[harsh21234i]

Summary

• Preserve and restore root dirty state around the temporary router dirtying used by linked/shared state patching.
• Prevent regular linked-state events from sending an unchanged root/router delta.
• Add regression coverage for the issue described in Events against SharedState should not mark root state dirty all the time #6392.

Tests

• uv run --frozen pytest tests/units/test_state.py::test_linked_state_event_does_not_dirty_root_state -q
• uv run --frozen pytest tests/units/test_state.py::test_linked_state_event_does_not_dirty_root_state tests/units/test_state.py::test_router_var_dep tests/units/test_state.py::test_computed_var_depends_on_parent_non_cached tests/units/test_state.py::test_async_computed_var_get_state -q
• uv run --frozen pytest tests/units/test_state.py -q
• uv run --frozen pre-commit run --files reflex/istate/shared.py tests/units/test_state.py news/6841.bugfix.md

Closes #6392

[chatgpt-codex-connector]

Codex usage limits have been reached for code reviews. Please check with the admins of this repo to increase the limits by adding credits.
Credits must be used to enable repository wide code reviews.

[codspeed]

Merging this PR will improve performance by 9.96%

2 warnings

⚠️ 22 benchmarks are dominated by system calls

System calls cannot be simulated consistently and are left out of the measure. The reported value understates the real cost. Switch to the Walltime instrument to measure them.

Learn more about system calls

See affected benchmarks

⚠️ Base and head ran on different runner types

Some benchmarks with significant performance changes were compared across different runner types, so the changes might not come from the code. Run base and head on the same runner type, or use CodSpeed Macro Runners for more stable results.

Learn more about Macro Runners

Open the report in CodSpeed to investigate

⚡ 5 improved benchmarks
✅ 143 untouched benchmarks

Performance Changes

	Benchmark	BASE	HEAD	Efficiency
⚡	test_modify_state[linked]	12.6 ms	11.1 ms	+13.03%
⚡	test_process_event[linked_8_clients]	57 ms	51.2 ms	+11.29%
⚡	test_process_event[linked_1_client]	11.7 ms	10.5 ms	+10.98%
⚡	test_process_event[linked]	6.3 ms	5.7 ms	+9.92%
⚡	test_hydration_snapshot[200]	8.9 ms	8.4 ms	+4.76%

Tip

Curious why performance improved? Comment @codspeedbot explain why performance improved on this PR, or directly use the CodSpeed MCP with your agent.

---

_{Comparing harsh21234i:fix/shared-state-root-dirty-6392 (ae45ca2) with main (be0c685)}

[greptile-apps]

Confidence Score: 5/5

[Medium impact] The PR appears safe to merge; no new blocking issues were found.

Summary

The PR prevents temporary router dirtiness from leaking into linked-state event updates.

• Linked-state events leave unchanged router values out of their deltas.
• The Redis lock test waits until both requests reach the queue.

The changes since the previous review only adjust Redis test synchronization. No new actionable issues were found. Tests were not run during this review.

_{Reviews (28) · Last reviewed commit: "test: synchronize redis lock contention ..." · Reviewed by Greptile}

[cubic-dev-ai]

All reported issues were addressed across 3 files

<sub>Reply with feedback, questions, or to request a fix.

Fix all with cubic | Re-trigger cubic</sub>

[harsh21234i]

hey @masenf The state-dirty cleanup changes and regression tests are now implemented, including descendant cleanup when delta resolution fails. Most CI checks are passing, but reflex-install-and-init is failing. Could you confirm whether that failure is infrastructure-related or requires a code change? Once it passes, could you please review the latest commit?

[masenf]

patch breaks delta calculation for computed vars in shared state that depend on root state.

see https://github.com/reflex-dev/reflex/blob/claude/substate-dirty-tracking-demo-rple7z/shared_state_router_delta_repro.py

[cubic-dev-ai]

All reported issues were addressed across 2 files (changes from recent commits).

<sub>Tip: Review your code locally with the cubic CLI to iterate faster.

Re-trigger cubic</sub>

[cubic-dev-ai]

All reported issues were addressed across 2 files (changes from recent commits).

<sub>Tip: Review your code locally with the cubic CLI to iterate faster.

Re-trigger cubic</sub>

[cubic-dev-ai]

All reported issues were addressed across 2 files (changes from recent commits).

<sub>Tip: Review your code locally with the cubic CLI to iterate faster.

Re-trigger cubic</sub>

[cubic-dev-ai]

All reported issues were addressed across 2 files (changes from recent commits).

<sub>Tip: Review your code locally with the cubic CLI to iterate faster.

Re-trigger cubic</sub>

[harsh21234i]

hey @mansef could you please review the latest commit?

[greptile-apps]

Comments Outside Diff

These findings could not be posted inline.

•  Dirty markers survive patch exit reflex/istate/shared.py:283 ▶

  If a computed var on another state depends on the restored private state, this call marks that var and its ancestor path dirty. Patch exit occurs after the event delta has been resolved and cleaned, so the marker survives into a later event and can emit an unrelated computed-var delta.

[cubic-dev-ai]

All reported issues were addressed across 3 files

<sub>Reply with feedback, questions, or to request a fix.

Re-trigger cubic</sub>

[harsh21234i]

hey @masenf please take a look!!