Python Shell
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
schul_cloud_resources_server_tests
scripts
.gitignore
.travis.yml
Dockerfile
LICENSE
MANIFEST.in
README.rst
docker-compose.yml
requirements.in
requirements.txt
setup.py
tox.ini

README.rst

Schul-Cloud Resources Server Tests

Build Status Python Package Index Dockerhub Automated Build Status First Timers

This repository is a test and reference implementation for the Schul-Cloud-Resources API. The purpose is to

  • have a reference to verify the API specification is actually working
  • find uncertainties in the API specification
  • provide tests to speed up an implementation of the API on other servers such as schulcloud-content
  • have people contribute tests as a means of communication between applications: one test, X server-implementations
  • have an easy-to-setup server for crawler implementations and tests

This repository contains

  • a server to test scrapers against
  • tests to test the server

The package works under Python 2 and 3.

You can view the live docker instance at http://scrst.quelltext.eu/

Installation

Using pip, you can install all dependencies like this:

pip install --user schul_cloud_resources_server_tests

When you are done, you can import the package.

import schul_cloud_resources_server_tests

Installation for Development

If you want to work on the module, you need to install it differently. If you set it up for development, you can add new tests and features.

First, you need to have git installed. Then, clone the repository:

git clone https://github.com/schul-cloud/schul_cloud_resources_server_tests.git
cd schul_cloud_resources_server_tests

If you have pip installed, you can now install the packages:

pip install --user -r requirements.txt pip-tools==1.6.5

Now, as long as you are in the schul_cloud_resources_server_tests folder, you can run the commands as mentioned below. Additionally, you can use git to create pull-requests with your edited code.

Usage

This section describes how to use the server and the tests.

Server

You can find the API definition. The server serves according to the API. It verifies the input and output for correctness.

To start the server, run

python -m schul_cloud_resources_server_tests.app

The server should appear at http://localhost:8080/v1.

Tests

You always test against the running server. Tests may delete everyting you can reach. If you test the running server, make sure to authenticate in a way that does not destroy the data you want to keep.

python -m schul_cloud_resources_server_tests.tests --url=http://localhost:8080/v1/

http://localhost:8080/v1/ is the default url.

Steps for Implementation

If you want to implement your server you can follow the TDD steps to implement one test after the other.

python -m schul_cloud_resources_server_tests.tests -m step1
python -m schul_cloud_resources_server_tests.tests -m step2
python -m schul_cloud_resources_server_tests.tests -m step3
...
  • step1 runs the first test
  • step2 runs the first and the second test
  • step3 runs the first, second and third test
  • ...

You can run a single test with

python -m schul_cloud_resources_server_tests.tests -m step3only

Test Authentication

The test server supports api key authentication and basic authentication. If you test authentication over the internet. Use https to protect the secrets. Thus, an example test call to your api could look like this:

python -m schul_cloud_resources_server_tests.tests  \
       --url=https://url.to/your/server               \
       --noauth=false --basic=username:password

If you have an api key, you can test that the server works.

python -m schul_cloud_resources_server_tests.tests   \
       --url=http://url.to/your/server                \
       --noauth=false --apikey=apikey

By default the test server accepts authentication with several credentials

  • no authentication
  • basic:
    • user valid1@schul-cloud.org password 123abc
    • user valid2@schul-cloud.org password supersecure
  • api key: abcdefghijklmn for the user valid1@schul-cloud.org. The client does not send the user name to the server.

To test these, you can add the --basic and --apikey parameters several times to the tests. The --noauth=true parameter is default. If the api only accepts authenticated requests, set --noauth=false.

python -m schul_cloud_resources_server_tests.tests    \
       --basic=valid1@schul-cloud.org:123abc           \
       --basic=valid2@schul-cloud.org:supersecure      \
       --apikey=valid1@schul-cloud.org:abcdefghijklmn  \
       --noauth=true

All tests are run with the different authentication options. If we have several ways to authenticate, the tests test if the user sees the other users' data.

It is assumed, that adding invalid to the password, user name and api key will make it invalid. Tests use the invalid credentials to test the server behavior in rejected cases.

Example Travis Configuration

If you want to implement a crawler or server, you can use Travis-CI to test it. An example travis configuration can be found in the test-example branch. You can view the output The configuration is generic. It will run under any other language you configure.

Usage in Crawler

The url-crawler uses the test server to test synchronization.

Use the server in pytest

You can use the sever in Python tests. There are fixtures available that start and stop the server.

from schul_cloud_resources_server_tests.tests.fixtures import *

def test_pytest(resources_server):
    """pytest using the server"""

The following attributes are available:

  • resources_server.url The url of the server.
  • resources_server.api A schul_cloud_resources_api_v1.ResourcesApi object connected to the server.
  • resources_server.get_resources() A function to return a list of resources on the server.

For more information, see the module schul_cloud_resources_server_tests.tests.fixtures. You can add support for more test frameworks.

Docker

You can build the this image with the following docker command:

docker build -t schulcloud/schul_cloud_resources_server_tests .

Or you can pull the docker container and run it.

docker run schulcloud/schul_cloud_resources_server_tests

This starts the server at the port 8080 as in the examples above.

Docker-Compose

There is a docker-compose.yml file, so you can use the docker-compose command. The server will be available at http://localhost:80/v1 It uses a memory limit of 60MB.


You can edit this document on Github and check it with this editor.