Lines of code count not available
Probably functional tests will fail because of missing orderables like ARV0001, Male Condom, IPV - 5 dose,injection and some other. You can find them here: https://github.com/OpenLMIS/openlmis-functional-tests/search?l=Gherkin&q=product. I'm not sure what should we do - adjust tests or restore those orderables.
I have not changed products that have a connection to commodity types or trade items. I have tried to not change products we use for testing but I will ask our QA's to check this.
Yes. The table can contain a lot of data so adding columns and set default values could drop performance
Is it because dropping and recreating is faster?
I wanted to separate loading csv files from services (as from demo-data directory in each service) from those which are in the demo-data repository. Just for the sake of clarity.
I don't think we should be settings this in the constructor. I might want to use this constructor in retrieval endpoints, for example Isn't doing this in PrePersist/Update annotated method enough?
Why do we need to keep it in a separate file? (as opposed to eg. load_data.sql)
In the following review, I only added a new field to the FTAP resource. I will adjust endpoints in next commits and reviews. Stay tuned.
I didn't find any place where we are comparing dates from the database to the current timestamp, but I replaced hardcoded zone with SQL function NOW().
could check how do we handle such cases in other parts of the system?
Got a suggestion or an issue?
Discuss it on Atlassian Answers or
visit Atlassian Support for other options.