Get all Unicode with names, including emoji.

uni.py

#!/usr/bin/env python3
import re
import csv
import unicodedata2
import urllib.request

EMOJI_TEST_URL = 'https://unicode.org/Public/emoji/latest/emoji-test.txt'
MIN_CHAR = 0x20
MAX_CHAR = 0x100000
OUTPUT_CSV = 'unicode.csv'

# Pattern for lines like:
# "1F469 200D 1F4BB ; fully-qualified     # 👩‍💻 E2.0 woman technologist"
LINE_RE = re.compile(
    r'^([0-9A-F ]+)\s*;\s*'
    r'(fully-qualified|minimally-qualified|unqualified)\s*#\s*'
    r'(\S+)\s+E\d+\.\d+\s*(.+)$'
)

# try to import tqdm; fallback to identity function
try:
    from tqdm import tqdm
except ImportError:
    def tqdm(x, **kwargs):
        return x

def fetch_emoji_test(url):
    """Download emoji-test.txt and return its lines."""
    print('Downloading emoji-test.txt…')
    with urllib.request.urlopen(url) as resp:
        lines = [line.decode('utf-8') for line in resp]
    print(f'Downloaded {len(lines)} lines')
    return lines

def parse_sequences(lines):
    """
    Parse all qualified and unqualified sequences.
    Returns dict: seq_id -> (char, name, qualification)
    """
    seqs = {}
    for line in tqdm(lines, desc='Parsing sequences'):
        m = LINE_RE.match(line)
        if not m:
            continue
        cps_str, qual, char, name = m.groups()
        seq_id = '_'.join(cp.lower() for cp in cps_str.split())
        # keep fully-qualified if duplicates exist
        if seq_id not in seqs or seqs[seq_id][2] != 'fully-qualified':
            seqs[seq_id] = (char, name.title(), qual)
    return seqs

def collect_singletons(min_cp, max_cp):
    """
    Iterate all codepoints, take those with a Unicode name.
    Returns dict: seq_id -> (char, name, '')
    """
    singles = {}
    for cp in tqdm(range(min_cp, max_cp), desc='Scanning codepoints'):
        try:
            ch = chr(cp)
            name = unicodedata2.name(ch)
        except ValueError:
            continue
        seq_id = format(cp, 'x').lower()
        singles[seq_id] = (ch, name.title(), '')
    return singles

def merge_and_write(seqs, singles, out_csv):
    """
    Merge two dicts and write to CSV sorted by sequence id.
    Returns count of merged entries.
    """
    merged = {**singles, **seqs}  # sequences overwrite singletons
    with open(out_csv, 'w', encoding='utf-8', newline='') as f:
        writer = csv.writer(f, quoting=csv.QUOTE_MINIMAL)
        writer.writerow(['Emoji', 'Name', 'Qualification', 'Sequence'])
        for seq_id in tqdm(
            sorted(merged, key=lambda s: (len(s.split('_')), s)),
            desc='Writing CSV'
        ):
            ch, name, qual = merged[seq_id]
            writer.writerow([ch, name, qual, seq_id])
    return len(merged)

def main():
    lines   = fetch_emoji_test(EMOJI_TEST_URL)
    seqs    = parse_sequences(lines)
    singles = collect_singletons(MIN_CHAR, MAX_CHAR)
    count   = merge_and_write(seqs, singles, OUTPUT_CSV)
    print(f'Wrote {count} rows into {OUTPUT_CSV}')

if __name__ == '__main__':
    main()