[T4rk1n]

Release 4.5.0.

• dash/version.py to 4.5.0, CHANGELOG Unreleased section dated.
• npm packages bumped past the rc2 versions already published: dcc/html 4.5.3, dash-table 7.5.3, dash-renderer 3.5.3 (plus the unpkg pin).

[github-actions]

Dash performance benchmarks

✅ all within thresholds

	scenario	metric	p90 (ms)	median	growth	baseline p90	note
✅	callback_chain	chain_ms	417.7	402.6	0.95x	499.1
✅	callback_chain	graph_ms	1.9	1.9	1.0x	2.4
✅	callback_fanout	fanout_ms	81.4	76.6	0.88x	92.5
✅	deep_nesting	render_ms	54.0	50.0	1.22x	56.8
✅	full_children_replace	replace_ms	5452.2	1984.0	19.19x	4697.1
✅	initial_render_large	render_ms	565.8	556.1	0.93x	694.4
✅	initial_render_small	render_ms	105.4	97.7	1.2x	104.0
✅	patch_append_nested	append_ms	123.1	95.9	2.42x	192.3
✅	patch_append_toplevel	append_ms	118.3	79.8	2.46x	140.2
✅	patch_scalar_update_large	update_ms	140.2	133.3	1.02x	202.9
✅	wildcard_all_resolve	wildcard_ms	282.2	268.2	0.94x	313.6
✅	wildcard_all_resolve	graph_ms	1.0	1.0	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: 1.02x - 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.

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud