litellm/docs/extras/integrations/document_loaders/recursive_url_loader.ipynb

Recursive URL Loader

We may want to process load all URLs under a root directory.

For example, let's look at the LangChain JS documentation.

This has many interesting child pages that we may want to read in bulk.

Of course, the WebBaseLoader can load a list of pages.

But, the challenge is traversing the tree of child pages and actually assembling that list!

We do this using the RecursiveUrlLoader.

This also gives us the flexibility to exclude some children (e.g., the api directory with > 800 child pages).

In [1]:

from langchain.document_loaders.recursive_url_loader import RecursiveUrlLoader

Let's try a simple example.

In [2]:

url = "https://js.langchain.com/docs/modules/memory/examples/"
loader = RecursiveUrlLoader(url=url)
docs = loader.load()

In [3]:

len(docs)

Out [3]:

12

In [4]:

docs[0].page_content[:50]

Out [4]:

'\n\n\n\n\nBuffer Window Memory | 🦜️🔗 Langchain\n\n\n\n\n\nSki'

In [5]:

docs[0].metadata

Out [5]:

{'source': 'https://js.langchain.com/docs/modules/memory/examples/buffer_window_memory',
 'title': 'Buffer Window Memory | 🦜️🔗 Langchain',
 'description': 'BufferWindowMemory keeps track of the back-and-forths in conversation, and then uses a window of size k to surface the last k back-and-forths to use as memory.',
 'language': 'en'}

Now, let's try a more extensive example, the docs root dir.

We will skip everything under api.

For this, we can lazy_load each page as we crawl the tree, using WebBaseLoader to load each as we go.

In [ ]:

url = "https://js.langchain.com/docs/"
exclude_dirs = ["https://js.langchain.com/docs/api/"]
loader = RecursiveUrlLoader(url=url, exclude_dirs=exclude_dirs)
# Lazy load each
docs = [print(doc) or doc for doc in loader.lazy_load()]

In [7]:

# Load all pages
docs = loader.load()

In [8]:

len(docs)

Out [8]:

188

In [9]:

docs[0].page_content[:50]

Out [9]:

'\n\n\n\n\nAgent Simulations | 🦜️🔗 Langchain\n\n\n\n\n\nSkip t'

In [10]:

docs[0].metadata

Out [10]:

{'source': 'https://js.langchain.com/docs/use_cases/agent_simulations/',
 'title': 'Agent Simulations | 🦜️🔗 Langchain',
 'description': 'Agent simulations involve taking multiple agents and having them interact with each other.',
 'language': 'en'}