1#!/usr/bin/env python
 2
 3import re
 4import json
 5
 6# http://mathiasbynens.be/notes/javascript-encoding#surrogate-formulae
 7# http://stackoverflow.com/a/13436167/96656
 8def unisymbol(codePoint):
 9	if codePoint >= 0x0000 and codePoint <= 0xFFFF:
10		return unichr(codePoint)
11	elif codePoint >= 0x010000 and codePoint <= 0x10FFFF:
12		highSurrogate = int((codePoint - 0x10000) / 0x400) + 0xD800
13		lowSurrogate = int((codePoint - 0x10000) % 0x400) + 0xDC00
14		return unichr(highSurrogate) + unichr(lowSurrogate)
15	else:
16		return 'Error'
17
18def hexify(codePoint):
19	return 'U+' + hex(codePoint)[2:].upper().zfill(6)
20
21def writeFile(filename, contents):
22	print filename
23	with open(filename, 'w') as f:
24		f.write(contents.strip() + '\n')
25
26data = []
27for codePoint in range(0x000000, 0x10FFFF + 1):
28	symbol = unisymbol(codePoint)
29	# http://stackoverflow.com/a/17199950/96656
30	bytes = symbol.encode('utf8').decode('latin1')
31	data.append({
32		'codePoint': codePoint,
33		'decoded': symbol,
34		'encoded': bytes
35	});
36
37jsonData = json.dumps(data, sort_keys=False, indent=2, separators=(',', ': '))
38# Use tabs instead of double spaces for indentation
39jsonData = jsonData.replace('  ', '\t')
40# Escape hexadecimal digits in escape sequences
41jsonData = re.sub(
42	r'\\u([a-fA-F0-9]{4})',
43	lambda match: r'\u{}'.format(match.group(1).upper()),
44	jsonData
45)
46
47writeFile('data.json', jsonData)