#!/usr/bin/python3

'''
 Unicode と CID の情報を元に、
 OpenType/CID (Adobe-Identity-0) を Adobe-Japan1-* にマッピングするための情報を出力。

 (1) 変換元の CID フォントから cmap の情報を取得
  $ spot -t cmap=7 [FONTFILE] > cmap.txt
 (2) https://github.com/adobe-type-tools/cmap-resources から、以下をダウンロード。
    - Adobe-Japan1-7/CMap/UniJIS2004-UTF32-H
 (3) https://github.com/adobe-type-tools/Adobe-Japan1 から、以下をダウンロード。
    - Adobe-Japan1_sequences.txt
 (3) 変換
  $ ./remap-unicode.py cmap.txt [UniJIS2004-UTF32-H] [Adobe-Japan1_sequences.txt] > out.txt
'''

import sys
import re

if len(sys.argv) < 2:
	print("[usage] " + sys.argv[0] + " <spot_map7_file> [<cmap_file>] [<sequence_file>]")
	sys.exit(0)

fname_spot = sys.argv[1]

if len(sys.argv) < 3:
	fname_cmap = 'UniJIS2004-UTF32-H'
else:
	fname_cmap = sys.argv[2]

if len(sys.argv) < 4:
	fname_seq = 'Adobe-Japan1_sequences.txt'
else:
	fname_seq = sys.argv[3]


class CmapConv:
	# srcmap 読み込み
	def read_src(self,fname):
		f = open(fname, 'rt')
		lines = f.readlines()
		f.close()

		self.src = {}
		mo = re.compile(r'\[(.+?)\]\=\s*<(.+?)>')

		for l in lines:
			l = l.strip()
			if (not l) or l[0] != '[': continue

			m = mo.match(l)
			if m:
				if ' ' in m.group(1):
					# 異体字
					uc = m.group(1)
				else:
					uc = int(m.group(1), base=16)

				if m.group(2)[0] == '\\':
					cid = int(m.group(2)[1:])
					self.src[uc] = cid

	# 変換用 CMap 読み込み
	def read_dst(self,fname):
		f = open(fname, 'rt')
		lines = f.readlines()
		f.close()

		self.dst = [None] * 65536
		flag = 0
		moc = re.compile(r'<(.+?)>\s+(\d+)$')
		mor = re.compile(r'<(.+?)>\s+<(.+?)>\s+(\d+)$')

		for l in lines:
			l = l.strip()
			if not l: continue

			if flag == 1:
				# cidchar
				if l == 'endcidchar':
					flag = 0
				else:
					m = moc.match(l)
					if m:
						uc = int(m.group(1), base=16)
						cid = int(m.group(2))
						if self.dst[cid]:
							self.dst[cid].append(uc)
						else:
							self.dst[cid] = [uc]

			elif flag == 2:
				# cidrange
				if l == 'endcidrange':
					flag = 0
				else:
					m = mor.match(l)
					if m:
						uc1 = int(m.group(1), base=16)
						uc2 = int(m.group(2), base=16)
						cid = int(m.group(3))

						for uc in range(uc1, uc2 + 1):
							if self.dst[cid]:
								self.dst[cid].append(uc)
							else:
								self.dst[cid] = [uc]
							cid += 1
				
			elif l.endswith('begincidchar'):
				flag = 1
			elif l.endswith('begincidrange'):
				flag = 2

		# 個数
		i = len(self.dst) - 1
		while i > 0:
			if self.dst[i]: break
			i -= 1

		self.dstnum = i + 1

	# dst に対してマッピング
	def remap(self):
		self.res = [None] * self.dstnum
		
		cid_dst = 1
		while cid_dst < self.dstnum:
			dlist = self.dst[cid_dst]
			cid_src = None

			# dst と同じ Unicode の CID を検索
			if dlist:
				for uc in dlist:
					if uc in self.src:
						cid_src = self.src[uc]
						break

			if cid_src:
				self.res[cid_dst] = cid_src

			cid_dst += 1

	# 異体字マッピング
	def remap_seq(self,fname):
		f = open(fname, 'rt')
		lines = f.readlines()
		f.close()

		mo = re.compile(r'(.+?)\; .+?\; CID\+(\d+)$')

		for l in lines:
			l = l.strip()
			if (not l) or l[0] == '#': continue

			m = mo.match(l)
			if m:
				uc = m.group(1)
				cid = int(m.group(2))
				self.dst[cid] = uc

				if uc in self.src:
					self.res[cid] = self.src[uc]

	# 書き込み
	def write(self):
		i = 1
		num = 0
		
		while i < self.dstnum:
			sid = self.res[i]
			uclist = self.dst[i]

			o = '{0:05d}'.format(i)

			# src 割り当て
			if sid:
				o += '[{0}]'.format(sid)
				num += 1

			# Unicode
			if uclist:
				o += '#'
				if type(uclist) is str:
					# 異体字
					o += uclist
				else:
					for uc in uclist:
						o += '{0:X}/'.format(uc)
					o = o[:-1]
			
			print(o)
			i += 1

		print('{0}/{1} ({2:.2%})'.format(num, self.dstnum, num / self.dstnum) ,file=sys.stderr)

#----

cv = CmapConv()
cv.read_src(fname_spot)
cv.read_dst(fname_cmap)
cv.remap()
cv.remap_seq(fname_seq)
cv.write()