ericmjl / minimal-flask-example

The simplest complex example that I can think of to show main Flask app concepts.

minimal-flask-example

The simplest complex example that I can think of to show main Flask app concepts.

All examples are in app.py. There are 3 view functions embedded, each showing a different example of how to build Flask apps.

Key concepts applicable throughout:

• View functions
• Routes
• Returning a template.

example 1

The first example is under /df. In this example, I show you how to embed a pandas.DataFrame() table inside an HTML page served up using Flask, essentially as a dump of the table.

Key concepts here:

• Jinja2 templating: ({{ variable_name }} syntax)
• Template inheritance: ({% extends "parent.html" %})
• Styling with Bootstrap CSS
• Marking strings as safe to render: ({{ string|safe }})

example 2

The second example is under /dfcustom. In this example, I show you how to apply custom formatting to a DataFrame rendered as an HTML table.

Key concepts here, building on top of example 1:

• Jinja2 syntax is very Python-like.
• Macros behave like Python functions.
• Looping is very Pythonic.

example 3

The second example is /bokehplot. In this example, I show you how to embed a Bokeh plot on an HTML page served up using Flask.

Key concepts here:

• Marking strings as safe to render (see above)
• Refactoring granular logic into a utility function, so that only application "business logic" remains.
• It is possible to cleverly pass strings around so that you can keep things like versioning automatically correct.

minimal-flask-example's Issues

With j2 extension autoescape in Jinja templates is turned off

https://github.com/pallets/flask/blob/8f422d2b5efe685bf67852b8f06672a32c1f07e6/src/flask/app.py#L808

Update the template schema to get version automatically

minimal-flask-example/templates/data.html

Line 4 in 43ab49e

<link href="http://cdn.pydata.org/bokeh/release/bokeh-0.12.13.min.css" rel="stylesheet" type="text/css">

Thank you for this example!
I would suggest to update the template with bokeh version variable and then supply it from python using return of bokeh.__version__ - this should make your example bokeh_version agnostic.