oehlschl

Just for the record, another solution I've seen encouraged for this problem is to set something in the Rails/request env in the controller and check it in the Strategy. Ex:...

I can also submit a PR if needed.

Spec to replicate in #20 FYI also @krainboltgreene

From a quick glance, any record that's skipped over here: https://github.com/godaddy/activerecord-delay_touching/blob/e71ade4ae4280859df8021fa0e56ef79b2ac02fd/lib/activerecord/delay_touching.rb#L90 Should not updated in the query here: https://github.com/godaddy/activerecord-delay_touching/blob/e71ade4ae4280859df8021fa0e56ef79b2ac02fd/lib/activerecord/delay_touching.rb#L98

It looks like ` ::RestClient::BadRequest` is also rescued twice (without being reraised), on both https://github.com/goshippo/shippo-ruby-client/blob/a36a024a302c83fc8d93166ec2c5817d6e40a28f/lib/shippo/api/request.rb#L62 and https://github.com/goshippo/shippo-ruby-client/blob/a36a024a302c83fc8d93166ec2c5817d6e40a28f/lib/shippo/api/request.rb#L72

It actually looks like InvalidInputError doesn't exist, possibly missed because the rescue code path is never hit. https://github.com/goshippo/shippo-ruby-client/search?utf8=%E2%9C%93&q=InvalidInputError&type=

Thanks @isnotajoke! Came across this while also looking for Resque optimizations and it seems like it could be helpful for us as well. I'm happy to pilot this branch for...

Just confirming that this is still happening in at least Resque `1.27.4`. At minimum it would be worth identifying a way to have these works fail loudly on exit.

We've been using the following pattern: ``` # for state machine on attribute 'status' and some event_name my_model.update_attributes!(status_event: event_name) ``` This raises an exception, essentially triggering the bang version of...

Still looking for the implementation, but there's some documentation here: https://github.com/pluginaweek/state_machine#explicit-vs-implicit-event-transitions