When we insert a record in PostgreSQL and we specify the ID, the
internal ID sequence for that table isn't updated.
In order to keep the original IDs so we didn't break any foreign keys,
we specified the IDs when copying the table, resulting in a table having
its ID sequence with a value of an existing record. When trying to
insert a new record, we got a `PG::UniqueViolation` exception.
Updating the sequence after the data migration might not be the most
elegant solution, but it's easy to do and it's already been tested on a
production environment.
Calling `load_tasks` in several files made rails load the tasks several
times, and so they were executed several times when called.
Since the milestone migration can't be executed twice in a row (it would
fail with duplicated ID records), loading the tasks several times made
the milestone migrations task specs fail.
Ruby can't have hyphens in method names, so sending something like
`record.title_pt-BR` would raise an exception.
Using globalize's `localized_attr_name_for` method fixes the bug.
Thanks Marko for the tip.
We think aborting the migration will generate more headaches to system
administrators, who will have to manually check and fix every invalid
record before anything can be migrated.
The db:dev_seed rake logs info as it progresses as information for the
developer. But that's not needed when ran from its tests file, and it
bloats the travis/rspec output with unnecessary information.
Now the task will always log info unless the rake task receives an
optional argument.
Why:
UserSegments are not only used for Newsletters or Email downloads, but
also for internal Global Notifications. We don't want to have that scope
hardcoded inside UserSegments as users that have opted-out from the
newsletter should still be recipients of global notifications.
How:
Removing the scope from the UserSegments `all_users` method that acts as
base for all the other segments. Including that `newsletter` scope only
on the places that is relevant:
* When listing recipients for a newsletter
* When downloading a listing emails that can be newsletter recipients
Also updated relevant tests
This cop tries to avoid state leaking between examples. The fixes done on the code don't follow the suggested path, since the usage of `before(:all)` was not really useful. By using RSpec's `let` method we achieve same goals but with much better and readable tests.
Check Cop description at http://www.rubydoc.info/gems/rubocop-rspec/RuboCop/Cop/RSpec/BeforeAfterAll
