graphrag icon indicating copy to clipboard operation
graphrag copied to clipboard

text unit sources are not deduplicated

Open volksen opened this issue 1 year ago • 1 comments

Hello,

when creating the text unit sources in the LocalSearchMixedContext, the sources are not deduplicated:

While there is a text_unit_ids_set in the code, it is never used: Line 335 of mixed_context.py

I fixed it by adding: text_unit_ids_set.add(text_id)

in line 337.

volksen avatar Dec 04 '24 13:12 volksen

tried to create a PR for the issue: https://github.com/microsoft/graphrag/pull/1468 My first PR, please be kind :)

volksen avatar Dec 04 '24 14:12 volksen