Repository navigation
Conversation
Lms24
added this pull request to stack #24904
October 2, 2026 12:58
Lms24
changed the title
Lms24
force-pushed
the
lms/fix-replay-cls-timestamp
branch
from
October 2, 2026 14:13
53c97f7 to
10bc6d6
Compare
Lms24
marked this pull request as ready for review
October 2, 2026 14:15
Lms24
requested review from
logaretm and
msonnb
and removed request for
a team
October 2, 2026 14:15
Contributor
size-limit report 📦
|
JPeer264
approved these changes
Oct 5, 2026
| name: string, | ||
| nodes: Node[] | undefined, | ||
| attributions?: WebVitalData['attributions'], | ||
| time = metric.value, |
Member
There was a problem hiding this comment.
q: Can't we set this inside this function? It seem we set INP and CLS both the "last entry", but they are both setting it to metric.entries[metric.entries.length - 1]?.startTime ?? 0. Couldn't this be potentially:
const time = metric.entries[metric.entries.length - 1] ?? metric.valueIt is a different logic than the current implemented, but asking since LCP doesn't make use of the "last entry". If LCP shouldn't make use of it, maybe we should document when this should be overwritten
Member
There was a problem hiding this comment.
m: +1, LCP's last entry startTime is its render time, so it works there too. metric.value is wrong as a time for soft navs, prerender and bfcache.
For bfcache there are no entries, so we could fall back to the navigation start:
const lastEntry = metric.entries[metric.entries.length - 1];
const time = lastEntry?.startTime ?? (metric.navigationStartTime ?? 0) + metric.value;This also would put a CLS of 0 on its soft nav which is correct.
Member
Author
There was a problem hiding this comment.
Good points, thanks! I went with your suggestion @logaretm
logaretm
reviewed
Oct 5, 2026
| // The CLS value is a score, not a time, so we place the event at the last layout shift. A CLS of 0 has no layout | ||
| // shift, so it goes at the time origin. | ||
| const lastEntry = metric.entries[metric.entries.length - 1]; | ||
| return getWebVital(metric, 'cumulative-layout-shift', nodes, layoutShifts, lastEntry?.startTime ?? 0); |
Member
There was a problem hiding this comment.
l: a soft nav with a CLS of 0 lands at page load here.
metric.navigationStartTime ?? 0 would put it on its navigation, like _sendClsSpan does.
Member
Author
There was a problem hiding this comment.
Good catch, thx! Went with the suggestion.
| name: string, | ||
| nodes: Node[] | undefined, | ||
| attributions?: WebVitalData['attributions'], | ||
| time = metric.value, |
Member
There was a problem hiding this comment.
m: +1, LCP's last entry startTime is its render time, so it works there too. metric.value is wrong as a time for soft navs, prerender and bfcache.
For bfcache there are no entries, so we could fall back to the navigation start:
const lastEntry = metric.entries[metric.entries.length - 1];
const time = lastEntry?.startTime ?? (metric.navigationStartTime ?? 0) + metric.value;This also would put a CLS of 0 on its soft nav which is correct.
Lms24
requested review from
chargome and
s1gr1d
and removed request for
a team
October 6, 2026 09:58
Lms24
force-pushed
the
lms/fix-replay-cls-timestamp
branch
from
October 6, 2026 09:58
10bc6d6 to
e7aa72f
Compare
Lms24
force-pushed
the
lms/fix-replay-cls-timestamp
branch
from
October 6, 2026 10:01
e7aa72f to
1825836
Compare
![cursor[bot]](https://avatars.githubusercontent.com/in/1210556?s=60&v=4){kind=small}
There was a problem hiding this comment.
Cursor Bugbot has reviewed your changes and found 1 potential issue.
❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.
Reviewed by Cursor Bugbot for commit 1825836. Configure here.
Lms24
force-pushed
the
lms/fix-replay-cls-timestamp
branch
2 times, most recently
from
October 6, 2026 16:04
3c0856f to
ddb99d3
Compare
Lms24
force-pushed
the
lms/fix-replay-cls-timestamp
branch
from
October 6, 2026 17:24
ddb99d3 to
1ae96f8
Compare
Lms24
force-pushed
the
lms/fix-replay-cls-timestamp
branch
from
October 7, 2026 08:34
1ae96f8 to
47e4946
Compare
Lms24
force-pushed
the
lms/fix-replay-cls-timestamp
branch
from
October 7, 2026 08:48
47e4946 to
4a8b3bc
Compare
![sentry[bot]](https://avatars.githubusercontent.com/in/12637?s=60&v=4){kind=small}
Lms24
force-pushed
the
lms/fix-replay-cls-timestamp
branch
2 times, most recently
from
October 7, 2026 11:58
7b429e3 to
c94f742
Compare
The replay CLS event used the CLS score as its timestamp, so it always landed a fraction of a millisecond after page load. Use the start time of the last layout shift instead, like the CLS span in tracing does. This also picks the time origin from when the shift happened. Co-Authored-By: Claude <noreply@anthropic.com>
Like CLS, the replay INP event used the metric value as its timestamp. For INP that value is a duration, so the event landed shortly after page load. Use the start time of the interaction instead. Co-Authored-By: Claude <noreply@anthropic.com>
Soft navigation and bfcache metrics are relative to the start of their navigation, not the document time origin. A CLS of 0 or a bfcache LCP has no entry to place the event at, so it landed at page load. Fall back to the navigation start plus the value instead, like tracing does. All three vitals now share this logic in getWebVital. LCP also uses its last entry now, which is the render time, so a soft navigation LCP is no longer placed relative to page load. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Lms24
force-pushed
the
lms/fix-replay-cls-timestamp
branch
from
October 7, 2026 15:02
f867ef9 to
c43e986
Compare
This branch has not been deployed
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
3 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.
The Replay integration adding web vital breadcrumbs used the metric value as the timestamp for CLS and INP web vitals. This is incorrect because that value is a score for CLS and a duration for INP, so both events always landed just after page load. This was a pre-existing bug, but it showed up during work on #23054. With this PR, CLS goes at the last layout shift and INP at the interaction, like we already do with spans in tracing. Both also get the time origin from when that happened, so they stay correct after a clock drift correction. A CLS of 0 has no layout shift, so it stays at the start of its navigation