You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
A cool setup for dumb-pypi would be to have it run in a lambda function triggered by changes to an S3 bucket (e.g. new object uploaded) and then export the archive into a new S3 bucket. This is neat because there's very little infrastructure to maintain and will have very high uptime (assuming you serve from S3 directly).
It's probably pretty easy to use dumb-pypi already in lambda, but it'd be nice to document how to do it.
Note that if you need to support old pip (<8.1.2), then you still won't be able to serve the index directly from S3 because the webserver needs to do some normalization of package names. However, your webserver could be as simple as proxy_pass to the right S3 URL, with the redirect trick in the README.
The text was updated successfully, but these errors were encountered:
Lambda supports python3.6 now!
A cool setup for dumb-pypi would be to have it run in a lambda function triggered by changes to an S3 bucket (e.g. new object uploaded) and then export the archive into a new S3 bucket. This is neat because there's very little infrastructure to maintain and will have very high uptime (assuming you serve from S3 directly).
It's probably pretty easy to use dumb-pypi already in lambda, but it'd be nice to document how to do it.
Note that if you need to support old pip (<8.1.2), then you still won't be able to serve the index directly from S3 because the webserver needs to do some normalization of package names. However, your webserver could be as simple as
proxy_pass
to the right S3 URL, with the redirect trick in the README.The text was updated successfully, but these errors were encountered: