When comparing Flask vs Django, the Slant community recommends Django for most people. In the question“What are the best backend web frameworks?” Django is ranked 2nd while Flask is ranked 7th. The most important reason people chose Django is:
Django's philosophy of batteries included means that experienced developers won't have to plan too much ahead on what kind of application infrastructure they need and instead just start developing web applications quickly.
Specs
Ranked in these QuestionsQuestion Ranking
Pros
Pro Minimalist without losing power
Flask is very easy to get up and going, with vanilla HTML or with bootstrap pieces. It doesn't take much lines of Python to load Flask to get headers working, etc, and since it's all modular you don't have to have something you don't want in your application.
Pro Lots of resources available online
Flask is one of the most popular Python web frameworks, if not the most popular one. As such, there's plenty of guides, tutorials, and libraries available for it. A large number of important Python libraries, such as SQLAlchemy have libraries for Flask, which add valuable bindings to make the development process and the integration between these libraries and Flask as easy as possible.
Pro Extremely easy to build a quick prototype
Even though it's pretty minimalistic out of the box, Flask still provides the necessary tools to build a quick prototype for a web app right after a fresh install. With all the main components pretty much packed in the flask
package, building a simple web app in a single Python file is as easy as it gets.
Pro Very flexible
Flask gives developers a lot of flexibility in how they develop their web applications.
For example, the choice of not having an ORM, but instead choosing one suited to the task, or another area where Flask gives a lot of options to developers is the templating. They can use Jinja2, Flask's default templating language or choose from a number of different templating languages they desire.
Pro Great documentation
The official documentation is very thorough and complete. Everything is explained in-depth and followed by extremely well-explained tutorials that tackle real-world problems.
Pro Able to use ORM or "true SQL"
Pro Developing a simple prototype can be very fast
Django's philosophy of batteries included means that experienced developers won't have to plan too much ahead on what kind of application infrastructure they need and instead just start developing web applications quickly.
Pro ORM support out of the box
Django supports Object-Relational Mapping. With models defined as Python classes which are actually subclasses of Django's django.db.models.Model
.
Each attribute of the model is then represented as a database field. Queries are lazily executed and Django gives developers an automatically-generated database-access API.
Pro Top notch documentation and help from community
The official Django documentation is probably some of the best around. Well written, thorough and they explain every little detail of the framework. Django is also a very popular tool, with an extensive community and a lot of experienced developers that have been using it for years. This means that there are a lot of guides and tutorials out there for new and experienced developers alike.
Pro Highly customizable
Django is in itself a highly customizable web framework. The database, template framework and ORM can all be swapped out.
Pro Has an admin panel out of the box
Django comes with a highly customizable admin panel and authentication out of the box. This makes the development and production of a simple CMS extremely easy.
Pro Mature software with many plugins developed over the years
Django was first released in 2005, it has had a lot of time to mature and become better with each release. It also has by far the largest community out of all python frameworks who have continuously over the years built and maintained many powerful plugins.
Pro Clear and defined MVC organization
Django follows some pretty well established MVC patterns. With everything in place and where requests follow a clear path through urlresolvers, middleware, view and context processors.
Pro Simple database management
Just a few lines of code can instruct Django to create all the tables and fields required in your database automatically. Schemas are managed with "migrations", that are also created automatically, and can be rolled out from your development box and implemented on production systems with just a single command.
This performs any database changes required, from table creation, indexes, renaming fields, and pre-populating initial data. Each migration builds on the previous migrations, so you can trace the evolution of your data and even recreate the layout of your database at any point in the lifecycle of your application.
Cons
Con Not async-friendly
Flask is explicitly not designed to handle async programming.
Con Setting up a large project requires some previous knowledge of the framework
Setting up a large project with Flask is not that easy considering how there's no "official" way of doing it. Blueprints are a useful tool in this regard but require some additional reading and are a bit tricky to get right for a beginner.
The lack of some defaults can also be problematic. Having to choose between different libraries for a certain task is never easy, especially if you have never worked with Flask before.
Con Threadlocals and globals used everywhere
The default way of creating applications in flask makes it hard to use reusable and clean code.
Con HTML-oriented, not API-oriented
Not necessarily designed for making APIs, though that is possible
Con Can feel bloated for small projects
Django's sheer scale and functionality can feel clunky and bloated for small applications. It has too many bells and whistles which can get in the way when developing a small scale application.
Con The documentation does not cover real-world scenarios
It is a larger documentation indeed, however is not deep and covers non real problems or even don't show any examples. You'll be better with Google or Stackoverflow
Con Routing requires some knowledge of regular expressions
Given a GET request for /topics/426/viewpoints/1/sections/create
, how does Django decide which bit of Python code is invoked to handle it?
It compares the request path to your giant pile of regular expressions.
And then if there's some other regular expression starts matching /top
and all your requests for /topics/
start going there, good luck figuring out why. You won't be informed of any conflict until you notice you seem to be getting the wrong pages back.
The structure of URL paths is almost universally hierarchal. There is no call to have anything as ridiculously flexible (and notoriously hard-to-read) as regular expressions to organize request routing.
Con Template errors fail silently by default
If you make a typo in a template variable, or change a view so that variable is no longer passed to the template, you won't get an error message pointing out that something has gone wrong. That reference will just be treated as if it is an empty string instead.
There is a way to configure this, but since so many templates have been written assuming this behavior, nobody ever enables template errors because it would break so much of the existing support tools (e.g. the built-in admin interface).