pygvariant 0.2.0

A library to serialize and deserialize gvariant values in gschema files.

GVariant Python Utilities

This project provides a comprehensive toolkit for working with GVariant type strings and value representations within Python. It allows for the bidirectional conversion between GVariant type signatures, GVariant text formats (gschema compatible), and native Python objects.

---

Features

• Type Parsing: Converts GVariant type strings (e.g., a{sv}) into standard Python typing hints.
• Value Conversion: Parses GVariant text literals into coerced Python types based on a provided signature.
• GSchema Serialization: Serializes Python objects into strings compatible with GVariant text format and glib-compile-schemas.
• Recursive Support: Handles nested containers including arrays, tuples, dictionaries, and "maybe" types up to the GVariant limit of 65 levels.

---

Core Components

1. GVariantParser

The parser implements a recursive descent strategy to interpret GVariant type signatures. It maps basic GVariant characters to Python types and handles complex container definitions.

GVariant Character	Python Type	Description
b	bool	Boolean
s, o, g	str	String, Object Path, or Signature
y, n, i, x	int	Various width signed integers
q, u, t	int	Various width unsigned integers
d	float	Double precision floating point
a	List[T]	Array of type T
m	Optional[T]	Maybe/Nullable type T
()	Tuple[...]	Fixed-size tuple
{}	Dict[K, V]	Dictionary entry (usually within an array)
v, *, ?	Any	Indefinite or variant types

2. GVariantValueConverter

This utility takes a string representation of a value and a GVariant type string. It uses ast.literal_eval for safe structural parsing and then recursively coerces the data into the target Python types. It specifically handles GVariant-specific keywords like nothing for null values and true/false for booleans.

3. GVariantSerializer

The serializer ensures that Python data structures are converted back into strings that the gschema compiler can understand. This includes converting Python's True to true, None to nothing, and ensuring correct bracket usage for arrays and dictionaries.

---

Usage Examples

Parsing Type Strings

from gvariant_utils import GVariantParser

parser = GVariantParser()
python_type = parser.parse("a{si}") 
# Result: typing.Dict[str, int]

Converting String Literals to Python Objects

from gvariant_utils import GVariantValueConverter

converter = GVariantValueConverter()
value = converter.parse_value_string("[1, 2, 3]", "ai")
# Result: [1, 2, 3] (as a list of integers)

maybe_val = converter.parse_value_string("nothing", "ms")
# Result: None

Serializing to GSchema Format

from gvariant_utils import GVariantSerializer

data = {"enabled": True, "timeout": 30}
gschema_string = GVariantSerializer.serialize(data)
# Result: '{"enabled": true, "timeout": 30}'

---

Technical Specifications

• Nesting Limit: Following the D-Bus and GVariant specifications, this implementation supports a recursion depth of up to 65 levels.
• Basic Type Restriction: In accordance with the GVariant spec, dictionary keys (the first element in {}) are restricted to basic types to ensure hashability.
• Indefinite Types: Types such as v (variant) and * (any) are mapped to Python's Any type, as their specific structure is determined at runtime rather than by the signature.

---

Implementation Notes

The implementation relies on the Python typing module for type representation and the ast module for secure evaluation of string literals. It avoids the use of eval(), making it safe for use with untrusted input strings.