Address #1902: Converts example/flaskr to have a setup.py (#1945)

* Converts example/flaskr to have a setup.py

Makes the flaskr app easier to run, ex. workflow:
- pip install --editable .
- export FLASK_APP=flaskr.flaskr
- flask initdb
- flask run

Testing is also easier now:
- python setup.py test

* Fixed an import error in flaskr/tests

- the statement `import flaskr` caused errors in python3
- `from . import flaskr` fixes the issue in 2.7.11 and 3.5.1

* Better project structure and updates the docs

- Re-factors *flaskr*'s project structure a bit
- Updates docs to make sense with the new structure
  - Adds a new step about installing Flask apps with setuptools
  - Switches first-person style writing to second-person (reads better IMO)
  - Adds segments in *testing.rst* for running tests with setuptools

* Remove __init__.py from tests

- py.test recommends not using __init__.py

* Fix testing import errors

docs/tutorial/css.rst

@@ -1,11 +1,11 @@
 .. _tutorial-css:
 
-Step 7: Adding Style
+Step 9: Adding Style
 ====================
 
 Now that everything else works, it's time to add some style to the
 application.  Just create a stylesheet called :file:`style.css` in the
-:file:`static` folder we created before:
+:file:`static` folder:
 
 .. sourcecode:: css
 

docs/tutorial/dbcon.rst

@@ -1,25 +1,23 @@
 .. _tutorial-dbcon:
 
-Step 3: Database Connections
+Step 4: Database Connections
 ----------------------------
 
-We have created a function for establishing a database connection with
-`connect_db`, but by itself, that's not particularly useful.  Creating and
-closing database connections all the time is very inefficient, so we want
-to keep it around for longer.  Because database connections encapsulate a
-transaction, we also need to make sure that only one request at the time
-uses the connection. How can we elegantly do that with Flask?
+You now have a function for establishing a database connection with
+`connect_db`, but by itself, it is not particularly useful.  Creating and
+closing database connections all the time is very inefficient, so you will
+need to keep it around for longer.  Because database connections
+encapsulate a transaction, you will need to make sure that only one
+request at a time uses the connection. An elegant way to do this is by
+utilizing the *application context*.
 
-This is where the application context comes into play, so let's start
-there.
-
-Flask provides us with two contexts: the application context and the
-request context.  For the time being, all you have to know is that there
+Flask provides two contexts: the *application context* and the
+*request context*.  For the time being, all you have to know is that there
 are special variables that use these.  For instance, the
 :data:`~flask.request` variable is the request object associated with
 the current request, whereas :data:`~flask.g` is a general purpose
-variable associated with the current application context.  We will go into
-the details of this a bit later.
+variable associated with the current application context.  The tutorial
+will cover some more details of this later on.
 
 For the time being, all you have to know is that you can store information
 safely on the :data:`~flask.g` object.
@@ -37,7 +35,7 @@ already established connection::
             g.sqlite_db = connect_db()
         return g.sqlite_db
 
-So now we know how to connect, but how do we properly disconnect?  For
+Now you know how to connect, but how can you properly disconnect?  For
 that, Flask provides us with the :meth:`~flask.Flask.teardown_appcontext`
 decorator.  It's executed every time the application context tears down::
 

docs/tutorial/dbinit.rst

@@ -1,6 +1,6 @@
 .. _tutorial-dbinit:
 
-Step 4: Creating The Database
+Step 5: Creating The Database
 =============================
 
 As outlined earlier, Flaskr is a database powered application, and more
@@ -16,13 +16,14 @@ Such a schema can be created by piping the ``schema.sql`` file into the
 
 The downside of this is that it requires the ``sqlite3`` command to be
 installed, which is not necessarily the case on every system.  This also
-requires that we provide the path to the database, which can introduce
+requires that you provide the path to the database, which can introduce
 errors.  It's a good idea to add a function that initializes the database
-for you to the application.
+for you, to the application.
 
-To do this, we can create a function and hook it into the :command:`flask`
-command that initializes the database.  Let me show you the code first.  Just
-add this function below the `connect_db` function in :file:`flaskr.py`::
+To do this, you can create a function and hook it into a :command:`flask`
+command that initializes the database.  For now just take a look at the
+code segment below.  A good place to add this function, and command, is
+just below the `connect_db` function in :file:`flaskr.py`::
 
     def init_db():
         db = get_db()
@@ -38,23 +39,23 @@ add this function below the `connect_db` function in :file:`flaskr.py`::
 
 The ``app.cli.command()`` decorator registers a new command with the
 :command:`flask` script.  When the command executes, Flask will automatically
-create an application context for us bound to the right application.
-Within the function, we can then access :attr:`flask.g` and other things as
-we would expect.  When the script ends, the application context tears down
+create an application context which is bound to the right application.
+Within the function, you can then access :attr:`flask.g` and other things as
+you might expect.  When the script ends, the application context tears down
 and the database connection is released.
 
-We want to keep an actual function around that initializes the database,
+You will want to keep an actual function around that initializes the database,
 though, so that we can easily create databases in unit tests later on.  (For
 more information see :ref:`testing`.)
 
 The :func:`~flask.Flask.open_resource` method of the application object
 is a convenient helper function that will open a resource that the
 application provides.  This function opens a file from the resource
-location (your ``flaskr`` folder) and allows you to read from it.  We are
-using this here to execute a script on the database connection.
+location (the :file:`flaskr/flaskr` folder) and allows you to read from it.
+It is used in this example to execute a script on the database connection.
 
-The connection object provided by SQLite can give us a cursor object.
-On that cursor, there is a method to execute a complete script.  Finally, we
+The connection object provided by SQLite can give you a cursor object.
+On that cursor, there is a method to execute a complete script.  Finally, you
 only have to commit the changes.  SQLite3 and other transactional
 databases will not commit unless you explicitly tell it to.
 

docs/tutorial/folders.rst

@@ -3,21 +3,25 @@
 Step 0: Creating The Folders
 ============================
 
-Before we get started, let's create the folders needed for this
+Before getting started, you will need to create the folders needed for this
 application::
 
     /flaskr
-        /static
-        /templates
+        /flaskr
+            /static
+            /templates
 
-The ``flaskr`` folder is not a Python package, but just something where we
-drop our files. Later on, we will put our database schema as well as main
-module into this folder. It is done in the following way. The files inside
-the :file:`static` folder are available to users of the application via HTTP.
-This is the place where CSS and Javascript files go.  Inside the
-:file:`templates` folder, Flask will look for `Jinja2`_ templates.  The
-templates you create later on in the tutorial will go in this directory.
+The application will be installed and run as Python package.  This is the
+recommended way to install and run Flask applications.  You will see exactly
+how to run ``flaskr`` later on in this tutorial.  For now go ahead and create
+the applications directory structure.  In the next few steps you will be
+creating the database schema as well as the main module.
 
-Continue with :ref:`tutorial-schema`.
+As a quick side note, the files inside of the :file:`static` folder are
+available to users of the application via HTTP.  This is the place where CSS and
+Javascript files go.  Inside the :file:`templates` folder, Flask will look for
+`Jinja2`_ templates.  You will see examples of this later on.
+
+For now you should continue with :ref:`tutorial-schema`.
 
 .. _Jinja2: http://jinja.pocoo.org/

docs/tutorial/index.rst

@@ -24,6 +24,7 @@ the `example source`_.
    folders
    schema
    setup
+   setuptools
    dbcon
    dbinit
    views

docs/tutorial/introduction.rst

@@ -3,8 +3,9 @@
 Introducing Flaskr
 ==================
 
-We will call our blogging application Flaskr, but feel free to choose your own
-less Web-2.0-ish name ;)  Essentially, we want it to do the following things:
+This tutorial will demonstrate a blogging application named Flaskr, but feel
+free to choose your own less Web-2.0-ish name ;)  Essentially, it will do the
+following things:
 
 1. Let the user sign in and out with credentials specified in the
    configuration.  Only one user is supported.
@@ -14,8 +15,8 @@ less Web-2.0-ish name ;)  Essentially, we want it to do the following things:
 3. The index page shows all entries so far in reverse chronological order
    (newest on top) and the user can add new ones from there if logged in.
 
-We will be using SQLite3 directly for this application because it's good
-enough for an application of this size.  For larger applications, however,
+SQLite3 will be used directly for this application because it's good enough
+for an application of this size.  For larger applications, however,
 it makes a lot of sense to use `SQLAlchemy`_, as it handles database
 connections in a more intelligent way, allowing you to target different
 relational databases at once and more.  You might also want to consider

docs/tutorial/schema.rst

@@ -3,10 +3,10 @@
 Step 1: Database Schema
 =======================
 
-First, we want to create the database schema. Only a single table is needed
-for this application and we only want to support SQLite, so creating the
-database schema is quite easy. Just put the following contents into a file
-named `schema.sql` in the just created `flaskr` folder:
+In this step, you will create the database schema.  Only a single table is
+needed for this application and it will only support SQLite.  All you need to do
+is put the following contents into a file named :file:`schema.sql` in the
+:file:`flaskr/flaskr` folder:
 
 .. sourcecode:: sql
 
@@ -17,7 +17,7 @@ named `schema.sql` in the just created `flaskr` folder:
       'text' text not null
     );
 
-This schema consists of a single table called ``entries``. Each row in
+This schema consists of a single table called ``entries``.  Each row in
 this table has an ``id``, a ``title``, and a ``text``.  The ``id`` is an
 automatically incrementing integer and a primary key, the other two are
 strings that must not be null.

docs/tutorial/setup.rst

@@ -3,15 +3,16 @@
 Step 2: Application Setup Code
 ==============================
 
-Now that we have the schema in place, we can create the application module.
-Let's call it ``flaskr.py``. We will place this file inside the ``flaskr``
-folder. We will begin by adding the imports we need and by adding the config
-section.  For small applications, it is possible to drop the configuration
-directly into the module, and this is what we will be doing here. However,
-a cleaner solution would be to create a separate ``.ini`` or ``.py`` file,
-load that, and import the values from there.
+Now that the schema is in place, you can create the application module,
+:file:`flaskr.py`.  This file should be placed inside of the
+:file:`flaskr/flaskr` folder.  The first several lines of code in the
+application module are the needed import statements.  After that there will be a
+few lines of configuration code. For small applications like ``flaskr``, it is
+possible to drop the configuration directly into the module.  However, a cleaner
+solution is to create a separate ``.ini`` or ``.py`` file, load that, and
+import the values from there.
 
-First, we add the imports in :file:`flaskr.py`::
+Here are the import statements (in :file:`flaskr.py`)::
 
     # all the imports
     import os
@@ -19,12 +20,13 @@ First, we add the imports in :file:`flaskr.py`::
     from flask import Flask, request, session, g, redirect, url_for, abort, \
          render_template, flash
 
-Next, we can create our actual application and initialize it with the
-config from the same file in :file:`flaskr.py`::
+The next couple lines will create the actual application instance and
+initialize it with the config from the same file in :file:`flaskr.py`:
 
-    # create our little application :)
-    app = Flask(__name__)
-    app.config.from_object(__name__)
+.. sourcecode:: python
+
+    app = Flask(__name__) # create the application instance :)
+    app.config.from_object(__name__) # load config from this file , flaskr.py
 
     # Load default config and override config from an environment variable
     app.config.update(dict(
@@ -35,8 +37,8 @@ config from the same file in :file:`flaskr.py`::
     ))
     app.config.from_envvar('FLASKR_SETTINGS', silent=True)
 
-The :class:`~flask.Config` object works similarly to a dictionary so we
-can update it with new values.
+The :class:`~flask.Config` object works similarly to a dictionary, so it can be
+updated with new values.
 
 .. admonition:: Database Path
 
@@ -55,10 +57,10 @@ can update it with new values.
 
 Usually, it is a good idea to load a separate, environment-specific
 configuration file.  Flask allows you to import multiple configurations and it
-will use the setting defined in the last import. This enables robust
+will use the setting defined in the last import.  This enables robust
 configuration setups.  :meth:`~flask.Config.from_envvar` can help achieve this.
 
-.. code-block:: python
+.. sourcecode:: python
 
    app.config.from_envvar('FLASKR_SETTINGS', silent=True)
 
@@ -74,15 +76,15 @@ that in all cases, only variable names that are uppercase are considered.
 The ``SECRET_KEY`` is needed to keep the client-side sessions secure.
 Choose that key wisely and as hard to guess and complex as possible.
 
-We will also add a method that allows for easy connections to the
-specified database. This can be used to open a connection on request and
+Lastly, you will add a method that allows for easy connections to the
+specified database.  This can be used to open a connection on request and
 also from the interactive Python shell or a script.  This will come in
-handy later.  We create a simple database connection through SQLite and
+handy later.  You can create a simple database connection through SQLite and
 then tell it to use the :class:`sqlite3.Row` object to represent rows.
-This allows us to treat the rows as if they were dictionaries instead of
+This allows the rows to be treated as if they were dictionaries instead of
 tuples.
 
-::
+.. sourcecode:: python
 
     def connect_db():
         """Connects to the specific database."""
@@ -90,29 +92,6 @@ tuples.
         rv.row_factory = sqlite3.Row
         return rv
 
-With that out of the way, you should be able to start up the application
-without problems.  Do this with the following commands::
-
-    export FLASK_APP=flaskr
-    export FLASK_DEBUG=1
-    flask run
-
-(In case you are on Windows you need to use `set` instead of `export`).
-The :envvar:`FLASK_DEBUG` flag enables or disables the interactive debugger.
-*Never leave debug mode activated in a production system*, because it will
-allow users to execute code on the server!
-
-You will see a message telling you that server has started along with
-the address at which you can access it.
-
-When you head over to the server in your browser, you will get a 404 error
-because we don't have any views yet.  We will focus on that a little later,
-but first, we should get the database working.
-
-.. admonition:: Externally Visible Server
-
-   Want your server to be publicly available?  Check out the
-   :ref:`externally visible server <public-server>` section for more
-   information.
+In the next section you will see how to run the application.
 
-Continue with :ref:`tutorial-dbcon`.
+Continue with :ref:`tutorial-setuptools`.

docs/tutorial/setuptools.rst

@@ -0,0 +1,89 @@
+.. _tutorial-setuptools:
+
+Step 3: Installing flaskr with setuptools
+=========================================
+
+Flask is now shipped with built-in support for `Click`_.  Click provides
+Flask with enhanced and extensible command line utilities.  Later in this
+tutorial you will see exactly how to extend the ``flask`` command line
+interface (CLI).
+
+A useful pattern to manage a Flask application is to install your app
+using `setuptools`_.  This involves creating a :file:`setup.py`
+in the projects root directory.  You also need to add an empty
+:file:`__init__.py` file to make the :file:`flaskr/flaskr` directory
+a package.  The code structure at this point should be::
+
+    /flaskr
+        /flaskr
+            __init__.py
+            /static
+            /templates
+        setup.py
+
+The content of the ``setup.py`` file for ``flaskr`` is:
+
+.. sourcecode:: python
+
+    from setuptools import setup
+
+    setup(
+        name='flaskr',
+        packages=['flaskr'],
+        include_package_data=True,
+        install_requires=[
+            'flask',
+        ],
+    )
+
+When using setuptools, it is also necessary to specify any special files
+that should be included in your package (in the :file:`MANIFEST.in`).
+In this case, the static and templates directories need to be included,
+as well as the schema. Create the :file:`MANIFEST.in` and add the
+following lines::
+
+    graft flaskr/templates
+    graft flaskr/static
+    include flaskr/schema.sql
+
+At this point you should be able to install the application.  As usual, it
+is recommended to install your Flask application within a `virtualenv`_.
+With that said, go ahead and install the application with::
+
+    pip install --editable .
+
+.. note:: The above installation command assumes that it is run within the
+    projects root directory, `flaskr/`.  Also, the `editable` flag allows
+    editing source code without having to reinstall the Flask app each time
+    you make changes.
+
+With that out of the way, you should be able to start up the application.
+Do this with the following commands::
+
+    export FLASK_APP=flaskr.flaskr
+    export FLASK_DEBUG=1
+    flask run
+
+(In case you are on Windows you need to use `set` instead of `export`).
+The :envvar:`FLASK_DEBUG` flag enables or disables the interactive debugger.
+*Never leave debug mode activated in a production system*, because it will
+allow users to execute code on the server!
+
+You will see a message telling you that server has started along with
+the address at which you can access it.
+
+When you head over to the server in your browser, you will get a 404 error
+because we don't have any views yet.  That will be addressed a little later,
+but first, you should get the database working.
+
+.. admonition:: Externally Visible Server
+
+   Want your server to be publicly available?  Check out the
+   :ref:`externally visible server <public-server>` section for more
+   information.
+
+Continue with :ref:`tutorial-dbcon`.
+
+.. _Click: http://click.pocoo.org
+.. _setuptools: https://setuptools.readthedocs.io
+.. _virtualenv: https://virtualenv.pypa.io

docs/tutorial/templates.rst

@@ -1,11 +1,12 @@
 .. _tutorial-templates:
 
-Step 6: The Templates
+Step 8: The Templates
 =====================
 
-Now we should start working on the templates.  If we were to request the URLs
-now, we would only get an exception that Flask cannot find the templates.  The
-templates are using `Jinja2`_ syntax and have autoescaping enabled by
+Now it is time to start working on the templates.  As you may have
+noticed, if you make requests with the app running, you will get
+an exception that Flask cannot find the templates.  The templates
+are using `Jinja2`_ syntax and have autoescaping enabled by
 default.  This means that unless you mark a value in the code with
 :class:`~flask.Markup` or with the ``|safe`` filter in the template,
 Jinja2 will ensure that special characters such as ``<`` or ``>`` are
@@ -57,9 +58,9 @@ show_entries.html
 
 This template extends the :file:`layout.html` template from above to display the
 messages.  Note that the ``for`` loop iterates over the messages we passed
-in with the :func:`~flask.render_template` function.  We also tell the
-form to submit to your `add_entry` function and use ``POST`` as HTTP
-method:
+in with the :func:`~flask.render_template` function.  Notice that the form is
+configured to to submit to the `add_entry` view function and use ``POST`` as
+HTTP method:
 
 .. sourcecode:: html+jinja
 

docs/tutorial/testing.rst

@@ -8,3 +8,81 @@ expected, it's probably not a bad idea to add automated tests to simplify
 modifications in the future.  The application above is used as a basic
 example of how to perform unit testing in the :ref:`testing` section of the
 documentation.  Go there to see how easy it is to test Flask applications.
+
+Adding Tests to flaskr
+======================
+
+Assuming you have seen the testing section above and have either written
+your own tests for ``flaskr`` or have followed along with the examples
+provided, you might be wondering about ways to organize the project.
+
+One possible and recommended project structure is::
+
+    flaskr/
+        flaskr/
+            __init__.py
+            static/
+            templates/
+        tests/
+            context.py
+            test_flaskr.py
+        setup.py
+        MANIFEST.in
+
+For now go ahead a create the :file:`tests/` directory as well as the
+:file:`context.py` and :file:`test_flaskr.py` files, if you haven't
+already. The context file is used as an import helper. The contents
+of that file are::
+
+    import sys, os
+
+    basedir = os.path.dirname(os.path.abspath(__file__))
+    sys.path.insert(0, basedir + '/../')
+
+    from flaskr import flaskr
+
+Testing + Setuptools
+====================
+
+One way to handle testing is to integrate it with ``setuptools``. All it
+requires is adding a couple of lines to the :file:`setup.py` file and
+creating a new file :file:`setup.cfg`. Go ahead and update the
+:file:`setup.py` to contain::
+
+    from setuptools import setup
+
+    setup(
+        name='flaskr',
+        packages=['flaskr'],
+        include_package_data=True,
+        install_requires=[
+            'flask',
+        ],
+    )
+        setup_requires=[
+            'pytest-runner',
+        ],
+        tests_require=[
+            'pytest',
+        ],
+    )
+Now create :file:`setup.cfg` in the project root (alongside
+:file:`setup.py`)::
+
+    [aliases]
+    test=pytest
+
+Now you can run::
+
+    python setup.py test
+
+This calls on the alias created in :file:`setup.cfg` which in turn runs
+``pytest`` via ``pytest-runner``, as the :file:`setup.py` script has
+been called. (Recall the `setup_requires` argument in :file:`setup.py`)
+Following the standard rules of test-discovery your tests will be
+found, run, and hopefully pass.
+
+This is one possible way to run and manage testing.  Here ``pytest`` is
+used, but there are other options such as ``nose``.  Integrating testing
+with ``setuptools`` is convenient because it is not necessary to actually
+download ``pytest`` or any other testing framework one might use.

docs/tutorial/views.rst

@@ -1,10 +1,10 @@
 .. _tutorial-views:
 
-Step 5: The View Functions
+Step 7: The View Functions
 ==========================
 
-Now that the database connections are working, we can start writing the
-view functions.  We will need four of them:
+Now that the database connections are working, you can start writing the
+view functions.  You will need four of them:
 
 Show Entries
 ------------
@@ -30,7 +30,7 @@ Add New Entry
 
 This view lets the user add new entries if they are logged in.  This only
 responds to ``POST`` requests; the actual form is shown on the
-`show_entries` page.  If everything worked out well, we will
+`show_entries` page.  If everything worked out well, it will
 :func:`~flask.flash` an information message to the next request and
 redirect back to the `show_entries` page::
 
@@ -45,8 +45,8 @@ redirect back to the `show_entries` page::
         flash('New entry was successfully posted')
         return redirect(url_for('show_entries'))
 
-Note that we check that the user is logged in here (the `logged_in` key is
-present in the session and ``True``).
+Note that this view checks that the user is logged in (that is, if the
+`logged_in` key is present in the session and ``True``).
 
 .. admonition:: Security Note
 
@@ -81,11 +81,11 @@ notified about that, and the user is asked again::
         return render_template('login.html', error=error)
 
 The `logout` function, on the other hand, removes that key from the session
-again.  We use a neat trick here: if you use the :meth:`~dict.pop` method
+again.  There is a neat trick here: if you use the :meth:`~dict.pop` method
 of the dict and pass a second parameter to it (the default), the method
 will delete the key from the dictionary if present or do nothing when that
-key is not in there.  This is helpful because now we don't have to check
-if the user was logged in.
+key is not in there.  This is helpful because now it is not necessary to
+check if the user was logged in.
 
 ::
 
@@ -94,11 +94,24 @@ if the user was logged in.
         session.pop('logged_in', None)
         flash('You were logged out')
         return redirect(url_for('show_entries'))
-        
-Note that it is not a good idea to store passwords in plain text. You want to
-protect login credentials if someone happens to have access to your database.
-One way to do this is to use Security Helpers from Werkzeug to hash the
-password. However, the emphasis of this tutorial is to demonstrate the basics
-of Flask and plain text passwords are used for simplicity.
+
+.. admonition:: Security Note
+
+    Passwords should never be stored in plain text in a production
+    system. This tutorial uses plain text passwords for simplicity. If you
+    plan to release a project based off this tutorial out into the world,
+    passwords should be both `hashed and salted`_ before being stored in a
+    database or file.
+
+    Fortunately, there are Flask extensions for the purpose of
+    hashing passwords and verifying passwords against hashes, so adding
+    this functionality is fairly straight forward. There are also
+    many general python libraries that can be used for hashing.
+
+    You can find a list of recommended Flask extensions
+    `here <http://flask.pocoo.org/extensions/>`_
+
 
 Continue with :ref:`tutorial-templates`.
+
+.. _hashed and salted: https://blog.codinghorror.com/youre-probably-storing-passwords-incorrectly/

examples/flaskr/.gitignore

@@ -1 +1,2 @@
 flaskr.db
+.eggs/

examples/flaskr/MANIFEST.in

@@ -0,0 +1,3 @@
+graft flaskr/templates
+graft flaskr/static
+include flaskr/schema.sql

examples/flaskr/README

@@ -13,15 +13,19 @@
          export an FLASKR_SETTINGS environment variable
          pointing to a configuration file.
 
-      2. Instruct flask to use the right application
+      2. install the app from the root of the project directory
 
-         export FLASK_APP=flaskr
+         pip install --editable .
 
-      3. initialize the database with this command:
+      3. Instruct flask to use the right application
+
+         export FLASK_APP=flaskr.flaskr
+
+      4. initialize the database with this command:
 
          flask initdb
 
-      4. now you can run flaskr:
+      5. now you can run flaskr:
 
          flask run
 
@@ -30,5 +34,5 @@
 
     ~ Is it tested?
 
-      You betcha.  Run the `test_flaskr.py` file to see
+      You betcha.  Run `python setup.py test` to see
       the tests pass.

examples/flaskr/setup.cfg

@@ -0,0 +1,2 @@
+[aliases]
+test=pytest

examples/flaskr/setup.py

@@ -0,0 +1,16 @@
+from setuptools import setup
+
+setup(
+    name='flaskr',
+    packages=['flaskr'],
+    include_package_data=True,
+    install_requires=[
+        'flask',
+    ],
+    setup_requires=[
+        'pytest-runner',
+    ],
+    tests_require=[
+        'pytest',
+    ],
+)

examples/flaskr/tests/context.py

@@ -0,0 +1,6 @@
+import sys, os
+
+basedir = os.path.dirname(os.path.abspath(__file__))
+sys.path.insert(0, basedir + '/../')
+
+from flaskr import flaskr

examples/flaskr/tests/test_flaskr.py

@@ -10,11 +10,10 @@
 """
 
 import pytest
-
 import os
-import flaskr
 import tempfile
 
+from context import flaskr
 
 @pytest.fixture
 def client(request):