Unify logic for disk and blob size / hash computation

codalab / codalab-worksheets

A collaborative platform for reproducible research (web interface and CLI).

Other

154 stars 82 forks source link

Open epicfaace opened 1 year ago

epicfaace commented 1 year ago

Unify logic for disk and blob size / hash computation for bundles. Currently, we use:

specific logic for disk that recursively looks at files in a directory -- https://github.com/codalab/codalab-worksheets/blob/master/codalab/model/bundle_model.py#L1097
other logic for blob

We should make this logic simpler / more efficient and potentially try to unify both code paths.

epicfaace commented 1 year ago

Other question: should data_size be compressed or uncompressed size?

pranavjain commented 1 month ago

@epicfaace @dma1dma1 Do we still this issue open?