[lanyue-llk]

Watchdog termination can make a captured WeakMap call return no value after V8 resets the isolate's termination flag. Wrap construction, get and set in local TryCatch scopes, recognize has_terminated(), rethrow termination and propagate the absent cross-origin Window/Location surface. Ordinary intrinsic failures still panic. Tests cover interrupted construction and insertion, recovery, observer identity and an ordinary throwing intrinsic.

The HTML-to-Markdown converter uses an empty anchor's own nonblank aria-label when it has no descendant content. Visible inline and block descendants remain primary, including code blocks, headings, lists and tables; ancestor and sibling labels are never borrowed. PermissionStatus now inherits EventTarget and has per-instance listener storage, as required by the Permissions IDL. Tests cover listener removal, once, target identity, instance isolation and the illegal constructor.

This branch contains only these fixes on main. It does not import the content/navigation stack. Office document MIME classification is already fixed on main and was validated without duplicating that change.

Validation on ARM64 Linux with rust:1.96.1-bookworm:

• cargo fmt --all and final format check passed.

• cargo clippy --workspace --all-targets --all-features -- -D warnings passed before the final TryCatch refinement; cargo clippy -p moli-renderer-v8 --lib --all-features -- -D warnings passed again on the final renderer code.

• Initial selected nextest tests: 7 passed (3 cross-origin cache tests, PermissionStatus events, link-label regression, binary response headers and document MIME classification).

• The same link-label regression failed against unmodified main ([](/report) instead of [Open report](/report)). The first termination tests also exposed the insufficient isolate-flag check and passed after the TryCatch correction.

• Additional link regressions: 3 passed, with the block-descendant regression first failing on the prior implementation. Package clippy passed for all moli-html2md targets.

Full workspace nextest did not reach test execution: building the renderer test binary was killed by the kernel at about 7.39 GiB resident anonymous memory plus 1.75 GiB swap in the original 8 GiB ARM environment. Validation was then explicitly narrowed to the affected functional tests. No CI result is claimed.

[github-actions]

Spider Bench A/B

✅ All benchmark browser service runs completed; results are informational.

Public HEAD: 184 / 240 rows; 38 / 48 sites produced rows; 10 unexpected empty sites.

✅ Deterministic fixture contract is clean.

Common ancestor 7a36b6a63f7b → HEAD db61fe82348c; benchmark order: head-first; workflow: success.

Public 48-site run · informational

Metric	Base	HEAD	Δ
Run status	success	success	—
Extracted rows / contract	184 / 240	184 / 240	0 (0.00%)
Contract fill	76.67%	76.67%	0.00 pp (0.00%)
Sites with rows	38 / 48	38 / 48	0 (0.00%)
Sites meeting row contract	36 / 48	36 / 48	0 (0.00%)
Unexpected empty sites	10	10	0 (0.00%)
Partial-row sites	2	2	0 (0.00%)
Site outcomes	extracted: 37; HTTP error + rows: 1; empty: 7; HTTP error + empty: 1; snapshot error + empty: 2	extracted: 36; HTTP error + rows: 2; empty: 7; HTTP error + empty: 1; snapshot error + empty: 2	—
Observed duration	45.59 s	55.08 s	+9.50 s (+20.83%)
Estimated CPU	47.12 s	48.90 s	+1.78 s (+3.77%)
Peak PSS	273.46 MiB	289.69 MiB	+16.23 MiB (+5.94%)
Peak RSS	276.63 MiB	292.89 MiB	+16.26 MiB (+5.88%)
Peak threads	27	27	0 (0.00%)
Mismatched sites	0	0	0
Timeout with items	0	0	0

Category rows	Base	HEAD	Δ
news	35 / 40	35 / 40	0
stocks	20 / 40	20 / 40	0
tech	35 / 40	35 / 40	0
sports	27 / 40	27 / 40	0
games	37 / 40	37 / 40	0
life	30 / 40	30 / 40	0

Public-site content, timing, and memory are noisy. This single A/B run reports evidence only; site outcome counts explain missing rows without treating them as a deterministic regression.

CPU and memory timelines

These are bounded 40-point views of the same complete process-tree samples used by the HTML report. CPU uses 100% per occupied logical core; memory lines are RSS first and PSS second when complete PSS samples are available.

Base

xychart-beta
    title "Base CPU"
    x-axis "Elapsed seconds" [1.539, 2.541, 4.044, 5.045, 6.546, 7.547, 9.05, 10.052, 11.554, 12.555, 14.057, 15.058, 16.561, 17.563, 19.065, 20.066, 21.568, 23.07, 24.072, 25.073, 25.574, 26.576, 28.076, 29.077, 30.578, 31.58, 33.082, 34.082, 35.584, 36.586, 38.089, 39.089, 39.589, 40.59, 41.592, 42.093, 43.094, 44.096]
    y-axis "CPU percent" 0 --> 500
    line [65.81, 133.68, 147.64, 117.79, 143.84, 175.78, 179.69, 159.72, 133.85, 167.75, 59.9, 2, 115.86, 83.82, 67.91, 45.94, 75.9, 123.84, 151.75, 101.89, 115.71, 101.79, 132.04, 115.87, 109.88, 141.73, 113.82, 124.04, 73.91, 149.74, 107.76, 23.99, 0, 109.98, 93.77, 201.54, 65.96, 127.75]

Loading

xychart-beta
    title "Base memory: RSS then PSS"
    x-axis "Elapsed seconds" [0.037, 1.539, 2.541, 4.044, 5.045, 6.546, 7.547, 9.05, 10.052, 11.554, 12.555, 14.057, 15.058, 16.561, 17.563, 19.065, 20.066, 21.568, 23.07, 24.072, 25.073, 25.574, 26.576, 28.076, 29.077, 30.578, 31.58, 33.082, 34.082, 35.584, 36.586, 38.089, 39.089, 39.589, 40.59, 41.592, 42.093, 43.094, 44.096, 45.587]
    y-axis "Memory MiB" 0 --> 500
    line [12.18, 94.16, 123.48, 132.15, 156.83, 168.18, 159.98, 196.46, 192.07, 177.83, 179.79, 174.36, 151.94, 183.77, 202.8, 149.76, 156.2, 182.32, 165.58, 152.21, 276.63, 249.08, 220.16, 158.97, 154.53, 203.28, 176.29, 157.34, 151.59, 164.68, 202.05, 178.12, 163.8, 163.8, 185.47, 187.4, 193.96, 164.11, 190.8, 0]
    line [9.2, 90.8, 120.08, 128.89, 153.97, 164.82, 156.57, 192.6, 189.97, 173.78, 176.73, 171, 148.51, 180.48, 199.39, 146.33, 152.77, 178.95, 162.08, 148.64, 273.46, 244.98, 216.62, 155.33, 151.32, 199.36, 172.56, 153.63, 147.83, 161.04, 198.65, 174.42, 160.1, 160.1, 181.78, 184.18, 190.52, 160.4, 187.14, 0]

Loading

HEAD

xychart-beta
    title "HEAD CPU"
    x-axis "Elapsed seconds" [1.547, 3.05, 4.051, 4.552, 5.553, 6.054, 7.556, 9.057, 10.559, 12.06, 14.062, 15.565, 17.065, 18.567, 20.069, 21.57, 23.072, 24.074, 24.575, 26.076, 27.578, 29.087, 30.601, 32.103, 33.605, 35.111, 36.613, 38.116, 39.616, 41.619, 43.119, 44.621, 46.123, 47.626, 49.128, 50.63, 52.13, 53.632]
    y-axis "CPU percent" 0 --> 200
    line [65.82, 131.79, 141.93, 117.85, 155.92, 165.9, 155.83, 93.97, 158, 57.95, 63.96, 143.7, 39.97, 91.91, 57.91, 37.97, 2, 0, 5.99, 51.99, 41.93, 93.76, 136.43, 109.88, 101.91, 149.6, 167.77, 127.71, 40.02, 127.8, 28, 133.79, 95.87, 0, 3.99, 139.8, 133.97, 133.83]

Loading

xychart-beta
    title "HEAD memory: RSS then PSS"
    x-axis "Elapsed seconds" [0.045, 1.547, 3.05, 4.051, 4.552, 5.553, 6.054, 7.556, 9.057, 10.559, 12.06, 14.062, 15.565, 17.065, 18.567, 20.069, 21.57, 23.072, 24.074, 24.575, 26.076, 27.578, 29.087, 30.601, 32.103, 33.605, 35.111, 36.613, 38.116, 39.616, 41.619, 43.119, 44.621, 46.123, 47.626, 49.128, 50.63, 52.13, 53.632, 55.084]
    y-axis "Memory MiB" 0 --> 500
    line [12.18, 93.79, 130.94, 161.61, 127.44, 155.17, 164.85, 175.02, 163.02, 168.81, 168.5, 166.54, 166.43, 143.97, 169.56, 195.38, 142.42, 140.29, 140.29, 140.28, 149.34, 166.88, 148.48, 162.91, 292.89, 209.1, 166.55, 163.31, 169.08, 150.14, 160.47, 151.15, 211.66, 169.63, 172.04, 158.54, 183.06, 150.63, 183.82, 0]
    line [9.26, 90.61, 127.93, 158.37, 124.14, 151.85, 161.79, 171.75, 159.69, 165.56, 165.5, 163.16, 163, 140.62, 166.59, 192.13, 139.12, 137.04, 136.99, 137.01, 146.21, 163.64, 145.49, 159.73, 289.69, 205.82, 163.34, 160.14, 165.8, 146.72, 157.35, 147.76, 208.55, 166.24, 168.65, 155.15, 179.44, 147.21, 180.35, 0]

Loading

Deterministic fixture · required contract

The fixture exercises 15 routes: 8 are expected to emit rows and 7 intentionally exercise empty, loading, or timeout behavior.

Metric	Base	HEAD	Δ
Run status	success	success	—
Extracted rows / contract	40 / 40	40 / 40	0 (0.00%)
Contract fill	100.00%	100.00%	0.00 pp (0.00%)
Sites with rows	8 / 15	8 / 15	0 (0.00%)
Sites meeting row contract	15 / 15	15 / 15	0 (0.00%)
Unexpected empty sites	0	0	0
Partial-row sites	0	0	0
Site outcomes	extracted: 7; HTTP error + rows: 1; empty: 3; timeout + empty: 4	extracted: 7; HTTP error + rows: 1; empty: 3; timeout + empty: 4	—
Observed duration	19.87 s	19.86 s	-0.01 s (-0.05%)
Estimated CPU	0.44 s	0.44 s	0.00 s (0.00%)
Peak PSS	66.56 MiB	66.70 MiB	+0.14 MiB (+0.21%)
Peak RSS	69.65 MiB	69.70 MiB	+0.05 MiB (+0.08%)
Peak threads	18	18	0 (0.00%)
Mismatched sites	0	0	0
Timeout with items	0	0	0

Full HTML, JSON, CSV, logs, and page snapshots: workflow run and full spider-bench-results artifact.

The public-web A/B is informational. The exact HEAD fixture contract runs as a separate required CI check; timing and public-site content deltas remain non-blocking.

[github-actions]

CI Regression Report

Source CI run · source state at render: in_progress · artifacts: 6/6

Check	Status	Signal
Release regression	✅	HEAD/base failures 0/0; raw binary 0.00%
Frontend differential	❌	1,019/1,020 cases matched; 1 issues
Agent episodes	✅	8/8 Moli episodes passed; 0 failures
Runtime/CDP contracts	✅	26 contract cases; 0 failures
CDP smoke	✅	52/52 groups passed; 621 scenarios
WebMainBench	✅	545/545 completed; 0 unexpected failures

Release regression — ✅ HEAD/base failures 0/0; raw binary 0.00%

Package and image size

Metric	Base	HEAD	Delta	Delta %
Raw binary	150.47 MiB	150.47 MiB	0 B	0.00%
Stripped binary	103.94 MiB	103.94 MiB	0 B	0.00%
gzip binary	51.74 MiB	51.74 MiB	+833 B	+0.001535%
Rootfs	157.37 MiB	157.37 MiB	0 B	0.00%
gzip rootfs	60.68 MiB	60.68 MiB	−194 B	−0.000305%

Startup latency and PSS

Case	Base p50	HEAD p50	Delta	Base PSS	HEAD PSS	Delta
Serve ready	27.23 ms	27.75 ms	+0.52 ms	3.88 MiB	5.00 MiB	+1.12 MiB
CDP first page	50.08 ms	51.13 ms	+1.05 ms	4.28 MiB	4.19 MiB	−87.00 KiB
CDP warm page	11.76 ms	12.59 ms	+0.83 ms	49.22 MiB	49.54 MiB	+335.00 KiB
CLI about:blank	14.44 ms	14.49 ms	+0.05 ms	1.41 MiB	4.55 MiB	+3.14 MiB
CLI local JS	20.19 ms	19.38 ms	−0.81 ms	4.64 MiB	2.85 MiB	−1.78 MiB

Concurrency matrix

Concurrency	Base p50	HEAD p50	Delta	Base PSS	HEAD PSS	Delta	HEAD failures	Unstable B/H
1	19.14 ms	18.03 ms	−1.11 ms	694.00 KiB	696.00 KiB	+2.00 KiB	0	0/0
5	20.97 ms	20.33 ms	−0.64 ms	12.89 MiB	18.38 MiB	+5.49 MiB	0	0/0
10	21.35 ms	20.21 ms	−1.14 ms	27.50 MiB	22.93 MiB	−4.56 MiB	0	0/0
25	21.81 ms	21.50 ms	−0.31 ms	15.71 MiB	13.06 MiB	−2.65 MiB	0	0/0
100	21.46 ms	21.96 ms	+0.49 ms	27.58 MiB	12.84 MiB	−14.73 MiB	0	0/0

Frontend differential — ❌ 1,019/1,020 cases matched; 1 issues

Match	DOM mismatch	Diagnostic mismatch	Moli error	Reference error	Infrastructure error	Mismatched frames	Duration
0	0	0	0	1	0	0	4m 17.90s

First failing cases

• react/service-worker-cache-boundaries/registration-ready-controller — reference_error · Chromium error fixture did not become ready: {'id': 'react/service-worker-cache-boundaries/registration-ready-controller', 'phase': 'timeout', 'observed': {'id': 'react/servic

Agent episodes · Moli vs Chromium — ✅ 8/8 Moli episodes passed; 0 failures

Fast contract	Episodes	Passed	Assertions	Ready p50	Episode p50
Moli	8	8	47/47	4.42 ms	26.44 ms
Chromium	8	8	47/47	278.86 ms	90.63 ms

Operation p50	Chromium	Moli	Delta	Delta %
navigate	44.62 ms	5.91 ms	−38.71 ms	−86.76%
observe	0.76 ms	1.07 ms	+0.31 ms	+40.83%
fill	3.91 ms	3.89 ms	−0.02 ms	−0.55%
click	4.80 ms	4.39 ms	−0.42 ms	−8.65%

Idle-resource episode	Peak RSS	Peak PSS	Average CPU	Peak CPU	Peak processes	Sampler
Moli	65.79 MiB	64.14 MiB	0.84%	10.00%	1	✅
Chromium	1325.42 MiB	433.56 MiB	2.53%	98.41%	13	⚠️ PSS partial

Runtime and CDP session contracts — ✅ 26 contract cases; 0 failures

Suite	Cases	Failures	Median case p50	Median case p95	Median peak PSS p50
Synthetic fetch	13	0	18.89 ms	19.64 ms	693.00 KiB
Long-lived CDP session	13	0	18.02 ms	18.55 ms	—

CDP smoke — ✅ 52/52 groups passed; 621 scenarios

Workers: 4 · cumulative group time: 139.26 s · failed groups: 0

Slowest group	Scenarios	Duration	Status
scrollbar-visibility	22	14.35 s	✅
target-lifecycle	24	11.40 s	✅
chromium-cdp	56	9.24 s	✅
dom-input	27	9.16 s	✅
multi-page	45	8.28 s	✅

WebMainBench · 545 pages — ✅ 545/545 completed; 0 unexpected failures

Completed	Successful	Expected DNS failure	Unexpected failures	Panics	Crashes	Timeouts	Empty output
545/545	544	1	0	0	0	0	0

Frozen HTML · Linux · --wait done · 45 seconds per page · no retries. Content quality is evaluated separately.

Diagnostics: webmainbench-results — every page's Markdown, stderr, and exit status. Source run and artifacts

All artifact fields are parsed by the trusted default-branch renderer; missing or invalid inputs remain visible as unavailable.

[github-actions]

Sequential Navigation Soak A/B

✅ HEAD completed the 200-navigation resilience and memory observation.

One browser process, one target, and one CDP session navigate CSDN → SegmentFault → Huaban → example.com repeatedly. This run issued 200 Page.navigate commands per binary.

Common ancestor 7a36b6a63f → HEAD db61fe8234; benchmark order: head-first; workflow: success.

Session resilience

Metric	Base	HEAD
Attempted / planned	200 / 200	200 / 200
Direct observable pass	200 / 200	200 / 200
Failures / recovered	0 / 0	0 / 0
Recovery failures	0	0
Lifecycle order violations	0	0
Network order violations	0	0
Wall time	480.1 s	509.5 s

A public-page failure is reported separately from an unrecoverable session. The soak requires all navigation attempts, zero failed recovery, zero lifecycle/network ordering violations, and complete resource evidence.

Process-tree memory

Metric	Base	HEAD	Δ
Boundary samples	200	200	0
Boundary RSS samples	200	200	0
Boundary PSS samples	200	200	0
Periodic peak RSS	263.0 MiB	264.0 MiB	+1.0 MiB
Periodic peak PSS	260.3 MiB	260.3 MiB	+0.0 MiB
RSS · first 10 avg	167.8 MiB	177.2 MiB	+9.4 MiB
RSS · last 10 avg	185.5 MiB	153.5 MiB	-32.1 MiB
RSS · last − first window	+17.7 MiB	-23.8 MiB	-41.5 MiB
RSS · warm slope / 100 nav	+7.3 MiB	-19.4 MiB	-26.7 MiB
PSS · first 10 avg	164.5 MiB	173.8 MiB	+9.3 MiB
PSS · last 10 avg	182.7 MiB	150.5 MiB	-32.2 MiB
PSS · last − first window	+18.2 MiB	-23.3 MiB	-41.5 MiB
PSS · warm slope / 100 nav	+7.5 MiB	-19.2 MiB	-26.7 MiB
Peak file descriptors	60	64	4
Peak threads	29	31	2

HEAD memory by 50-navigation quarter

HEAD range	Avg RSS	Final RSS	Peak RSS	Avg PSS	Final PSS	Peak FD
1–50	175.5 MiB	165.3 MiB	236.6 MiB	172.2 MiB	161.6 MiB	52
51–100	174.8 MiB	121.0 MiB	233.7 MiB	171.5 MiB	117.5 MiB	54
101–150	163.9 MiB	132.1 MiB	222.3 MiB	160.4 MiB	129.0 MiB	54
151–200	150.4 MiB	146.8 MiB	194.8 MiB	147.5 MiB	143.8 MiB	53

Raw reports and per-navigation boundary samples: workflow run and full sequential-navigation-soak-results artifact.

Public-site timing and memory are observational. Use the A/B deltas and trend shape as evidence, not as a deterministic performance threshold.

[lanyue-llk]

The Web Fetch portions of this PR are superseded by #1501. This PR remains open because it also contains an independent PermissionStatus EventTarget fix that is outside #1501's scope; closing it now would discard that unrelated fix.

[lanyue-llk]

Superseded cleanly by two scoped PRs:

• Web Fetch correctness fixes: split review index #1501 contains the Web Fetch fixes for V8 termination, accessible link labels, frame Markdown, frame stripping, and renderer lifecycle delivery.
• fix(permissions): expose PermissionStatus as EventTarget #1503 contains the independent PermissionStatus: EventTarget fix and its focused behavior test.

No unique code remains in this mixed PR.