jyson

h1. How jyson maps JSON data to jython data

When decoding a JSON document, jyson maps JSON data types and constants to jython data types and objects as described below.

h2. JSON null

The JSON constant *null* is decoded as a jython *None* (specifically, the jython singleton object *Py.None* which is of type *org.python.core.PyObject*).

<pre>

>>> from com.xhaus.jyson import JysonCodec as json

>>> json.loads('[null]')

[None]

>>>

</pre>

h2. JSON true and false.

The JSON constants *true* and *false* are decoded as jython *True* and *False* respectively (specifically as the jython singleton objects *Py.True* and *Py.False* respectively).

<pre>

>>> from com.xhaus.jyson import JysonCodec as json

>>> json.loads('[true,false]')

[True, False]

>>>

</pre>

h2. JSON integers.

JSON integers are decoded as either a jython *int* (*org.python.core.PyInteger*) or *long* (*org.python.core.PyLong*)

<pre>

>>> from com.xhaus.jyson import JysonCodec as json

>>> json.loads('[42]')

[42]

>>>

</pre>

If an JSON integer is greater than the maximum value representable by a jython integer, then it will instead be returned as a jython long. The following illustrates

<pre>

>>> import sys

>>> json_ints = "[%d,%d]" % (sys.maxint, sys.maxint+1)

>>> json_ints

'[2147483647,2147483648]'

>>> json.loads(json_ints)

[2147483647, 2147483648L]

>>>

</pre>

h2. JSON floats

JSON floats are decoded as jython *float* (*org.python.core.PyFloat*)

<pre>

>>> from com.xhaus.jyson import JysonCodec as json

>>> json.loads('[42.0]')

[42.0]

>>>

</pre>

h2. JSON strings

JSON strings are decoded as jython *strings* (*org.python.core.PyString*)

<pre>

>>> from com.xhaus.jyson import JysonCodec as json

>>> json.loads('["Hello World!"]')

['Hello World!']

>>>

</pre>

h2. JSON objects

JSON objects are decoded as jython *dictionaries* (*org.python.core.PyStringMap*)

<pre>

>>> from com.xhaus.jyson import JysonCodec as json

>>> json.loads('{"hello": 1, "world": 2.0}')

{'world': 2.0, 'hello': 1}

>>>

</pre>

*NOTE* Only objects with string keys can be decoded, as according to the JSON spec. Any attempt to decode an object with a non-string key will cause a JSONDecodeError. The values in an object can be any JSON type, including arrays and objects.

h2. JSON arrays

JSON arrays are decoded as jython *lists* (*org.python.core.PyList*)

<pre>

>>> from com.xhaus.jyson import JysonCodec as json

>>> json.loads('[1, 42, "Hello world!"]')

[1, 42, 'Hello world!']

>>>

</pre>

Array elements may be of any JSON type, including arrays and objects.