nestedContext builds one EvaluationContext per recursive expression step. A conditional object spread forces V8 through the generic property-copy path even when the spread adds no fields, and prevents a stable shape. It was the top self-time frame (25.4%) in a profile of semantic indexing.
Use two explicit object literals selected by capturePoint, keeping the same field set, order, values and optionality. Completed analyses are byte-identical across 12 representative sources (36 IR/facts/coverage digests), and member-mutation indexing is 1.47-1.65x faster on the profiled alias-chain workload.
Co-authored-by: Zhenyang Han <mshzy@users.noreply.github.com>
Co-authored-by: morluto <76467478+morluto@users.noreply.github.com>
Use the same APT word reader for positional fields and option assignments. Carry authored and effective repository facts through capture and replay.
Fixes#12694.
* fix(desktop): verify single-stanza APT source identity
* fix(desktop): verify single-stanza APT source identity
* fix(desktop): match APT suites using its configured architecture
* fix(desktop): match APT URI architecture templates
* fix(desktop): normalize empty APT URI ports
* test(desktop): regress empty-port APT repository identity
Assert entryForTarget and repository facts keep the authored empty-port URI while matching the portless Repo-URI APT 2.8.3 emits, including the signing key.
Co-authored-by: morluto <morluto@users.noreply.github.com>
* fix(desktop): parse APT port suffixes with atoi
Match deb822 Repo-URI identity to APT's unsigned port: the host's last colon suffix is atoi'd, zero is omitted, and IPv6 brackets stay intact while the authored URI is kept.
Co-authored-by: morluto <morluto@users.noreply.github.com>
* fix(desktop): accept GNU architecture names and bound port scans
Keep x86_64 as an APT architecture token for $(ARCH) identity, and stop growing a port magnitude once it passes the signed 64-bit limit.
Co-authored-by: morluto <morluto@users.noreply.github.com>
* fix(desktop): keep file:// as APT's empty-path Repo-URI
An empty host with no path stays scheme://, matching APT 2.8.3. A path after that host still drops it, so file:///srv remains file:/srv.
Co-authored-by: morluto <morluto@users.noreply.github.com>
* fix(desktop): normalize APT scheme:host URIs in one host pass
APT 2.8.3 rewrites http:127.0.0.1:18743/archive to an http:// Repo-URI. Identity comparison follows that form and keeps the authored URI. Bracket removal copies each host character once.
Co-authored-by: morluto <morluto@users.noreply.github.com>
* fix(desktop): replay the APT suite selected during capture
---------
Co-authored-by: morluto <76467478+morluto@users.noreply.github.com>
Co-authored-by: ShuaixinHuang <ShuaixinHuang@users.noreply.github.com>
Co-authored-by: Cursor Agent <cursoragent@cursor.com>
Co-authored-by: morluto <morluto@users.noreply.github.com>
* fix(native): record inspected plist bytes and refuse files outside the bundle
An open app session keeps the executable as the Evidence subject for
inspect_plist, same as the other bundle-resource tools, but the result had
no digest of the plist bytes and any absolute path was sealed under that
subject. Record source_sha256 for the bytes read, and refuse a selection
that resolves outside the opened bundle.
Fixes#9479
* fix(native)!: scope plist selection to the opened bundle
BREAKING CHANGE: To inspect a plist outside an opened app bundle or resource root, open the plist as a separate target.
* fix(native): report plist snapshot filesystem effects
---------
Co-authored-by: Cursor Agent <cursoragent@cursor.com>
Co-authored-by: morluto <76467478+morluto@users.noreply.github.com>
* docs(agents): state limit, test, verifier and feature-record conventions
Record ceiling, runtime-setting, assertion, verifier, and feature-record
rules for agents, and make the pull request title pattern explicit.
* docs(agents): defer limit controls to their owning guide
---------
Co-authored-by: Cursor Agent <cursoragent@cursor.com>
REQUIRED_RUNTIME_LIMITS still named three limits whose PRs never merged.
The checker skipped a required name absent from source, so the stale
entries passed. Remove them and fail that case with a typed result that
names the limit.
Co-authored-by: Cursor Agent <cursoragent@cursor.com>
* test(apktool): give Windows a drive-rooted authored launcher
The apktool unit and CLI boundary tests configure REA_APKTOOL_COMMAND
as /authored/apktool, /opt/apktool/apktool or /nonexistent/apktool.
On win32 hosts REA admits a configured launcher only when it is drive-
or UNC-rooted, so 17 cases there failed with the absolute-path refusal
before reaching the behavior they test. On Windows they now use the
same paths under C:\; other hosts keep the original strings.
* test(adb): give Windows a drive-rooted missing adb in the CLI refusals
The two "/nonexistent/adb names the host requirement" cases hit the
same win32 absolute-path refusal as the apktool launcher cases before
reaching the not-found check. On Windows they now use C:\nonexistent\adb.
Name the verifier command owner, run helpers, and cleanup order so new
verifier scripts do not start compilers, readers, or the REA CLI with bare
execFile.
Co-authored-by: Cursor Agent <cursoragent@cursor.com>
Keep the audited-base citation of src/gpu/Lz4Block.ts and record the
current src/compression/Lz4Block.ts location. Point the obligation-ledger
verifier example at an existing MCP cancellation test.
Co-authored-by: Cursor Agent <cursoragent@cursor.com>
List the six feature records missing from the spec index, drop the open
ReFS owner-review task now that admission is on main, and state when a
record is required.
Co-authored-by: Cursor Agent <cursoragent@cursor.com>
Co-authored-by: morluto <morluto@users.noreply.github.com>
* docs(limits): refresh the limit inventory and state the runtime-setting rule
Recompute the inventory from ANALYSIS_SETTINGS and the numeric-limit
check, and make a hardcoded limit a runtime setting unless a fixed
ceiling has a stated format, protocol, or representation reason.
* docs(limits): distinguish runtime settings from per-call overrides
---------
Co-authored-by: Cursor Agent <cursoragent@cursor.com>
* fix(v8): read the code cache header size from the container
V8 sizes the header with `kHeaderSize = POINTER_SIZE_ALIGN(kUnalignedHeaderSize)` and
`kUnalignedHeaderSize` is 28, so the header is padded *after* its words: a 64-bit producer writes
32 header bytes and a 32-bit one writes 28. The reader pinned 32 and required the declared payload
to equal `bytes.length - 32`, so every container from a 32-bit producer failed as "truncated or
corrupt".
Measured here on Node 24.22.1 (V8 in x64): take a real cache and drop the four alignment bytes at
offset 28. V8 loads it unchanged — `cachedDataRejected === false` — while the reader refused it:
before: Header declares a 392-byte payload but 388 bytes follow the 32-byte header
after: ACCEPT header_bytes=28 words=7 payload.offset=28
The payload-length word sits at byte 20 in both layouts, so the declared length already states how
much trailing padding the container carries. The header size is derived from it now: `28` and `32`
are admitted, the reported word table covers exactly the header the container holds (7 or 8 words),
and the schema requires that table to agree with `header_bytes`.
Nothing is loosened. A header that is neither 28 nor 32 bytes is still `malformed`, and the
16 existing cases pass unchanged — dropping only the derivation back to the constant fails exactly
the new narrow-header case.
`src/v8` — 26 passing. `npm run lint` and `tsc --noEmit` clean.
* fix(v8): refuse a wide cache that is only four bytes short
A review finding on the previous commit, and a real one: deriving the header size from the declared
payload length makes a 32-byte-header container truncated by four bytes resolve to 28, because
`bytes.length - declaredPayload` loses exactly those four bytes. The reader then accepted a
truncated cache and read its payload one word long.
The wide header's last word is `POINTER_SIZE_ALIGN` padding that V8 leaves zero, and a narrow
container has payload where that word sits. Measured on Node 24.22.1 across six different sources:
the padding word is always `0x00000000` and the first payload word is always non-zero. So the zero
word is the only evidence that separates the two, and it is now required to be non-zero for the
28-byte layout:
| container | before this commit | after |
| --- | --- | --- |
| real 32-byte header | accept | accept, `header_bytes=32` |
| real 28-byte header | accept | accept, `header_bytes=28` |
| 32-byte header truncated by 4 | **accept** | refused as `malformed` |
The guide also described one pinned 32-byte header; it now describes both layouts, how the size is
resolved, and why an ambiguous container is refused.
`src/v8` — 27 passing. Removing the guard accepts the truncated container again.
* fix(v8): read the padding word through the header accessor
A Static checks finding: `verify/architecture-guards.mjs` counts raw Buffer reads in parser folders
and allows only a shrinking baseline, and the previous commit added one:
```
raw Buffer or DataView numeric read (use BinaryReader ...); new copies are not allowed:
src/v8/V8CodeCacheParser.ts: 2 (baseline 1)
```
The read is the narrow-header discriminator — the alignment word a 64-bit producer leaves zero at
offset 28. It now goes through the file's existing `field(index)` accessor instead of calling
`bytes.readUInt32LE` directly, so the count returns to the baseline while the check itself is
unchanged: `field(7)` is byte 28, and `V8_CODE_CACHE_NARROW_HEADER_BYTES / 4` spells that index.
`node scripts/verify/architecture-guards.mjs` — passed across 2 843 source files.
`src/v8` + `tests/boundary/rust` — 68 passing. `tsc --noEmit` 0 errors.
Setting the condition to `false` fails "refuses a wide container that is only four bytes short",
so the check is still load-bearing.
* fix(v8): ground the narrow-header rule in the producer, and test the word table
Two review findings.
**The rule was stated as measurement, not as format.** The previous comment justified the zero-word
discriminator with "measured on Node 24.22.1 every real payload begins with a non-zero word", which
makes support for other producers depend on one host. The producer zeroes that padding rather than
copying payload over it: `SerializedData::AllocateWithPayload` in `src/snapshot/snapshot-data.cc`
runs `std::fill_n(data_, header_size, 0)` before writing any header word, and
`kHeaderSize = POINTER_SIZE_ALIGN(kUnalignedHeaderSize)` in `src/snapshot/code-serializer.h` puts the
four padding bytes inside `header_size`. So a zero word at byte 28 is a fact about the producer, not
about one Node build.
The refusal is now worded as the reservation it is: a narrow container whose payload happened to
begin with four zero bytes would match the same test, and no reading of the header alone separates
the two cases. The message says both readings instead of asserting corruption.
**The word-table rule had no test that could fail.** `header.fields.length * 4 === header.header_bytes`
cannot disagree on a parsed container — the parser derives both from `payloadOffset` — so it is only
reachable through the schema, which is what a stored or projected report also goes through. The new
case builds four reports directly: seven words with 32 bytes and eight words with 28 are refused on
`header.fields`, while the two consistent pairs are accepted. Setting the rule to `false` fails it
with `7 words / 32 bytes: expected true to be false`.
`node scripts/verify/architecture-guards.mjs` — passed across 2 843 source files.
`src/v8` + `tests/boundary/rust` — 69 passing. `tsc --noEmit` 0 errors, `npm run lint` 0 errors.
* refactor(v8): share decoded code cache header words
* fix(v8): validate narrow header evidence without assuming runtime compatibility
---------
Co-authored-by: morluto <76467478+morluto@users.noreply.github.com>