Help Page

Revision Control

Any time the code or inputs on a utility get changed, a new revision is created. If you don't specify a revision when editing or running a utility, the latest revision is used. Changing the description or writeup won't create a new revision.

You can view a list of revisions and compare revisions on the utilities' revisions page.

A sample revision page

Comparing two revisions

API

Every utility has a RESTful XML and JSON API for your pleasure. You can also get a listing of all of the utilities.

Learn more about the API

Available Libraries for Your Code

You can import any library included with Python 2.5 for Linux.

Additionally for your programming pleasure you also have these third party libraries available for use in your utilities:

BeautifulSoup

Beautiful Soup(Home/Documentation) is a Python HTML/XML parser designed for quick turnaround projects like screen-scraping. Most importantly it won't choke if you give it bad markup.

Crypto (Python Cryptography Toolkit) and ezPyCrypto

The Python Cryptography Toolkit(Home/Documentation) is a collection of cryptographic algorithms and protocols, implemented for use from Python.

ezPyCrypto(Home/Documentation) is a wrapper around Crypto which makes its more common functions MUCH easier to use.

dateutil

The dateutil module(Home/Documentation) provides powerful extensions to the standard datetime module such as generic parsing of dates in almost any string format, and relative time deltas for units such as weeks and months.

PyFacebook

PyFacebook(Home/Documentation) is a thin wrapper for accessing Facebook's RESTful API through Python.

Universal Feed Parser

Parse RSS and Atom feeds in Python.(Home/Documentation)

GData Python Client Library

The Google data Python Client Library(Home/Documentation) provides a library and source code that make it easy to access data through Google Data APIs.

NumPy

NumPy(Home/Documentation) includes a powerful N-dimensional array object, basic linear algebra functions, basic Fourier transforms, and more.

PIL (Python Imaging Library)

The Python Imaging Library(Home/Documentation) adds image processing capabilities to your Python interpreter. This library supports many file formats, and provides powerful image processing and graphics capabilities.

pyExcelerator

Generate Excel files directly from Python.

Home/Documentation

A Useful Example

Pyparsing

The Pyparsing module(Home/Documentation) is an alternative approach to creating and executing simple grammars, vs. the traditional lex/yacc approach, or the use of regular expressions. The pyparsing module provides a library of classes that client code uses to construct the grammar directly in Python code.

pyPdf

pyPdf(Home/Documentation) is a is a PDF toolkit capable of extracting document information (title, author, ...), splitting documents page by page, merging documents page by page, cropping pages, merging multiple pages into a single page, and encrypting and decrypting PDF files. Also see the Reportlab library for PDF generation.

PyRSS2Gen

Beautiful Soup(Home/Documentation) is a Python HTML/XML parser designed for quick turnaround projects like screen-scraping. Most importantly it won't choke if you give it bad markup.

ReportLab (PDF Generation)

The ReportLab PDF library(Home/Documentation) is a proven industry-strength PDF generating solution, that you can use for meeting your requirements and deadlines in enterprise reporting systems.

Reverend Bayesian Classifier,

Reverend(Home/Documentation) is a general purpose Bayesian classifier, Use the Reverend to quickly add Bayesian smarts to your app. Bayesian fun has never been so quick and easy.

SciPy

Scientific, statistical, and machine learning functions for use in Python. (Home/Documentation)

simplejson

simplejson(Home/Documentation) is a simple, fast, extensible JSON encoder/decoder for Python.

Python Excel Tools: xlrd, xlwt, xlutils

These tools (Home/Documentation) let you read, write, and manipulate Excel files. They're a huge improvement over pyExcelerator.

PyYAML

PyYAML(Home/Documentation) is a YAML parser and emitter for the Python programming language.

Biopython

Biopython(Home/Documentation) is a set of freely available tools for biological computation written in Python.

FAQ

How do I make a utility output a file such as an image or a spreadsheet?

Normally all of the printed output from a utility will end up in the output text area on the utility's page. To return a file instead, simply make your first print statement provide a content type. For example to return a PNG image file, the first print statement run in your utility should be:

print 'Content-Type: image/PNG'

All of your print statements after that would make up the contents of that image.

If it's still not clear, here is a working example of a utility that outputs an image file.

What file formats can I return?

Anything you want! Here's a handy list of common content/MIME types.

What if my utility needs to store a static file?

Sometimes a utility will need to access a static file to do its work. For example the keyboard heatmap utility needs an image of a keyboard which it can mark up with the heat map data.
The way to do this is to store the file right in your code. Use this utility to get the code for your file and follow the instructions on that page to use the encoded file in your code. (I don't recommend doing this for a file larger than say 500KB.)

I have an idea for a utility but I don't want to program

You can request a utility here.

I want to make a utility but I don't have any ideas

You see requested a utilities here.

How can I suggest changes to the site or view planned changes?

You can view and vote on Utility Mill enhancements here.

How does the Python code get executed safely?

This is by far the most popular question. Running Python code somewhat securely is no easy feat. Here are some more resources on this topic if you're curious.

Here is the basic idea for security on Utility Mill:

I have a Python daemon that maintains N child processes.
Each child process runs in a chroot jail and executes the Python code sent to it.
If a child process hits an error, runs too long, or uses too much memory the daemon will kill it and start a new child process to replace it.

The website code (web.py stuff, etc) sends code to be executed to the daemon over a socket and gets back the printed result, or an error message.

The current limitations on utilities are 2 real seconds of run time, and a memory limit of 28MB (including the interpreter session).
I'm considering offering premium accounts with higher limits to offset the cost. Email me if you're interested in purchasing one. If I hear from over 20 people I'll definately do it.

Of course I'm no security expert, so I'm still waiting for someone to find a hole in the design :-) Best to email me privately if you find one.

Have a frequently asked question?

If so, kindly Ask me.