Directory loader for bulk documents in LangChain - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ component_behavior

intermediate

What does this DirectoryLoader code output?

Given this code snippet using Langchain's DirectoryLoader, what will be the output type of documents?

LangChain

from langchain.document_loaders import DirectoryLoader
loader = DirectoryLoader('data/docs', glob='*.txt')
documents = loader.load()

A A list of Document objects representing each text file in 'data/docs'

B A dictionary with filenames as keys and file contents as values

C A single string concatenating all text files in 'data/docs'

D A generator object yielding file paths from 'data/docs'

💡 Hint

Think about what the DirectoryLoader.load() method returns in Langchain.

DirectoryLoader.load() returns a list of Document objects, each representing one loaded file.

📝 Syntax

intermediate

Which option correctly filters files with DirectoryLoader?

You want to load only PDF files from a folder using DirectoryLoader. Which code snippet is correct?

A loader = DirectoryLoader('files', glob='*.pdf')

B loader = DirectoryLoader('files', filetype='pdf')

C loader = DirectoryLoader('files', pattern='*.pdf')

D loader = DirectoryLoader('files', filter='*.pdf')

💡 Hint

Check the parameter name used for file matching in DirectoryLoader.

The DirectoryLoader uses the 'glob' parameter to filter files by pattern.

🔧 Debug

advanced

Why does this DirectoryLoader code raise an error?

This code raises a FileNotFoundError. What is the most likely cause?

LangChain

from langchain.document_loaders import DirectoryLoader
loader = DirectoryLoader('docs', glob='*.md')
docs = loader.load()

A DirectoryLoader requires absolute paths only

B The glob pattern '*.md' is invalid syntax

C The 'docs' folder does not exist or is misspelled

D The load() method must be awaited as it is async

💡 Hint

Check if the folder path is correct and exists.

FileNotFoundError means the folder path is wrong or missing, not a glob or async issue.

❓ state_output

advanced

What is the length of documents after loading?

If the 'articles' folder contains 5 text files and you run this code, what is the length of documents?

LangChain

from langchain.document_loaders import DirectoryLoader
loader = DirectoryLoader('articles', glob='*.txt')
documents = loader.load()
print(len(documents))

A 1

B Number of lines in all files combined

C 0

D 5

💡 Hint

Each file matching the glob creates one Document object.

DirectoryLoader.load() returns one Document per file matched, so length equals number of files.

🧠 Conceptual

expert

Which statement about DirectoryLoader is true?

Select the correct statement about how DirectoryLoader handles files.

A DirectoryLoader only loads file names, not contents, until accessed

B DirectoryLoader reads all files into memory as Document objects immediately on load()

C DirectoryLoader streams files lazily and returns a generator on load()

D DirectoryLoader requires manual file reading after load()

💡 Hint

Think about when file contents are read by DirectoryLoader.

DirectoryLoader.load() reads all matching files immediately and returns a list of Document objects with content.