Commit Graph
817 Commits
Author SHA1 Message Date
66c21572de fix(bun): report the real runtime and enforce worker memory budgets on Windows (#1730)
* fix(bun): report the real runtime and enforce worker memory budgets on Windows

Under Bun, REA recorded Node/V8 identity it did not observe and its JavaScript helper processes ran without a memory bound, because Bun ignores the V8 heap flags REA passes. Source-map trace Evidence claimed runtime node-v8 with Bun's synthetic V8 heap limit (the decoder then rejected that limit, so trace_web_source_location always failed), analysis resource failures reported heap limits Bun does not enforce, the HAR decoder claimed a fixed 192 MiB heap, and doctor reported Bun's Node.js compatibility version as the Node version.

Identify the runtime once from process.versions. Source-map reports record id bun with the Bun and Node.js compatibility versions and a null v8_heap_limit_bytes; the decoder checks the heap budget only for V8 reports and rejects a report from a runtime other than the launching one. The analysis worker reports a null heap limit under Bun, HAR limitations and failures omit heap sizes under Bun, and the doctor node check reports Bun and passes when its Node.js compatibility version is supported.

On Windows under Bun, each helper's owned Job Object now carries a commit limit of its heap budget plus a measured 512 MiB Bun runtime reserve (704 MiB for the codec and HAR decoder, 1536 MiB for the default analyzer). The native addon accepts an optional page-aligned job memory limit on process_spawn, verifies it by readback, associates a private completion port before the child is created, and reports a sticky memoryLimitReached and peak job commit on each poll; inspect reports jobMemoryLimit. A requested limit with an artifact lacking the control refuses the launch, and non-Windows launchers refuse a limit.

Reaching the limit is classified by the job notification rather than Bun's exit status: the analyzer fails the current source with a memory resource failure reporting worker_process_memory_limit_bytes, the codec returns a memory resource failure, and the HAR decoder returns its memory failure with process_memory_limit_bytes. Bun source-map Evidence records process_memory_limit_bytes, verified against the launcher. Node.js launches and outputs are unchanged.

* fix(windows): drain memory notifications after observing exit

---------

Co-authored-by: morluto <76467478+morluto@users.noreply.github.com>
2026-10-11 19:20:00 +08:00
d412def722 fix(setup): register a stable launcher when run through bunx (#1731)
Setup run through bunx or bun x registered the bin inside bun's temporary
package cache. Detect bun's package-runner launch from npm_command=exec
with a bun user agent and register the pinned bunx launcher, wrapped in
cmd.exe on Windows like the npx launcher. Doctor, update maintenance and
uninstall recognize it as REA-owned, and setup's printed commands name
the runner that launched it.

Read the runner variables case-insensitively on Windows, where the CLI
environment snapshot upper-cases variable names and setup run through
npx also fell back to the cache path.

Co-authored-by: morluto <76467478+morluto@users.noreply.github.com>
2026-10-11 19:17:15 +08:00
40d9c7b836 test(native): skip the EACCES directory diagnostic as root (#1394)
Root bypasses the mode-0 file permission this test relies on, so
inventory succeeds and the expected EACCES failure never occurs. Match
the root guard used by the other permission-dependent boundary tests.


Claude-Session: https://claude.ai/code/session_01XPY2GmufMNGQKoEgRUx1pd

Co-authored-by: Claude <noreply@anthropic.com>
2026-10-11 19:12:37 +08:00
Jason CatandGitHub 951c2aae3b fix(native): raise default PE resource entry budget to 16384 (#1703)
inspect-pe-resources rejected stock Windows resource images because their
tables exceed the 4096-entry default. Windows 11 25H2 SystemResources
measures imageres.dll.mun at 6476 examined entries, shell32.dll.mun at 5903
and wmploc.DLL.mun at 5440, so the default budget now admits each of them
with headroom while the maximum stays 65536. The independent 8 MiB metadata
budget still bounds allocation, and exceeding any budget still fails without
partial results.

The stable filesystem boundary test now emits a 6476-entry table through a
larger fixture section and asserts that the default budgets report complete
coverage, covering stock-image scale without a real Windows file.
2026-10-11 19:01:33 +08:00
Yann RobertoandGitHub 3ad5782dfd fix(javascript): honor package exports before sibling files
Resolve package exports before sibling-file fallback, preserve explicit export blocking, and retain precise CommonJS package self-reference behavior and caller provenance.

Fixes #1761.

Co-authored-by: Yann Roberto <165009858+shoyann@users.noreply.github.com>
2026-10-11 18:57:58 +08:00
morluto 847a1a3f25 test: share native keyed archive integer fixture generation 2026-10-11 18:54:55 +08:00
12c416005f feat(javascript): recover code-split bundles from multiple inputs (#1686)
* feat(javascript): recover code-split bundles from multiple inputs

recover_javascript_sources accepted a single path, and
validateModuleSet bound every module's provenance to that one snapshot.
A code-split app (runtime + chunks, or an Electron main/preload/renderer
set) had to be recovered one file per call, and modules recovered alone
could not resolve imports across chunks: cross-chunk require(id)
references stayed as numeric ids with property reads.

Add an optional additional_paths input. The listed files are
concatenated, in order after path, into the single engine snapshot, so
one Wakaru run sees every module and cross-chunk imports resolve.
validateModuleSet's single-snapshot binding is unchanged. Module
provenance is attributed back to the input file its first extraction
range starts in, with byte ranges rebased to that file's own offsets;
a module whose ranges cross an input boundary keeps the concatenated
snapshot identity and offsets, recorded as an explicit limitation.
Single-path behavior is byte-for-byte unchanged (one segment,
offset 0).

Fixes #1605

* fix(javascript): preserve source identity in multi-input recovery

---------

Co-authored-by: JingHao-Leon <w15909130000@gmail.com>
Co-authored-by: morluto <76467478+morluto@users.noreply.github.com>
2026-10-11 18:39:49 +08:00
morluto 044c0a92d5 test: reuse application analysis across comparison directions
Reuse immutable source graphs for both export comparison orientations and obtain rest-argument seeds from the existing public CLI Evidence. Retain all direction-specific assertions while removing nine repeated application analyses.

Validation: 75 focused tests; covered locations and hit status unchanged across all 1137 source files in controlled before/after coverage runs. Comparison file execution fell from 19.2s to 15.3s locally. Static and generated-document checks passed.
2026-10-11 18:15:57 +08:00
Aravind RaoandGitHub 262efe6c54 test(native): skip the DMG EACCES diagnostic as root (#1584) 2026-10-11 18:05:24 +08:00
morluto 6bcfeb0e55 test(cli): cache subprocess compilation within each worker 2026-10-11 18:01:32 +08:00
YsopwnableandGitHub 817aef4068 fix(apple): validate build-version tool records within cmdsize (#1784) 2026-10-11 17:50:56 +08:00
morluto 42951bbbbe test: reuse identical JavaScript analysis fixtures 2026-10-11 17:43:28 +08:00
N0zoM1z0 6e38529333 test(docs): type README drift scenarios as fixed pairs 2026-10-11 17:24:46 +08:00
N0zoM1z0 c9136eef71 fix(docs): validate translated README website destinations
Use the website language inventory when comparing README links. Accept the matching localized page while retaining checks for wrong locales, changed destinations, fragments and query parameters.
2026-10-11 17:24:46 +08:00
N0zoM1z0 a2d3338e90 feat(website): add multilingual editions and translation workflow 2026-10-11 17:24:46 +08:00
morluto a4dd4e8e87 perf(cli): defer command-specific analysis implementations
Load application and Electron workflow implementations only when their commands execute. Keep command schemas and catalog registration eager, and extend the startup import boundary to cover the deferred services.
2026-10-11 17:23:41 +08:00
Immanuel TikhonovandGitHub 914ba833de fix(javascript): report JSON status for imported package manifests
Use the existing JSON-text status checker for explicit package.json imports while retaining separate package-metadata validity. Preserve malformed/unavailable status and partial facts after cancellation.

Co-authored-by: Immanuel Tikhonov <122638311+immanuwell@users.noreply.github.com>
2026-10-11 17:07:57 +08:00
YsopwnableandGitHub 4daea941bf fix(cli): retry retained browser scenario cleanup
Retry retained browser scenario cleanup before the one-shot CLI exits. Preserve primary failures, partial observations and completed Evidence when shutdown fails.

Closes #1771.
2026-10-11 16:51:23 +08:00
morluto 40b37aeee1 ci: balance test shards and reuse package download caches 2026-10-11 16:40:17 +08:00
e8528cedcd fix(hopper): let failed one-shot commands exit while retaining a lease (#1763)
* fix(hopper): let failed one-shot commands exit while retaining a lease

A Hopper target lease retained after unconfirmed cleanup kept its
listening socket referenced, so a one-shot CLI emitted its final
cleanup_incomplete result but never exited, and its live lease rejected
every retry for the same target as already open.

Unreference the lease listener. A live MCP process still holds the
lease for as long as it runs; a completed CLI exits with status 1, and
the stale socket it leaves is recovered by the next acquirer.

Fixes #1751

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* test(hopper): remove only this test's stale lease sockets

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-11 01:38:34 -07:00
YsopwnableandGitHub 713db4dc26 fix(process): align capture Evidence provider admission
Use the canonical current and supported legacy process-provider rule for authenticated capture Evidence at both CLI and MCP admission boundaries. Preserve successful Evidence and distinguish provider input errors from digest authentication failures.

Closes #1757.
2026-10-11 16:14:42 +08:00
58e91e14bd fix(setup): preserve unrelated Hermes YAML line formatting
Preserve unrelated long YAML scalars when editing an existing Hermes configuration.

Co-authored-by: Zayd <iam@zayd.wtf>
2026-10-11 16:07:44 +08:00
YsopwnableandGitHub 857a1b7b04 fix(process): aggregate incomplete trace-spec diagnostics
Summarize incomplete trace-specification pair declarations with exact counts and the first undeclared pair. Preserve complete pair coverage, successful Evidence and independent validation errors; remove temporary pair-scan copies and sorting.

Closes #1753.
2026-10-11 16:04:46 +08:00
YsopwnableandGitHub e5699d23b1 fix(mcp): cancel evidence imports before session mutation
Propagate request cancellation through immutable JSON bundle reading and
check it before committing session state. Preserve failed-close ownership.

Fixes #1746

Assisted-by: Suika (gensokyo-skills:suika-scatter-gather-planning)
2026-10-11 15:59:21 +08:00
clydeandGitHub 062723e44e fix(browser): retain effective source-map response urls
Preserve source-map response identity and legacy omission, admit identities before body reading, and retain completed results and cleanup causes within the existing output budget. Price fixed envelopes once.

Closes #1729.
Closes #1738.

Co-authored-by: allisaid <curruptdwastaken@gmail.com>
2026-10-11 15:35:43 +08:00
YsopwnableandGitHub 21650455a9 fix(evidence): normalize unknown revisions and preserve mutation intent (#1733)
Normalize generated revision fields before hashing and retain complete typed
creation/update requests in a shared mutation Evidence producer. Keep imported
revision authentication and canonical persisted identity/digests unchanged.

Fixes #1723
Fixes #1724

Assisted-by: Suika (gensokyo-skills:suika-scatter-gather-planning)
2026-10-11 14:56:46 +08:00
YsopwnableandGitHub e36c82becd fix(artifacts): preserve emitted ZIP entry identity
Reject duplicate ZIP filenames before an emitted entry source lookup can be overwritten. Preserve valid repeated enumeration, case-distinct paths and typed inventory path failures.

Closes #1725. Related: #1480.
2026-10-11 14:25:03 +08:00
YsopwnableandGitHub 092de8f4b3 perf(snapshot): restore large legacy caches within live memory bounds
Stream legacy snapshot cache rows independently, authenticate complete Evidence once, and restore exact immutable payloads without assembling duplicate full graphs. Preserve the existing file format, provenance, resource admission, cancellation and cache export isolation.

Verified through real Linux SDK MCP import of the original 83 MB Ghidra snapshot and full 41 MB bundle hash, actual Windows CLI/MCP PE recovery and the installed Windows large-file decoder boundary, focused tests and static checks.

Fixes #1680. Refs #1065.
2026-10-11 13:59:40 +08:00
YsopwnableandGitHub 966dc3acde perf(evidence): reuse authenticated file snapshots during session import (#1721)
Seal file-backed MCP imports at the session boundary while preserving mutable
ordinary file reads. Validate internally constructed ledger candidates directly
with the existing immutable bundle constructor.

Assisted-by: Suika (gensokyo-skills:suika-scatter-gather-planning)
2026-10-11 13:53:01 +08:00
YsopwnableandGitHub d9b4ae182d fix(native): retain failed command cleanup owners
Retain failed native command owners through both xcrun resolution and command capture. Retry all owned cleanup failures before new commands and after consumers join their launched work at close; preserve partial output and every resource diagnostic.

Refs #740.
2026-10-11 13:39:42 +08:00
morluto 01762b10c6 perf(tests): consolidate TypeScript dialect producer checks 2026-10-11 13:29:50 +08:00
morluto 693ed332b5 perf(tests): remove timed output and synchronize browser fixtures 2026-10-11 13:20:16 +08:00
morluto a82d14e4db perf(cli): defer command-specific modules until dispatch 2026-10-11 13:20:15 +08:00
morluto 5f4c7ff6df perf(tests): reuse stateless workflows and immutable fixtures 2026-10-11 13:20:15 +08:00
YsopwnableandGitHub 13e03e9436 perf(evidence): fuse streamed JSON token handling (#1694)
Fuse upstream JSON parsing, scalar packing, heap admission, and value assembly into synchronous token stages while retaining asynchronous file reads and existing capacity, cancellation, decoding, identity, and typed-error semantics. Declare stream-chain 4.2.6 directly.

Validation: 30 focused tests; check:fast, Knip and formatting; actual Flameox paired full-content oracles; real Linux evidence replay and Windows source-built package/Ghidra snapshot recovery; final-head CI 31/31 success.

Refs #1065, #1680.
2026-10-11 12:36:35 +08:00
0ecf3f2438 perf(mcp): keep Playwright and type-only imports off MCP startup (#1691)
* fix(build): emit no runtime imports for type-only bindings

Under verbatimModuleSyntax, an import whose specifiers all carry inline
`type` modifiers compiles to `import {} from "..."`, a side-effect import
that still resolves and evaluates the module. 78 compiled modules carried
one, including `import {} from "playwright-core"` in the browser scenario
session, which loaded Playwright's module graph at MCP startup.

Use top-level `import type` for these bindings and enable oxlint's
typescript/no-import-type-side-effects rule so the pattern cannot return.

Refs #1685

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* perf(mcp): load Playwright when a scenario launches, not at startup

The browser scenario and Electron active providers are constructed before
the stdio server serves, and both bound Playwright's launchers at module
load. Resolve `chromium` and `_electron` from playwright-core when a
capture opens instead, as the web module resolver already does.

MCP startup now resolves 1,312 modules (4,902 resolutions) instead of
1,598 (6,076) before reading the first request, and no longer loads
playwright-core. Provider construction, the tool catalog, and capability
inventory are unchanged.

A boundary test spawns the stdio server, initializes, lists tools, and
asserts playwright-core and isomorphic-git were not resolved; the CLI
probe now exercises the real launch path to prove detection.

Closes #1685

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-10 21:04:05 -07:00
morluto a99857df23 test(release): keep credits attached to their contribution 2026-10-11 11:51:17 +08:00
morluto 6b043dc3de fix(release): avoid duplicated contributor thanks 2026-10-11 11:49:49 +08:00
ad3e41d46e fix(browser): resolve XLink and SVG prefixes in XHTML dom_urls (#1683)
* fix(browser): resolve XLink and SVG prefixes in XHTML dom_urls

DOMSnapshot has no namespace URI. HTML documents still use the parser's
local names. In XHTML and other XML documents, resolve in-scope xmlns
declarations so an element is SVG only in the SVG namespace, and report
any XLink href as href when plain href is absent.

Refs #1586

* fix(browser): bound XML namespace resolution and reject cyclic ancestry

---------

Co-authored-by: Cursor Agent <cursoragent@cursor.com>
Co-authored-by: morluto <76467478+morluto@users.noreply.github.com>
2026-10-11 11:26:26 +08:00
YsopwnableandGitHub 17f704050f feat(evidence): discover retained records without full payloads (#1682)
* feat(evidence): discover retained records without full payloads

* fix(evidence): guide oversized results to retained discovery

* style(tests): format retained discovery recovery assertion
2026-10-11 11:18:26 +08:00
c8820e03db fix(apple): decode binary app Info.plist files without plutil (#1476)
* fix(apple): decode binary app Info.plist files without plutil

* test(apple): use Windows-legal names in the binary plist bundle test

* fix(apple): ignore __proto__ entries in binary app Info.plist files

---------

Co-authored-by: morluto <76467478+morluto@users.noreply.github.com>
2026-10-11 11:14:15 +08:00
morluto b3e977e6d0 fix(browser): bound screenshot comparison memory 2026-10-11 11:11:38 +08:00
morlutoandYou Wish a99b2fd3b5 test: retain focused CommonJS export regressions
Keep uncovered assignment and binding regressions in the shared CLI and MCP runtime oracle, with a POSIX escaped-path fixture.

Co-authored-by: You Wish <29498139+Y0UW1SH@users.noreply.github.com>
2026-10-11 11:07:15 +08:00
Rudy Celekliandmorluto 4e4b15b825 fix(browser): validate PNG chunks and preserve unsupported encodings
Validate chunk checksums, type bytes, ordering and end boundaries before pixel comparison. Distinguish malformed PNGs and corrupt DEFLATE streams from recognized unsupported encodings through CLI and MCP error projection.

Adapted from #1375 with additional decoder and public boundary regressions. Eighteen decoder tests and six compiled CLI/MCP cases passed.

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

Co-authored-by: morluto <76467478+morluto@users.noreply.github.com>
2026-10-11 11:07:15 +08:00
morluto 21dbcf7a14 perf(cli): defer binary provider composition 2026-10-11 11:07:15 +08:00
a67dbf6a48 fix: reject directory paths for the Wakaru limiter and mitmdump
`REA_JAVASCRIPT_PRLIMIT_COMMAND` and `REA_MITMDUMP_COMMAND` were admitted
after an access check alone, which a searchable directory passes. Admission
then succeeded, and the failure surfaced later as a launch-time adapter
error instead of the capability refusal that names the configured path.

Require a regular file, as #751 did for the firmware engine and limiter.
Linux-only tests mirror that change; both checks only run on Linux hosts.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-11 10:43:48 +08:00
morlutoandClaude Opus 5.5 bd3a9bbbaf fix(native): classify PE resource input failures
Report missing and non-file selections as input issues, malformed PE images as format errors, and unstable reads as artifact changes. Keep existing limit, permission, cancellation, and other I/O classifications.

Co-authored-by: Kaoru0822 <208281322+kaoru0822-kitauji@users.noreply.github.com>

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-11 10:43:48 +08:00
b212e92f9b ci: select semantic tests and fixture lanes for their guarded inputs (#1675)
Edits to `release.yml` or `real-ghidra-windows.yml` selected only the
workflow lane, so the shards holding releaseWorkflow, releaseCheckpoint
and realGhidraWindowsWorkflow tests were skipped. Actionlint checks
syntax only, so a PR could drop the package-before-publish gate, the
registry canary or the default-branch-only self-hosted runner restriction
without CI noticing.

Conformance fixture sources under tests/ set only the code lane, so the
conformance-fixtures and reconstruction-readiness jobs, the only callers
of `verify:fixtures` and `verify:readiness`, never ran for them. Route
the directories that build-conformance-fixtures.mjs compiles to both
lanes, and readiness applications to the readiness lane.

Both were raised by the Codex review of #1660 after it merged.

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-10 19:41:40 -07:00
morluto 76d8f68cb9 feat(ghidra): inspect NativeAOT PE metadata read-only 2026-10-11 10:31:49 +08:00
morluto 83e52559b8 style(reference-source): format loader binding changes 2026-10-11 10:20:04 +08:00