Optimize state updates with structural sharing on frontend

[FarhanAliRaza]

Type of change

• New feature (non-breaking change which adds functionality)
• This change requires a documentation update

Changes To Core Features

• Have you added an explanation of what your changes do and why you'd like us to include them?
• Have you written new tests for your core changes, as applicable?
• Have you successfully ran tests with your changes locally?

Description

This PR improves frontend performance by implementing structural sharing in the applyDelta function. When state updates are applied, unchanged objects and arrays now retain their previous identity, allowing memoized React components to skip re-rendering and enabling identity-based caches to continue hitting.

Key changes:

1. Added isPlainObject helper — Distinguishes plain objects from arrays, Dates, and class instances to determine which values should participate in structural sharing.

2. Added replaceEqualDeep function — Recursively compares previous and next values, preserving object/array identity when contents are unchanged. This is the core of the structural sharing optimization:

   • For arrays: shares unchanged elements and handles length changes efficiently
   • For plain objects: shares unchanged properties and handles added/removed keys
   • For other types: returns the new value as-is

3. Refactored applyDelta — Now uses replaceEqualDeep to build a new state object only when necessary. If no keys changed, returns the original state object unchanged, preventing unnecessary re-renders.

Benefits:

• Memoized components (via React.memo) skip re-rendering when their props haven't actually changed
• Identity-based caches and optimizations continue to work across updates
• An update that changes nothing returns the same state object, skipping the render entirely
• Handles edge cases like __proto__ keys from JSON safely

Test Plan

Added comprehensive Node.js-based tests in test_apply_delta_shares_unchanged_values that verify:

• Identity preservation for unchanged nested objects and arrays
• Proper handling of array length changes
• Key additions, removals, and renames
• Type changes and non-plain values
• Security: __proto__ keys remain as own data properties

All tests pass with Node.js runtime validation.

[chatgpt-codex-connector]

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-09-22T20:24:01.296357Z	9873f0b	PR opened

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

[greptile-apps]

Confidence Score: 5/5

[Medium risk] Optimizes state update logic on the frontend.

This PR appears safe to merge based on the reviewed changes.

Summary

The PR adds structural sharing to frontend state updates and tests identity preservation, key ordering, numeric equality, and nested __proto__ handling.

_{Reviews (2) · Last reviewed commit: "Assert the extracted state.js helpers co..."}

[cubic-dev-ai]

1 issue found across 3 files

Prompt for AI agents (unresolved issues)

Check if these issues are valid — if so, understand the root cause of each and fix them. When an issue isn't valid or won't be fixed in this PR, reply in its thread with the reason and then resolve the thread. If appropriate, use sub-agents to investigate and fix each issue separately.


<file name="tests/units/compiler/test_state_js_template.py">

<violation number="1" location="tests/units/compiler/test_state_js_template.py:131">
P3: This test extracts the template helpers by slicing the file between literal marker strings (`content.index("const isPlainObject =")` / `content.index("export const evalReactComponent =")`). Any future edit that mentions `evalReactComponent` before its definition, renames it, or restructures the file silently truncates the window (dropping `applyDelta`, so node fails with an unhelpful ReferenceError) or raises a bare ValueError. The repo already flagged delimiter-based source splitting in previous reviews; follow that guidance and locate `applyDelta` by parsing state.js with the installed Rolldown parser instead. At minimum, assert the sliced chunk actually contains `export const applyDelta` so a moved marker fails loudly at the extraction point.</violation>
</file>

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

Re-trigger cubic</sub>

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 9873f0b1f7

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[codspeed]

Merging this PR will not alter performance

✅ 147 untouched benchmarks
⏩ 17 skipped benchmarks¹

---

_{Comparing farhan/sleepy-hawking-bhua3z (b8727b1) with main (02efed7)}

Footnotes

1. 17 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩

[cubic-dev-ai]

All reported issues were addressed

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

Re-trigger cubic</sub>

[masenf]

this is definitely a behavior change and i'm not 100% sure it's what we want. in #7084 (not merged yet) i added a by_value kwarg to rx.memo which opts in to this kind of behavior at the memo level -- so the delta will still cause a re-render when it comes in, but opted in memo components won't re-render when their props have the same value.

this approaches it at a higher layer, so the optimization applies to all stateful components, but it requires deep checking every delta that comes in to make that determination.

there could also theoretically be apps out there that implicitly depend on re-renders for retriggering certain effects which could subtly break.