Add clickable metadata support to embedded Victory bar charts

[inimaga]

Held on:

• https://github.com/Expensify/Web-Expensify/pull/54516
• https://github.com/Expensify/Web-Expensify/pull/54677

Explanation of Change

This PR adds per-point metadata parsing for embedded Victory chart data, preserving label and searchQuery fields and carrying them through the Victory chart renderer context.
Vertical embedded bar charts should now show web hover tooltips from that metadata and navigate to Search when a bar has a searchQuery.

Fixed Issues

$ #92149
PROPOSAL: N/A

Tests

• Verify that no errors appear in the JS console

Automated Tests

Covered by added automated tests

Manual Test

Pre-requisite (Skip to Test Steps if testing in Staging/Prod):

1. Ensure you have AZ/Concierge is setup in dev
2. Checkout and run https://github.com/Expensify/Web-Expensify/pull/54516 in Web-E
3. Checkout and run Add clickable metadata support to embedded Victory bar charts #96065 (current branch) in App and login with a test account

Test Steps:

1. Ensure the selected test account has a bunch of expenses across several months this year. (If not, create some dummy expenses).
2. Start a 1-on-1 DM with Concierge.
3. Ask it something like “Give me a spend bar chart of my monthly expenses for this year.”
4. Wait for the response with an embedded vertical Victory bar chart.
5. On the App PR build, hover a bar.
6. Verify the tooltip appears.
7. Click a bar.
8. Verify it navigates to the corresponding Search results.

Issue.92149.mp4

Offline tests

QA Steps

Same as manual test steps above

• Verify that no errors appear in the JS console

PR Author Checklist

• I linked the correct issue in the ### Fixed Issues section above
• I wrote clear testing steps that cover the changes made in this PR
  • I added steps for local testing in the Tests section
  • I added steps for the expected offline behavior in the Offline steps section
  • I added steps for Staging and/or Production testing in the QA steps section
  • I added steps to cover failure scenarios (i.e. verify an input displays the correct error message if the entered data is not correct)
  • I turned off my network connection and tested it while offline to ensure it matches the expected behavior (i.e. verify the default avatar icon is displayed if app is offline)
  • I tested this PR with a High Traffic account against the staging or production API to ensure there are no regressions (e.g. long loading states that impact usability).
• I included screenshots or videos for tests on all platforms
• I ran the tests on all platforms & verified they passed on:
  • Android: Native
  • Android: mWeb Chrome
  • iOS: Native
  • iOS: mWeb Safari
  • MacOS: Chrome / Safari
• I verified there are no console errors (if there's a console error not related to the PR, report it or open an issue for it to be fixed)
• I followed proper code patterns (see Reviewing the code)
  • I verified that comments were added to code that is not self explanatory
  • I verified that any new or modified comments were clear, correct English, and explained "why" the code was doing something instead of only explaining "what" the code was doing.
  • I verified any copy / text that was added to the app is grammatically correct in English. It adheres to proper capitalization guidelines (note: only the first word of header/labels should be capitalized), and is either coming verbatim from figma or has been approved by marketing (in order to get marketing approval, ask the Bug Zero team member to add the Waiting for copy label to the issue)
• If a new code pattern is added I verified it was agreed to be used by multiple Expensify engineers
• I followed the guidelines as stated in the Review Guidelines
• I tested other components that can be impacted by my changes (i.e. if the PR modifies a shared library or component like Avatar, I verified the components using Avatar are working as expected)
• If a new CSS style is added I verified that:
  • A similar style doesn't already exist
  • The style can't be created with an existing StyleUtils function (i.e. StyleUtils.getBackgroundAndBorderStyle(theme.componentBG))
• If new assets were added or existing ones were modified, I verified that:
  • The assets are optimized and compressed (for SVG files, run npm run compress-svg)
  • The assets load correctly across all supported platforms.
• If the PR modifies code that runs when editing or sending messages, I tested and verified there is no unexpected behavior for all supported markdown - URLs, single line code, code blocks, quotes, headings, bold, strikethrough, and italic.
• If the PR modifies a generic component, I tested and verified that those changes do not break usages of that component in the rest of the App (i.e. if a shared library or component like Avatar is modified, I verified that Avatar is working as expected in all cases)
• If the PR modifies a component related to any of the existing Storybook stories, I tested and verified all stories for that component are still working as expected.
• If the PR modifies a component or page that can be accessed by a direct deeplink, I verified that the code functions as expected when the deeplink is used - from a logged in and logged out account.
• If the PR modifies the UI (e.g. new buttons, new UI components, changing the padding/spacing/sizing, moving components, etc) or modifies the form input styles:
  • I verified that all the inputs inside a form are aligned with each other.
  • I added Design label and/or tagged @Expensify/design so the design team can review the changes.
• I added unit tests for any new feature or bug fix in this PR to help automatically prevent regressions in this user flow.
• If the main branch was merged into this PR after a review, I tested again and verified the outcome was still expected according to the Test steps.

Screenshots/Videos

Android: Native

Android: mWeb Chrome

iOS: Native

iOS: mWeb Safari

MacOS: Chrome / Safari

[codecov]

Codecov Report

✅ Changes either increased or maintained existing code coverage, great job!

Files with missing lines	Coverage Δ
.../VictoryChartRenderer/BaseVictoryChartRenderer.tsx	0.00% <ø> (ø)
...ryChartRenderer/parsers/processVictoryChartTree.ts	75.00% <100.00%> (+1.78%)	⬆️
...oryChartRenderer/utils/getChartPointMetadataKey.ts	100.00% <100.00%> (ø)
...rs/VictoryChartRenderer/utils/parseRawChartData.ts	100.00% <100.00%> (ø)
src/components/Charts/hooks/useLabelHitTesting.ts	0.00% <0.00%> (ø)
.../components/VictoryChartContainer/index.native.tsx	9.09% <0.00%> (ø)
...ctoryChartRenderer/context/VictoryChartContext.tsx	5.55% <0.00%> (ø)
...ctoryChartContainer/VictoryChartContainerFixed.tsx	0.00% <0.00%> (ø)
...hartRenderer/context/VictoryChartLayoutContext.tsx	40.00% <40.00%> (ø)
...ictoryChartRenderer/parsers/victorySeriesParser.ts	90.00% <86.36%> (+5.38%)	⬆️
... and 8 more
... and 31 files with indirect coverage changes

[melvin-bot]

@hoangzinh Please copy/paste the Reviewer Checklist from here into a new comment on this PR and complete it. If you have the K2 extension, you can simply click: [this button]

[situchan]

If this requires C+ review, I'll be reviewing this as part of project.

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 3c467b0a5c

ℹ️ About Codex in GitHub

Codex has been enabled to automatically 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 👍.

When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".

[trjExpensify]

What's up with the axis here? On the spend page the values are dynamic based on the data set. This looks like it puts all of the bars below the first data point on the axis?

For the data on hover, why aren't we showing the same data as the Spend page?

CC: @Expensify/design @luacmartins

[shawnborton]

Yeah agree, I would think the y axis would use smaller amounts so the bars would be taller? That feels odd for sure.

[melvin-bot]

@mollfpr Please copy/paste the Reviewer Checklist from here into a new comment on this PR and complete it. If you have the K2 extension, you can simply click: [this button]

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: b35f125d54

ℹ️ About Codex in GitHub

Codex has been enabled to automatically 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 👍.

When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".

[MelvinBot]

🤖 Review — clickable metadata on embedded Victory bar charts

Solid, well-structured PR with good test coverage (parser metadata, geometry math, coordinate normalization). I dug into the two riskiest areas — the web coordinate-scaling and the new worklet hit-testing — and both check out. No blocking bugs found. A few non-blocking questions below.

Verified correct:

• Coordinate scaling (coordinateScale divide): correct direction on web. The transform scale is applied to the inner content view while the chart is laid out at design size, so RNGH reports gesture coords in rendered space (getBoundingClientRect-based) while pointOX/chartBounds stay in design space. e.x / scaledScale maps rendered → design, matching the point positions. The tooltip has no space mismatch either — onLayout width is offsetWidth (layout/design space), consistent with initialTooltipPosition.
• Hit-test geometry in getVictoryBarInteractionGeometry.ts matches victory-native's useBarWidth/useBarGroupPaths formulas (ungrouped + grouped, incl. BAR_INNER_PADDING and offset), and divide-by-zero is guarded (pointCount<=0 → 0, Math.max(1, …)).
• Index-space alignment: xs/ys/widths/searchQueryFlags are all built in one pass over interactiveBars, so matchedIndex ↔ interactiveBars[i] ↔ pointWidth[i] stay consistent — no off-by-one.
• Negative-index safety: every .at() on a possibly--1 index is guarded (JS .at(-1) returning the last element is a classic trap, and it's avoided here).
• Metadata key parity: getChartPointMetadataKey embeds typeof, and write (point.x) vs read (row[X_KEY]) use the same value on the vertical path (horizontal is excluded from interactivity), so lookups never miss.
• Percentage denominator: interactiveBars ⊆ visibleVerticalBarValues, so the active bar's value is always part of the total — % of total is well-defined.

Non-blocking questions / notes:

1. Expand/fullscreen modal isn't interactive. BaseVictoryChartRenderer now renders VictoryChartInteractiveContent, but VictoryChartExpandModal still renders the non-interactive VictoryChartContent (L129, L145). So hover tooltips / tap-to-search won't work in the expanded view. Is that intended (inline-only), or should the modal get the same treatment?

2. Bars with only searchQuery (no label/currency) still show a tooltip of just the raw x value, since label defaults to String(xValue) (useVictoryBarInteractions.ts:106) with an empty amount. Is a bare-x-value tooltip desired here, or should those bars be click-only?

3. Minor / cosmetic: hasTooltipLabels (useVictoryBarInteractions.ts:268) is effectively always true because label is defaulted above — the guard is dead. Not worth a change unless you're touching this area.

4. Reactive → imperative flag refactor: switching isTooltipActive/isCursorOverClickable/isCursorOverTarget from useDerivedValue to plain useSharedValue updated only inside gesture callbacks is fine for all current consumers (they read via useAnimatedStyle/useAnimatedReaction, which re-run on any write). The only behavioral delta: chartBottom.set(...) runs during relayout outside a gesture and no longer refreshes the flags until the next pointer event. Only observable if a relayout happens while the pointer sits perfectly still over a bar — acceptable, just flagging it.

_{Findings above were verified against victory-native and react-native-gesture-handler/-web internals; confidence is high, but I didn't run the app — the manual/QA steps still stand.}

[MelvinBot]

🤖 Regression scan

I compared old vs. new behavior across every chart type that flows through the shared render/parse/interaction paths this PR touched (line, area, scatter, vertical bar, horizontal bar, pie, donut — both the embedded Concierge charts and the standalone analytics charts). No hard regressions found. Two behavioral changes are worth a conscious sign-off:

1. Existing vertical bar charts with a labels attribute now become interactive (verify this is intended).
The parser fills point metadata from the labels attribute as a fallback (victorySeriesParser.ts:27-31), so any pre-existing vertical bar chart that already emits labels= — even with no searchQuery/currency — will now trip hasInteractiveBars (VictoryChartCartesianInteractive.tsx:51), get wrapped in a GestureDetector, and show a label-only hover tooltip (non-clickable, since no searchQuery). Line/area/scatter/horizontal-bar/pie/donut are unaffected (barSeriesConfig is {} for them, so hasInteractiveBars stays false). If some existing bar charts already ship a labels attribute, they'll silently gain hover tooltips — likely intended, just confirm it's desired.

2. Standalone BarChart hover tooltip / pointer-cursor no longer auto-clears on re-layout without mouse movement (low severity, web-only).
The refactor changed isTooltipActive / isCursorOverClickable / isCursorOverTarget from reactive useDerivedValue worklets to plain shared values updated only inside gesture callbacks. The standalone BarChartContent writes barWidth/yZero/chartBottom outside gestures during layout (:130-131, :164, :200) — its hit-test (checkIsOverBar) depends on those. Before, a re-layout/live-data refresh while the cursor sat still over a bar would auto-recompute the derived flags and hide the tooltip / reset the cursor; now the flags stay frozen until the next pointer move. Requires an active hover concurrent with a re-render and a stationary cursor, so it's an edge case, and any micro mouse movement re-syncs it — but it is a real difference for the existing bar chart. The equivalent line-chart path is not affected (its hit-test ignores chartBottom/barWidth/yZero).

Verified clean (no regression):

• Line / area / scatter / horizontal-bar / pie / donut — render through the same plain path as before; the new useVictoryBarInteractions() runs but produces hasInteractiveBars = false with no gesture overlay and no hook-order issue.
• victorySeriesParser rewrite — the data/yKeys output is byte-equivalent (same keys, order, number coercion); only the separate additive pointMetadata field is new.
• parseRawChartData — metadata copies are typeof-guarded; metadata-less points are identical, malformed metadata is dropped.
• ChartTooltip.getTooltipContent — the new if (!amount) return label branch only differs for empty amount; every existing caller (Line/Bar/Pie via useTooltipData) always passes a formatted non-empty amount, so their output is unchanged.
• pointMetadata threading (processVictoryChartTree / context / types) — purely additive; can't alter data/categories/labelItems/legendItems and adds no re-renders (context value was already a fresh literal each render).
• normalizeChartCoordinate — defaults to coordinateScale = 1 for the non-embedded consumers, so e.x / 1 === e.x (no-op); tap/handlePress decision and onEnd cleanup are behaviorally equivalent to before for line/bar.

_{Findings verified from code (PR head b35f125) against victory-native/reanimated internals — I did not run the app. Item 2 is best confirmed with a quick manual check: hover a bar on the analytics dashboard while it re-renders (e.g. resize / data refresh) and watch whether the tooltip clears without moving the mouse.}

[mollfpr]

Codex and Melvin pointed out that the expand/fullscreen modal still renders the non-interactive chart, so the tooltips and tap-to-search don't work there. Is that intentional for this PR or out of scope?

[situchan]

Codex and Melvin pointed out that the expand/fullscreen modal still renders the non-interactive chart, so the tooltips and tap-to-search don't work there. Is that intentional for this PR or out of scope?

I suggest expand modal support to be out of scope.

The last commit has bug. Modal doesn't dismiss.

Screen.Recording.2026-08-03.at.2.41.57.PM.mov

[situchan]

Minor: tooltip triangle is separated (need 🦅's 👁️)

[inimaga]

I suggest expand modal support to be out of scope.

Yeah, this is getting drawn out with increasing scope creep and fixing that is long. Going to undo last commit for now

[inimaga]

cc @mollfpr

[shawnborton]

Minor: tooltip triangle is separated

Agree, that's a bug. Are you zoomed in or zoomed out by chance? Sometimes that adds the small gap there.

[situchan]

Minor: tooltip triangle is separated

Agree, that's a bug. Are you zoomed in or zoomed out by chance? Sometimes that adds the small gap there.

No zoom. Normal view

[mollfpr]

Got it! Thanks @inimaga @situchan!

[github-actions]

🚧 mollfpr has triggered a test Expensify/App build. You can view the workflow run here.

[OSBotify]

✋ This PR was not deployed to staging yet because QA is ongoing. It will be automatically deployed to staging after the next production release.

[github-actions]

🧪🧪 Use the links below to test this adhoc build on Android, iOS, and Web. Happy testing! 🧪🧪
Built from App PR #96065.

Android 🤖	iOS 🍎
https://ad-hoc-expensify-cash.s3.us-east-1.amazonaws.com/rock-artifacts/ad-hoc/rock-android-Adhoc-23680e91b0d-2c5d336-9216baa9999aad3fe5c67483f4aaafb8fc5be802/index.html	https://ad-hoc-expensify-cash.s3.us-east-1.amazonaws.com/rock-artifacts/ad-hoc/rock-ios-device-AdHoc-23680e9-2c5d336-b15e513ed89588e1c674e76cebb67272ae6ce00b/index.html

Web 🕸️
https://96065.pr-testing.expensify.com

---

👀 View the workflow run that generated this build 👀

[OSBotify]

🚀 Deployed to staging by https://github.com/mollfpr in version: 9.4.49-0 🚀

platform	result
🕸 web 🕸	success ✅
🤖 android 🤖	success ✅
🍎 iOS 🍎	success ✅

[MelvinBot]

Help site review — changes required ✅

This PR adds interactive metadata to the embedded Victory bar charts Concierge renders in chat: on web you can now hover a bar for a tooltip and click a bar to open the matching expenses in Search. That's a new user-facing capability, and the help site didn't document it.

Where it belongs: How Concierge Analyzes Spend covers asking Concierge about your expense data, but never mentioned that Concierge can return an interactive chart — so I added a short section there rather than creating a new article.

Draft help site PR: #97738 — adds a task-based ## How to explore a Concierge spend chart section (interactive chart in chat; hover a bar for a tooltip; select a bar to open its expenses on the Search page) plus retrieval keywords.

Scope & authoring notes

• Reused the existing in-scope article (one workflow per article) instead of adding a new one.
• Followed HELP_AUTHORING_GUIDELINES.md, HELPSITE_NAMING_CONVENTIONS.md, and TEMPLATE.md: task-based ## How to… heading, exact UI label (Search), #/## only.
• Scoped the interactive behavior to web, matching what this PR implemented and tested (web hover tooltips) — no unverified mobile parity claims. If click-to-Search is also intended on mobile, flag it and I'll adjust.
• The in-app Insights reports (Spend tab bar/line/pie views) are a separate feature and already documented; no changes needed there.

---

@inimaga, please review the linked help site PR and confirm it reflects the current behavior. Then mark the linked help site PR Ready for review.

[OSBotify]

🚀 Deployed to production by https://github.com/arosiclair in version: 9.4.49-3 🚀

platform	result
🕸 web 🕸	success ✅
🤖 android 🤖	success ✅
🍎 iOS 🍎	success ✅

Bundle Size Analysis (Sentry):

• Android