Removal of 20 second delay in overwriting index

[chris1ance]

Hello,

When I run

RAG = RAGPretrainedModel.from_pretrained("colbert-ir/colbertv2.0")
processor = CorpusProcessor()
my_documents = processor.process_corpus(segments)
index_path = RAG.index(index_name="my_index", collection=my_documents, split_documents=True, max_document_length = 256*4)

where segments is a list of strings, if my_index already exists, I get the message:

Jan 18, 17:36:43] #> Note: Output directory .ragatouille/colbert/indexes/my_index already exists


[Jan 18, 17:36:43] #> Will delete 10 files already at .ragatouille/colbert/indexes/my_index in 20 seconds...

I was wondering if I could eliminate this 20 second delay?

[bclavie]

Hey, there's actually an argument that can be used internally to bypass the 20 seconds delay, but it's not currently exposed because I wanted to avoid people accidentally deleting indices! If this is useful, I'll make it a passable arg in an upcoming release!

[chris1ance]

Thanks for your reply! Makes sense. I would find it useful to be able to switch it on/off, or alternatively to set the number of seconds myself.

[bclavie]

Done in 0.0.6a1! You can pass overwrite_index="force_silent_overwrite" to an index() call to immediately (&permanently) delete any existing index at the target path.

[chris1ance]

Thank you!