Welcome to py-ulid’s documentation!

ulid logo

The py-ulid library is a minimal, self-contained implementation of the ULID (Universally Unique Lexicographically Sortable Identifier) specification in Python. For more information, please refer to the official specification.

Installation

You can install the py-ulid library from PyPi

pip install py-ulid

The py-ulid library can be used in any version of python >= 3.5 and does not require any additional packages or modules.

How to use

To generate a ULID, simple run the generate() function

from ulid import ULID

#Instantiate the ULID class
ulid = ULID()
ulid.generate()  #01BX5ZZKBKACTAV9WEVGEMMVRZ

Seeding Time

You can instantiate the instance of the ULID class with a seed time which will output the same string for the time component. This could be useful when migrating to ulid

from ulid import ULID

#Instantiate the ULID class
ulid = ULID(1469918176385)
ulid.generate()  #01ARYZ6S41TSV4RRFFQ69G5FAV

Monotonic ULIDs

from ulid import Monotonic

#Instantiate the Monotonic Class
ulid = Monotonic()

# Same timestamp when calls are made within the same
# millisecond and least-significant random bit is incremented by 1
ulid.generate()  #01DC8Y7RBV4RSXX0437Z1RQR11
ulid.generate()  #01DC8Y7RBV4RSXX0437Z1RQR12
ulid.generate()  #01DC8Y7RBV4RSXX0437Z1RQR13
ulid.generate()  #01DC8Y7RBV4RSXX0437Z1RQR14
ulid.generate()  #01DC8Y7RBV4RSXX0437Z1RQR15
ulid.generate()  #01DC8Y7RBV4RSXX0437Z1RQR16
ulid.generate()  #01DC8Y7RBV4RSXX0437Z1RQR17
ulid.generate()  #01DC8Y7RBV4RSXX0437Z1RQR18
ulid.generate()  #01DC8Y7RBV4RSXX0437Z1RQR19

ulid package

ulid module

This module provides immutable ULID objects (class ULID) according to the ULID spec and the functions generate() to generate ulids according to the specifications, encode() to transform a given integer to the canonical string representation of an ULID, and decode() to take a canonically encoded string and break it down into it’s timestamp and randomness components. The module also provides Monotonic sort order guarantee for ULIDs via the Monotonic class and it’s associated generate() function.

class ulid.ulid.Monotonic(seed=None)[source]

Bases: ulid.ulid.ULID

The Monotonic class represent an extension of the base ULID class with the addition of a monotonic sort order (correctly detects and handles the same millisecond)

MAX_EPOCH_TIME = 281474976710655
crockford_base = '0123456789ABCDEFGHJKMNPQRSTVWXYZ'
decode(s: str) → Tuple[int, int]

Given a properly formed ULID, return a tuple containing the timestamp and the randomness component as ints

Parameters:s (str) – ULID string to decode
Returns:(timestamp, randomness)
Return type:tuple(int, int)
Raises:TypeError, ValueError
>>> from ulid import ULID
>>> ulid = ULID()
>>> ulid.decode('01BX5ZZKBKACTAV9WEVGEMMVRY')
(1508808576371, 392928161897179156999966)
encode(i: int) → str

Convert a given integer into the canonical ULID string format

Parameters:i (int) – The integer to convert
Returns:Canonically encoded ULID string
Return type:str
Raises:TypeError, ValueError
>>> from ulid import ULID
>>> ulid = ULID()
>>> ulid.encode(340282366920938463463374607431768167)
00864KEJY6MZQSVCHD1SB08637
encode_timestamp(t: int) → str

Convert a given unix timestamp into the canonical ULID string format

Parameters:t (int) – The unix timestamp to convert
Returns:Canonically encoded ULID string
Return type:str
Raises:TypeError, ValueError
>>> from ulid import ULID
>>> ulid = ULID()
>>> ulid.encode(281474976710655)
7ZZZZZZZZZ
generate() → str[source]

Generate a 26 character ULID string encoded in Crockford’s Base32

Returns:Canonically encoded ULID string
Return type:str
Raises:ValueError
>>> from ulid import Monotonic
>>> ulid = Monotonic()
>>> ulid.generate()
01BX5ZZKBKACTAV9WEVGEMMVRZ
pretty_print(s: str) → None

Print the given ULID string in a binary layout

>>> from ulid import ULID
>>> ulid = ULID()
>>> ulid.pretty_print(ulid_str)
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|                      32_bit_uint_time_high                    |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|     16_bit_uint_time_low      |       16_bit_uint_random      |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|                       32_bit_uint_random                      |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|                       32_bit_uint_random                      |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
class ulid.ulid.ULID(seed=None)[source]

Bases: object

Instances of the ULID class represent ULIDS as specified in the ULID spec. ULIDS have 128-bit compatibility with UUID, are Lexicographically sortable, case insensitive and URL safe

MAX_EPOCH_TIME = 281474976710655
crockford_base = '0123456789ABCDEFGHJKMNPQRSTVWXYZ'
decode(s: str) → Tuple[int, int][source]

Given a properly formed ULID, return a tuple containing the timestamp and the randomness component as ints

Parameters:s (str) – ULID string to decode
Returns:(timestamp, randomness)
Return type:tuple(int, int)
Raises:TypeError, ValueError
>>> from ulid import ULID
>>> ulid = ULID()
>>> ulid.decode('01BX5ZZKBKACTAV9WEVGEMMVRY')
(1508808576371, 392928161897179156999966)
encode(i: int) → str[source]

Convert a given integer into the canonical ULID string format

Parameters:i (int) – The integer to convert
Returns:Canonically encoded ULID string
Return type:str
Raises:TypeError, ValueError
>>> from ulid import ULID
>>> ulid = ULID()
>>> ulid.encode(340282366920938463463374607431768167)
00864KEJY6MZQSVCHD1SB08637
encode_timestamp(t: int) → str[source]

Convert a given unix timestamp into the canonical ULID string format

Parameters:t (int) – The unix timestamp to convert
Returns:Canonically encoded ULID string
Return type:str
Raises:TypeError, ValueError
>>> from ulid import ULID
>>> ulid = ULID()
>>> ulid.encode(281474976710655)
7ZZZZZZZZZ
generate() → str[source]

Generate a 26 character ULID string encoded in Crockford’s Base32

Returns:Canonically encoded ULID string
Return type:str
>>> from ulid import ULID
>>> ulid = ULID()
>>> ulid.generate()
01BX5ZZKBKACTAV9WEVGEMMVRZ
pretty_print(s: str) → None[source]

Print the given ULID string in a binary layout

>>> from ulid import ULID
>>> ulid = ULID()
>>> ulid.pretty_print(ulid_str)
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|                      32_bit_uint_time_high                    |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|     16_bit_uint_time_low      |       16_bit_uint_random      |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|                       32_bit_uint_random                      |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|                       32_bit_uint_random                      |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

Module contents

ULID class

class ulid.ULID(seed=None)[source]

Instances of the ULID class represent ULIDS as specified in the ULID spec. ULIDS have 128-bit compatibility with UUID, are Lexicographically sortable, case insensitive and URL safe

decode(s: str) → Tuple[int, int][source]

Given a properly formed ULID, return a tuple containing the timestamp and the randomness component as ints

Parameters:s (str) – ULID string to decode
Returns:(timestamp, randomness)
Return type:tuple(int, int)
Raises:TypeError, ValueError
>>> from ulid import ULID
>>> ulid = ULID()
>>> ulid.decode('01BX5ZZKBKACTAV9WEVGEMMVRY')
(1508808576371, 392928161897179156999966)
encode(i: int) → str[source]

Convert a given integer into the canonical ULID string format

Parameters:i (int) – The integer to convert
Returns:Canonically encoded ULID string
Return type:str
Raises:TypeError, ValueError
>>> from ulid import ULID
>>> ulid = ULID()
>>> ulid.encode(340282366920938463463374607431768167)
00864KEJY6MZQSVCHD1SB08637
encode_timestamp(t: int) → str[source]

Convert a given unix timestamp into the canonical ULID string format

Parameters:t (int) – The unix timestamp to convert
Returns:Canonically encoded ULID string
Return type:str
Raises:TypeError, ValueError
>>> from ulid import ULID
>>> ulid = ULID()
>>> ulid.encode(281474976710655)
7ZZZZZZZZZ
generate() → str[source]

Generate a 26 character ULID string encoded in Crockford’s Base32

Returns:Canonically encoded ULID string
Return type:str
>>> from ulid import ULID
>>> ulid = ULID()
>>> ulid.generate()
01BX5ZZKBKACTAV9WEVGEMMVRZ
pretty_print(s: str) → None[source]

Print the given ULID string in a binary layout

>>> from ulid import ULID
>>> ulid = ULID()
>>> ulid.pretty_print(ulid_str)
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|                      32_bit_uint_time_high                    |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|     16_bit_uint_time_low      |       16_bit_uint_random      |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|                       32_bit_uint_random                      |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|                       32_bit_uint_random                      |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

Monotonic class

class ulid.Monotonic(seed=None)[source]

The Monotonic class represent an extension of the base ULID class with the addition of a monotonic sort order (correctly detects and handles the same millisecond)

decode(s: str) → Tuple[int, int]

Given a properly formed ULID, return a tuple containing the timestamp and the randomness component as ints

Parameters:s (str) – ULID string to decode
Returns:(timestamp, randomness)
Return type:tuple(int, int)
Raises:TypeError, ValueError
>>> from ulid import ULID
>>> ulid = ULID()
>>> ulid.decode('01BX5ZZKBKACTAV9WEVGEMMVRY')
(1508808576371, 392928161897179156999966)
encode(i: int) → str

Convert a given integer into the canonical ULID string format

Parameters:i (int) – The integer to convert
Returns:Canonically encoded ULID string
Return type:str
Raises:TypeError, ValueError
>>> from ulid import ULID
>>> ulid = ULID()
>>> ulid.encode(340282366920938463463374607431768167)
00864KEJY6MZQSVCHD1SB08637
encode_timestamp(t: int) → str

Convert a given unix timestamp into the canonical ULID string format

Parameters:t (int) – The unix timestamp to convert
Returns:Canonically encoded ULID string
Return type:str
Raises:TypeError, ValueError
>>> from ulid import ULID
>>> ulid = ULID()
>>> ulid.encode(281474976710655)
7ZZZZZZZZZ
generate() → str[source]

Generate a 26 character ULID string encoded in Crockford’s Base32

Returns:Canonically encoded ULID string
Return type:str
Raises:ValueError
>>> from ulid import Monotonic
>>> ulid = Monotonic()
>>> ulid.generate()
01BX5ZZKBKACTAV9WEVGEMMVRZ
pretty_print(s: str) → None

Print the given ULID string in a binary layout

>>> from ulid import ULID
>>> ulid = ULID()
>>> ulid.pretty_print(ulid_str)
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|                      32_bit_uint_time_high                    |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|     16_bit_uint_time_low      |       16_bit_uint_random      |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|                       32_bit_uint_random                      |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|                       32_bit_uint_random                      |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

Inspiration

UUID can be suboptimal for many uses-cases because:

  • It isn’t the most character efficient way of encoding 128 bits of randomness
  • UUID v1/v2 is impractical in many environments, as it requires access to a unique, stable MAC address
  • UUID v3/v5 requires a unique seed and produces randomly distributed IDs, which can cause fragmentation in many data structures
  • UUID v4 provides no other information than randomness which can cause fragmentation in many data structures

Instead, herein is proposed ULID:

  • 128-bit compatibility with UUID
  • 1.21e+24 unique ULIDs per millisecond
  • Lexicographically sortable!
  • Canonically encoded as a 26 character string, as opposed to the 36 character UUID
  • Uses Crockford’s base32 for better efficiency and readability (5 bits per character)
  • Case insensitive
  • No special characters (URL safe)
  • Monotonic sort order (correctly detects and handles the same millisecond)

Prior Art

Partly inspired by:

README

ulid logo

The py-ulid library is a minimal, self-contained implementation of the ULID (Universally Unique Lexicographically Sortable Identifier) specification in Python. For more information, please refer to the official specification.

Installation

You can install the py-ulid library from PyPi

pip install py-ulid

The py-ulid library can be used in any version of python >= 3.5 and does not require any additional packages or modules.

How to use

To generate a ULID, simple run the generate() function

from ulid import ULID

#Instantiate the ULID class
ulid = ULID()
ulid.generate()  #01BX5ZZKBKACTAV9WEVGEMMVRZ

Seeding Time

You can instantiate the instance of the ULID class with a seed time which will output the same string for the time component. This could be useful when migrating to ulid

from ulid import ULID

#Instantiate the ULID class
ulid = ULID(1469918176385)
ulid.generate()  #01ARYZ6S41TSV4RRFFQ69G5FAV

Monotonic ULIDs

from ulid import Monotonic

#Instantiate the Monotonic Class
ulid = Monotonic()

# Same timestamp when calls are made within the same
# millisecond and least-significant random bit is incremented by 1
ulid.generate()  #01DC8Y7RBV4RSXX0437Z1RQR11
ulid.generate()  #01DC8Y7RBV4RSXX0437Z1RQR12
ulid.generate()  #01DC8Y7RBV4RSXX0437Z1RQR13
ulid.generate()  #01DC8Y7RBV4RSXX0437Z1RQR14
ulid.generate()  #01DC8Y7RBV4RSXX0437Z1RQR15
ulid.generate()  #01DC8Y7RBV4RSXX0437Z1RQR16
ulid.generate()  #01DC8Y7RBV4RSXX0437Z1RQR17
ulid.generate()  #01DC8Y7RBV4RSXX0437Z1RQR18
ulid.generate()  #01DC8Y7RBV4RSXX0437Z1RQR19

Indices and tables