Ability to auto-generate data based on model spec

[thibaudcolas]

In GitLab by @todd-dembrey on Jan 15, 2018, 12:14

Instead of specifying the data within a yaml file, have the ability to auto spec the data based on the model field definitions similar to how factory boy works.

A constant seed could be used as required to prevent rapidly changing text between refreshes. But the ability to generate long/short content for testing and visual validation would be useful.

Something like faker could be used to map field type -> default faker class and custom field definitions could also be used for properties etc.
https://github.com/joke2k/faker

[thibaudcolas]

In GitLab by @bcdickinson on Jan 24, 2018, 16:01

I'm not sure about this, I think it's not really a pattern library's job to know anything about your database schema. I could definitely get behind some YAML tags for lorem ipsum and some of the common faker providers, though.

[thibaudcolas]

@bcdickinson this feels related/similar to the discussion we had about writing stories in Python:

Investigate creating stories in Python, akin to unit tests (w/ DB awareness & factories)

[thibaudcolas]

Closing this in favour of #106.