general autocomplete solution

[LawyZheng]

🚀	This description was created by Ellipsis for commit d0c63d1

Summary:

This PR implements a comprehensive solution for handling autocomplete actions in Skyvern, including new exception classes, prompt templates, and extensive updates to action handlers, DOM utilities, and scraper logic.

Key points:

• Added new exception classes in skyvern/exceptions.py for handling autocomplete errors.
• Introduced templates in skyvern/forge/prompts/skyvern/ for autocomplete prompts.
• Implemented AutoCompletionResult class in skyvern/webeye/actions/handler.py.
• Modified handle_input_text_action to handle autocomplete inputs.
• Added choose_auto_completion_dropdown and input_or_auto_complete_input functions for handling autocomplete logic.
• Updated domUtils.js to support new DOM operations related to autocomplete.
• Enhanced scraper.py to handle incremental scraping for autocomplete elements.
• Updated SkyvernElement class in dom.py to identify autocomplete inputs.
• Adjusted SkyvernFrame in page.py to support new JS functions for autocomplete.

---

Generated with ❤️ by ellipsis.dev

[ellipsis-dev]

👍 Looks good to me! Reviewed everything up to d0c63d1 in 52 seconds

More details

• Looked at 1064 lines of code in 9 files
• Skipped 0 files when reviewing.
• Skipped posting 4 drafted comments based on config settings.

1. skyvern/webeye/scraper/scraper.py:418

• Draft comment:
  Remove async from the JavaScript function in frame.evaluate. It is not needed and can be directly executed.
• Reason this comment was not posted:
  Confidence changes required: 50%
  The function get_incremental_element_tree in IncrementalScrapePage class is using frame.evaluate with an async function, which is not necessary. The frame.evaluate method can directly execute the JavaScript function without the async keyword.

2. skyvern/webeye/scraper/domUtils.js:971

• Draft comment:
  Remove async from buildTreeFromBody and getIncrementElements functions as they do not perform asynchronous operations.
• Reason this comment was not posted:
  Confidence changes required: 50%
  The buildTreeFromBody and getIncrementElements functions in domUtils.js are using async keyword unnecessarily. These functions are not performing any asynchronous operations and can be simplified by removing the async keyword.

3. skyvern/webeye/scraper/scraper.py:376

• Draft comment:
  Remove async from buildTreeFromBody function as it does not perform asynchronous operations.
• Reason this comment was not posted:
  Confidence changes required: 50%
  The buildTreeFromBody function in scraper.py is using async keyword unnecessarily. This function is not performing any asynchronous operations and can be simplified by removing the async keyword.

4. skyvern/webeye/scraper/scraper.py:415

• Draft comment:
  Remove async from the JavaScript function in frame.evaluate. It is not needed and can be directly executed.
• Reason this comment was not posted:
  Confidence changes required: 50%
  The get_incremental_element_tree method in IncrementalScrapePage class is using frame.evaluate with an async function, which is not necessary. The frame.evaluate method can directly execute the JavaScript function without the async keyword.

Workflow ID: wflow_w9enw1QZF5yCdMTp

---

You can customize Ellipsis with 👍 / 👎 feedback, review rules, user-specific overrides, quiet mode, and more.