Spaces:
Runtime error
Runtime error
File size: 1,852 Bytes
ed4d993 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 |
"""Load Documents from Docusarus Documentation"""
from typing import Any, List, Optional
from langchain_community.document_loaders.sitemap import SitemapLoader
class DocusaurusLoader(SitemapLoader):
"""Load from Docusaurus Documentation.
It leverages the SitemapLoader to loop through the generated pages of a
Docusaurus Documentation website and extracts the content by looking for specific
HTML tags. By default, the parser searches for the main content of the Docusaurus
page, which is normally the <article>. You can also define your own
custom HTML tags by providing them as a list, for example: ["div", ".main", "a"].
"""
def __init__(
self,
url: str,
custom_html_tags: Optional[List[str]] = None,
**kwargs: Any,
):
"""Initialize DocusaurusLoader
Args:
url: The base URL of the Docusaurus website.
custom_html_tags: Optional custom html tags to extract content from pages.
kwargs: Additional args to extend the underlying SitemapLoader, for example:
filter_urls, blocksize, meta_function, is_local, continue_on_failure
"""
if not kwargs.get("is_local"):
url = f"{url}/sitemap.xml"
self.custom_html_tags = custom_html_tags or ["main article"]
super().__init__(
url,
parsing_function=kwargs.get("parsing_function") or self._parsing_function,
**kwargs,
)
def _parsing_function(self, content: Any) -> str:
"""Parses specific elements from a Docusaurus page."""
relevant_elements = content.select(",".join(self.custom_html_tags))
for element in relevant_elements:
if element not in relevant_elements:
element.decompose()
return str(content.get_text())
|