Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Corel #7011

Draft
wants to merge 121 commits into
base: next
Choose a base branch
from
Draft

Corel #7011

wants to merge 121 commits into from

Conversation

juice49
Copy link
Contributor

@juice49 juice49 commented Jun 24, 2024

Description

What to review

Testing

Notes for release

Copy link

vercel bot commented Jun 24, 2024

The latest updates on your projects. Learn more about Vercel for Git ↗︎

Name Status Preview Comments Updated (UTC)
page-building-studio ✅ Ready (Inspect) Visit Preview 💬 Add feedback Oct 7, 2024 9:32am
performance-studio ✅ Ready (Inspect) Visit Preview 💬 Add feedback Oct 7, 2024 9:32am
test-compiled-studio ✅ Ready (Inspect) Visit Preview 💬 Add feedback Oct 7, 2024 9:32am
test-next-studio ✅ Ready (Inspect) Visit Preview 💬 Add feedback Oct 7, 2024 9:32am
test-studio ✅ Ready (Inspect) Visit Preview 💬 Add feedback Oct 7, 2024 9:32am
1 Skipped Deployment
Name Status Preview Comments Updated (UTC)
studio-workshop ⬜️ Ignored (Inspect) Visit Preview Oct 7, 2024 9:32am

Copy link
Contributor

No changes to documentation

pedrobonamin and others added 23 commits October 7, 2024 10:17
…nts (#7404)

* refactor(sanity): add changes to allow for comments in version documents

* refactor(sanity): send version via prop to commentsListItemLayout

* docs(sanity): add comments for todo for readonly

* refactor(sanity): fix comments

* refactor(sanity): reverts changes to the potential ui for the future

* chore(sanity): remove "comments support" comment

* refactor(sanity): rename document ids and revert change in CommentsProvider for useEditState

* refactor(sanity): refactor solution for re-rendering panes when switching between versions
* fix: draftId correct set in timeline for version docs

* refactor: timeline uses documentId rather than publishedId for agnostic use with versions

* refactor(core): defining versionId rather than draftId

* refactor(core): refactor to better reflect use of version nomenclature in history

* fix(core): resolving tsdoc issue with destructure

* fix(core): resolving tsdoc issue with destructure
* fix(sanity): issue where the createVersionSuccess was never resolving

* feat(sanity): add telemetry when adding version

* refactor(sanity): rename bundles to releases telemetry

* feat(sanity): add telemetry when creating release

* feat(sanity): add telemetry when updating release

* feat(sanity): add telemetry when deleting release

* feat(sanity): add telemetry when publishing release

* feat(sanity): add telemetry when archive and unarchive release

* chore(sanity): fix dependencies

* feat(sanity): add telemetry when creating a draft

* refactor(sanity): add telemetry for tracking drafts and include version

* refactor(sanity): update telemetry request for creating release

* refactor(sanity): reuse interface for OriginInfo

* docs(sanity): clarify unarchive / archive telemetry track
…ts (#7463)

* fix(core): update useReleaseHistory to fetch with the version documents

* fix(core): rename publishedIds to versionIds in useReleaseHistory
* fix(core): update intent link for releases tool

* chore(core): rename releases tool bundleId param to releaseId

* chore(core): update DocumentPerspectiveMenu test
* feat(sanity): update restore button

* chore(sanity): update translations

* feat(sanity): add left side information on footer (revision status line)

* refactor(sanity): set up button action to discard version

* feat(sanity): add badge to footer on versions

* refactor(sanity): remove unused translations (moved them to core)

* docs(sanity): add to do on discard documents?

* refactor(sanity): move bundleactions props to const

* refactor(sanity): fix issues with translation packages
…undle to release (#7469)

* refactor(sanity): update bundles metada to releases metadata

* refactor(sanity): rename bundlebadge to releasebadge

* refactor(sanity): rename bundlesMenu to ReleasesMenu

* refactor(sanity): rename BundleDetailsDialog to ReleaseDetailDialog

* refactor(sanity): rename bundleForm to ReleaseForm

* refactor(sanity): rename ReleaseDetailsDialog test

* refactor(sanity): rename BundleIconEditorPicker to ReleaseIconEditorPicker

* fix(sanity): fix issues with renaming within ReleaseDetailsDialog

* refactor(sanity): rename BundleActions to ReleaseActions

* refactor(sanity): fix translation

* refactor(sanity): remove workshop story

* refactor(sanity): rename bundlePerspective to perspective on getPReviewStateObservable

* refactor(sanity): rename BundleMenuButton to perspective on ReleaseMenuButton

* test(sanity): fix test for ReleaseMenuButton

* test(sanity): clean up (remove duplicates and clean up button)

* chore(sanity): remove unused bundle translations
* chore(releases): moving releases/components to releases/tool/components

* chore(releases): moving bundles/components to releases/components

* chore(releases): moving bundles/hooks to releases/hooks

* chore(releases): moving rest of bundles to releases

* chore(releases): fixing various test imports

* chore(releases): fixing various test imports

* chore(releases): fixing various test imports

* chore(releases): further experimentation on SanityDocumentPreview test failures

* chore(releases): fixing ReleaseSummary and Review by mocking components

* chore(releases): reverting irrelevent changes
Copy link
Contributor

github-actions bot commented Oct 7, 2024

⚡️ Editor Performance Report

Updated Mon, 07 Oct 2024 09:44:00 GMT

Benchmark reference
latency of sanity@latest
experiment
latency of this branch
Δ (%)
latency difference
article (title) 16.3 efps (62ms) 11.6 efps (86ms) +25ms (+39.8%) 🔴
article (body) 54.6 efps (18ms) 53.8 efps (19ms) +0ms (+1.6%)
article (string inside object) 17.2 efps (58ms) 12.9 efps (78ms) +20ms (+33.6%) 🔴
article (string inside array) 13.7 efps (73ms) 10.6 efps (94ms) +21ms (+28.8%) 🔴
recipe (name) 29.4 efps (34ms) 20.0 efps (50ms) +16ms (+47.1%) 🔴
recipe (description) 32.8 efps (31ms) 21.7 efps (46ms) +16ms (+50.8%) 🔴
recipe (instructions) 99.9+ efps (7ms) 99.9+ efps (7ms) +0ms (-/-%)
synthetic (title) 14.3 efps (70ms) 5.8 efps (174ms) +104ms (+147.9%) 🔴
synthetic (string inside object) 14.9 efps (67ms) 6.0 efps (167ms) +100ms (+149.3%) 🔴

efps — editor "frames per second". The number of updates assumed to be possible within a second.

Derived from input latency. efps = 1000 / input_latency

Detailed information

🏠 Reference result

The performance result of sanity@latest

Benchmark latency p75 p90 p99 blocking time test duration
article (title) 62ms 69ms 78ms 215ms 1490ms 15.5s
article (body) 18ms 21ms 30ms 131ms 277ms 6.0s
article (string inside object) 58ms 62ms 67ms 146ms 1220ms 9.1s
article (string inside array) 73ms 79ms 104ms 192ms 2151ms 10.4s
recipe (name) 34ms 36ms 40ms 72ms 89ms 10.4s
recipe (description) 31ms 34ms 37ms 120ms 20ms 6.6s
recipe (instructions) 7ms 9ms 10ms 10ms 0ms 3.3s
synthetic (title) 70ms 74ms 92ms 289ms 2355ms 18.6s
synthetic (string inside object) 67ms 73ms 79ms 332ms 2161ms 10.6s

🧪 Experiment result

The performance result of this branch

Benchmark latency p75 p90 p99 blocking time test duration
article (title) 86ms 92ms 108ms 229ms 2856ms 17.8s
article (body) 19ms 20ms 26ms 257ms 428ms 5.9s
article (string inside object) 78ms 83ms 95ms 281ms 2500ms 11.3s
article (string inside array) 94ms 101ms 116ms 296ms 3375ms 12.7s
recipe (name) 50ms 52ms 57ms 84ms 832ms 11.7s
recipe (description) 46ms 48ms 52ms 83ms 632ms 7.9s
recipe (instructions) 7ms 9ms 10ms 39ms 0ms 3.4s
synthetic (title) 174ms 186ms 293ms 511ms 9508ms 28.6s
synthetic (string inside object) 167ms 176ms 189ms 610ms 9524ms 20.7s

📚 Glossary

column definitions

  • benchmark — the name of the test, e.g. "article", followed by the label of the field being measured, e.g. "(title)".
  • latency — the time between when a key was pressed and when it was rendered. derived from a set of samples. the median (p50) is shown to show the most common latency.
  • p75 — the 75th percentile of the input latency in the test run. 75% of the sampled inputs in this benchmark were processed faster than this value. this provides insight into the upper range of typical performance.
  • p90 — the 90th percentile of the input latency in the test run. 90% of the sampled inputs were faster than this. this metric helps identify slower interactions that occurred less frequently during the benchmark.
  • p99 — the 99th percentile of the input latency in the test run. only 1% of sampled inputs were slower than this. this represents the worst-case scenarios encountered during the benchmark, useful for identifying potential performance outliers.
  • blocking time — the total time during which the main thread was blocked, preventing user input and UI updates. this metric helps identify performance bottlenecks that may cause the interface to feel unresponsive.
  • test duration — how long the test run took to complete.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

5 participants