vodorod/Dorod-Sky
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity

refactor context tree (#212)

Browse Source
This commit is contained in:
LawyZheng
2024-04-21 22:30:37 +08:00
committed by GitHub
parent 02db2a90e6
commit cc6ae8bae0
4 changed files with 217 additions and 63 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
skyvern/webeye/scraper/scraper.py
View File

@@ -184,7 +184,7 @@ async def scrape_web_unsafe(
await remove_bounding_boxes(page)
await scroll_to_top(page, drow_boxes=False)
elements, element_tree = await get_interactable_element_tree(page)
elements, element_tree = await get_interactable_element_tree(page, browser_state.new_context_tree)
element_tree = cleanup_elements(copy.deepcopy(element_tree))
_build_element_links(elements)
@@ -211,15 +211,15 @@ async def scrape_web_unsafe(
)
async def get_interactable_element_tree(page: Page) -> tuple[list[dict], list[dict]]:
async def get_interactable_element_tree(page: Page, new_context_tree: bool) -> tuple[list[dict], list[dict]]:
"""
Get the element tree of the page, including all the elements that are interactable.
:param page: Page instance to get the element tree from.
:return: Tuple containing the element tree and a map of element IDs to elements.
"""
await page.evaluate(JS_FUNCTION_DEFS)
js_script = "() => buildTreeFromBody()"
elements, element_tree = await page.evaluate(js_script)
js_script = "(new_ctx) => buildTreeFromBody(new_ctx)"
elements, element_tree = await page.evaluate(js_script, new_context_tree)
return elements, element_tree