Skip to main content
This is a pre-production deployment of Warehouse. Changes made here affect the production instance of PyPI (
Help us improve Python packaging - Donate today!

Basic Python-based data server for exposing flat table and object hierarchy files via REST-ful queries

Project Description


You know what’s awesome? Setting up a REST-ful server you can start with a single command, out-of-the-box.

You know what’s even cooler? Adding file resources to that server just by dragging and dropping said files into a folder.

Welcome to pyroclast.

Get Started

To start a Pyroclast instance, simply point the Python interpreter to the file:


Alternatively, you can import server and start an instance manually from within Python:

from pyroclast import server

Once the server is started, you will see a blank HTTP response for This means it’s working!

Get Served

Within the Pyroclast package, the ‘data/’ folder contains all files your server will host. This can include:

  • JavaScript Object Notation files (.JSON)
  • Comma-Seperated Value files (.CSV)
  • Excel files (.XLS, .XLSX)
  • SQLite database files (.SQL)
  • UnQLite database files (.UNQ)
  • eXtensible Markup Language files (.XML)

To add a file, simply copy it into the Pyroclast package’s ‘data/’ folder and the Pyroclast server will do the rest.

Alternatively, you can serve a file procedurally using the data module’s serve() function, which will copy directory contents into ‘data/’:

from pyroclast import data

Get Your Data

Once your server is up and running, you can query files from a URL based on the file path relative to ‘data/’–i.e., if ‘test.xlsx’ is in ‘data/’, it is accessed by the URL ‘’.

All data queried will be returned in one of two formats:

  • A CSV-formatted text table (for flat data table formats, like .CSV, .XLS/.XLSX, and .SQL)
  • A JSON-formatted text tree (for object hierarchy formats, like .JSON, .XML, and .UNQ)

Specific subsets of data from a source can be selected using the query segment of the URL (i.e., ?key=value&key2=value2). In addition to down-selecting the returned data set based on attribute key-value pairs, there are several format-specific keys that can be used. (Note that there are no format-specific keys for .JSON, .XML, and .CSV formats).


Excel spreadsheets can contain multiple sheets. Pyroclast assumes each queried sheet is a flat table starting from the first row (a header) and first column. Specific sheets can be selected in one of two ways:

  • Passing the name of the desired sheet as the value of the ‘_sheet’ key
  • Passing the index of the desired sheet (starting with 0) as the value of the ‘_sheet’ key

In cases where no ‘_sheet’ value is given, Pyroclast will default to the first sheet in the file.

In cases where sheets are given integer values as names, Pyroclast will default to the string interpretation.


SQLite database files require a query indicate a specific table. This is done by passing the desired table name as a value of the ‘_table’ key. If the ‘_table’ key is not specified, an error will be raised.


UnQLite is a funny beast. .UNQ database files can contain both key-value pairs AND collections of related object hierarchies.

  • The ‘_collection’ key can be used to indicate a specific collection.
  • The ‘_key’ key can be used to indicate the key of a specific key-value pair.

If neither ‘_collection’ or ‘_key’ is indicated, Pyroclast will return all contents of the database. Key-value pairs will be included in the response twice:

  • Once as single-entry dictionaries at the root level of the response
  • Once as entries to a dictionary assigned to the ‘_root’ key

Release History

This version
History Node


History Node


History Node


History Node


History Node


History Node


History Node


History Node


Download Files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Filename, Size & Hash SHA256 Hash Help File Type Python Version Upload Date
(51.8 kB) Copy SHA256 Hash SHA256
Source None Dec 4, 2015

Supported By

Elastic Elastic Search Pingdom Pingdom Monitoring Dyn Dyn DNS Sentry Sentry Error Logging CloudAMQP CloudAMQP RabbitMQ Heroku Heroku PaaS Kabu Creative Kabu Creative UX & Design Fastly Fastly CDN DigiCert DigiCert EV Certificate Google Google Cloud Servers