Minimal, prototype RESTful server for basic CRUD transactions
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.

License Status


  1. What is tiddly
  2. Project Details
  3. Installation
  4. Feature Roadmap
  5. License
  6. Stack

What is tiddly

tiddly is a minimal, prototype RESTful server app for basic CRUD transactions. It is database agnostic and uses sqlalchemy on the backend to talk to RDBMS. The database is sqlite by default, but you can switch to anything you want by making this small change in the file:

#TODO: Change as needed:
engine = create_engine("sqlite:///tiddly.db", echo=True)
Base = declarative_base()

Project Details

You can host this app and use it as your own replacement for any cloud based backend such as firebase or CouchDB as long as you follow the CRUD based I/O approach. The RESTful methods are versatile and compatible with the data needs of typical Single Page Apps developed in frameworks like angular, backbone or ember. For example:

Request url body Action
GET /books Get all rows from the books table.
GET /books/1 Fetch the book record where the id field equals 1.
POST /books {"name":"Harry Potter and the order of the Phoenix."} Insert record into books table from posted JSON data.
PUT /books/1 {"name":"Some new title."} Update the record where id equals 1 with posted JSON data.
DELETE /books/1 Delete the record where id equals 1.

Additional tables like books in this example need to be defined in By default, tiddly comes with only two tables called dual and user to play around. Of course, you can customize and change this logic in if you really want. There is also the case when the GET method that gets all or one row by id isn't sufficient and we need advanced fetching using the where or orderby clauses. For this, I've introduced the FETCH http method which does the job:

Request url body Action
FETCH /dual {"where": {"text":"foobar"}} Fetch all rows from the books table where the text field matches the string "foobar".
FETCH /user {"where": {"name":"admin"}, "orderby": "email"} Fetch all rows from the user table where the name matches "admin" and sort results by email column.
FETCH /user {"where": {"name":"admin"}, "orderby": "email desc"} Fetch all rows from the user table where the name matches "admin" and sort results by email column in descending order.
FETCH /user {"where": {"name":"admin"}, "orderby": "email desc", "limit":10, "offset": 2} Fetch at most 10 rows, starting at offset 2 from the user table where the name matches "admin". Sort results by email column in descending order.


To install and run this project, just clone or download this github repository and just start the tiddly app:


I've tested this on Python 3.6, but it should ideally run on other versions too. Once you run it, you can open your browser and browse the following url to make sure its running:

After that, you can make the following POST request to insert some data into the dual table.'http://localhost:5000/dual', json={"text":"chai and biscuit."})

You can use the accompanying or any other tool like curl to make these requests. Now, open up the browser again and visit below url to list all the records in the dual table:

After running the above command, you should be able to see something like this:

  "data": [
	  "id": 1, 
	  "text": "chai and biscuit."
  "status": "success", 
  "verb": "GET"

Note that the app only works with json data, so the usual POST requests won't work. Similarly, you can try out other verbs - PUT and DELETE to update and delete records respectively. To define your own new table, open the in a text-editor and define your own class based on the sqlalchemy declarative Base class. Write the new class code around the other defined Model classes, for example the Dual class:

class Dual(Base):
	__tablename__ = "dual"
	id = Column(Integer, primary_key=True)
	text = Column(String)
	def repr(self):
		return "<Dual(id=%s, text=%s, )>" % (id, text)

Finally, you can also install tiddly using pip just to play around with:

pip install tiddly

Feature Roadmap

I intend to add the following features to this template project soon:

  • Sorting using order by clause (done).
  • Pagination.
  • Fetching summaries using group by clause.
  • User sign-in and authentication using session or tokens.


tiddly is free and open source, and available under MIT license.


tiddly is built using the robust flask framework and sqlalchemy library.