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
It's useful to run CPAN Testers processes locally for testing purposes. But, to make useful tests, we need real data to work with.
For example, in order for the backend processes to run, we need the uploads table. It's too onerous to require the user to mirror CPAN and run the upload backend process. To test some backend processes, we would need some rows from the test_reports table and the uploads table.
We should have a method on the schema object, fetch, that syncs data from a remote API server. It should take a list of arguments of API path parts and an optional hashref of query parameters for that API. It should make the request, and be able to map the data that comes back from the API into the correct CPAN::Testers::Schema ResultSet class to write to the local database.
This method should be able to fetch:
upload - The CPAN Uploads table
report - Test reports
summary - Test report summaries
release - Test report releases
There should be an option to automatically fulfill dependencies in the data: summary rows require report and upload rows. release rows require summary rows (and all summary row dependencies).
It would be nice if we could search using MetaCPAN's database and then fill our database with the distributions we find. This would make it easy to, for example, create a local database full of the test reports for the current releases of modules high in the CPAN river.
The text was updated successfully, but these errors were encountered:
This is useful for development instances to give them some real data for
testing purposes.
This does not yet download full reports. That should be completed before
the ticket is considered done.
Refs #20
It's useful to run CPAN Testers processes locally for testing purposes. But, to make useful tests, we need real data to work with.
For example, in order for the backend processes to run, we need the uploads table. It's too onerous to require the user to mirror CPAN and run the upload backend process. To test some backend processes, we would need some rows from the test_reports table and the uploads table.
We should have a method on the schema object, fetch, that syncs data from a remote API server. It should take a list of arguments of API path parts and an optional hashref of query parameters for that API. It should make the request, and be able to map the data that comes back from the API into the correct CPAN::Testers::Schema ResultSet class to write to the local database.
This method should be able to fetch:
upload
- The CPAN Uploads tablereport
- Test reportssummary
- Test report summariesrelease
- Test report releasesThere should be an option to automatically fulfill dependencies in the data:
summary
rows requirereport
andupload
rows.release
rows requiresummary
rows (and all summary row dependencies).It would be nice if we could search using MetaCPAN's database and then fill our database with the distributions we find. This would make it easy to, for example, create a local database full of the test reports for the current releases of modules high in the CPAN river.
The text was updated successfully, but these errors were encountered: