docker run -p 80:8000 -d zellerede/mockserver

docker run -p 443:443 -e HTTPS=True -d zellerede/mockserver

pip install django djangorestframework django-url-filter django-sslserver
git clone https://github.com/zellerede/mockserver.git
cd mockserver

Create/adjust the SQlite3 database by the migrate command, then run unittests to see if all works as intended:

python manage.py migrate
python manage.py test

python manage.py runserver

or, for external reachability,

python manage.py runserver 0.0.0.0:8000

python manage.py runsslserver 0.0.0.0:443

For the following examples, set the environment variable MS to the actual base URL where Mock Server is listening. E.g.

export MS="localhost:8000"

For HTTPS, as the certificates are self signed, the curl commands require additional -k option. So that, in that case, set

export MS="-k https://localhost" 

Check out url localhost:8000/__mock/ in your browser. (Use the actual address and port of the running Mock Server.)

You can post a new mock answer record there by typing in the values for the fields. All fields are optional, though the url should be given.

Note: All answer bodies must be in valid json format.

The mock answers preset this way can also be tested in the browser: Say, a record with url: abc/ and ans_body: "success" was posted, then getting the path /abc/ should show the json string "success" as result.

curl $MS/__mock/
curl -v $MS/abc/
curl -X POST -d "url=abc/&ans_body=\"success\"" $MS/__mock/
curl -v $MS/abc/

Note that records can also be posted/patched in json format, using header parameter -H "content-type: application/json" with curl.

The mock answer records also remember the query parameters and request body of the last handled request. The id of a record is returned upon creation. After its mock happened, curl $MS/__mock/id will return with the details of the actual request that the mock server received.

The records can also be searched by url and req_method, e.g. curl $MS/__mock/?url=abc/&req_method=PATCH.

Bulk creation, deletion and partial update are supported, and can be achieved either on webpage, or by curl of /__mock/bulk/, using possibly filters for fields url and req_method.

 curl -X POST -d "@example.json" -H "content-type: application/json" $MS/__mock/bulk/
 curl -X DELETE $MS/__mock/bulk/
 curl -X DELETE $MS/__mock/bulk/?req_method=POST
 curl -X PATCH -d "ans_status=402" $MS/__mock/bulk/
 curl -X PATCH -d "ans_status=500" $MS/__mock/bulk/?url__icontains=error

The field use_up can be set to an integer to indicate how many times the answer given in the record should be used. When set, use_up will simply decrease until reaches 0 when the record will be deleted and a next record with same url+method can act.

curl -X POST -d "url=poll/&ans_status=404&use_up=2" $MS/__mock/
curl -X POST -d "url=poll/&ans_body=\"appeared\"&use_up=1" $MS/__mock/
curl -X POST -d "url=poll/&ans_body=\"done\"" $MS/__mock/
curl -v $MS/poll/   # 404
curl -v $MS/poll/   # 404
curl -v $MS/poll/   # appeared
curl -v $MS/poll/   # done
curl -v $MS/poll/   # done

We can create new MockAnswer records also using YAML format. It's practical, since it can display the json inside the data in a more readable way. Use the following command and study the example yaml files.

./manage.py loadyaml example.yaml example2.yaml

The mock server is assumed to run in the background when test execution starts. Import the library as

Resource  |  path/to/mockserver/robot_keywords/mock_server.robot

The following robot keywords are implemented:

Init Mockserver  |  [json file]  |  [json file2]  |  [...]

-- clears all MockAnswer records and loads the given json fixtures if given.

Delete All MockAnswers

-- clears all MockAnswer records.

Load MockAnswers  |  YAML/JSON  |  file   |  [file2]  |  [...]

-- loads MockAnswer records from the given json/yaml file(s).

Prepare MockAnswer  |  url=<url>  |  [ans_status=<status>]    [ans_body=<ans_body>]  |  [...]

-- creates a MockAnswer record based on the given parameters.