In this example, we implement a very basic test on a Github API endpoint:
# suites/github.py import lemoncheesecake.api as lcc from lemoncheesecake.matching import * from lemoncheesecake_requests import Session, is_2xx @lcc.test() def get_org(): session = Session(base_url="https://api.github.com") resp = session.get("/orgs/lemoncheesecake") resp.require_status_code(is_2xx()) check_that_in( resp.json(), "id", is_integer(), "name", equal_to("lemoncheesecake") )
We run the test:
$ lcc run =================================== github ==================================== OK 1 # github.get_org Statistics : * Duration: 0.214s * Tests: 1 * Successes: 1 (100%) * Failures: 0
And here are the report details :
Install through pip:
$ pip install lemoncheesecake-requests
lemoncheesecake-requests is compatible with Python 3.7-3.10.
This logger which can be set session-wide (it is set to
session = Session(base_url="https://api.github.com", logger=Logger.no_headers()) session.get("/orgs/lemoncheesecake")
or overridden on a per request basis:
session = Session(base_url="https://api.github.com") session.get("/orgs/lemoncheesecake", logger=Logger.no_headers())
As you might guess from the previous examples, the logger can be fine tuned to control exactly what HTTP request/response details that will be logged and how through the following logger boolean attributes:
If you want for instance to create a logger that only logs data coming from the response:
logger = Logger( request_line_logging=False request_headers_logging=False request_body_logging=False )
and then pass this logger to a session or to a specific HTTP method call.
debug attribute controls whether or not logging is done using the “debug” level or the default “info” level.
Please note that with lemoncheesecake 1.10.0 and higher debug logs are hidden by default in HTML reports and
lcc report (with an option to show them) which make them convenient to log less-important data without overloading the report.
lemoncheesecake_requests.Logger class also provide class methods to easily create instances for common
HTTP request bodies and especially response bodies might be very large and make the final report unreadable.
That’s why the logger will log the request/response bodies as attachment if their (serialized) content size
exceed a certain size. This size can be configured through the
max_inlined_body_size logger attribute.
The various request-methods (
post(), etc…) of
lemoncheesecake_requests.Session return a
lemoncheesecake_requests.Response which extends
requests.Response and provides several
extra methods to check the response status code.
As you may already know, lemoncheesecake provides three different way (for three different behaviors) to perform a matching operation:
lemoncheesecake_requests.Response follows the same logic by offering the corresponding three methods:
resp.check_status_code(200) resp.check_status_code(equal_to(200)) resp.check_status_code(any_of(200, 201))
There is also an alternative way to check status code, that is not built uppon lemoncheesecake’s logging facilities:
StatusCodeMismatch exception if the condition is not met
and then interrupt the test (unless the exception is explicitly caught):
(this function can also take a
Matcher instance as argument).
All these methods have a corresponding shortcut method that directly check a 2xx status code:
It means that those two method calls are equivalent:
Please note that all these extra methods return the
Response instance itself, meaning that they can be chained like
resp = session.get("/orgs/lemoncheesecake").require_ok()
See the API Reference for full details about the lemoncheesecake-requests API.
The Changelog will tell you about features, improvements and fixes of each version.