S3: zip archives
bemoody opened this issue · comments
We have to decide whether and how we are going to upload zip archives to S3.
Personally I would rather fix issue #1674 first, but that may take some time and it would probably be wiser to upload the zip files in their current imperfect state.
Currently, pull #2086 would upload the archive for mitdb/1.0.0 as s3://XXXXXX/mitdb/mit-bih-arrhythmia-database-1.0.0.zip and the files in that archive would be named mit-bih-arrhythmia-database-1.0.0/100.hea etc.
Preferably the archive would be named s3://XXXXXX/mitdb/mitdb-1.0.0.zip and a file inside the archive would be named mitdb-1.0.0/100.hea.
Should we:
-
keep the file as
s3://XXXXXX/mitdb/mit-bih-arrhythmia-database-1.0.0.zip, plan to remove and replace it withs3://XXXXXX/mitdb/mitdb-1.0.0.zipeventually? -
upload the existing file as
s3://XXXXXX/mitdb/mitdb-1.0.0.zip, plan to replace it eventually with a different version at the same URL? -
not upload zip files to S3 at all until issue #1674 is resolved?
not upload zip files to S3 at all until issue #1674 is resolved?
This is tempting, but as I don't want to be responsible for holding @Chrystinne up, I think the best solution is:
upload the existing file as s3://XXXXXX/mitdb/mitdb-1.0.0.zip, plan to replace it eventually with a different version at the same URL?