llama_index icon indicating copy to clipboard operation
llama_index copied to clipboard

Way to exclude directories and files using SimpleDirectoryReader

Open ShakesB33r opened this issue 1 year ago • 1 comments

Sometimes directories contain both relevant and irrelevant subdirectories and/or files, it would be nice to have a keyword argument for excluding them, perhaps even with some matching syntax, e.g.

documents = SimpleDirectoryReader('...', recursive=True, exclude=["*/misc/*, "note.txt"])

ShakesB33r avatar Mar 09 '23 11:03 ShakesB33r

nice feature

Minweiwangaaaa avatar Mar 10 '23 08:03 Minweiwangaaaa