Releases: web-infra-dev/midscene
Releases · web-infra-dev/midscene
v0.5.0
What's Changed
New Features 🎉
- feat(web): input supports additions, deletions, and changes by @zhoushaw in #78
- feat(web-extract): extract some s as container by @yuyutaotao in #80
- feat(app): supports control of iOS and Android devices through appium by @quanru in #82
- feat(ai-model): support JSON 100% limit capability for gpt-4o-2024-08-06 model by @zhoushaw in #86
- feat(cache): The cache is generalized to support puppeteers and mobile terminals by @zhoushaw in #85
- feat(web-extract): collect svg elements on page, trim attribute values to save token by @yuyutaotao in #83
- feat(model): support azure open ai by @zhoushaw in #90
Bug Fixes 🐞
- fix(web-extract): fix the extractor may fail if there is no by @yuyutaotao in #76
- fix(ai-plan): optimize KeyboardPress plan type reduce hallucinations by @zhoushaw in #79
- fix(ci): fix build error for Cli, remove used package by @yuyutaotao in #89
- fix: type error in AI test by @yuyutaotao in #91
- fix(cache): app should not crash when it fails to get the cache id by @yuyutaotao in #93
Other Changes
- chore: optimize code style by @zhoushaw in #77
- refactor(plan): optimize input prompt by @zhoushaw in #81
- chore: delete unless debug code and fix ci cache logic by @zhoushaw in #84
- chore(web): merge the Page implementation of Puppeteer and Playwright by @yuyutaotao in #87
- workflow(ci): fix ai e2e test by @zhoushaw in #94
New Contributors
Full Changelog: v0.4.0...v0.5.0
Contributors
quanru, zhoushaw, and yuyutaotao
Assets 2
v0.4.0
What's Changed
Bug Fixes 🐞
- fix(cli): the flashing issue in cli headed mode by @yuyutaotao in #73
- fix(web-extract): fix the extractor for form item like by @yuyutaotao in #65
Other Changes
- refactor(shared): migrate sharp to jimp and migrate common img logci to shared lib by @zhoushaw in #74
- workflow(ci): optimize ci command by @zhoushaw in #75
Full Changelog: v0.3.4...v0.4.0
Contributors
zhoushaw and yuyutaotao
Assets 2
v0.3.1
What's Changed
Feature ✨
- feat: add 'aiWaitFor' by @yuyutaotao in #61
- feat(playwright): optimize the visualization and extraction of playwright ai process data by @zhoushaw in #66
Bug Fixes 🐞
- fix(prompt): change my to may by @fengzi2016 in #60
- fix(html-extract): optimize the logic for extracting HTML content to improve extraction accuracy by @zhoushaw in #62
Document 📖
Other Changes
- workflow(ci): optimize ai unit test by @zhoushaw in #56
- workflow(ci): fix the web extraction logic test by @zhoushaw in #63
- workflow(ci): optimize web extract unit test by @zhoushaw in #64
- workflow(ci): optimize ai e2e test by @zhoushaw in #67
- chore: update license by @chenjiahan in #59
New Contributors
- @chenjiahan made their first contribution in #59
- @fengzi2016 made their first contribution in #60
Full Changelog: v0.3.0...v0.3.1
Contributors
chenjiahan, fengzi2016, and 2 other contributors
Assets 2
v0.3.0
What's Changed
Feature ✨
- feat(report): optimize report content by @yuyutaotao @zhoushaw in #55
Bug Fixes 🐞
- fix: scroll issue of detail panel, exclude from extractor by @yuyutaotao @zhoushaw in #51
Other Changes
Full Changelog: v0.2.2...v0.3.0
Contributors
zhoushaw and yuyutaotao
Assets 2
v0.2.0
What's Changed
Features
- feat: update visualizer, and cache info field by @yuyutaotao in #24
- feat: update theme color by @yuyutaotao in #30
- feat: correct the captitalization by @yuyutaotao in #33
- feat: implement .aiAssert, update some docs by @yuyutaotao in #38
Document 📖
- docs: fix ch quick start by @zhoushaw in #34
- docs: update docs and logo by @yuyutaotao in #35
- docs: add cache introduce by @zhoushaw in #36
- docs: optimize words by @zhoushaw in #37
- docs: add introduce video and optimize pkg readme by @zhoushaw in #39
Other Changes
- workflow(cd): add pr label yml file by @zhoushaw in #23
- chore(deps-dev): bump minimist from 1.2.5 to 1.2.6 by @dependabot in #27
- chore(deps-dev): bump semver from 7.3.7 to 7.5.2 by @dependabot in #28
- workflow(ci): add contributing guide and migrate eslint to biome by @zhoushaw in #29
- workflow(ci): optimize biome lint by @zhoushaw in #31
- workflow(ci): optimize biome lint by @zhoushaw in #32
New Contributors
- @dependabot made their first contribution in #27
Full Changelog: v0.1.4...v0.2.0
Contributors
dependabot, zhoushaw, and yuyutaotao
Assets 2
v0.1.4
v0.1.2
v0.1.1
What's Changed
Features
- feat(web): export puppeteer integration by @yuyutaotao in #11
- feat(report): supports playwright test report generation by @zhoushaw in #15
- feat(web): added ai cache capability by @zhoushaw in #18
Bug Fixes 🐞
Document 📖
- doc: update API docs by @yuyutaotao in #17
Other Changes
- chore: delete unless file and optimize ci/cd command by @zhoushaw in #14
- workflow(cd): fix core 'langsmith/wrappers' deps error by @zhoushaw in #19
Full Changelog: v0.0.1...v0.1.1
Contributors
zhoushaw and yuyutaotao
Assets 2
v0.0.1
MidScene.js v0.0.1 - Initial Release 🎉
We are excited to announce the first release of MidScene.js! MidScene.js is a brand-new framework that leverages Natural Language Query (NLQ) and Artificial Intelligence (AI) technologies to simplify complex data queries and user interface interactions. Here are the main features and highlights of this release:
Key Features ✨
- Natural Language Page Control Operate pages using natural language, including actions like clicking and typing. 🗣️💻
- Natural Language Query Locate page elements using natural language, eliminating the need for DOM selectors. 🔍🗂️
- JSON Responses Prompt AI to generate the required data structures, ensuring the predictability of JSON structures and values. 📊📋
- TypeScript Friendly Automatically assign types and access data easily using dot notation.📝🔍
- Visualization ToolsEasily debug prompts and reasoning processes with visualization tools. 🛠️👀
- New Experience Enjoy a new world of automated development. 🌟🚀
- Ready-to-Use AI Models Utilize GPT-4o without any custom training. 🤖🔧
Detailed Update
- Natural Language Page Control: Users can directly perform actions like clicking and typing on web pages using natural language commands, greatly simplifying user interface interactions.
- Natural Language Query: Users can describe the location of page elements using natural language, eliminating the need to write complex DOM selector code.
- JSON Response Generation: AI can generate the required JSON data structures based on user needs, ensuring data accuracy and consistency.
- TypeScript Support: The framework automatically assigns types to generated data, making it easier and safer to use in TypeScript.
- Visualization Tools: Powerful visualization tools are provided to help users debug natural language prompts and AI reasoning processes.
- Ready-to-Use AI Models: The built-in GPT-4o model allows users to leverage powerful AI capabilities without additional training.
New Contributors
- @zhoushaw made their first contribution in #1
- @yuyutaotao made their first contribution in #5
Full Changelog: https://github.com/web-infra-dev/midscene/commits/v0.0.1
Contributors
zhoushaw and yuyutaotao