The Safety Net

Programmers spend more time reading code than writing it (a fact well known by most programmers who tend not to publicise this to their employers).  It therefore stands to reason that (most?) programming languages should be designed as much for human consumption as for machine consumption and should be as readable as possible.

Python is a very readable language (a fact which contributes to its popularity) and has been termed “executable pseudocode” on account of its readability.  An aspect of Python which makes it readable is its avoidance of syntactic fluff, extraneous words and symbols which add nothing to the code’s meaning but serve to detract from it.

In the past I’ve felt somewhat negative about Python’s terminal colon “:”, the symbol used to terminate if, while, def and class statements and to signify the start of a new block of indented code.  For example:

if a == 1:
    b = do_something_cool()

def do_something_cool():
    return 'Doing something cool'

Even without the colon, it’s quite clear that we’re starting a new block of indented code because (a) the statement starts with the keyword if, while, def or class and (b) the next line of code is indented. For comparison, Ruby gets on just fine without the colon after its def statement. So why the need for a colon in Python? Is it syntactic fluff?

The Python FAQ explains that the colon enhances readability and helps editors with syntax highlighting and code indentation. Lets face it, any self respecting editor should be capable of parsing a line beginning with an if, while, def or class, so the “helps editors” argument is bogus. I do however buy the argument that the code is visibly more readable. But how does it enhance readability?

I’ve already mentioned that a programmer spends more time reading than writing code. What I haven’t yet suggested is that a programmer will often reread and scan the same code repeatedly to form a mental picture of a larger codebase. It’s what the eyes do when they’re scanning code that’s key to the importance of the colon. There is some evidence to suggest that the eyes linger at the beginning and at the end of a sentence when reading text and draw especially from visual cues at those locations. Let’s assume for the moment that this holds true for a line of code. So the visual cue heralding an indented block of code is clear at the beginning of a line of code, namely an if, while, def or class followed by an indented line. The only visual cue at the end of a line of Python code is the colon, and without the colon there would be no cue. So even though the colon is not strictly necessary, there is an argument that its existence is there for human consumption and aids readability.

When all’s said and done, the advantage of the colon is probably slight at best, and then probably only for a newcomer to the language. (This sort of advantage possibly completely vanishes for experienced users of any language). Never-the-less, on balance, I’m now happy it’s there!

safe :: 10 Aug 2011 :: Posted in Programming :: :: No Comments »
Tags: python

A recent thread on the Python Northwest mailing list asked for opinions on Marty Alchin‘s book Pro Python.  I thought I’d reproduce the answer I gave and expand on it a little.

I’ve owned Marty Alchin’s first book, Pro Django, for some time and was very happy with that purchase.  Based on that, I decided to buy his Pro Python book last year.  Pro Python is targeted at readers who are proficient with basic Python but are looking to push their skills further.  Quite naturally there’s a large number of beginners’ Python books out there but a shortage of more advanced books so it was nice to see this published.

Marty Alchin starts his book with a refreshing approach.  Rather than regurgitating Python facts to the reader, he takes a step by step tour of The Zen of Python discussing how it’s philosophy can be practically applied to make your programming more Pythonic.  He then delves into traditional topics such as classes, objects and strings as well as development topics such as packaging and testing.

I like Marty Alchin’s style of writing and find it to be clear and concise.  Even if you’re reasonably knowledgeable about the advanced topics he covers such as metaclasses, descriptors, introspection and multiple inheritance, I think the book benefits from the fact that these topics are backed up with good examples of how they work, and just as importantly, how they might usefully be used in ways you might not have seen before.  In fact, Chapter 11 walks through the building of a real world Python library which can be found on PyPI (try pip install Sheets) using the principles outlined in the previous chapters.

The other aspect of the book I find very useful is the fact that it is based on Python 3, however all examples are annotated and compared with the “legacy” Python 2 equivalent where relevant.  I’ve gotten a lot more comfortable with Python 3 by reading this book and better understand the improvements in the language from Python 2 to Python 3.

This isn’t a book aimed at newcomers to Python, even if you have a lot of programming experience, as it expects a reasonable amount of basic Python proficiency.  It’s also a “thin” book in the sense that it gives each topic a light treatment rather than aiming to be a complete reference.  This may or may not suit your needs, however there’s plenty of reference material elsewhere both online (e.g. the official Python documentation) and in print.

By comparison, the other advanced Python book I’ve read (and reread!) is Python In a Nutshell by Alex Martelli.  It’s based on Python 2.5 and getting a bit out of date, but much of it is still very relevant for all Python 2.x versions.  (I think a Python 3 version might be in the works).  It’s a much heftier and more detailed book and acts as much a reference text as well as being a book you’d enjoy reading from cover to cover.

In summary, I’d recommend Pro Python to any intermediate level Python programmer who’d like to advance their Python skills with a clear and concise text.

N.B. I am in no way associated with Pro Python, Apress or Marty Alchin … except of course for owning the book!

safe :: 14 Mar 2011 :: Posted in Programming :: :: No Comments »
Tags: book, python, python3

Over the last few days I’ve been playing with Hessian, “a compact binary protocol for connecting web services”. In my previous company we used Hessian extensively for communicating between a Java thick client and a Java Apache Tomcat HTTP server with good success. These days we talk of JSON and REST and peer our noses down at thick clients so Hessian might seem irrelevant, however around the time we were implementing our client-server communications (2004 / 2005), we were bathing in the waters of SOAP, WSDL and so-called heavyweight web services. The beauty of Hessian was our ability to take our Plain Old Java Objects which we had already implemented on our thick client and send them down the pipe unchanged to our server. Hessian took care of the marshalling and unmarshalling of data. In fact, because we took advantage of Hessian integration with the Spring Framework, a declarative application framework which encourages defining objects and their relationships and dependencies in configuration files, all it took was a bit of code and a bit of configuration to get everything working.

So does it now make more sense to use JSON / REST? One of the advantages of JSON / REST includes the inherent decoupling of client and server. The client fires a JSON string to the server at the correct URL using an HTTP POST and the server parses what it needs from that string and happily replies. This process is platform agnostic as HTTP and JSON libraries are available for many programming languages and platforms, not least including Javascript in the web browser. This model is widely used by service providers such as Google and Amazon whereby they can provide and update REST interfaces to their services without having to deliver and maintain multiple API client libraries. A drawback of this model is the need to hand code the marshalling and unmarshalling of JSON data by both client and server, though this can also be seen as an advantage as it decouples an application’s internal representation of data from the wire format.

Hessian compares well with the JSON / REST model. Hessian is also designed around HTTP POST whereby a client connects to a URL on the server and sends data, however Hessian goes one step further and encodes an RPC call i.e. a function name and arguments. In fact the Hessian library makes this process transparent by proxying the server i.e. it provides an object on which the client makes function calls without knowing that the call will be sent to a server. Note that there is no “contract” or abstract interface which you are forced to code to – client and server ensure they’re sending and receiving the correct function arguments by “unwritten agreement” much like the JSON / REST model. Unlike JSON, Hessian is a binary protocol meaning that the data exchanged between client and server is very compact. It also encodes type information, in fact, entire object structures are maintained when unmarshalled on either client or server. Hessian is also cross platform and libraries exist for many programming languages including Javascript.

So what’s not to like?  Well binary communication and the concept of RPC function calls in general seems to have gained a bad reputation, possibly due to the extra complexity and library support needed over simple JSON / REST and possibly because of the increased coupling an RPC call implies.  Experience at my previous company taught us that the communication can be little brittle if the definitions of objects sent over the wire are not kept in step on both client and server. If an object sent from the client to the server has an extra unknown field, there will be an error when the Hessian library on the server tries to unmarshall that data to create an object.  (The reverse, however, is not true – any fields missing from data over the wire will simply end up unset on the unmarshalled object).

Passing JSON over HTTP is much more forgiving in that the client or server will blissfully ignore any field it doesn’t know how to handle, though of course if a field that is expected is not found, the server must know handle that.  Ordinarily, keeping the client and server in step shouldn’t be a problem, however we had many clients in the field with different versions of our software all connecting to the same server.

It has only recently occurred to me that the brittleness described above is peculiar to statically typed languages such as Java where an Exception is thrown at any attempt to apply a value to a field where that field not been defined in an object’s class. The same is not true of dynamically typed languages such as Python which is forgiving when applying values to arbitrary fields on an object. For many years, hessianlib.py has been the standard Python implementation of Hessian. It has been little unmaintained over that time and includes a Hessian client implementation but no Hessian server implementation. The code is also a little impenetrable. Happily, earlier this year a fork of hessianlib.py called Mustaine has appeared. It doesn’t (yet) contain a server implementation, but the code is more penetrable so I submitted a patch with an implementation of a Hessian WSGI server.

Let’s see some code based on the proposed mustaine.server module. (Please note that Mustaine server support is in flux so this example is subject to change). An object can be served via WSGI by wrapping it with mustaine.server.WsgiApp. An object’s methods are only exposed if decorated with the mustaine.server.exposed decorator. For example:

from mustaine.server import exposed

class Calculator(object):
    @exposed
    def add(self, a, b):
        return a + b

    @exposed
    def subtract(self, a, b):
        return a - b

The following code will serve a Calculator() object on port 8080 using the Python reference WSGI server:

from wsgiref import simple_server
from mustaine.server import WsgiApp
s = simple_server.make_server('', 8080, WsgiApp(Calculator()))
s.serve_forever()

This object can now be accessed over the network using the Hessian client:

>>> from mustaine.client import HessianProxy
>>> h = HessianProxy('http://localhost:8080/')
>>> h.add(2, 3)
5

As a result of providing server support to Mustaine, I’ve started developing django-hessian, a library which serves Hessian objects in Django. Objects can be served using djangohessian.Dispatcher at a given URL with an entry in urls.py. The Calculator() object described above can be served at the URL http://localhost:8000/rpc/calculator/ in the Django development server as follows:

# mysite/urls.py:

from django.conf.urls.defaults import *

urlpatterns = patterns('',
    (r'^rpc/', include('mysite.myapp.urls')),
)

# mysite/myapp/urls.py:

from django.conf.urls.defaults import *
from djangohessian import Dispatcher
from server import Calculator

urlpatterns = patterns('',
    url(r'^calculator/', Dispatcher(Calculator())),
)

Full source can be found at http://bitbucket.com/safehammad/django-hessian/.

I can’t help wondering whether the Hessian protocol is getting attention it deserves, particularly in environments where both client and server are delivered and maintained by a single provider. Have you implemented JSON / REST systems which would have benefited from using Hessian? Do you have good arguments as to why the use of Hessian is to be discouraged?

safe :: 16 Oct 2010 :: Posted in Programming :: :: No Comments »
Tags: hessian, java, json, python, rpc

I’m currently involved in PyWeek, a game writing competition where entrants are given exactly one week to write a game from scratch in Python.  To be honest I’m not much of a gamer, but I couldn’t resist the creative challenge of writing a playable game in one week and I’m loving every minute of it.  As the front page of http://www.pyweek.org states:

The PyWeek challenge:

1. Invites entrants to write a game in one week from scratch either as an individual or in a team,
2. Is intended to be challenging and fun,
3. Will hopefully increase the public body of game tools, code and expertise,
4. Will let a lot of people actually finish a game, and
5. May inspire new projects (with ready made teams!)

This is the eleventh iteration of PyWeek and the first one I’ve entered.  The theme for this iteration is the word ‘caught’.  My entry is called Superfly Funky Stuff though the game itself is really called Voices Under Water.  I’ll write more about what I’m coding in due course, but here’s a sneak preview:

Voices Under Water

And here’s a picture of my other half, Annie, recording splashing sounds in the bath for some DIY sound effects!

safe :: 27 Aug 2010 :: Posted in Programming :: :: 2 Comments »
Tags: annie, games, python, pyweek

My 10 year old nephew, Bob, loves computer games. In fact, I think it would be fair to say that most self respecting 10 year olds love computer games and wonder what else a computer is for apart from computer games, Facebook, Bebo, and the occasional use of that legacy communications medium called email.

My nephew, Bob!

There was a brief period in the 1980′s as the home PC revolution was picking up speed when we thought we would all have to learn how to program a computer. Computers were taking over the workplace and without this skill we would all be unemployable. That brief period ended with the rise of the office suite and the advent of email. For many jobs we’re now expected to have familiarity with word processors, spreadsheets, email and web browsing. You only really have to know how to program if you’re a programmer, whether hobbyist or professional, but that goes without saying. One could argue that the evolution of our technological society where programming is mainly restricted to dedicated programmers is a natural division of labour. Why should a secretary or an advertiser or a salesperson need to program? Why would a 10 year old need to program?

Bob can often be found discussing the merits of the computer games he plays and how they might be altered or improved. For example, earlier this week he described how he’d like to play a game which was a first person shooter where the enemy stood still and only fired back several seconds after being discovered. The game description was accompanied with much hand waving and shooting noises. There were further descriptions of the types of guns that should be available and when they should be made available to the player.

Bob’s vision of a new computer game was born from a creative process. What Bob needs now is a medium of creative expression to bring that vision to life. In the same way a musical instrument can be used to create an infinite variety of music, or a word processor can be used to create an infinite variety of novels, it is through programming that you can create an infinite variety of games, and programming is the medium of creative expression.

Michael Sparks asked the question: If you were 7 again, what would you expect to find in a book on beginning programming? Apart from the usual conditionals and loops (I can hear the writer of the fictional Functional Programming for 7 Year Olds groaning) it would make sense to fulfil our junior members of society’s primal need to play, adapt and create games.

Armed with the knowledge that learning to program was the path to learning to create computer games, Bob was all too happy to sit down with me and walk through some basic Python programming. We started with a simple quiz game. The first decision was to whether to choose Python 2 or Python 3. I figured that extensive library support wasn’t necessary for writing a quiz so I plumped for Python 3.  With Python 3, you also get to avoid being asked sticky questions such as:

Why is it called raw_input() and not input()?

…  and …

Why does raw_input() have brackets and print doesn’t?

The quiz went down very well and we got to touch on several Python constructs. Here’s a cut down version of what we wrote together:

score = 0

fvcolour = input('What is my favourite colour? ')
if fvcolour == 'red' or fvcolour == 'blue':
    print('Correct!')
    score = score + 1
else:
    print('Incorrect!')

sport = input('What is my favourite sport to watch? ')
if sport == 'football' or sport == 'Football':
    print('Correct!')
    score = score + 1
else:
    print('Incorrect!')

print('You scored', score, 'out of 2')

Bob was very excited by this.  He was particularly excited about the fact that he could get the computer to ask any question and respond to any answer in any way he wanted it to.  After helping him install Python 3 on his own computer and showing him how to use IDLE, he sat about creating more intriguing and inventive quizzes.

I was conscious of the fact that I only see Bob once in a while so I set about looking for an online tutorial he could follow.  I was also conscious of the fact that he really wanted to write games, specifically first person shooters with advanced sound and graphics!  A few weeks ago I attended the informative and entertaining tutorial Introduction To Game Programming given by Richard Jones at Europython.  Richard pointed us to Invent With Python, an online (and dead tree) book “written to be understandable by kids as young as 10 to 12 years old” teaching them how to program games using PyGame.  And it has been updated to Python 3.  Perfect.  Bob is busy working his way through the book and I can’t wait to play Bob’s first person shooter!

How would you introduce your 10 year old nephew / niece / daughter / son to programming?

safe :: 13 Aug 2010 :: Posted in Programming :: :: 1 Comment »
Tags: games, python, python3