Table attribute can be removed when become empty after migration from annotations to attributes
yguedidi opened this issue ยท 8 comments
If Table annotation was using only indexes and uniqueConstraints, then after a migration to attributes it will become empty.
And an empty Table attribute is useless.
Feel free to contribute such rule ๐
Will try when I'll have time for sure! Issue opened more so that the idea is not forgotten :)
We keep mostly bugs and soon-to-be plans in here, to avoid issue creep :)
So I close it. Send whenever you're ready ๐
OK good to know, I understand.
Just a question about this one, about the philosophy, how it should be implemented?
is it:
- a dedicated removal rector?
- an improvement to annotation to attribute rector to support removing when empty?
- a post rector?
- something else I don't know about?
a dedicated removal rector?
That would be best, as it does not depend on another rule, but on empty state.
Thank you for the guidance
We keep mostly bugs and soon-to-be plans in here, to avoid issue creep
I'm curious about this statement, I'd like to understand the reasoning behind it if you have the time to answer.
Otherwise I'll stay with issue creep reason haha.
I'm questioning because a see a drawback doing it: there is no "ready to be taken backlog" of accepted ideas to be implemented, for new comers for example or who would like to train writing rectors.
Maybe GitHub Project could help here?
Hope I don't hurt you, I'm just curious :)
We used to have such backlog, but it attracted passer-by developers who came with idea, dropped it and never integrated the rule.
Instead, for the long-term project it's healthier to attract contributors, who open PR before creating issue and keep on improving the project ๐
I understand better, thank you! I'll try not to be such a passer-by developer ๐