Metadata-Version: 2.1
Name: apitist
Version: 1.0.0
Summary: Brand new way to test your API
Home-page: https://github.com/zeburek/python-apitist
Author: Parviz Khavari
Author-email: csctclan@gmail.com
License: gpl3
Description: [![Build Status](https://github.com/zeburek/python-apitist/workflows/Python%20package/badge.svg)](https://github.com/zeburek/python-apitist/actions) [![PyPI version](https://badge.fury.io/py/apitist.svg)](https://badge.fury.io/py/apitist) [![Downloads](https://pepy.tech/badge/apitist)](https://pepy.tech/project/apitist)
        
        # apitist
        
        Brand new way to test your API
        
        ## Installation
        
        Run the following command in your command line::
        ```bash
        pip install apitist
        ```
        
        ## Default hooks
        
          - RequestDebugLoggingHook
          - RequestInfoLoggingHook
          - PrepRequestDebugLoggingHook
          - PrepRequestInfoLoggingHook
          - ResponseDebugLoggingHook
          - ResponseInfoLoggingHook
        
        ### Example usage
        
        ```python
        from apitist.hooks import PrepRequestInfoLoggingHook, ResponseInfoLoggingHook
        from apitist.requests import session
        
        
        s = session()
        PrepRequestInfoLoggingHook.formatter = "Best formatter {req.method} {req.url}"
        
        s.add_prep_request_hook(PrepRequestInfoLoggingHook)
        s.add_response_hook(ResponseInfoLoggingHook)
        
        s.post("https://httpbin.org/post", params={"q": "test"})
        ```
        
        ## Custom Hooks
        
        ```python
        from requests import Request, PreparedRequest, Response
        
        from apitist.requests import session, RequestHook, PreparedRequestHook, ResponseHook
        
        s = session()
        
        class ReqHook(RequestHook):
        
            def run(self, request: Request) -> Request:
                print(request.url)
                return request
        
        class PrepReqHook(PreparedRequestHook):
        
            def run(self, request: PreparedRequest) -> PreparedRequest:
                print(request.url)
                return request
        
        
        class RespHook(ResponseHook):
        
            def run(self, response: Response) -> Response:
                print(response.url)
                return response
        
        s.add_request_hook(ReqHook)
        s.add_prep_request_hook(PrepReqHook)
        s.add_response_hook(RespHook)
        
        s.get("https://ya.ru", params={"q": "test"})
        ```
        
        ## Working with constructor
        
        ```python
        import attr
        import typing
        
        from apitist.constructor import converter
        from apitist.hooks import RequestConverterHook, ResponseConverterHook
        from apitist.requests import session
        
        
        class ExampleType:
            test = None
        
        @attr.s
        class ExampleStructure:
            test: ExampleType = attr.ib()
        
        @attr.s
        class TestResponse:
            args: typing.Dict = attr.ib()
            data: str = attr.ib()
            files: typing.Dict = attr.ib()
            form: typing.Dict = attr.ib()
            headers: typing.Dict = attr.ib()
            json: ExampleStructure = attr.ib()
            origin: str = attr.ib()
            url: str = attr.ib()
        
        s = session()
        s.add_hook(RequestConverterHook)
        s.add_hook(ResponseConverterHook)
        
        def structure_example_type(data, type_):
            example = ExampleType()
            example.test = data
            return example
        
        def unstructure_example_type(data):
            return data.test
        
        converter.register_hooks(
            ExampleType, structure_example_type, unstructure_example_type
        )
        
        t = ExampleType()
        t.test = "test"
        
        struc = ExampleStructure(t)
        
        res = s.post("https://httpbin.org/post", data=struc).structure(TestResponse)
        print(res.structured.json.test.test) # test
        ```
        
        ## Using random data generator
        
        First of all create an instance of random class:
        
        ```python
        from apitist.random import Randomer
        rand = Randomer()
        ```
        
        Now, you can add custom hooks for different types:
        
        ```python
        rand.add_type(str, lambda: str(random.random()))
        rand.add_type(float, lambda: random.random())
        ```
        
        Or using `add_types`:
        
        ```python
        types = {
            str: lambda: str(random.random()),
            float: lambda: random.random()
        }
        rand.add_types(types)
        ```
        
        Now you can create random object for given type or any attrs class with
        defined types:
        
        ```python
        import attr
        import typing
        
        rand.object(str) # '0.6147789314561384'
        rand.object(float) # 0.4664297665239271
        
        @attr.s
        class Data:
            value1: str = attr.ib()
            value2: typing.List[str] = attr.ib()
            value3: typing.Tuple[float] = attr.ib()
        
        print(rand.object(Data))
        # Data(
        #   value1='0.491058956716827',
        #   value2=['0.6568036485871975'],
        #   value3=(0.8603579349502298,)
        # )
        ```
        
        It is better to use it with [Faker](https://github.com/joke2k/faker).
        Just define different subclasses for `str` and add different hooks for them.
        By this you could create different data for different `str` fields.
        
        Also, using with `RequestConverterHook` and `ResponseConverterHook`
        you could easily create random json objects which would be send to server.
        
Platform: any
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: GNU General Public License v3 (GPLv3)
Classifier: Programming Language :: Python
Classifier: Programming Language :: Python :: 3.6
Classifier: Programming Language :: Python :: 3.7
Classifier: Programming Language :: Python :: 3.8
Classifier: Topic :: Software Development :: Quality Assurance
Classifier: Topic :: Software Development :: Testing
Classifier: Typing :: Typed
Requires-Python: >=3.6
Description-Content-Type: text/markdown; charset=UTF-8
Provides-Extra: testing
