Skip to content

Version 4.5.0rc2 - #4083

Merged
T4rk1n merged 3 commits into
devfrom
release/4.5.0rc2
Oct 23, 2026
Merged

T4rk1n merged 3 commits into
devfrom
release/4.5.0rc2

Conversation

@T4rk1n

@T4rk1n T4rk1n commented Oct 23, 2026

Copy link
Copy Markdown
Contributor

Third release candidate for 4.5.0.

  • dash/version.py -> 4.5.0rc2
  • npm sub-packages bumped to the next patch, since rc1 already published their x.5.1 versions to npm: dash-core-components 4.5.2, dash-html-components 4.5.2, dash-table 7.5.2, dash-renderer 3.5.2 (package.json + lockfile version fields + _dash_renderer.py __version__/external_url). Renderer rebuilt.
  • CHANGELOG: added the missing entry for fix(renderer): clear paths for null children #4052.

Since rc1 (see CHANGELOG [Unreleased] for the full list)

Changed

Removed

Fixed

Plus dependency and CI updates (#4072, #4079).

Before final 4.5.0

  • Rename CHANGELOG [Unreleased] -> [4.5.0] - <date>
  • Drop the rc2 from dash/version.py

rc1 published dcc/html 4.5.1, dash-table 7.5.1 and dash-renderer 3.5.1 to
npm, so the packages move to the next patch for rc2.
@sonarqubecloud

Copy link
Copy Markdown

@github-actions

Copy link
Copy Markdown
Contributor

Dash performance benchmarks

⚠️ regressions to review

scenario metric p90 (ms) median growth baseline p90 note
⚠️ full_children_replace replace_ms 6239.9 2179.6 20.44x 4697.1 p90 6239.9ms > warn 6000ms
✅ callback_chain chain_ms 507.0 476.1 1.0x 499.1
✅ callback_chain graph_ms 2.4 2.4 1.0x 2.4
✅ callback_fanout fanout_ms 88.9 83.5 0.88x 92.5
✅ deep_nesting render_ms 72.2 65.1 1.28x 56.8
✅ initial_render_large render_ms 816.4 650.6 1.31x 694.4
✅ initial_render_small render_ms 107.5 102.1 1.08x 104.0
✅ patch_append_nested append_ms 180.8 121.9 2.25x 192.3
✅ patch_append_toplevel append_ms 156.1 115.9 2.28x 140.2
✅ patch_scalar_update_large update_ms 169.0 157.9 0.88x 202.9
✅ wildcard_all_resolve wildcard_ms 308.7 293.8 0.95x 313.6
✅ wildcard_all_resolve graph_ms 1.5 1.5 1.0x 1.3

growth = late-third / early-third per-op time; ~1 is flat, a large value means the per-op cost scales with accumulated state.

machine scale vs baseline: 1.06x - divided out of the baseline ratios so they compare like for like (the absolute warn/fail ceilings are left un-scaled); calibrated on initial_render_small.

@T4rk1n
T4rk1n merged commit a553a10 into dev Oct 23, 2026
105 of 108 checks passed
@T4rk1n
T4rk1n deleted the release/4.5.0rc2 branch October 23, 2026 15:11
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.

1 participant