Open
Description
We want to detach the API from expecting to be on a machine with access to all of CPAN.
This will make it more portable.
Considerations
- how we create index jobs in the first place - we have the
backpan syncer
server which has log files that know when files are update? - timing of availability when indexing... the package file will be ok as it will be a new URL - but if we are fetching
modules/06perms.txt
over HTTPS then we may want a mechanism to ensure it is not cached... (we could create a special URL for this?) but should also consider the timing of when that file gets updated could be a few mins after the package .tar.gz is available
Probably worth moving as many pieces as possible to fetching via http and can then see what is left and have a discussion?
Metadata
Metadata
Assignees
Labels
No labels