PYSON is the PYthon Statement and Object Notation. It is a lightweight domain specific language for the general representation of statements. PYSON is used to encode statements which can be evaluated in different programming languages, serving for the communication between trytond and any third party software. A PYSON parser can easily be implemented in other programming languages. So third party softwares do not need to depend on Python to be able to fully communicate with the Tryton server.
PYSON is a deterministic algorithm which will always succeed to evaluate statements. There is a default behavior for unknown values. It is statically typed and checked on instantiation.
There is also a reference documentation of the API.
The syntax of a PYSON statement follows this pattern:
Statement(argument1[, argument2[, ...]])
where arguments can be another statement or a value. The evaluation direction is inside out, deepest first.
Given the PYSON statement:
Eval('active_id', -1)
Eval() checks the evaluation context for the variable active_id
and returns its value or -1 if not defined. A similar expression
in Python looks like this:
'active_id' in locals() and active_id or -1
Given the PYSON statement:
Not(Bool(Eval('active')))
Eval() checks the evaluation context for a variable active and
returns its value to Bool() or '' if not defined. Bool()
returns the corresponding boolean value of the former result to Not().
Not() returns the boolean negation of the previous result. A similar
expression in Python looks like this:
'active' in locals() and active == False
Given the PYSON statement:
Or(Not(Equal(Eval('state'), 'draft')), Bool(Eval('lines')))
In this example are the results of two partial expressions
Not(Equal(Eval('state'), 'draft')) and Bool(Eval('lines'))
evaluated by a logical OR operator. The first expression part is
evaluated as follow: When the value of Eval('state') is equal to
the string 'draft' then return true, else false. Not() negates
the former result. A similar expression in Python looks like this:
'states' in locals() and 'lines' in locals() \
and state != 'draft' or bool(lines)
Given the PYSON statement:
If(In('company', Eval('context', {})), '=', '!=')
In this example the result is determined by an if-then-else condition.
In('company', Eval('context', {})) is evaluated like this: When
the key 'company' is in the dictionary context, returns
true, otherwise false. If() evaluates the former result and returns
the string '=' if the result is true, otherwise returns the
string '!='. A similar expression in Python looks like this:
'context' in locals() and isinstance(context, dict) \
and 'company' in context and '=' or '!='
Given the PYSON statement:
Get(Eval('context', {}), 'company', 0))
Eval() checks the evaluation context for a variable context if
defined, return the variable context, otherwise return an empty
dictionary {}. Get() checks the former resulting dictionary
and returns the value of the key 'company', otherwise it returns
the number 0. A similar expression in Python looks like this:
'context' in locals() and context.get('company', 0) or 0