Working with the data model

The Message class has a datamodel property which gives you access to the message data model which is based on the MF2 spec. The data model nodes are documented here.

There are two classes to work with the datamodel - DataModelVisitor & DataModelTransformer.

Inspecting the data model

All the data model nodes have implemented __str__ and __repr__ for easier debugging and inspection.

from messageformat2 import Message
from messageformat2.datamodel import dump

msg = Message("Hello, {$name}!")
dump(msg.datamodel)  # -> 'Hello, {$name}!'
# Same as calling .dump()
str(msg.datamodel)  # -> 'Hello, {$name}!'
repr(msg.datamodel)  # -> "PatternMessage(declarations=[], pattern=['Hello, ', VariableExpression(arg=VariableRef(name='name'), annotation=None, attributes=[]), '!'])"

Walking the data model

Here is how you can walk the parsed data model syntax tree:

from messageformat2 import Message
from messageformat2.datamodel import DataModelVisitor

class VariableVisitor(DataModelVisitor):
    def visit_VariableRef(self, node):
        print(f"${node.name}")
        self.generic_visit(node)

message = Message("Hello, {$name}!")
visitor = VariableVisitor()
visitor.visit(message.datamodel)
# -> "$name"

Transforming the data model

Use the transformer class to modify the data model:

from messageformat2 import Message
from messageformat2.datamodel import Literal, DataModelTransformer

class LiteralTransformer(DataModelTransformer):
    def visit_Literal(self, node):
        return Literal(node.value.upper())

message = Message("{|foo|}")
ast = LiteralTransformer().visit(message.datamodel)
str(ast)
# -> "{|FOO|}"