ignorant alternatives and similar packages
Based on the "Testing" category.
Alternatively, view ignorant alternatives based on common mentions on social networks and blogs.
-
bypass
Bypass provides a quick way to create a custom plug that can be put in place instead of an actual HTTP server to return prebaked responses to client requests. -
mix_test_interactive
Interactive watch mode for Elixir's mix test. https://hexdocs.pm/mix_test_interactive/ -
mecks_unit
A simple Elixir package to elegantly mock module functions within (asynchronous) ExUnit tests using Erlang's :meck library
CodeRabbit: AI Code Reviews for Developers
Do you think we are missing an alternative of ignorant or a related project?
Popular Comparisons
README
Ignorant
Because sometimes ignorance is bliss.
Ignorant defines a protocol for data structures that may selectively ignore specific values they contain, usually to simplify partial comparison in tests.
Motivation
Sometimes it's useful to compare data structures in tests while ignoring pieces of data that are not deterministic - auto-generated primary keys, calculated timestamps, and so on. Ignoring certain bits ensures the shape of the data is correct when it's not possible to enforce its values.
Example
An implementation for Map
is provided with that package for the purpose described above. Suppose
you have an API endpoint that fetches a record from the database and returns JSON data which
translates to this response
map:
%{
id: 37,
name: "Jim"
}
Now let's say you want to write a test to ensure that it returns the right data. Problem is, the
id
field is auto-generated and therefore has no deterministic value between test runs. You could
use pattern matching...
assert %{id: _, name: "Jim"} = response
...but you can't assign the value on the left side to a variable (or module attribute) and reuse it,
and if the match fails you don't get a nice diff message. That's where Ignorant
comes in:
assert %{id: :ignored, name: "Jim"} == Ignorant.ignore(response, [:id])
The left side is now a proper value that can be put in a variable or attribute and reused. Also note
the ==
(equality) assertion instead of =
(match) assertion. That gives us a pretty diff that
highlights exactly what doesn't look the same on both sides, and is also stricter (which is a good
idea in tests).
It quickly becomes annoying and error-prone to tag things as :ignored
in one side of the
comparison and include the corresponding field on the second parameter in the call to ignore/2
on
the other side, so we can clean that up by using merge_ignored/2
:
expected = %{id: :ignored, name: "Jim"}
assert expected == Ignorant.merge_ignored(response, expected)
merge_ignored/2
will walk through the expected
map extracting all fields tagged as :ignored
, and
then apply those to response
while keeping all other values intact. The resulting map should be
equal to expected
. If not, you'll get a pretty diff explaining what's different.
Installation
- Add
ignorant
to your list of dependencies inmix.exs
:
def deps do
[{:ignorant, "~> 0.1.0"}]
end
- Ensure
ignorant
is started before your application:
def application do
[applications: [:ignorant]]
end