[Bug]: Blob storage loads all files regardless of `base_dir`

[Robotuks]

Do you need to file an issue?

• I have searched the existing issues and this bug is not already filed.
• My model is hosted on OpenAI or Azure. If not, please look at the "model providers" issue and don't file a new one here.
• I believe this is a legitimate bug, not just a question. If this is a question, please use the Discussions area.

Describe the bug

Using blob storage for input. But when running indexing noticed that it tries to read more files than it should. It gives me all files within a container.

input config example:

input:
  storage:
    type: blob # [file, blob]
    base_dir: "input/folder_1/"
    container_name: "graphrag-container"
    storage_account_blob_url: "https://graphragexample.blob.core.windows.net"
  file_type: text

Using this example config, getting warning messages during indexing like:
2025-10-24 11:30:38.0647 - WARNING - graphrag.storage.blob_pipeline_storage - Error getting key input/folder_1/input/folder_2/text.txt

In code, noticed you are getting all blobs from the container:

graphrag/graphrag/storage/blob_pipeline_storage.py

Line 146 in fdb7e38

all_blobs = list(container_client.list_blobs())

Couldn't pin point exact place why other directories are not being filtered out but you could just get files under a base dir to avoid this issue:
all_blobs = list(container_client.list_blobs(name_starts_with=base_dir))

Steps to reproduce

No response

Expected Behavior

No response

GraphRAG Config Used

No response

Logs and screenshots

No response

Additional Information

• GraphRAG Version: 2.6.0
• Operating System: Ubuntu 22
• Python Version: 3.12

[Robotuks]

Did some testing. Noticed that base_dir is not passed to .find() method here

By adding base_dir argument fixed my issue:

    files = list(
        storage.find(
            re.compile(config.file_pattern),
            base_dir=config.storage.base_dir,
            file_filter=config.file_filter,
        )
    )

[natoverse]

This is resolved on our v3 branch, with expected release to main and pypi in the next couple of weeks