Google Git
Sign in
apache/echarts/refs/heads/codex/fix-toolbox-stack-21111
commit
7d0292f58f7ac2c90036c5622fe0fc39c6ec4fdb
[log]
author
susiwen8 <susiwen8@gmail.com>
Sun May 03 00:41:44 2026 +0800
committer
susiwen8 <susiwen8@gmail.com>
Sun May 03 00:41:44 2026 +0800
tree
aab9d4c48d939af6082a9b6f761c5a011e1bb36b
parent
53edc47511b5d15709ac832eb0ec9eadc7f2a23b [diff]
Assert empty dataZoom preserves in-window raw stack values

Copilot suggested using raw axis dimensions for empty mode. A temporary verification showed that change turns raw y values for in-window stacked points into NaN, even though their stack result remains visible. Preserve the current implementation and cover that invariant explicitly.

Constraint: mapDimensionsAll(axisDim) includes both raw value and stack result dimensions for stacked series.

Rejected: Apply Copilot empty-mode suggestion | it corrupts raw values for in-window stacked points.

Confidence: high

Scope-risk: narrow

Directive: Do not special-case empty mode to mapDimensionsAll without preserving raw stacked values.

Tested: ./node_modules/.bin/jest --config test/ut/jest.config.cjs --coverage=false test/ut/spec/component/dataZoom --runInBand

Tested: npm run checktype

Tested: ./node_modules/.bin/eslint test/ut/spec/component/dataZoom/AxisProxy.test.ts

Tested: git diff --check -- test/ut/spec/component/dataZoom/AxisProxy.test.ts
1 file changed
tree: aab9d4c48d939af6082a9b6f761c5a011e1bb36b
  1. .github/
  2. .husky/
  3. .vscode/
  4. asset/
  5. build/
  6. dist/
  7. extension-src/
  8. i18n/
  9. licenses/
  10. src/
  11. ssr/
  12. test/
  13. theme/
  14. .asf.yaml
  15. .editorconfig
  16. .eslintignore
  17. .eslintrc-common.yaml
  18. .gitattributes
  19. .gitignore
  20. .headerignore
  21. .jshintrc-dist
  22. .lgtm.yml
  23. .npmignore
  24. CONTRIBUTING.md
  25. index.d.ts
  26. KEYS
  27. LICENSE
  28. NOTICE
  29. package-lock.json
  30. package.json
  31. package.README.md
  32. README.md
  33. tsconfig.json
README.md

Apache ECharts

Apache ECharts is a free, powerful charting and visualization library offering easy ways to add intuitive, interactive, and highly customizable charts to your commercial products. It is written in pure JavaScript and based on zrender, which is a whole new lightweight canvas library.

中文官网 | ENGLISH HOMEPAGE

License Latest npm release NPM downloads Contributors

Build Status

Get Apache ECharts

You may choose one of the following methods:

Docs

Get Help

Build

Build echarts source code:

Execute the instructions in the root directory of the echarts: (Node.js is required)

# Install the dependencies from NPM:
npm install

# Rebuild source code immediately in watch mode when changing the source code.
# It opens the `./test` directory, and you may open `-cases.html` to get the list
# of all test cases.
# If you wish to create a test case, run `npm run mktest:help` to learn more.
npm run dev

# Check the correctness of TypeScript code.
npm run checktype

# If intending to build and get all types of the "production" files:
npm run release

Then the “production” files are generated in the dist directory.

Contribution

Please refer to the contributing document if you wish to debug locally or make pull requests.

Resources

Awesome ECharts

https://github.com/ecomfe/awesome-echarts

Extensions

License

ECharts is available under the Apache License V2.

Code of Conduct

Please refer to Apache Code of Conduct.

Paper

Deqing Li, Honghui Mei, Yi Shen, Shuang Su, Wenli Zhang, Junting Wang, Ming Zu, Wei Chen. ECharts: A Declarative Framework for Rapid Construction of Web-based Visualization. Visual Informatics, 2018.