Serialization

Bag supports multiple formats for saving and loading data:

Format	Best For
XML	Human-readable configs, interop
JSON	Web APIs, simple data
TYTX	Full type preservation

XML

Writing

>>> from genro_bag import Bag

>>> bag = Bag()
>>> bag['name'] = 'Alice'
>>> bag['age'] = 30

>>> bag.to_xml()
'<name>Alice</name><age>30</age>'

With formatting:

>>> from genro_bag import Bag

>>> bag = Bag()
>>> bag['config.debug'] = True
>>> bag['config.port'] = 8080

>>> print(bag.to_xml(pretty=True))
<config>
  <debug>True</debug>
  <port>8080</port>
</config>

With XML declaration:

>>> from genro_bag import Bag

>>> bag = Bag({'test': 'value'})
>>> bag.to_xml(doc_header=True)
"<?xml version='1.0' encoding='UTF-8'?>..."

Reading

>>> from genro_bag import Bag

>>> xml = '<root><name>Test</name><count>42</count></root>'
>>> bag = Bag.from_xml(xml)
>>> bag['root.name']
'Test'

Attributes in XML

>>> from genro_bag import Bag

>>> bag = Bag()
>>> bag.set_item('user', 'Alice', role='admin', active='true')

>>> bag.to_xml()
'<user role="admin" active="true">Alice</user>'

Limitations

XML doesn’t preserve Python types — everything becomes a string:

>>> from genro_bag import Bag

>>> bag = Bag({'count': 42, 'active': True})
>>> restored = Bag.from_xml(f'<root>{bag.to_xml()}</root>')
>>> restored['root.count']
'42'
>>> type(restored['root.count'])
<class 'str'>

JSON

Writing

>>> from genro_bag import Bag

>>> bag = Bag({'name': 'Alice', 'age': 30})
>>> bag.to_json()
'[{"label":"name","value":"Alice","attr":{}},{"label":"age","value":30,"attr":{}}]'

Reading

>>> from genro_bag import Bag

>>> json_str = '{"name": "Test", "value": 42}'
>>> bag = Bag.from_json(json_str)
>>> bag['name']
'Test'
>>> bag['value']
42

Limitations

Standard JSON doesn’t preserve:

Node attributes
Complex types (datetime, Decimal, bytes)

TYTX (Typed Exchange)

TYTX preserves Python types exactly. Two transports available:

JSON (default): Human-readable
MessagePack: Compact binary

Writing

>>> from genro_bag import Bag
>>> from decimal import Decimal

>>> bag = Bag()
>>> bag['count'] = 42
>>> bag['price'] = Decimal('19.99')
>>> bag['active'] = True

>>> tytx = bag.to_tytx()  # JSON transport
>>> mp = bag.to_tytx(transport='msgpack')  # Binary
>>> type(mp)
<class 'bytes'>

Reading

>>> from genro_bag import Bag
>>> from decimal import Decimal

>>> bag = Bag()
>>> bag['price'] = Decimal('19.99')
>>> tytx = bag.to_tytx()

>>> restored = Bag.from_tytx(tytx)
>>> restored['price']
Decimal('19.99')
>>> type(restored['price'])
<class 'decimal.Decimal'>

Full Type Preservation

>>> from genro_bag import Bag
>>> from decimal import Decimal

>>> original = Bag()
>>> original['count'] = 42
>>> original['price'] = Decimal('19.99')
>>> original.set_item('user', 'Alice', role='admin')

>>> restored = Bag.from_tytx(original.to_tytx())

>>> restored['count']
42
>>> type(restored['count'])
<class 'int'>
>>> restored['price']
Decimal('19.99')
>>> restored['user?role']
'admin'

Supported Types

Type	Example
`int`	`42`
`float`	`3.14`
`Decimal`	`Decimal('19.99')`
`bool`	`True`
`None`	`None`
`str`	`'hello'`
`bytes`	`b'\x00\x01'`
`datetime`	`datetime(2025, 1, 1)`
`date`	`date(2025, 1, 1)`
`time`	`time(12, 30)`
`list/tuple`	`[1, 2, 3]`

File Operations

Save to File

# XML
bag.to_xml('/path/to/file.xml')

# TYTX JSON
bag.to_tytx('/path/to/data.bag.json')

# TYTX MessagePack
bag.to_tytx('/path/to/data.bag.mp', transport='msgpack')

Load from File

# Auto-detected from extension
bag = Bag()
bag.fill_from('/path/to/data.bag.json')
bag.fill_from('/path/to/data.bag.mp')

Format Comparison

Feature	XML	JSON	TYTX
Human readable	✓	✓	JSON: ✓
Type preservation	✗	Partial	✓
Attributes	✓	✗	✓
Binary data	✗	✗	✓
File size	Large	Medium	Small (MP)

Best Practices

Use Case	Format
Configuration files	XML or JSON
Data exchange with types	TYTX JSON
Storage/cache	TYTX MessagePack
Web APIs	JSON