Patterned Date and Time¶

This feature, introduced in v0.35.0, allows parsing custom date and time formats into Python’s date, time, and datetime objects. For example, strings like November 2, 2021 can now be parsed using customizable patterns – specified as format codes.

Key Features:

Supports standard, timezone-aware, and UTC patterns.
Annotate fields using DatePattern, TimePattern, or DateTimePattern.
Retains ISO 8601 serialization for compatibility.

Supported Patterns:

Naive Patterns (default)

DatePattern, DateTimePattern, TimePattern

Timezone-Aware Patterns

AwareDateTimePattern, AwareTimePattern

UTC Patterns

UTCDateTimePattern, UTCTimePattern

Pattern Comparison¶

The following table compares the different types of date-time patterns: Naive, Timezone-Aware, and UTC patterns. It summarizes key features and example use cases for each.

Pattern Type	Key Characteristics	Example Use Cases
Naive Patterns	No timezone info	`DatePattern` (local date) `TimePattern` (local time) `DateTimePattern` (local datetime)
Timezone-Aware Patterns	Specifies a timezone	`AwareDateTimePattern` (e.g., ‘Europe/London’) `AwareTimePattern` (timezone-aware time)
UTC Patterns	Interprets as UTC time	`UTCDateTimePattern` (UTC datetime) `UTCTimePattern` (UTC time)

Standard Date-Time Patterns¶

Hint

Note that the “naive” implementations TimePattern and DateTimePattern do not store timezone information – or tzinfo – on the de-serialized object (as explained in the Naive datetime concept). However, Timezone-Aware Date and Time Patterns do store this information.

Additionally, date does not have any timezone-related data, nor does its counterpart DatePattern.

To use, simply annotate fields with DatePattern, TimePattern, or DateTimePattern with supported format codes. These patterns support the most common date formats.

from dataclass_wizard import DataclassWizard
from dataclass_wizard.patterns import DatePattern, TimePattern


class MyClass(DataclassWizard):
    date_field: DatePattern['%b %d, %Y']
    time_field: TimePattern['%I:%M %p']


data = {'date_field': 'Jan 3, 2022', 'time_field': '3:45 PM'}
c1 = MyClass.from_dict(data)
print(c1)
print(c1.to_dict())
assert c1 == MyClass.from_dict(c1.to_dict())  # > True

Timezone-Aware Date and Time Patterns¶

Hint

Timezone-aware date-time objects store timezone information, as detailed in the Timezone-aware section. This is accomplished using the built-in zoneinfo module in Python 3.9+.

Tip

On Windows, install tzdata with the tz extra:

pip install dataclass-wizard[tz]

This is required because Windows does not ship IANA time zone data.

To handle timezone-aware datetime and time values, use the following patterns:

AwareDateTimePattern
AwareTimePattern
AwarePattern (with typing.Annotated)

These patterns allow you to specify the timezone for the date and time, ensuring that the values are interpreted correctly relative to the given timezone.

Example: Using Timezone-Aware Patterns

from dataclasses import dataclass
from pprint import pprint
from typing import Annotated

from dataclass_wizard import Alias, fromdict, asdict
from dataclass_wizard.patterns import AwareTimePattern, AwareDateTimePattern


@dataclass
class MyClass:
    my_aware_dt: AwareTimePattern['Europe/London', '%H:%M:%S']
    my_aware_dt2: Annotated[AwareDateTimePattern['Asia/Tokyo', '%m-%Y-%H:%M-%Z'], Alias('key')]



d = {'my_aware_dt': '6:15:45', 'key': '10-2020-15:30-UTC'}
c = fromdict(MyClass, d)

pprint(c)
print(asdict(c))
assert c == fromdict(MyClass, asdict(c))  # > True

UTC Date and Time Patterns¶

Hint

For UTC-specific time, use UTC patterns, which handle Coordinated Universal Time (UTC) as described in the UTC article.

For UTC-specific datetime and time values, use the following patterns:

UTCDateTimePattern
UTCTimePattern
UTCPattern (with typing.Annotated)

These patterns are used when working with date and time in Coordinated Universal Time (UTC), and ensure that timezone data – or tzinfo – is correctly set to UTC.

Example: Using UTC Patterns

from typing import Annotated

from dataclass_wizard import Alias, DataclassWizard
from dataclass_wizard.patterns import UTCTimePattern, UTCDateTimePattern


class MyClass(DataclassWizard):
    my_utc_time: UTCTimePattern['%H:%M:%S']
    my_utc_dt: Annotated[UTCDateTimePattern['%m-%Y-%H:%M-%Z'], Alias('key')]


d = {'my_utc_time': '6:15:45', 'key': '10-2020-15:30-UTC'}
c = MyClass.from_dict(d)
print(c)
print(c.to_dict())

Containers of Date and Time¶

For more complex annotations like list[date], you can use typing.Annotated with one of Pattern, AwarePattern, or UTCPattern to specify custom date-time formats.

Tip

The typing.Annotated type is used to apply additional metadata (like timezone information) to a field. When combined with a date-time pattern, it tells the library how to interpret the field’s value in terms of its format or timezone.

Example: Using Pattern with Annotated

from datetime import time
from typing import Annotated
from dataclass_wizard import DataclassWizard
from dataclass_wizard.patterns import Pattern


class MyTime(time):
    def get_hour(self):
        return self.hour


class MyClass(DataclassWizard):
    time_field: Annotated[list[MyTime], Pattern['%I:%M %p']]


data = {'time_field': ['3:45 PM', '1:20 am', '12:30 pm']}
c1 = MyClass.from_dict(data)
print(c1)  # > MyClass(time_field=[MyTime(15, 45), MyTime(1, 20), MyTime(12, 30)])

Multiple Date and Time Patterns¶

You can also use multiple date and time patterns (format codes) to parse and serialize your date and time fields. This feature allows for flexibility when handling different formats, making it easier to work with various date and time strings.

Example: Using Multiple Patterns¶

In the example below, the DatePattern and TimePattern are configured to support multiple formats. The class MyClass demonstrates how the fields can accept different formats for both dates and times.

from dataclass_wizard import DataclassWizard
from dataclass_wizard.patterns import DatePattern, UTCTimePattern


class MyClass(DataclassWizard):
    date_field: DatePattern['%b %d, %Y', '%I %p %Y-%m-%d']
    time_field: UTCTimePattern['%I:%M %p', '(%H)+(%S)']


# Using the first date pattern format: 'Jan 3, 2022'
data = {'date_field': 'Jan 3, 2022', 'time_field': '3:45 PM'}
c1 = MyClass.from_dict(data)

print(c1)
print(c1.to_dict())
assert c1 == MyClass.from_dict(c1.to_dict())  # > True
print()

# Using the second date pattern format: '3 PM 2025-01-15'
data = {'date_field': '3 PM 2025-01-15', 'time_field': '(15)+(45)'}
c2 = MyClass.from_dict(data)
print(c2)
print(c2.to_dict())
assert c2 == MyClass.from_dict(c2.to_dict())  # > True
print()

# ERROR! The date is not a valid format for the available patterns.
data = {'date_field': '2025-01-15 3 PM', 'time_field': '(15)+(45)'}
_ = MyClass.from_dict(data)

How It Works¶

DatePattern and TimePattern: These are special types that support multiple patterns (format codes). Each pattern is tried in the order specified, and the first one that matches the input string is used for parsing or formatting.
DatePattern Usage: The date_field in the example accepts two formats:
- %b %d, %Y (e.g., ‘Jan 3, 2022’)
- %I %p %Y-%m-%d (e.g., ‘3 PM 2025-01-15’)
TimePattern Usage: The time_field accepts two formats:
- %I:%M %p (e.g., ‘3:45 PM’)
- (%H)+(%S) (e.g., ‘(15)+(45)’)
Error Handling: If the input string doesn’t match any of the available patterns, an error will be raised.

This feature is especially useful for handling date and time formats from various sources, ensuring flexibility in how data is parsed and serialized.

Key Points¶

Multiple patterns are specified as a list of format codes in DatePattern and TimePattern.
The system automatically tries each pattern in the order provided until a match is found.
If no match is found, an error is raised, as shown in the example with the invalid date format '2025-01-15 3 PM'.

—

Serialization:

Hint

ISO 8601: Serialization of all date-time objects follows the ISO 8601 standard, a widely-used format for representing date and time.

All date-time objects are serialized as ISO 8601 format strings by default. This ensures compatibility with other systems and optimizes parsing.

Note: Parsing uses datetime.fromisoformat for ISO 8601 strings, which is much faster than datetime.strptime.

Patterned Date and Time¶

Pattern Comparison¶

Standard Date-Time Patterns¶

Timezone-Aware Date and Time Patterns¶

UTC Date and Time Patterns¶

Containers of Date and Time¶

Multiple Date and Time Patterns¶

Example: Using Multiple Patterns¶

How It Works¶

Key Points¶

Dataclass Wizard

Useful Links

Table of Contents

Related Topics

API Reference

This Page