mirror of https://github.com/hwchase17/langchain
parent
5565f56273
commit
07d7096de6
@ -0,0 +1,87 @@
|
|||||||
|
"""Loader that uses Playwright to load a page, then uses unstructured to load the html.
|
||||||
|
"""
|
||||||
|
import logging
|
||||||
|
from typing import List, Optional
|
||||||
|
|
||||||
|
from langchain.docstore.document import Document
|
||||||
|
from langchain.document_loaders.base import BaseLoader
|
||||||
|
|
||||||
|
logger = logging.getLogger(__file__)
|
||||||
|
|
||||||
|
|
||||||
|
class PlaywrightURLLoader(BaseLoader):
|
||||||
|
"""Loader that uses Playwright and to load a page and unstructured to load the html.
|
||||||
|
This is useful for loading pages that require javascript to render.
|
||||||
|
|
||||||
|
Attributes:
|
||||||
|
urls (List[str]): List of URLs to load.
|
||||||
|
continue_on_failure (bool): If True, continue loading other URLs on failure.
|
||||||
|
headless (bool): If True, the browser will run in headless mode.
|
||||||
|
"""
|
||||||
|
|
||||||
|
def __init__(
|
||||||
|
self,
|
||||||
|
urls: List[str],
|
||||||
|
continue_on_failure: bool = True,
|
||||||
|
headless: bool = True,
|
||||||
|
remove_selectors: Optional[List[str]] = None,
|
||||||
|
):
|
||||||
|
"""Load a list of URLs using Playwright and unstructured."""
|
||||||
|
try:
|
||||||
|
import playwright # noqa:F401
|
||||||
|
except ImportError:
|
||||||
|
raise ValueError(
|
||||||
|
"playwright package not found, please install it with "
|
||||||
|
"`pip install playwright`"
|
||||||
|
)
|
||||||
|
|
||||||
|
try:
|
||||||
|
import unstructured # noqa:F401
|
||||||
|
except ImportError:
|
||||||
|
raise ValueError(
|
||||||
|
"unstructured package not found, please install it with "
|
||||||
|
"`pip install unstructured`"
|
||||||
|
)
|
||||||
|
|
||||||
|
self.urls = urls
|
||||||
|
self.continue_on_failure = continue_on_failure
|
||||||
|
self.headless = headless
|
||||||
|
self.remove_selectors = remove_selectors
|
||||||
|
|
||||||
|
def load(self) -> List[Document]:
|
||||||
|
"""Load the specified URLs using Playwright and create Document instances.
|
||||||
|
|
||||||
|
Returns:
|
||||||
|
List[Document]: A list of Document instances with loaded content.
|
||||||
|
"""
|
||||||
|
from playwright.sync_api import sync_playwright
|
||||||
|
from unstructured.partition.html import partition_html
|
||||||
|
|
||||||
|
docs: List[Document] = list()
|
||||||
|
|
||||||
|
with sync_playwright() as p:
|
||||||
|
browser = p.chromium.launch(headless=self.headless)
|
||||||
|
for url in self.urls:
|
||||||
|
try:
|
||||||
|
page = browser.new_page()
|
||||||
|
page.goto(url)
|
||||||
|
|
||||||
|
for selector in self.remove_selectors or []:
|
||||||
|
element = page.locator(selector)
|
||||||
|
if element.is_visible():
|
||||||
|
element.evaluate("element => element.remove()")
|
||||||
|
|
||||||
|
page_source = page.content()
|
||||||
|
elements = partition_html(text=page_source)
|
||||||
|
text = "\n\n".join([str(el) for el in elements])
|
||||||
|
metadata = {"source": url}
|
||||||
|
docs.append(Document(page_content=text, metadata=metadata))
|
||||||
|
except Exception as e:
|
||||||
|
if self.continue_on_failure:
|
||||||
|
logger.error(
|
||||||
|
f"Error fetching or processing {url}, exception: {e}"
|
||||||
|
)
|
||||||
|
else:
|
||||||
|
raise e
|
||||||
|
browser.close()
|
||||||
|
return docs
|
Loading…
Reference in New Issue