Merge branch 'master' of github.com:miLibris/flask-rest-jsonapi into develop

Author: akira-dev

README.rst

@@ -11,14 +11,14 @@
 Flask-REST-JSONAPI
 ##################
 
-Flask-REST-JSONAPI is a flask extension to build REST api. It combines the power of `Flask-Restless <https://flask-restless.readthedocs.io/en/stable/>`_ and the flexibility of `Flask-RESTful <http://flask-restful-cn.readthedocs.io/en/0.3.5/a>`_ around a strong specification `JSONAPI 1.0 <http://jsonapi.org/>`_. This framework is design to quickly build REST api and to fit the complexity of real life projects with legacy data and multiple data storages.
+Flask-REST-JSONAPI is a flask extension for building REST APIs. It combines the power of `Flask-Restless <https://flask-restless.readthedocs.io/en/stable/>`_ and the flexibility of `Flask-RESTful <http://flask-restful-cn.readthedocs.io/en/0.3.5/a>`_ around a strong specification `JSONAPI 1.0 <http://jsonapi.org/>`_. This framework is designed to quickly build REST APIs and fit the complexity of real life projects with legacy data and multiple data storages.
 
 Install
 =======
 
     pip install Flask-REST-JSONAPI
 
-A minimal api
+A minimal API
 =============
 
 .. code-block:: python
@@ -67,19 +67,19 @@ A minimal api
         data_layer = {'session': db.session,
                       'model': Person}
 
-    # Create the api
+    # Create the API object
     api = Api(app)
     api.route(PersonList, 'person_list', '/persons')
     api.route(PersonDetail, 'person_detail', '/persons/<int:id>')
 
-    # start the flask loop
+    # Start the flask loop
     if __name__ == '__main__':
         app.run()
 
-This example provides this api:
+This example provides the following API structure:
 
 ========================  ======  =============  ===========================
-url                       method  endpoint       Usage
+URL                       method  endpoint       Usage
 ========================  ======  =============  ===========================
 /persons                  GET     person_list    Get a collection of persons
 /persons                  POST    person_list    Create a person
@@ -91,18 +91,18 @@ url                       method  endpoint       Usage
 Flask-REST-JSONAPI vs `Flask-RESTful <http://flask-restful-cn.readthedocs.io/en/0.3.5/a>`_
 ==========================================================================================
 
-* Instead of Flask-RESTful, Flask-REST-JSONAPI provides a default implementation of get, post, patch and delete methods around a strong specification JSONAPI 1.0. Thanks to this you can build REST api very quickly.
-* Flask-REST-JSONAPI is as flexible as Flask-RESTful. You can rewrite every default methods implementation to make custom work like distributing object creation.
+* Instead of Flask-RESTful, Flask-REST-JSONAPI provides a default implementation of get, post, patch and delete methods around a strong specification JSONAPI 1.0. Thanks to this you can build REST API very quickly.
+* Flask-REST-JSONAPI is as flexible as Flask-RESTful. You can rewrite every default method implementation to make custom work like distributing object creation.
 
 Flask-REST-JSONAPI vs `Flask-Restless <https://flask-restless.readthedocs.io/en/stable/>`_
 ==========================================================================================
 
-* Flask-REST-JSONAPI is a real implementation of JSONAPI 1.0 specification. So instead of Flask-Restless, Flask-REST-JSONAPI force you to create a real logical abstration over your data models with `Marshmallow <https://marshmallow.readthedocs.io/en/latest/>`_. So you can create complex resource over your data.
-* Instead of Flask-Restless, Flask-REST-JSONAPI can use any ORM or data storage through the data layer concept, not only `SQLAlchemy <http://www.sqlalchemy.org/>`_. A data layer is a CRUD interface between your resource and one or more data storage so you can fetch data from any data storage of your choice or create resource that use multiple data storage.
-* Like i said previously, Flask-REST-JSONAPI is a real implementation of JSONAPI 1.0 specification. So instead of Flask-Restless you can manage relationships via REST. You can create dedicated url to create a CRUD api to manage relationships.
+* Flask-REST-JSONAPI is a real implementation of JSONAPI 1.0 specification. So instead of Flask-Restless, Flask-REST-JSONAPI forces you to create a real logical abstration over your data models with `Marshmallow <https://marshmallow.readthedocs.io/en/latest/>`_. So you can create complex resource over your data.
+* Instead of Flask-Restless, Flask-REST-JSONAPI can use any ORM or data storage through the data layer concept, not only `SQLAlchemy <http://www.sqlalchemy.org/>`_. A data layer is a CRUD interface between your resource and one or more data storage so you can fetch data from any data storage of your choice or create resource that use multiple data storages.
+* Like I said previously, Flask-REST-JSONAPI is a real implementation of JSONAPI 1.0 specification. So instead of Flask-Restless you can manage relationships via REST. You can create dedicated URL to create a CRUD API to manage relationships.
 * Plus Flask-REST-JSONAPI helps you to design your application with strong separation between resource definition (schemas), resource management (resource class) and route definition to get a great organization of your source code.
-* Instead of Flask-Restless, Flask-REST-JSONAPI is highly customizable. For example you can entirely customize you urls, define multiple urls for the same resource manager,control serialization parameters of each methods and lot of very useful parameters.
-* Finally instead of Flask-Restless, Flask-REST-JSONAPI provides a great error handling system according to JSONAPI 1.0. Plus the exception handling system really helps the api developer to quickly find missing resources requirements.
+* Instead of Flask-Restless, Flask-REST-JSONAPI is highly customizable. For example you can entirely customize your URLs, define multiple URLs for the same resource manager, control serialization parameters of each method and lots of very useful parameters.
+* Finally instead of Flask-Restless, Flask-REST-JSONAPI provides a great error handling system according to JSONAPI 1.0. Plus the exception handling system really helps the API developer to quickly find missing resources requirements.
 
 Documentation
 =============
@@ -112,4 +112,4 @@ Documentation available here: http://flask-rest-jsonapi.readthedocs.io/en/latest
 Thanks
 ======
 
-Flask, marshmallow, marshmallow_jsonapi, sqlalchemy, Flask-RESTful and Flask-Restless are awesome project. These librairies give me inspiration to create Flask-REST-JSONAPI so huge thanks to authors and contributors.
+Flask, marshmallow, marshmallow_jsonapi, sqlalchemy, Flask-RESTful and Flask-Restless are awesome projects. These libraries gave me inspiration to create Flask-REST-JSONAPI, so huge thanks to authors and contributors.

docs/index.rst

@@ -3,7 +3,7 @@ Flask-REST-JSONAPI
 
 .. module:: flask_rest_jsonapi
 
-**Flask-REST-JSONAPI** is an extension for Flask that adds support for quickly building REST APIs with hudge flexibility around JSONAPI 1.0 specification. It is design to fit the complexity of real life environnement so Flask-REST-JSONAPI helps you to create a logicial abstraction of your data called "resource" and can interface any kind of ORMs or data storage through data layer concept.
+**Flask-REST-JSONAPI** is an extension for Flask that adds support for quickly building REST APIs with huge flexibility around the JSONAPI 1.0 specification. It is designed to fit the complexity of real life environments so Flask-REST-JSONAPI helps you to create a logical abstraction of your data called "resource" and can interface any kind of ORMs or data storage through data layer concept.
 
 Main concepts
 -------------
@@ -12,11 +12,11 @@ Main concepts
    :width: 900px
    :alt: Architecture
 
-| * `JSON API 1.0 specification <http://jsonapi.org/>`_: it is a very popular specification about client server interactions for REST JSON API. It helps you to work in team because it is very precise and sharable. Thanks to this specification your api offers lot of features like a strong structure of request and response, filtering, pagination, sparse fieldsets, including related objects, great error formatting etc.
+| * `JSON API 1.0 specification <http://jsonapi.org/>`_: it is a very popular specification about client server interactions for REST JSON API. It helps you work in a team because it is very precise and sharable. Thanks to this specification your API offers lots of features like a strong structure of request and response, filtering, pagination, sparse fieldsets, including related objects, great error formatting etc.
 | 
-| * **Logical data abstration**: you usually need to expose resources to clients that don't fit your data table architecture. For example sometimes you don't want to expose all attributes of a table, compute additional attributes or create a resource that use data from multiple data storage. Flask-REST-JSONAPI helps you to create a logical abstraction of your data with `Marshmallow <https://marshmallow.readthedocs.io/en/latest/>`_ / `marshmallow-jsonapi <https://marshmallow-jsonapi.readthedocs.io/>`_ so you can expose your data through a very flexible way.
+| * **Logical data abstration**: you usually need to expose resources to clients that don't fit your data table architecture. For example sometimes you don't want to expose all attributes of a table, compute additional attributes or create a resource that uses data from multiple data storages. Flask-REST-JSONAPI helps you create a logical abstraction of your data with `Marshmallow <https://marshmallow.readthedocs.io/en/latest/>`_ / `marshmallow-jsonapi <https://marshmallow-jsonapi.readthedocs.io/>`_ so you can expose your data through a very flexible way.
 | 
-| * **Data layer**: the data layer is a CRUD interface between your resource manager and your data. Thanks to it you can use any data storage or ORMs. There is an already full featured data layer that use SQLAlchemy ORM but you can create and use your own custom data layer to use data from your data storage(s). You can even create a data layer that use multiple data storage and ORMs, send notifications or make any custom work during CRUD operations.
+| * **Data layer**: the data layer is a CRUD interface between your resource manager and your data. Thanks to it you can use any data storage or ORMs. There is an already full featured data layer that uses the SQLAlchemy ORM but you can create and use your own custom data layer to use data from your data storage(s). You can even create a data layer that uses multiple data storages and ORMs, send notifications or make any custom work during CRUD operations.
 
 Features
 --------

tests/test_sqlalchemy_data_layer.py

@@ -10,12 +10,19 @@
 from flask import Blueprint, make_response
 from marshmallow_jsonapi.flask import Schema, Relationship
 from marshmallow_jsonapi import fields
+from marshmallow import ValidationError
 
 from flask_rest_jsonapi import Api, ResourceList, ResourceDetail, ResourceRelationship, JsonApiException
-from flask_rest_jsonapi.exceptions import RelationNotFound, InvalidSort
+from flask_rest_jsonapi.pagination import add_pagination_links
+from flask_rest_jsonapi.exceptions import RelationNotFound, InvalidSort, InvalidFilters, InvalidInclude, BadRequest
 from flask_rest_jsonapi.querystring import QueryStringManager as QSManager
 from flask_rest_jsonapi.data_layers.alchemy import SqlalchemyDataLayer
 from flask_rest_jsonapi.data_layers.base import BaseDataLayer
+from flask_rest_jsonapi.data_layers.filtering.alchemy import Node
+from flask_rest_jsonapi.schema import get_relationships
+import flask_rest_jsonapi.decorators
+import flask_rest_jsonapi.resource
+import flask_rest_jsonapi.schema
 
 
 @pytest.fixture(scope="module")
@@ -315,11 +322,115 @@ class get_object(object):
                 })()
             })()
         })()
+
         def __init__(self, kwargs):
             pass
     return get_object
 
 
+def test_add_pagination_links():
+    qs = {'page[number]': '15', 'page[size]': '10'}
+    qsm = QSManager(qs, None)
+    add_pagination_links(dict(), 1000, qsm, str())
+
+
+def test_Node(person_model, person_schema, monkeypatch):
+    from copy import deepcopy
+    filt = {
+        'val': '0000',
+        'field': True,
+        'not': dict(),
+        'name': 'name',
+        'op': 'eq',
+        'strip': lambda: 's'
+    }
+    filt['not'] = deepcopy(filt)
+    del filt['not']['not']
+    n = Node(person_model,
+             filt,
+             None,
+             person_schema)
+    with pytest.raises(TypeError):
+        # print(n.val is None and n.field is None)
+        # # n.column
+        n.resolve()
+    with pytest.raises(AttributeError):
+        n.model = None
+        n.column
+    with pytest.raises(InvalidFilters):
+        n.model = person_model
+        n.filter_['op'] = ''
+        n.operator
+    with pytest.raises(InvalidFilters):
+        n.related_model
+    with pytest.raises(InvalidFilters):
+        n.related_schema
+
+
+def test_check_method_requirements(monkeypatch):
+    self = type('self', (object,), dict())
+    request = type('request', (object,), dict(method='GET'))
+    monkeypatch.setattr(flask_rest_jsonapi.decorators, 'request', request)
+    with pytest.raises(Exception):
+        flask_rest_jsonapi.\
+            decorators.check_method_requirements(lambda: 1)(self())
+
+
+def test_json_api_exception():
+    JsonApiException(None, None, title='test', status='test')
+
+
+def test_query_string_manager(person_schema):
+    query_string = {'page[slumber]': '3'}
+    qsm = QSManager(query_string, person_schema)
+    with pytest.raises(BadRequest):
+        qsm.pagination
+    qsm.qs['sort'] = 'computers'
+    with pytest.raises(InvalidSort):
+        qsm.sorting
+
+
+def test_resource(person_model, person_schema, session, monkeypatch):
+    def schema_load_mock(*args):
+        raise ValidationError(dict(errors=[dict(status=None, title=None)]))
+    query_string = {'page[slumber]': '3'}
+    app = type('app', (object,), dict(config=dict(DEBUG=True)))
+    headers = {'Content-Type': 'application/vnd.api+json'}
+    request = type('request', (object,), dict(method='POST',
+                                              headers=headers,
+                                              get_json=dict,
+                                              args=query_string))
+    dl = SqlalchemyDataLayer(dict(session=session, model=person_model))
+    rl = ResourceList()
+    rd = ResourceDetail()
+    rl._data_layer = dl
+    rl.schema = person_schema
+    rd._data_layer = dl
+    rd.schema = person_schema
+    monkeypatch.setattr(flask_rest_jsonapi.resource, 'request', request)
+    monkeypatch.setattr(flask_rest_jsonapi.resource, 'current_app', app)
+    monkeypatch.setattr(flask_rest_jsonapi.decorators, 'request', request)
+    monkeypatch.setattr(rl.schema, 'load', schema_load_mock)
+    r = super(flask_rest_jsonapi.resource.Resource, ResourceList)\
+        .__new__(ResourceList)
+    with pytest.raises(Exception):
+        r.dispatch_request()
+    rl.post()
+    rd.patch()
+
+
+def test_compute_schema(person_schema):
+    query_string = {'page[number]': '3', 'fields[person]': list()}
+    qsm = QSManager(query_string, person_schema)
+    with pytest.raises(InvalidInclude):
+        flask_rest_jsonapi.schema.compute_schema(
+            person_schema, dict(), qsm, ['id']
+        )
+    s = flask_rest_jsonapi.schema.compute_schema(
+        person_schema, dict(only=list()), qsm, list()
+    )
+
+
 # test good cases
 def test_get_list(client, register_routes, person, person_2):
     with client: