- Sep 22, 2020
-
-
Jean-Paul Calderone authored
Do a better job generating sensible error results from the payment endpoint
-
Jean-Paul Calderone authored
-
Jean-Paul Calderone authored
Stripe CardError is now the only error to receive a "Stripe charge didn't succeed" response. Other Stripe errors can now fail in other ways. This is a better client interface and makes HTTP status codes more accurately reflect the failure which makes them more usable as a monitorable metric.
-
Jean-Paul Calderone authored
Add a metric for signatures issued
-
Jean-Paul Calderone authored
-
Jean-Paul Calderone authored
-
Jean-Paul Calderone authored
Record first successful redemption of any voucher, counter pair
-
Jean-Paul Calderone authored
So, let it. Just demonstrate we at least see the test metric.
-
- Sep 21, 2020
-
-
Jean-Paul Calderone authored
-
Jean-Paul Calderone authored
Add general HTTP metrics and particular Stripe charge metrics
-
Jean-Paul Calderone authored
-
Jean-Paul Calderone authored
-
Jean-Paul Calderone authored
-
- Sep 18, 2020
-
-
Jean-Paul Calderone authored
-
Jean-Paul Calderone authored
-
Jean-Paul Calderone authored
-
- Sep 17, 2020
-
-
Jean-Paul Calderone authored
-
Jean-Paul Calderone authored
-
Jean-Paul Calderone authored
It was passing but only after ~57 seconds because the busy_timeout setting in the test definition was being applied to a different connection than was being used by the implementation. With this change we hook into the connection function and apply the shorter busy_timeout after the normal setup work is done but before handing the connection off to application code.
-
- Sep 11, 2020
-
-
Jean-Paul Calderone authored
Have fewer SQLite3 "busy" errors
-
Jean-Paul Calderone authored
-
Jean-Paul Calderone authored
-
Jean-Paul Calderone authored
-
Jean-Paul Calderone authored
Handle exceptions during SQLite3 schema initialization and close the connection if necessary
-
Jean-Paul Calderone authored
-
Jean-Paul Calderone authored
Speed up the blinded token fingerprinting function
-
- Sep 10, 2020
-
-
Jean-Paul Calderone authored
-
Jean-Paul Calderone authored
-
Jean-Paul Calderone authored
-
Jean-Paul Calderone authored
-
Jean-Paul Calderone authored
-
Jean-Paul Calderone authored
It seems to be faster to hash one huge string than many smaller ones that total to the same length. Also by only showing one final hash we remove O(N) show calls. The old and new fingerprints are not compatible but I haven't deployed this in production anywhere yet so that doesn't bother me. Had I, we'd need some kind of versioning scheme for fingerprint computation I suppose.
-
- Jun 03, 2020
-
-
Jean-Paul Calderone authored
Delay redemption response if it would be a "not paid" error Internally try repeatedly to try to be able to return a success response.
-
Jean-Paul Calderone authored
-
Jean-Paul Calderone authored
I should really learn how to do some real logging...
-
Jean-Paul Calderone authored
-
Jean-Paul Calderone authored
-
Jean-Paul Calderone authored
-
Jean-Paul Calderone authored
-
- May 30, 2020
-
-
Jean-Paul Calderone authored
Fix intermittent "Database is locked" errors
-