HUnit is a command-line tool that runs tests against HTTP APIs.
Since tests and documentation are maintained in parallel, HUnit combines these two highly-related tasks into one. Descriptions can be added to your tests and documentation of your endpoints can be automatically generated.
Tests can be run by pointing hunit to a test suite document.
$ hunit test.yml
====> test.yml
----> GET https://raw.githubusercontent.com/bww/hunit/master/example/test.txt
#1 Unexpected status code:
expected: (int) 404
actual: (int) 200
#2 Entities do not match:
--- Expected
+++ Actual
@@ -1,2 +1,2 @@
-Heres a simple
+Here's a simple
response from the
When needed, documentation for the endpoints in your tests can be generated by passing the -gendoc flag. In this case tests are run and their input and output are automatically incorporated into documentation.
Refer to example/docs for an example of the docs generated for the example tests.
$ hunit -gendoc test.yml
====> test.yml
# ...
Tests are described by a simple YAML-based document format. Tests and documentation are managed together.
The test suite document format is straightforward and highly readible. Advanced features are also available if you need them, like:
- Comparing entities semantically to ignore insignificant differences like whitespace,
- Referencing the output of previously-run tests from the currently-executing one,
- Some useful functions which can be invoked in string interpolation.
Refer to the full test.yml file for a more complete illustration of test cases.
-
doc: |
Fetch a document from our [Github repo](github.com/bww/hunit).
request:
method: GET
url: https://raw.githubusercontent.com/bww/hunit/master/example/test.txt
headers:
Origin: localhost
response:
status: 200
headers:
Content-Type: text/plain; charset=utf-8
entity: |
Heres a simple
response from the
server.