Ways to get the name of a Unicode block for a character in Python

get-unicode-blocks.py

#!/usr/bin/env PYTHONIOENCODING=utf-8  python
# encoding: utf-8

from __future__ import absolute_import, print_function, unicode_literals

import os
import re

import requests


def get_block_for_codepoint(cp):
    """Return the Unicode block name for the provided numeric codepoint"""

    for start, end, block_name in UNICODE_BLOCKS:
        if start <= cp <= end:
            return block_name

    return 'No_Block'


def load_unicode_blocks_from_file(f):
    file_contents = f.read().decode('utf-8')

    blocks = []
    for start, end, block_name in re.findall(r'([0-9A-F]+)\.\.([0-9A-F]+);\ (\S.*\S)', file_contents):
        if block_name == 'No_Block':
            continue

        blocks.append((int(start, 16), int(end, 16), block_name))

    return blocks


def load_unicode_blocks(block_filename):
    if not os.path.exists(block_filename):
        print('Unicode block file %s does not exist. Downloading…' % block_filename)
        r = requests.get('http://unicode.org/Public/UNIDATA/Blocks.txt')
        r.raise_for_status()

        with open(block_filename, 'wb') as f:
            for chunk in r.iter_content():
                f.write(chunk)

    with open(block_filename, 'rb') as f:
        blocks = load_unicode_blocks_from_file(f)

    return blocks

UNICODE_BLOCKS = load_unicode_blocks('UNIDATA-Blocks.txt')

pyicu-unicode-block-names.py

import icu 

# Astoundingly, PyICU has no documented way to get a Unicode block name.
# There are two ways to get the offset into the UCodeBlock enum
# – icu.Char.ublock_getCode and icu.Char.getIntPropertyValue(…, icu.UProperty.BLOCK) –
# so we'll build a lookup table to turn that into a human-readable string:

UNICODE_BLOCKS = {getattr(icu.UBlockCode, i): i for i in icu.UBlockCode.__dict__ if i.isupper()}

If you are using PyICU to get a dictionary of Unicode blocks, it may be easier to just use PyICU to identify the name of the block for a given character:

def get_enum_property(char, property):
    if len(char) != 1:
        print("Please specify a single character.")
        return None
    value = icu.Char.getIntPropertyValue(char, property)
    return icu.Char.getPropertyValueName(property, value, icu.UPropertyNameChoice.LONG_PROPERTY_NAME)

get_enum_property('𞤀', icu.UProperty.BLOCK)
# 'Adlam'

get_enum_property("Ɛ", icu.UProperty.BLOCK)
# 'Latin_Extended_B'
get_enum_property("Ɛ", icu.UProperty.SCRIPT)
# 'Latin'

Alternatively, use unicodedataplus, a drop-in replacement for unicodedata:

pip install -U unicodedataplus

import unicodedataplus as ud
ud.block('𞤀')
# 'Adlam'

ud.block("Ɛ")
# 'Latin Extended-B'

If you are using PyICU to get a dictionary of Unicode blocks, it may be easier to just use PyICU to identify the name of the block for a given character:

Thanks for the update - amusingly, it looks like LONG_PROPERTY_NAME was added 9 years ago, right around when I wrote that little utility script so if I'd been only a few months later it would have had a simple answer!

https://gitlab.pyicu.org/main/pyicu/-/commit/47f2f2858aba6d6e5de21d41a809d2e46e50e0f4

My goal for this initially was to have something for a demo which didn't use anything more than stdlib Python but I definitely would recommend one of the dedicated libraries for serious use.