from the case count

Verified by mutation: a 1,000-entry import cap, a 500-frame capture cap, a
100-framework projection cap, a 100,000-call budget cap, and a 200-character
alias cap each fail the corresponding suite.

* fix(process): require a quiet period before declaring the target lost

Returning on the first unchanged journal sample granted a single event-loop
turn of grace. A chunk queued on a later iteration therefore still let
`target_lost` be submitted before the final terminal observation, discarding a
satisfiable reactive trigger under the same load-dependent exit scenario the
drain was added for, and `maxTicks` never bounded that because the loop exited
early every time.

Require several consecutive unchanged turns so a callback arriving on any of
the next few iterations still lands first, while keeping the overall tick bound
so a pathological stream cannot stall exit.

* fix: stop a 4 GiB fixture and measure the drain quiet period in time

The reference import fixture cycled its five sizes across all 5,000 entries,
which made a thousand 4 MiB members and nearly 4 GiB of fixture written
concurrently. It happened to fit locally but cannot fit a runner's disk. Cross
the two boundaries independently instead: entry count by volume of tiny
members, byte size by a single 4 MiB member, for about 4.2 MiB total. Verified
by mutation that a 1,000-entry cap and a 1 MiB per-member cap both still fail.

The drain check counted scheduler turns, but consecutive setImmediate turns
elapse in microseconds, so four of them finish long before a pty callback
delayed under host contention, which is when the trailing chunk matters.
Measure the quiet interval in elapsed time instead, so a late callback within
the window still lands before target loss. Verified by simulation: the
turn-count version missed any delay beyond a few microseconds, the timed
version observes delays up to the quiet interval.
2026-10-03 23:22:07 +08:00
morluto c34efca4cd refactor(files): use caller paths for local evidence files 2026-09-29 14:53:20 +08:00
morluto 2160d4fae5 fix: consolidate boundary ownership and cleanup failure paths (#477) 2026-09-08 10:59:13 +08:00
morluto 513d9b1555 test: overhaul deterministic suite architecture 2026-07-29 22:00:38 +00:00