Add block info to unicodedata

[flying-sheep]

BPO	22612
Nosy	@vstinner, @ezio-melotti, @flying-sheep

^{Note: these values reflect the state of the issue at the time it was migrated and might not reflect the current state.}

Show more details

GitHub fields:

assignee = None
closed_at = None
created_at = <Date 2014-10-11.17:58:45.231>
labels = ['type-feature', 'expert-unicode']
title = 'Add block info to unicodedata'
updated_at = <Date 2015-10-17.07:47:47.945>
user = 'https://github.com/flying-sheep'

bugs.python.org fields:

activity = <Date 2015-10-17.07:47:47.945>
actor = 'Denis Jacquerye'
assignee = 'none'
closed = False
closed_date = None
closer = None
components = ['Unicode']
creation = <Date 2014-10-11.17:58:45.231>
creator = 'flying sheep'
dependencies = []
files = []
hgrepos = []
issue_num = 22612
keywords = []
message_count = 2.0
messages = ['229101', '230498']
nosy_count = 4.0
nosy_names = ['vstinner', 'ezio.melotti', 'flying sheep', 'Denis Jacquerye']
pr_nums = []
priority = 'normal'
resolution = None
stage = 'needs patch'
status = 'open'
superseder = None
type = 'enhancement'
url = 'https://bugs.python.org/issue22612'
versions = ['Python 3.5']

Linked PRs

• gh-66802: Add unicodedata.block() function #145042

[flying-sheep]

See also bpo-6331.

The repo https://github.com/nagisa/unicodeblocks contains pretty much the functionality i’d like to see: a way to get access to information about all blocks, and a way to get the block name a char is in.

I propose to include something very similar to those two APIs in unicodedata:

unicodedata.Block: class with start, end, and name property.

its __contains__ should work for single-char-strings (which tests if that char is in the block) and for ints (which tests if the codepoint is in the block)

maybe make it iterable over its chars?

unicodedata.blocks: OrderedDict of str (block name) → Block object mappings ordered by Block.start.

then blocks.keys() would yield the names in order, and blocks.values() the block objects in order.

unicodedata.block_of(chr, name_only=False): returns the Block object for which “chr in block” is True, or its name.

---

alternative: make the Block class an unfancy namedtuple without __contains__ method.

---

Together with bpo-18234, fixing this bug will complete UnicodeData support in python, i guess.

[ezio-melotti]

I needed this in the past and had to implement it myself, so adding it to unicodedata might be OK. A script to generate the list of blocks from the official Unicode files should probably be created and used whenever we update the version of the Unicode database.