jyson

h1. Jyson FAQ

{{toc}}

h2. Why use Jyson instead of a java codec?

The advantage of using jyson instead of a java codec is that jyson produces a hierarchy of jython objects, namely "org.python.core.PyObject":http://www.jython.org/docs/javadoc/org/python/core/PyObject.html . These objects are custom designed to be manipulated by jython code. 

On the other hand, a java codec might produce a "java.util.Map":http://java.sun.com/j2se/1.5.0/docs/api/java/util/Map.html or "java.util.List":http://java.sun.com/j2se/1.5.0/docs/api/java/util/List.html or other java objects, which it is possible to manipulate from jython code, but would not be natural jython objects.

Consider the following

<pre>

Jython 2.5b2+ (trunk, Mar 8 2009, 20:40:45)

[Java HotSpot(TM) Client VM (Sun Microsystems Inc.)] on java1.5.0_17

Type "help", "copyright", "credits" or "license" for more information.

>>> from com.xhaus.jyson import JysonCodec

>>> l = JysonCodec.loads('[3,1,4,1,5,9,2,7]')

>>> type(l)

<type 'list'>

>>> l

[3, 1, 4, 1, 5, 9, 2, 7]

>>> l.count(1)

2

>>> import java

>>> jl = java.util.ArrayList([3,1,4,1,5,9,2,7])

>>> type(jl)

<type 'java.util.ArrayList'>

>>> jl

[3, 1, 4, 1, 5, 9, 2, 7]

>>> jl.count(1)

Traceback (most recent call last):

  File "<stdin>", line 1, in <module>

AttributeError: 'java.util.ArrayList' object has no attribute 'count'

>>>

</pre>

h2. Why use jyson instead of a python codec?

Simple answer: speed.

When you run a pure python program under jython, there is an "interpretive overhead":http://en.wikipedia.org/wiki/Interpreter_(computing)#Efficiency which makes it likely that the pure python code will run slower, and use more memory, than the equivalent code written in java (although "advances in JVM optimization":http://blogs.sun.com/jrose/entry/bravo_for_the_dynamic_runtime mean that this performance difference is not guaranteed).

Jyson is written in pure java, and is thus highly likely to be significantly faster and more resource efficient than a pure python codec.