perf(replay): reduce DOM accessor overhead and add benchmarks - #4801
Open
marandaneto wants to merge 1 commit into
Open
perf(replay): reduce DOM accessor overhead and add benchmarks#4801marandaneto wants to merge 1 commit into
marandaneto wants to merge 1 commit into
Conversation
Contributor
|
Contributor
Replay incident risk checkThis diff touches code involved in past incidents. This is a heads-up, not a verdict: read the matched sections of INCIDENTS.md and answer their review questions before merging. For a judgment on whether this diff has the same failure mode, run the |
marandaneto
marked this pull request as ready for review
September 5, 2026 16:27
marandaneto
requested review from
TueHaulund,
arnohillen,
fasyy612 and
ksvat
and removed request for
a team
September 5, 2026 16:27
![github-advanced-security[bot]](https://avatars.githubusercontent.com/in/57789?s=60&v=4){kind=small}
| if (!untaintedAccessor) return instance[accessor]; | ||
|
|
||
| untaintedAccessorCache[cacheKey] = untaintedAccessor; | ||
| cache[accessor as string] = untaintedAccessor; |
Contributor
Prompt To Fix All With AI### Issue 1
packages/browser/scripts/benchmark-replay.md:4-5
**Performance thresholds are missing**
The new benchmark explicitly provides no acceptance thresholds or statistical criteria, so it collects measurements but cannot automatically identify a performance regression. This violates the repository requirement that performance tests define clear thresholds, such as average duration and standard deviation. That requirement must be satisfied before merging.
---
For each issue above, determine whether it is valid and should be fixed. If so, fix it directly.Reviews (1): Last reviewed commit: "perf(replay): reduce DOM accessor overhe..." | Re-trigger Greptile |
![greptile-apps[bot]](https://avatars.githubusercontent.com/in/867647?s=60&v=4){kind=small}
Comment on lines
+4
to
+5
| mocked replay requests. Uses the installed Playwright Chromium (no external sites | ||
| or new dependencies). Timing numbers are evidence, not machine-independent CI |
Contributor
There was a problem hiding this comment.
The new benchmark explicitly provides no acceptance thresholds or statistical criteria, so it collects measurements but cannot automatically identify a performance regression. This violates the repository requirement that performance tests define clear thresholds, such as average duration and standard deviation. That requirement must be satisfied before merging.
Rule Used: Performance tests should have clear thresholds, su... (source)
Learned From
PostHog/posthog#32520
Prompt To Fix With AI
This is a comment left during a code review.
Path: packages/browser/scripts/benchmark-replay.md
Line: 4-5
Comment:
**Performance thresholds are missing**
The new benchmark explicitly provides no acceptance thresholds or statistical criteria, so it collects measurements but cannot automatically identify a performance regression. This violates the repository requirement that performance tests define clear thresholds, such as average duration and standard deviation. That requirement must be satisfied before merging.
**Rule Used:** Performance tests should have clear thresholds, su... ([source](https://app.greptile.com/posthog-org-19734/-/custom-context?memory=2fe31af0-f290-43bc-b879-4c1a73b3ac28))
**Learned From**
[PostHog/posthog#32520](https://github.com/PostHog/posthog/pull/32520)
---
For each issue above, determine whether it is valid and should be fixed. If so, fix it directly.Note: If this suggestion doesn't match your team's coding style, reply to this and let me know. I'll remember it for next time!
Contributor
Contributor
|
Size Change: +7.71 kB (+0.04%) Total Size: 20.5 MB 📦 View Changed
ℹ️ View Unchanged
|
Contributor
posthog-js Compliance ReportDate: 2026-09-05 16:35:06 UTC ✅ All Tests Passed!26/26 tests passed Capture Tests✅ 26/26 tests passed View Details
|
Member
Author
|
not a considerable improvement but its still worth since its a small change |
23 tasks
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Problem
Large session replay snapshots and structural mutations still block the main thread on node-dense pages. Profiling the built SDK found that rrweb repeatedly constructs composite string keys when looking up its cached native DOM getters.
This is a small first improvement for #4217, not a complete fix. DOM capture and parts of payload encoding remain synchronous.
Changes
pnpm benchmark:replaycommand. It exercises the built SDK through mocked replay transport with compression on and off, checks intermediate replay state and privacy sentinels, and reports long tasks, frame gaps, CPU, heap deltas, drops and extra full snapshots.Performance results
Three unprofiled repetitions per arm on a roughly 50k-node fixture in Playwright Chromium 136.0.7103.25, with no CPU throttling. Both builds use the same generation-aware benchmark. These are descriptive local samples, not statistically significant estimates or customer-level guarantees.
Median longest observed main-thread task:
Uncompressed rebuild performance was essentially unchanged. No oversized mutation drops were observed in this comparison, and replay-state and privacy assertions passed. The recorder artifact increased by 110 bytes before compression and 27 bytes with gzip.
The benchmark records artifact hashes and environment metadata. It deliberately separates profiled runs from timing comparisons. A zero longest-task result means no observed task reached the Long Tasks API's 50 ms threshold, not zero blocking.
Validation
e9058295e8b060a9c847bdb4f0872fa144e041be, with no actionable findings.No recorder configuration, event ordering, masking policy, sampling or session lifecycle behavior changes. No recording-volume change is expected. Encoding-tail and asynchronous snapshot work will be evaluated separately.
Release info Sub-libraries affected
Libraries affected
Checklist
If releasing new changes
posthog-jspatch changeset🤖 Agent context
Autonomy: Human-driven (agent-assisted)
Implemented with Pi using repository inspection, Chromium profiling, focused tests and an independent review. The work keeps synchronous recording semantics and avoids importing the time-sliced snapshot implementation from the earlier closed PRs.
The pre-commit code formatter attempted to reformat entire rrweb files, although the root formatting command excludes rrweb. Only that hook was skipped to retain the existing rrweb style. Other hooks and targeted formatting checks passed. No changes were made to hook configuration.