This is a Tornado web app that integrates with Github Gists so that people in the Tornado community can share snippets of code with each other.

Anybody can log in (using their Github account) and start adding gists and discuss them. Not only their own but anybody's gist.

All the code is of course Open Source and is licensed under the Apache Licence, Version 2.0

It's by no means perfect and absolutely not the ideal way of writing a Tornado app (because there is no such way).

To run the code you simply need a MongoDB (of a recentish version) and install MongoKit which glues MongoDB to Tornado.

To run the tests simply run this command:

./bin/run_tests.sh

If you want to run tests just for one single app, gists run this:

./bin/run_tests.sh apps.gists.tests

And to run a specific test:

./bin/run_tests.sh  \
  apps.main.tests.test_handlers.HandlersTestCase.test_homepage

If you're doing development and want to run the same test(s) over and over just add --autoreload like this:

./bin/run_tests.sh --autoreload apps.gists.tests

To run the coverage tests, make sure you have coverage installed and run this script:

./bin/run_coverage_tests.sh

At the moment, this covers 87% of all the code on my computer.

MongoDB has no problem with some documents having different structure even if they're in the same collection. However, the glue code we use has. MongoKit expects every document in a collection to have the same structure. At least, every document must have everything that is entered in the class attribute structure in the document classes.

To add migration code you can do two things. For simple changes create a file called <myapp>/migrations/always_<whatever>.py these files are executed whenever you run the script that starts it:

./bin/run_migrations.py

If you have a specific change you want to make create a file that starts with a number like this for example 003.gender_on_user.py and stick it in the migrations directory. Next time you run run_migrations it will execute the file and create a file called 003.gender_on_user.py.done so that you don't accidentally run the file again. All *.done files are supposed to be ignored by git and not added to the repo.

Indexing can be a precious operation and has therefore been removed from the code as an automated task. Instead, has to be executed manually. It's only really necessary to run once since MongoDB is smart enough to maintain the indexes for new documents once everything is set up. If you haven't already done so, to ensure all the indexes run this:

./bin/run_migrations.py

Adding more indexing is basically the same as running migrations.

To update the user details, run the script bin/update-user-details.py periodically. You can specify how many updates to make maximum. Default is 100. Example usage:

./bin/update-user-details.py