How to change column to nullable with modify in Ecto migration

Sooner or later you'll have to change the null constraint in one of your DB relations. How to do it easily in Ecto?
Although I came across many different examples where a raw SQL has been used to perform this type of operation in Ecto, it's actually super easy to do it with modify/3 function.
Let's assume that your migration looks like this:
create table(:blog_posts) do
add :title, :string, null: false
add :intro, :text, null: false
add :body, :text, null: false
add :category_id, references(:blog_categories, on_delete: :delete_all)
end
... and at some point, you realize that you don't want to force passing the intro
column value.
You can change it easily this way:
alter table(:blog_posts) do
modify :intro, :text, null: true, from: :text
end
It's also worth mentioning that it's possible to modify foreign keys with modify:
alter table(:blog_posts) do
modify :category_id,
references(:blog_categories, on_delete: :delete_all),
null: false,
from: references(:blog_categories, on_delete: :delete_all)
end
Related posts
Dive deeper into this topic with these related posts
You might also like
Discover more content from this category
Each of us had a situation, where we had to invoke a few, same commands each time. Making it once is not that problematic, but when we are supposed to repeat given operations, it may be better just to create one function that will handle it all.
Did you ever create a commit that you wish never happened? Let's be honest - we all did. There is an easy way to revert it in Git.
Warnings in Elixir are usually an important sign of a problem in the codebase. There is an easy way to make them gone.