casio_doc/fontcharacter/tools/makebin.py

#!/usr/bin/env python3
""" Utility to make the set binary files. """

import os, shutil
from argparse import ArgumentParser
from fontcharacter import Reference

# ---
# Function for encoding.
# ---

def tobytes(num, length):
	return num.to_bytes(length, byteorder='big', signed=False)
def charbytes(char):
	return tobytes(char, 2) if char > 0xFF else tobytes(char, 1)

def make_pool(pool):
	# TODO: make this function optimize space, some day?

	bdata = b''
	offsets = []
	for data in pool:
		offsets += [len(bdata)]
		bdata += bytes(data)
	return bdata, offsets

def encode_set(ref, fset):
	global args
	bheader = bytes()
	blead   = bytes()
	bchars  = bytes()
	bdata   = bytes()

	# Prepare characters data
	leading = {}
	char_count = 0
	dchars = []
	pool = []
	for char_id, (code, char) in enumerate(fset['characters'].items()):
		char_count = char_id + 1

		# Check if we should inherit.
		if char['_pr'] > 0:
			char = ref.get(char['inherit'])['characters'][code]

		# Check the leader.
		lead = code >> 8
		if not lead in leading:
			leading[lead] = char_id

		# Get the multi size, feed the pool if needed.
		mul_sz = 0
		if char.get('multi') and char['multi']:
			mul = b''.join(map(charbytes, char['multi']))
			pool += [mul]
			mul_sz = len(mul)

		# Get the unicode size, feed the pool if needed.
		uni_sz = 0
		if not args.no_unicode and char.get('unicode') and char['unicode']:
			uni = ''.join(map(chr, char['unicode'])).encode('utf-8')
			pool += [uni]
			uni_sz  = len(uni)

		# Get the CAT token size, feed the pool if needed.
		cat_sz = 0

		# Get the Newcat token size
		newcat_sz = 0

		# Get the CTF token size
		ctf_sz = 0

		# Get the Casemul token size
		casemul_sz = 0

		# Add all of these elements to the chars data tab.
		dchars += [{
			'code': code,
			'mul_sz': mul_sz,
			'uni_sz': uni_sz,
			'cat_sz': cat_sz,
			'newcat_sz': newcat_sz,
			'ctf_sz': ctf_sz,
			'casemul_sz': casemul_sz
		}]

	# Make the pool.
	bdata, offsets = make_pool(pool)

	# Prepare pool.
	offsets = iter(offsets)
	for char in dchars:
		ent = tobytes(char['code'], 2) + tobytes(char['mul_sz'], 1) \
			+ tobytes(char['uni_sz'], 1) + tobytes(char['cat_sz'], 1) \
			+ tobytes(char['newcat_sz'], 1) + tobytes(char['ctf_sz'], 1) \
			+ tobytes(char['casemul_sz'], 1)

		# Offsets
		ent += tobytes(next(offsets) if char['mul_sz'] > 0 else 0, 4)
		if not args.no_unicode:
			ent += tobytes(next(offsets) if char['uni_sz'] > 0 else 0, 4)
		if not args.no_cat:
			ent += tobytes(next(offsets) if char['cat_sz'] > 0 else 0, 4)
		if not args.no_newcat:
			ent += tobytes(next(offsets) if char['newcat_sz'] > 0 else 0, 4)
		if not args.no_ctf:
			ent += tobytes(next(offsets) if char['ctf_sz'] > 0 else 0, 4)
		if not args.no_casemul:
			ent += tobytes(next(offsets) if char['casemul_sz'] > 0 else 0, 4)

		bchars += ent

	# Correct the leading.
	for lead in fset['leading']:
		if not lead in leading:
			leading[lead] = char_count

	# Make the leading tab.
	for lead, off in leading.items():
		ent = bytes([lead, 0]) + tobytes(off, 2)
		blead += bytes(ent)

	# Make the flags.
	flags = 0x1F
	if args.no_unicode: flags &= ~0x01
	if args.no_cat: flags &= ~0x02
	if args.no_newcat: flags &= ~0x04
	if args.no_ctf: flags &= ~0x08
	if args.no_casemul: flags &= ~0x10

	# Make the checksum.
	csum = (sum(blead) + sum(bchars) + sum(bdata)) & 0xFFFFFFFF

	# Make the lengths.
	datalen = len(bdata)
	filesize = 32 + len(blead) + len(bchars) + datalen

	# Finish making the main header.
	bheader = bytes(list(map(ord, "CASIOFC\x7F")) + [0x01]) \
		+ tobytes(len(leading), 1) + tobytes(char_count, 2) \
		+ tobytes(flags, 1) + tobytes(0, 1) + tobytes(0, 2) + tobytes(0, 4) \
		+ tobytes(csum, 4) + tobytes(filesize, 4) + tobytes(datalen, 4)

	return bheader + blead + bchars + bdata

# ---
# Main function.
# ---

if __name__ == '__main__':
	ap = ArgumentParser(description="FONTCHARACTER reference binary generator")

	ap.add_argument('--no-unicode', help='No Unicode equivalents?',
		action="store_true")
	ap.add_argument('--no-cat', help='No CAT tokens?', action="store_true")
	ap.add_argument('--no-newcat', help='No Newcat tokens?',
		action="store_true")
	ap.add_argument('--no-ctf', help='No CTF tokens?', action="store_true")
	ap.add_argument('--no-casemul', help="No Casemul tokens?",
		action="store_true")
	ap.add_argument('--output', '-o', help='The output directory path.',
		default=os.path.join(os.getcwd(), 'generated_sets'))
	ap.add_argument('--refpath', help='The reference path.',
		default=os.getcwd())
	args = ap.parse_args()

	# Obtain the reference.
	ref = Reference(args.refpath)

	# Make the directory.
	if   os.path.isdir(args.output):
		shutil.rmtree(args.output)
	elif os.path.exists(args.output):
		os.remove(args.output)
	os.makedirs(args.output)

	# For each set, make the file.
	for set_name, set_val in map(lambda x: (x, ref.get(x)), ref.list()):
		with open(os.path.join(args.output, set_name + '.set'), 'wb') as f:
			f.write(encode_set(ref, set_val))

# End of file.