Errors with DirectoryLoader

[tanayvarshney]

from langchain.document_loaders import DirectoryLoader
loader = DirectoryLoader('./server', glob="**/*.md")
data = loader.load()

Error

    from pdfminer.utils import open_filename
ImportError: cannot import name 'open_filename' from 'pdfminer.utils' (/usr/local/lib/python3.8/dist-packages/pdfminer/utils.py)

Langchain version: '0.0.152'

[ysato]

Ran into the same issue this morning. I had to downgrade unstructured to version 0.6.1.

[botchagalupe]

from langchain.document_loaders import DirectoryLoader
loader = DirectoryLoader('./server', glob="**/*.md")
data = loader.load()

Error

    from pdfminer.utils import open_filename
ImportError: cannot import name 'open_filename' from 'pdfminer.utils' (/usr/local/lib/python3.8/dist-packages/pdfminer/utils.py)

Langchain version: '0.0.152'

I have the same problem with loading certain pdfs. Is there a way to turn on a trace/debug option when the loader is running so I can see what file if fails on?

[eminmtas]

from getpass import getpass
from langchain.document_loaders import GitHubIssuesLoader
ACCESS_TOKEN = getpass()

loader = GitHubIssuesLoader(
    repo="eminmtas/check-license",
    access_token=ACCESS_TOKEN,
    creator="eminmtas",
)

Error

from langchain.document_loaders import GitHubIssuesLoader
ImportError: cannot import name 'GitHubIssuesLoader' from 'langchain.document_loaders'

I am getting this error too. I have the required libraries on my venv. What to do?

[dosubot[bot]]

Hi, @tanayvarshney. I'm Dosu, and I'm helping the LangChain team manage their backlog. I wanted to let you know that we are marking this issue as stale.

From what I understand, the issue titled "Errors with DirectoryLoader" is about an import error when using the DirectoryLoader from the langchain.document_loaders module. The error seems to be related to the open_filename function from the pdfminer.utils module. This issue was encountered with version 0.0.152 of LangChain. Another user mentioned that downgrading the unstructured library to version 0.6.1 resolved the issue. There is also a user who encountered a similar error with the GitHubIssuesLoader and is seeking guidance on how to resolve it.

Before we close this issue, we wanted to check with you if it is still relevant to the latest version of the LangChain repository. If it is, please let us know by commenting on the issue. Otherwise, feel free to close the issue yourself, or it will be automatically closed in 7 days.

Thank you for your understanding and contribution to the LangChain project. Let us know if you have any further questions or concerns.

[advaithgit]

ImportError: cannot import name 'open_filename' from 'pdfminer.utils' Tried all the mentioned methods in the net but still getting the same issue.

[dosubot[bot]]

@baskaryan Could you please help @advaithgit with the issue titled "Errors with DirectoryLoader"? They are encountering an import error when using the DirectoryLoader from the langchain.document_loaders module. The error seems to be related to the open_filename function from the pdfminer.utils module. They have tried the mentioned methods in the net but are still facing the same issue. Thank you!

[Amaresh078724]

yes, please help with this issue

[lakshman-11]

# Install package

pip install "unstructured[all-docs]"

this worked for me

[mkmohangb]

Just upgrade your pdfminer.six package. See here