tinydb-serialization 2.1.0

Serialization for objects that TinyDB otherwise couldn't handle

tinydb-serialization provides serialization for objects that TinyDB otherwise couldn’t handle.

Usage

General Usage

To use a serializer, create a SerializationMiddleware instance with the storage class you want to use and register the serializers you want to use. Then you pass the middleware instance as the storage to TinyDB:

>>> from tinydb import TinyDB, Query
>>> from tinydb_serialization import SerializationMiddleware
>>> from tinydb_serialization.serializers import DateTimeSerializer
>>>
>>> from datetime import datetime
>>>
>>> serialization = SerializationMiddleware(JSONStorage)
>>> serialization.register_serializer(DateTimeSerializer(), 'TinyDate')
>>>
>>> db = TinyDB('db.json', storage=serialization)
>>> db.insert({'date': datetime(2000, 1, 1, 12, 0, 0)})
>>> db.all()
[{'date': datetime.datetime(2000, 1, 1, 12, 0)}]
>>> query = Query()
>>> db.insert({'date': datetime(2010, 1, 1, 12, 0, 0)})
>>> db.search(query.date > datetime(2005, 1, 1))
[{'date': datetime.datetime(2010, 1, 1, 12, 0)}]

Provided Serializers

• tinydb_serialization.serializers.DateTimeSerializer: serializes datetime objects as ISO 8601 formatted strings

Creating Custom Serializers

In this example we implement a serializer for datetime objects (like the one provided by this package):

from datetime import datetime
from tinydb_serialization import Serializer

class DateTimeSerializer(Serializer):
    OBJ_CLASS = datetime  # The class this serializer handles

    def encode(self, obj):
        return obj.isoformat()

    def decode(self, s):
        return datetime.fromisoformat(s)

Changelog

v2.1.0 (2021-01-23)

• Include the DateTimeSerializer in this package (see issue #10)
• Drop Python 3.6 support (as 3.7 is needed for date parsing)

v2.0.0 (2020-05-26)

• Add TinyDB v4.0.0 support (see pull request #9)

v1.0.4 (2017-03-27)

• Don’t modify the original element if it contains a list (see pull request #5)

v1.0.3 (2016-02-11)

• Handle nested data (nested dicts, lists) properly when serializing/deserializing (see pull request #3)

v1.0.2 (2016-01-04)

• Don’t destroy original data when serializing (see pull request #2)

v1.0.1 (2015-11-17)

• Fix installation via pip (see issue #1)

v1.0.0 (2015-09-27)

• Initial release on PyPI