use preprocessor

18 files changed, 1766 insertions, 327 deletions

diff --git a/Makefile b/Makefile
index ce077789c..8c36fb831 100644
--- a/Makefile
+++ b/Makefile
@@ -10,6 +10,8 @@ GFX := @tools/gbagfx/gbagfx
 
 SCANINC := tools/scaninc/scaninc
 
+PREPROC := tools/preproc/preproc
+
 # Clear the default suffixes.
 .SUFFIXES:
 
@@ -20,19 +22,22 @@ SCANINC := tools/scaninc/scaninc
 
 .PHONY: rom tools gbagfx scaninc clean compare deps
 
-CSRCS := $(wildcard src/*.c)
-OBJS := asm/crt0.o asm/rom1.o asm/rom2.o asm/rom3.o asm/rom4.o asm/rom5.o \
-	asm/libgcnmultiboot.o asm/libmks4agb.o asm/libagbsyscall.o asm/libgcc.o \
-	src/string_util.o src/rtc.o src/play_time.o src/task.o \
-	src/agb_flash.o src/agb_flash_1m.o src/agb_flash_mx.o src/siirtc.o \
-	data/data1.o data/data2.o
+C_SRCS := $(wildcard src/*.c)
+C_OBJS := $(C_SRCS:%.c=%.o)
+
+ASM_OBJS := asm/crt0.o asm/rom1.o asm/rom2.o asm/rom3.o asm/rom4.o asm/rom5.o \
+	asm/libgcnmultiboot.o asm/libmks4agb.o asm/libagbsyscall.o asm/libgcc.o
+
+DATA_ASM_OBJS := data/data1.o data/data2.o
+
+OBJS := $(C_OBJS) $(ASM_OBJS) $(DATA_ASM_OBJS)
 
 ROM := pokeruby.gba
 ELF := $(ROM:.gba=.elf)
 
 rom: $(ROM)
 
-tools: gbagfx scaninc
+tools: gbagfx scaninc preproc
 
 gbagfx:
 	cd tools/gbagfx && make
@@ -40,6 +45,9 @@ gbagfx:
 scaninc:
 	cd tools/scaninc && make
 
+preproc:
+	cd tools/preproc && make
+
 # For contributors to make sure a change didn't affect the contents of the ROM.
 compare: $(ROM)
 	@$(SHA1) rom.sha1
@@ -59,22 +67,25 @@ include tilesets.mk
 %.gbapal: %.pal ; $(GFX) $< $@
 %.lz: % ; $(GFX) $< $@
 
-$(OBJS): $(CSRCS:src/%.c=src/%.s)
+src/siirtc.o: CFLAGS := -mthumb-interwork -Iinclude
 
-src/siirtc.s: CFLAGS := -mthumb-interwork -Iinclude
+src/agb_flash.o: CFLAGS := -O -mthumb-interwork -Iinclude
+src/agb_flash_1m.o: CFLAGS := -O -mthumb-interwork -Iinclude
+src/agb_flash_mx.o: CFLAGS := -O -mthumb-interwork -Iinclude
 
-src/agb_flash.s: CFLAGS := -O -mthumb-interwork -Iinclude
-src/agb_flash_1m.s: CFLAGS := -O -mthumb-interwork -Iinclude
-src/agb_flash_mx.s: CFLAGS := -O -mthumb-interwork -Iinclude
+$(C_OBJS): %.o : %.c
+	$(CC) $(CFLAGS) -o $*.s $< -S
+	echo -e ".text\n\t.align\t2, 0\n" >> $*.s
+	$(AS) $(ASFLAGS) -o $@ $*.s
 
-src/%.s: src/%.c
-	$(CC) $(CFLAGS) -o $@ $< -S
-	echo -e ".text\n\t.align\t2, 0\n" >> $@
+%.o : dep = $(shell $(SCANINC) $*.s)
 
-%.o: dep = $(shell $(SCANINC) $(@D)/$*.s)
-%.o: %.s $$(dep)
+$(ASM_OBJS): %.o: %.s $$(dep)
 	$(AS) $(ASFLAGS) -o $@ $<
 
+$(DATA_ASM_OBJS): %.o: %.s $$(dep)
+	$(PREPROC) $< charmap.txt | $(AS) $(ASFLAGS) -o $@
+
 # Link objects to produce the ROM.
 $(ROM): $(OBJS)
 	./pokeld -T ld_script.txt -T iwram_syms.txt -T ewram_syms.txt -o $(ELF) $(OBJS)
diff --git a/asm/charmap.s b/asm/charmap.s
deleted file mode 100644
index 4b0e19917..000000000
--- a/asm/charmap.s
+++ /dev/null
@@ -1,280 +0,0 @@
-	.charmap " ", 0x00
-	.charmap "é", 0x1B
-	.charmap "&", 0x2D
-	.charmap "%", 0x5B
-	.charmap "(", 0x5C
-	.charmap ")", 0x5D
-	.charmap "0", 0xA1
-	.charmap "1", 0xA2
-	.charmap "2", 0xA3
-	.charmap "3", 0xA4
-	.charmap "4", 0xA5
-	.charmap "5", 0xA6
-	.charmap "6", 0xA7
-	.charmap "7", 0xA8
-	.charmap "8", 0xA9
-	.charmap "9", 0xAA
-	.charmap "!", 0xAB
-	.charmap "?", 0xAC
-	.charmap ".", 0xAD
-	.charmap "-", 0xAE
-	.charmap "·", 0xAF
-	.charmap "…", 0xB0
-	.charmap "“", 0xB1
-	.charmap "”", 0xB2
-	.charmap "‘", 0xB3
-	.charmap "’", 0xB4
-	.charmap "♂", 0xB5
-	.charmap "♀", 0xB6
-	.charmap "¥", 0xB7
-	.charmap ",", 0xB8
-	.charmap "×", 0xB9
-	.charmap "/", 0xBA
-	.charmap "A", 0xBB
-	.charmap "B", 0xBC
-	.charmap "C", 0xBD
-	.charmap "D", 0xBE
-	.charmap "E", 0xBF
-	.charmap "F", 0xC0
-	.charmap "G", 0xC1
-	.charmap "H", 0xC2
-	.charmap "I", 0xC3
-	.charmap "J", 0xC4
-	.charmap "K", 0xC5
-	.charmap "L", 0xC6
-	.charmap "M", 0xC7
-	.charmap "N", 0xC8
-	.charmap "O", 0xC9
-	.charmap "P", 0xCA
-	.charmap "Q", 0xCB
-	.charmap "R", 0xCC
-	.charmap "S", 0xCD
-	.charmap "T", 0xCE
-	.charmap "U", 0xCF
-	.charmap "V", 0xD0
-	.charmap "W", 0xD1
-	.charmap "X", 0xD2
-	.charmap "Y", 0xD3
-	.charmap "Z", 0xD4
-	.charmap "a", 0xD5
-	.charmap "b", 0xD6
-	.charmap "c", 0xD7
-	.charmap "d", 0xD8
-	.charmap "e", 0xD9
-	.charmap "f", 0xDA
-	.charmap "g", 0xDB
-	.charmap "h", 0xDC
-	.charmap "i", 0xDD
-	.charmap "j", 0xDE
-	.charmap "k", 0xDF
-	.charmap "l", 0xE0
-	.charmap "m", 0xE1
-	.charmap "n", 0xE2
-	.charmap "o", 0xE3
-	.charmap "p", 0xE4
-	.charmap "q", 0xE5
-	.charmap "r", 0xE6
-	.charmap "s", 0xE7
-	.charmap "t", 0xE8
-	.charmap "u", 0xE9
-	.charmap "v", 0xEA
-	.charmap "w", 0xEB
-	.charmap "x", 0xEC
-	.charmap "y", 0xED
-	.charmap "z", 0xEE
-	.charmap ":", 0xF0
-	.charmap "$", 0xFF
-
-@ Hiragana
-	.charmap "あ", 0x01
-	.charmap "い", 0x02
-	.charmap "う", 0x03
-	.charmap "え", 0x04
-	.charmap "お", 0x05
-	.charmap "か", 0x06
-	.charmap "き", 0x07
-	.charmap "く", 0x08
-	.charmap "け", 0x09
-	.charmap "こ", 0x0A
-	.charmap "さ", 0x0B
-	.charmap "し", 0x0C
-	.charmap "す", 0x0D
-	.charmap "せ", 0x0E
-	.charmap "そ", 0x0F
-	.charmap "た", 0x10
-	.charmap "ち", 0x11
-	.charmap "つ", 0x12
-	.charmap "て", 0x13
-	.charmap "と", 0x14
-	.charmap "な", 0x15
-	.charmap "に", 0x16
-	.charmap "ぬ", 0x17
-	.charmap "ね", 0x18
-	.charmap "の", 0x19
-	.charmap "は", 0x1A
-	.charmap "ひ", 0x1B
-	.charmap "ふ", 0x1C
-	.charmap "へ", 0x1D
-	.charmap "ほ", 0x1E
-	.charmap "ま", 0x1F
-	.charmap "み", 0x20
-	.charmap "む", 0x21
-	.charmap "め", 0x22
-	.charmap "も", 0x23
-	.charmap "や", 0x24
-	.charmap "ゆ", 0x25
-	.charmap "よ", 0x26
-	.charmap "ら", 0x27
-	.charmap "り", 0x28
-	.charmap "る", 0x29
-	.charmap "れ", 0x2A
-	.charmap "ろ", 0x2B
-	.charmap "わ", 0x2C
-	.charmap "を", 0x2D
-	.charmap "ん", 0x2E
-	.charmap "ぁ", 0x2F
-	.charmap "ぃ", 0x30
-	.charmap "ぅ", 0x31
-	.charmap "ぇ", 0x32
-	.charmap "ぉ", 0x33
-	.charmap "ゃ", 0x34
-	.charmap "ゅ", 0x35
-	.charmap "ょ", 0x36
-	.charmap "が", 0x37
-	.charmap "ぎ", 0x38
-	.charmap "ぐ", 0x39
-	.charmap "げ", 0x3A
-	.charmap "ご", 0x3B
-	.charmap "ざ", 0x3C
-	.charmap "じ", 0x3D
-	.charmap "ず", 0x3E
-	.charmap "ぜ", 0x3F
-	.charmap "ぞ", 0x40
-	.charmap "だ", 0x41
-	.charmap "ぢ", 0x42
-	.charmap "づ", 0x43
-	.charmap "で", 0x44
-	.charmap "ど", 0x45
-	.charmap "ば", 0x46
-	.charmap "び", 0x47
-	.charmap "ぶ", 0x48
-	.charmap "べ", 0x49
-	.charmap "ぼ", 0x4A
-	.charmap "ぱ", 0x4B
-	.charmap "ぴ", 0x4C
-	.charmap "ぷ", 0x4D
-	.charmap "ぺ", 0x4E
-	.charmap "ぽ", 0x4F
-	.charmap "っ", 0x50
-
-@ Katakana
-	.charmap "ア", 0x51
-	.charmap "イ", 0x52
-	.charmap "ウ", 0x53
-	.charmap "エ", 0x54
-	.charmap "オ", 0x55
-	.charmap "カ", 0x56
-	.charmap "キ", 0x57
-	.charmap "ク", 0x58
-	.charmap "ケ", 0x59
-	.charmap "コ", 0x5A
-	.charmap "サ", 0x5B
-	.charmap "シ", 0x5C
-	.charmap "ス", 0x5D
-	.charmap "セ", 0x5E
-	.charmap "ソ", 0x5F
-	.charmap "タ", 0x60
-	.charmap "チ", 0x61
-	.charmap "ツ", 0x62
-	.charmap "テ", 0x63
-	.charmap "ト", 0x64
-	.charmap "ナ", 0x65
-	.charmap "ニ", 0x66
-	.charmap "ヌ", 0x67
-	.charmap "ネ", 0x68
-	.charmap "ノ", 0x69
-	.charmap "ハ", 0x6A
-	.charmap "ヒ", 0x6B
-	.charmap "フ", 0x6C
-	.charmap "ヘ", 0x6D
-	.charmap "ホ", 0x6E
-	.charmap "マ", 0x6F
-	.charmap "ミ", 0x70
-	.charmap "ム", 0x71
-	.charmap "メ", 0x72
-	.charmap "モ", 0x73
-	.charmap "ヤ", 0x74
-	.charmap "ユ", 0x75
-	.charmap "ヨ", 0x76
-	.charmap "ラ", 0x77
-	.charmap "リ", 0x78
-	.charmap "ル", 0x79
-	.charmap "レ", 0x7A
-	.charmap "ロ", 0x7B
-	.charmap "ワ", 0x7C
-	.charmap "ヲ", 0x7D
-	.charmap "ン", 0x7E
-	.charmap "ァ", 0x7F
-	.charmap "ィ", 0x80
-	.charmap "ゥ", 0x81
-	.charmap "ェ", 0x82
-	.charmap "ォ", 0x83
-	.charmap "ャ", 0x84
-	.charmap "ュ", 0x85
-	.charmap "ョ", 0x86
-	.charmap "ガ", 0x87
-	.charmap "ギ", 0x88
-	.charmap "グ", 0x89
-	.charmap "ゲ", 0x8A
-	.charmap "ゴ", 0x8B
-	.charmap "ザ", 0x8C
-	.charmap "ジ", 0x8D
-	.charmap "ズ", 0x8E
-	.charmap "ゼ", 0x8F
-	.charmap "ゾ", 0x90
-	.charmap "ダ", 0x91
-	.charmap "ヂ", 0x92
-	.charmap "ヅ", 0x93
-	.charmap "デ", 0x94
-	.charmap "ド", 0x95
-	.charmap "バ", 0x96
-	.charmap "ビ", 0x97
-	.charmap "ブ", 0x98
-	.charmap "ベ", 0x99
-	.charmap "ボ", 0x9A
-	.charmap "パ", 0x9B
-	.charmap "ピ", 0x9C
-	.charmap "プ", 0x9D
-	.charmap "ペ", 0x9E
-	.charmap "ポ", 0x9F
-	.charmap "ッ", 0xA0
-
-	.charmap "ー", 0xAE
-
-	.charmap_const PK, 0x53
-	.charmap_const PKMN, 0x53, 0x54
-	.charmap_const POKEBLOCK, 0x55, 0x56, 0x57, 0x58, 0x59
-
-	@ indicates the end of a town/city name (before " TOWN" or " CITY")
-	.charmap_const NAME_END, 0xFC, 0x00
-
-@ string placeholders
-	@ unknown (0xFD, 0x00)
-	.charmap_const PLAYER,             0xFD, 0x01
-	.charmap_const STR_VAR_1,          0xFD, 0x02
-	.charmap_const STR_VAR_2,          0xFD, 0x03
-	.charmap_const STR_VAR_3,          0xFD, 0x04
-	@ unknown (0xFD, 0x05)
-	.charmap_const RIVAL,              0xFD, 0x06
-@ version-dependent strings
-	.charmap_const VERSION,            0xFD, 0x07 @ "RUBY"    / "SAPPHIRE"
-	.charmap_const EVIL_TEAM,          0xFD, 0x08 @ "MAGMA"   / "AQUA"
-	.charmap_const GOOD_TEAM,          0xFD, 0x09 @ "AQUA"    / "MAGMA"
-	.charmap_const EVIL_TEAM_LEADER,   0xFD, 0x0A @ "MAXIE"   / "ARCHIE"
-	.charmap_const GOOD_TEAM_LEADER,   0xFD, 0x0B @ "ARCHIE"  / "MAXIE"
-	.charmap_const CUR_VERSION_MASCOT, 0xFD, 0x0C @ "GROUDON" / "KYOGRE"
-	.charmap_const OPP_VERSION_MASCOT, 0xFD, 0x0D @ "KYOGRE"  / "GROUDON"
-
-	.charmap_escape 'l', 0xFA @ scroll up window text
-	.charmap_escape 'p', 0xFB @ new paragraph
-	.charmap_escape 'n', 0xFE @ new line
diff --git a/charmap.txt b/charmap.txt
new file mode 100644
index 000000000..0bd2b7c9d
--- /dev/null
+++ b/charmap.txt
@@ -0,0 +1,280 @@
+' ' = 00
+'é' = 1B
+'&' = 2D
+'%' = 5B
+'(' = 5C
+')' = 5D
+'0' = A1
+'1' = A2
+'2' = A3
+'3' = A4
+'4' = A5
+'5' = A6
+'6' = A7
+'7' = A8
+'8' = A9
+'9' = AA
+'!' = AB
+'?' = AC
+'.' = AD
+'-' = AE
+'·' = AF
+'…' = B0
+'“' = B1
+'”' = B2
+'‘' = B3
+'’' = B4
+'♂' = B5
+'♀' = B6
+'¥' = B7
+',' = B8
+'×' = B9
+'/' = BA
+'A' = BB
+'B' = BC
+'C' = BD
+'D' = BE
+'E' = BF
+'F' = C0
+'G' = C1
+'H' = C2
+'I' = C3
+'J' = C4
+'K' = C5
+'L' = C6
+'M' = C7
+'N' = C8
+'O' = C9
+'P' = CA
+'Q' = CB
+'R' = CC
+'S' = CD
+'T' = CE
+'U' = CF
+'V' = D0
+'W' = D1
+'X' = D2
+'Y' = D3
+'Z' = D4
+'a' = D5
+'b' = D6
+'c' = D7
+'d' = D8
+'e' = D9
+'f' = DA
+'g' = DB
+'h' = DC
+'i' = DD
+'j' = DE
+'k' = DF
+'l' = E0
+'m' = E1
+'n' = E2
+'o' = E3
+'p' = E4
+'q' = E5
+'r' = E6
+'s' = E7
+'t' = E8
+'u' = E9
+'v' = EA
+'w' = EB
+'x' = EC
+'y' = ED
+'z' = EE
+':' = F0
+'$' = FF
+
+@ Hiragana
+'あ' = 01
+'い' = 02
+'う' = 03
+'え' = 04
+'お' = 05
+'か' = 06
+'き' = 07
+'く' = 08
+'け' = 09
+'こ' = 0A
+'さ' = 0B
+'し' = 0C
+'す' = 0D
+'せ' = 0E
+'そ' = 0F
+'た' = 10
+'ち' = 11
+'つ' = 12
+'て' = 13
+'と' = 14
+'な' = 15
+'に' = 16
+'ぬ' = 17
+'ね' = 18
+'の' = 19
+'は' = 1A
+'ひ' = 1B
+'ふ' = 1C
+'へ' = 1D
+'ほ' = 1E
+'ま' = 1F
+'み' = 20
+'む' = 21
+'め' = 22
+'も' = 23
+'や' = 24
+'ゆ' = 25
+'よ' = 26
+'ら' = 27
+'り' = 28
+'る' = 29
+'れ' = 2A
+'ろ' = 2B
+'わ' = 2C
+'を' = 2D
+'ん' = 2E
+'ぁ' = 2F
+'ぃ' = 30
+'ぅ' = 31
+'ぇ' = 32
+'ぉ' = 33
+'ゃ' = 34
+'ゅ' = 35
+'ょ' = 36
+'が' = 37
+'ぎ' = 38
+'ぐ' = 39
+'げ' = 3A
+'ご' = 3B
+'ざ' = 3C
+'じ' = 3D
+'ず' = 3E
+'ぜ' = 3F
+'ぞ' = 40
+'だ' = 41
+'ぢ' = 42
+'づ' = 43
+'で' = 44
+'ど' = 45
+'ば' = 46
+'び' = 47
+'ぶ' = 48
+'べ' = 49
+'ぼ' = 4A
+'ぱ' = 4B
+'ぴ' = 4C
+'ぷ' = 4D
+'ぺ' = 4E
+'ぽ' = 4F
+'っ' = 50
+
+@ Katakana
+'ア' = 51
+'イ' = 52
+'ウ' = 53
+'エ' = 54
+'オ' = 55
+'カ' = 56
+'キ' = 57
+'ク' = 58
+'ケ' = 59
+'コ' = 5A
+'サ' = 5B
+'シ' = 5C
+'ス' = 5D
+'セ' = 5E
+'ソ' = 5F
+'タ' = 60
+'チ' = 61
+'ツ' = 62
+'テ' = 63
+'ト' = 64
+'ナ' = 65
+'ニ' = 66
+'ヌ' = 67
+'ネ' = 68
+'ノ' = 69
+'ハ' = 6A
+'ヒ' = 6B
+'フ' = 6C
+'ヘ' = 6D
+'ホ' = 6E
+'マ' = 6F
+'ミ' = 70
+'ム' = 71
+'メ' = 72
+'モ' = 73
+'ヤ' = 74
+'ユ' = 75
+'ヨ' = 76
+'ラ' = 77
+'リ' = 78
+'ル' = 79
+'レ' = 7A
+'ロ' = 7B
+'ワ' = 7C
+'ヲ' = 7D
+'ン' = 7E
+'ァ' = 7F
+'ィ' = 80
+'ゥ' = 81
+'ェ' = 82
+'ォ' = 83
+'ャ' = 84
+'ュ' = 85
+'ョ' = 86
+'ガ' = 87
+'ギ' = 88
+'グ' = 89
+'ゲ' = 8A
+'ゴ' = 8B
+'ザ' = 8C
+'ジ' = 8D
+'ズ' = 8E
+'ゼ' = 8F
+'ゾ' = 90
+'ダ' = 91
+'ヂ' = 92
+'ヅ' = 93
+'デ' = 94
+'ド' = 95
+'バ' = 96
+'ビ' = 97
+'ブ' = 98
+'ベ' = 99
+'ボ' = 9A
+'パ' = 9B
+'ピ' = 9C
+'プ' = 9D
+'ペ' = 9E
+'ポ' = 9F
+'ッ' = A0
+
+'ー' = AE
+
+PK        = 53
+PKMN      = 53 54
+POKEBLOCK = 55 56 57 58 59
+
+@ indicates the end of a town/city name (before " TOWN" or " CITY")
+NAME_END = FC 00
+
+@ string placeholders
+@ unknown           (FD 00)
+PLAYER             = FD 01
+STR_VAR_1          = FD 02
+STR_VAR_2          = FD 03
+STR_VAR_3          = FD 04
+@ unknown           (FD 05)
+RIVAL              = FD 06
+@ version-dependent strings
+VERSION            = FD 07 @ "RUBY"    / "SAPPHIRE"
+EVIL_TEAM          = FD 08 @ "MAGMA"   / "AQUA"
+GOOD_TEAM          = FD 09 @ "AQUA"    / "MAGMA"
+EVIL_TEAM_LEADER   = FD 0A @ "MAXIE"   / "ARCHIE"
+GOOD_TEAM_LEADER   = FD 0B @ "ARCHIE"  / "MAXIE"
+CUR_VERSION_MASCOT = FD 0C @ "GROUDON" / "KYOGRE"
+OPP_VERSION_MASCOT = FD 0D @ "KYOGRE"  / "GROUDON"
+
+'\l' = FA @ scroll up window text
+'\p' = FB @ new paragraph
+'\n' = FE @ new line
diff --git a/data/data1.s b/data/data1.s
index ebfa367a5..6428dbac6 100644
--- a/data/data1.s
+++ b/data/data1.s
@@ -2,7 +2,6 @@
 
 	.include "asm/macros.s"
 	.include "constants/constants.s"
-	.include "asm/charmap.s"
 
 	.section script_data, "aw", %progbits
 
diff --git a/data/data2.s b/data/data2.s
index cdddccb8f..d0a3611c5 100644
--- a/data/data2.s
+++ b/data/data2.s
@@ -2,7 +2,6 @@
 
 	.include "asm/macros.s"
 	.include "constants/constants.s"
-	.include "asm/charmap.s"
 
 	.section .rodata
 
diff --git a/fix_local_labels.pl b/fix_local_labels.pl
deleted file mode 100644
index d3258c766..000000000
--- a/fix_local_labels.pl
+++ /dev/null
@@ -1,28 +0,0 @@
-use strict;
-use warnings;
-
-open(IN_FILE, $ARGV[0]);
-open(OUT_FILE, ">", $ARGV[1]);
-
-my @labels = ();
-
-while (<IN_FILE>) {
-	if ($_ =~ /^\.(.+):/) {
-		push(@labels, $1);
-	}
-}
-
-seek IN_FILE, 0, 0;
-
-while (<IN_FILE>) {
-	for (my $i = 0; $i < scalar(@labels); $i++) {
-		my $find = quotemeta '.' . $labels[$i];
-		my $replace = '$' . $labels[$i];
-		$_ =~ s/$find/$replace/;
-	}
-
-	print OUT_FILE $_;
-}
-
-close(IN_FILE);
-close(OUT_FILE);
diff --git a/tools/preproc/.gitignore b/tools/preproc/.gitignore
new file mode 100644
index 000000000..eb3470879
--- /dev/null
+++ b/tools/preproc/.gitignore
@@ -0,0 +1 @@
+preproc
diff --git a/tools/preproc/LICENSE b/tools/preproc/LICENSE
new file mode 100644
index 000000000..534d15349
--- /dev/null
+++ b/tools/preproc/LICENSE
@@ -0,0 +1,19 @@
+Copyright (c) 2016 YamaArashi
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in
+all copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+THE SOFTWARE.
diff --git a/tools/preproc/Makefile b/tools/preproc/Makefile
new file mode 100644
index 000000000..10f930ebd
--- /dev/null
+++ b/tools/preproc/Makefile
@@ -0,0 +1,15 @@
+CXX := g++
+
+CXXFLAGS := -std=c++14 -O2
+
+SRCS := asm_file.cpp charmap.cpp preproc.cpp utf8.cpp
+
+HEADERS := asm_file.h char_util.h charmap.h preproc.h utf8.h
+
+.PHONY: clean
+
+preproc: $(SRCS) $(HEADERS)
+	$(CXX) $(CXXFLAGS) $(SRCS) -o $@
+
+clean:
+	$(RM) preproc preproc.exe
diff --git a/tools/preproc/asm_file.cpp b/tools/preproc/asm_file.cpp
new file mode 100644
index 000000000..0b42d9ab2
--- /dev/null
+++ b/tools/preproc/asm_file.cpp
@@ -0,0 +1,563 @@
+// Copyright(c) 2016 YamaArashi
+//
+// Permission is hereby granted, free of charge, to any person obtaining a copy
+// of this software and associated documentation files (the "Software"), to deal
+// in the Software without restriction, including without limitation the rights
+// to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+// copies of the Software, and to permit persons to whom the Software is
+// furnished to do so, subject to the following conditions:
+//
+// The above copyright notice and this permission notice shall be included in
+// all copies or substantial portions of the Software.
+//
+// THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+// IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+// FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+// AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+// LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+// OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+// THE SOFTWARE.
+
+#include <cstdio>
+#include <cstdarg>
+#include "preproc.h"
+#include "asm_file.h"
+#include "char_util.h"
+#include "utf8.h"
+
+AsmFile::AsmFile(std::string filename) : m_filename(filename)
+{
+    FILE *fp = std::fopen(filename.c_str(), "rb");
+
+    if (fp == NULL)
+        FATAL_ERROR("Failed to open \"%s\" for reading.\n", filename.c_str());
+
+    std::fseek(fp, 0, SEEK_END);
+
+    m_size = std::ftell(fp);
+
+    m_buffer = new char[m_size + 1];
+
+    std::rewind(fp);
+
+    if (std::fread(m_buffer, m_size, 1, fp) != 1)
+        FATAL_ERROR("Failed to read \"%s\".\n", filename.c_str());
+
+    m_buffer[m_size] = 0;
+
+    std::fclose(fp);
+
+    m_pos = 0;
+    m_lineNum = 1;
+    m_lineStart = 0;
+
+    RemoveComments();
+}
+
+AsmFile::AsmFile(AsmFile&& other) : m_filename(std::move(other.m_filename))
+{
+    m_buffer = other.m_buffer;
+    m_pos = other.m_pos;
+    m_size = other.m_size;
+    m_lineNum = other.m_lineNum;
+    m_lineStart = other.m_lineStart;
+
+    other.m_buffer = nullptr;
+}
+
+AsmFile::~AsmFile()
+{
+    delete[] m_buffer;
+}
+
+// Removes comments to simplify further processing.
+// It stops upon encountering a null character,
+// which may or may not be the end of file marker.
+// If it's not, the error will be caught later.
+void AsmFile::RemoveComments()
+{
+    long pos = 0;
+    bool inString = false;
+    char stringChar;
+
+    for (;;)
+    {
+        if (m_buffer[pos] == 0)
+            return;
+
+        if (inString)
+        {
+            if (m_buffer[pos] == '\\' && m_buffer[pos + 1] == stringChar)
+            {
+                pos += 2;
+            }
+            else
+            {
+                if (m_buffer[pos] == stringChar)
+                    inString = false;
+                pos++;
+            }
+        }
+        else if (m_buffer[pos] == '@' && (pos == 0 || m_buffer[pos - 1] != '\\'))
+        {
+            while (m_buffer[pos] != '\n' && m_buffer[pos] != 0)
+                m_buffer[pos++] = ' ';
+        }
+        else if (m_buffer[pos] == '/' && m_buffer[pos + 1] == '*')
+        {
+            m_buffer[pos++] = ' ';
+            m_buffer[pos++] = ' ';
+
+            bool inCommentString = false;
+            char commentStringChar;
+
+            for (;;)
+            {
+                if (m_buffer[pos] == 0)
+                    return;
+
+                if (inCommentString)
+                {
+                    if (m_buffer[pos] == '\\' && m_buffer[pos + 1] == commentStringChar)
+                    {
+                        m_buffer[pos++] = ' ';
+                        m_buffer[pos++] = ' ';
+                    }
+                    else
+                    {
+                        if (m_buffer[pos] == commentStringChar)
+                            inCommentString = false;
+                        if (m_buffer[pos] != '\n')
+                            m_buffer[pos] = ' ';
+                        pos++;
+                    }
+                }
+                else
+                {
+                    if (m_buffer[pos] == '*' && m_buffer[pos + 1] == '/')
+                    {
+                        m_buffer[pos++] = ' ';
+                        m_buffer[pos++] = ' ';
+                        break;
+                    }
+                    else
+                    {
+                        if (m_buffer[pos] == '"' || m_buffer[pos] == '\'')
+                        {
+                            commentStringChar = m_buffer[pos];
+                            inCommentString = true;
+                        }
+                        if (m_buffer[pos] != '\n')
+                            m_buffer[pos] = ' ';
+                        pos++;
+                    }
+                }
+            }
+        }
+        else
+        {
+            if (m_buffer[pos] == '"' || m_buffer[pos] == '\'')
+            {
+                stringChar = m_buffer[pos];
+                inString = true;
+            }
+            pos++;
+        }
+    }
+}
+
+// Checks if we're at a particular directive and if so, consumes it.
+// Returns whether the directive was found.
+bool AsmFile::CheckForDirective(std::string name)
+{
+    long i;
+    long length = static_cast<long>(name.length());
+
+    for (i = 0; i < length && m_pos + i < m_size; i++)
+        if (name[i] != m_buffer[m_pos + i])
+            return false;
+
+    if (i < length)
+        return false;
+
+    m_pos += length;
+
+    return true;
+}
+
+// Checks if we're at a known directive and if so, consumes it.
+// Returns which directive was found.
+Directive AsmFile::GetDirective()
+{
+    SkipWhitespace();
+
+    if (CheckForDirective(".include"))
+        return Directive::Include;
+    else if (CheckForDirective(".string"))
+        return Directive::String;
+    else
+        return Directive::Unknown;
+}
+
+// Skips tabs and spaces.
+void AsmFile::SkipWhitespace()
+{
+    while (m_buffer[m_pos] == '\t' || m_buffer[m_pos] == ' ')
+        m_pos++;
+}
+
+// Reads include path.
+std::string AsmFile::ReadPath()
+{
+    SkipWhitespace();
+
+    if (m_buffer[m_pos] != '"')
+        RaiseError("expected file path");
+
+    m_pos++;
+
+    int length = 0;
+    long startPos = m_pos;
+
+    while (m_buffer[m_pos] != '"')
+    {
+        unsigned char c = m_buffer[m_pos++];
+
+        if (c == 0)
+        {
+            if (m_pos >= m_size)
+                RaiseError("unexpected EOF in include string");
+            else
+                RaiseError("unexpected null character in include string");
+        }
+
+        if (!IsAsciiPrintable(c))
+            RaiseError("unexpected character '\\x%02X' in include string", c);
+
+        // Don't bother allowing any escape sequences.
+        if (c == '\\')
+        {
+            c = m_buffer[m_pos];
+            RaiseError("unexpected escape '\\%c' in include string", c);
+        }
+
+        length++;
+
+        if (length > kMaxPath)
+            RaiseError("path is too long");
+    }
+
+    m_pos++; // Go past the right quote.
+
+    ExpectEmptyRestOfLine();
+
+    return std::string(&m_buffer[startPos], length);
+}
+
+// Reads a charmap char or escape sequence.
+std::string AsmFile::ReadCharOrEscape()
+{
+    std::string sequence;
+
+    bool isEscape = (m_buffer[m_pos] == '\\');
+
+    if (isEscape)
+    {
+        m_pos++;
+
+        if (m_buffer[m_pos] == '"')
+        {
+            sequence = g_charmap->Char('"');
+
+            if (sequence.length() == 0)
+                RaiseError("no mapping exists for double quote");
+
+            return sequence;
+        }
+        else if (m_buffer[m_pos] == '\\')
+        {
+            sequence = g_charmap->Char('\\');
+
+            if (sequence.length() == 0)
+                RaiseError("no mapping exists for backslash");
+
+            return sequence;
+        }
+    }
+
+    unsigned char c = m_buffer[m_pos];
+
+    if (c == 0)
+    {
+        if (m_pos >= m_size)
+            RaiseError("unexpected EOF in UTF-8 string");
+        else
+            RaiseError("unexpected null character in UTF-8 string");
+    }
+
+    if (IsAscii(c) && !IsAsciiPrintable(c))
+        RaiseError("unexpected character U+%X in UTF-8 string", c);
+
+    UnicodeChar unicodeChar = DecodeUtf8(&m_buffer[m_pos]);
+    m_pos += unicodeChar.encodingLength;
+    std::int32_t code = unicodeChar.code;
+
+    if (code == -1)
+        RaiseError("invalid encoding in UTF-8 string");
+
+    if (isEscape && code >= 128)
+        RaiseError("escapes using non-ASCII characters are invalid");
+
+    sequence = isEscape ? g_charmap->Escape(code) : g_charmap->Char(code);
+
+    if (sequence.length() == 0)
+    {
+        if (isEscape)
+            RaiseError("unknown escape '\\%c'", code);
+        else
+            RaiseError("unknown character U+%X", code);
+    }
+
+    return sequence;
+}
+
+// Reads a charmap constant, i.e. "{FOO}".
+std::string AsmFile::ReadConstant()
+{
+    m_pos++; // Assume we're on the left curly bracket.
+
+    long startPos = m_pos;
+
+    while (IsIdentifierChar(m_buffer[m_pos]))
+        m_pos++;
+
+    if (m_buffer[m_pos] != '}')
+    {
+        if (m_buffer[m_pos] == 0)
+        {
+            if (m_pos >= m_size)
+                RaiseError("unexpected EOF in identifier");
+            else
+                RaiseError("unexpected null character in identifier");
+        }
+
+        RaiseError("unexpected character in identifier");
+    }
+
+    std::string sequence = g_charmap->Constant(std::string(&m_buffer[startPos], m_pos - startPos));
+
+    if (sequence.length() == 0)
+    {
+        m_buffer[m_pos] = 0;
+        RaiseError("unknown constant '%s'", &m_buffer[startPos]);
+    }
+
+    m_pos++; // Go past the right curly bracket.
+
+    return sequence;
+}
+
+// Reads a charmap string.
+int AsmFile::ReadString(unsigned char* s)
+{
+    SkipWhitespace();
+
+    if (m_buffer[m_pos] != '"')
+        RaiseError("expected UTF-8 string literal");
+
+    m_pos++;
+
+    int length = 0;
+
+    while (m_buffer[m_pos] != '"')
+    {
+        std::string sequence = (m_buffer[m_pos] == '{') ? ReadConstant() : ReadCharOrEscape();
+
+        for (const char& c : sequence)
+        {
+            if (length == kMaxStringLength)
+                RaiseError("mapped string longer than %d bytes", length);
+
+            s[length++] = c;
+        }
+    }
+
+    m_pos++; // Go past the right quote.
+
+    SkipWhitespace();
+
+    if (ConsumeComma())
+    {
+        SkipWhitespace();
+        int padLength = ReadPadLength();
+
+        while (length < padLength)
+        {
+            s[length++] = 0;
+        }
+    }
+
+    ExpectEmptyRestOfLine();
+
+    return length;
+}
+
+// If we're at a comma, consumes it.
+// Returns whether a comma was found.
+bool AsmFile::ConsumeComma()
+{
+    if (m_buffer[m_pos] == ',')
+    {
+        m_pos++;
+        return true;
+    }
+
+    return false;
+}
+
+// Converts digit character to numerical value.
+static int ConvertDigit(char c, int radix)
+{
+    int digit;
+
+    if (c >= '0' && c <= '9')
+        digit = c - '0';
+    else if (c >= 'A' && c <= 'F')
+        digit = 10 + c - 'A';
+    else if (c >= 'a' && c <= 'f')
+        digit = 10 + c - 'a';
+    else
+        return -1;
+
+    return (digit < radix) ? digit : -1;
+}
+
+// Reads the pad length for a charmap string.
+int AsmFile::ReadPadLength()
+{
+    if (!IsAsciiDigit(m_buffer[m_pos]))
+        RaiseError("expected integer");
+
+    int radix = 10;
+
+    if (m_buffer[m_pos] == '0' && m_buffer[m_pos + 1] == 'x')
+    {
+        radix = 16;
+        m_pos += 2;
+    }
+
+    int n = 0;
+    int digit;
+
+    while ((digit = ConvertDigit(m_buffer[m_pos], radix)) != -1)
+    {
+        n = n * radix + digit;
+
+        if (n > kMaxStringLength)
+            RaiseError("pad length greater than maximum length (%d)", kMaxStringLength);
+
+        m_pos++;
+    }
+
+    return n;
+}
+
+// Outputs the current line and moves to the next one.
+void AsmFile::OutputLine()
+{
+    while (m_buffer[m_pos] != '\n' && m_buffer[m_pos] != 0)
+        m_pos++;
+
+    if (m_buffer[m_pos] == 0)
+    {
+        if (m_pos >= m_size)
+        {
+            RaiseWarning("file doesn't end with newline");
+            puts(&m_buffer[m_lineStart]);
+        }
+        else
+        {
+            RaiseError("unexpected null character");
+        }
+    }
+    else
+    {
+        m_buffer[m_pos] = 0;
+        puts(&m_buffer[m_lineStart]);
+        m_buffer[m_pos] = '\n';
+        m_pos++;
+        m_lineStart = m_pos;
+        m_lineNum++;
+    }
+}
+
+// Asserts that the rest of the line is empty and moves to the next one.
+void AsmFile::ExpectEmptyRestOfLine()
+{
+    SkipWhitespace();
+
+    if (m_buffer[m_pos] == 0)
+    {
+        if (m_pos >= m_size)
+            RaiseWarning("file doesn't end with newline");
+        else
+            RaiseError("unexpected null character");
+    }
+    else if (m_buffer[m_pos] == '\n')
+    {
+        m_pos++;
+        m_lineStart = m_pos;
+        m_lineNum++;
+    }
+    else if (m_buffer[m_pos] == '\r')
+    {
+        RaiseError("only Unix-style LF newlines are supported");
+    }
+    else
+    {
+        RaiseError("junk at end of line");
+    }
+}
+
+// Checks if we're at the end of the file.
+bool AsmFile::IsAtEnd()
+{
+    return (m_pos >= m_size);
+}
+
+// Output the current location to set gas's logical file and line numbers.
+void AsmFile::OutputLocation()
+{
+    printf("# %ld \"%s\"\n", m_lineNum, m_filename.c_str());
+}
+
+// Reports a diagnostic message.
+void AsmFile::ReportDiagnostic(const char* type, const char* format, std::va_list args)
+{
+    const int bufferSize = 1024;
+    char buffer[bufferSize];
+    std::vsnprintf(buffer, bufferSize, format, args);
+    std::fprintf(stderr, "%s:%ld: %s: %s\n", m_filename.c_str(), m_lineNum, type, buffer);
+}
+
+#define DO_REPORT(type)                   \
+do                                        \
+{                                         \
+    std::va_list args;                    \
+    va_start(args, format);               \
+    ReportDiagnostic(type, format, args); \
+    va_end(args);                         \
+} while (0)
+
+// Reports an error diagnostic and terminates the program.
+void AsmFile::RaiseError(const char* format, ...)
+{
+    DO_REPORT("error");
+    exit(1);
+}
+
+// Reports a warning diagnostic.
+void AsmFile::RaiseWarning(const char* format, ...)
+{
+    DO_REPORT("warning");
+}
diff --git a/tools/preproc/asm_file.h b/tools/preproc/asm_file.h
new file mode 100644
index 000000000..1c137d7d3
--- /dev/null
+++ b/tools/preproc/asm_file.h
@@ -0,0 +1,71 @@
+// Copyright(c) 2016 YamaArashi
+//
+// Permission is hereby granted, free of charge, to any person obtaining a copy
+// of this software and associated documentation files (the "Software"), to deal
+// in the Software without restriction, including without limitation the rights
+// to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+// copies of the Software, and to permit persons to whom the Software is
+// furnished to do so, subject to the following conditions:
+//
+// The above copyright notice and this permission notice shall be included in
+// all copies or substantial portions of the Software.
+//
+// THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+// IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+// FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+// AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+// LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+// OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+// THE SOFTWARE.
+
+#ifndef ASM_FILE_H
+#define ASM_FILE_H
+
+#include <cstdarg>
+#include <cstdint>
+#include <string>
+#include "preproc.h"
+
+enum class Directive
+{
+    Include,
+    String,
+    Unknown
+};
+
+class AsmFile
+{
+public:
+    AsmFile(std::string filename);
+    AsmFile(AsmFile&& other);
+    AsmFile(const AsmFile&) = delete;
+    ~AsmFile();
+    Directive GetDirective();
+    std::string ReadPath();
+    int ReadString(unsigned char* s);
+    bool IsAtEnd();
+    void OutputLine();
+    void OutputLocation();
+
+private:
+    char* m_buffer;
+    long m_pos;
+    long m_size;
+    long m_lineNum;
+    long m_lineStart;
+    std::string m_filename;
+
+    bool ConsumeComma();
+    int ReadPadLength();
+    void RemoveComments();
+    bool CheckForDirective(std::string name);
+    std::string ReadCharOrEscape();
+    std::string ReadConstant();
+    void SkipWhitespace();
+    void ExpectEmptyRestOfLine();
+    void ReportDiagnostic(const char* type, const char* format, std::va_list args);
+    void RaiseError(const char* format, ...);
+    void RaiseWarning(const char *format, ...);
+};
+
+#endif // ASM_FILE_H
diff --git a/tools/preproc/char_util.h b/tools/preproc/char_util.h
new file mode 100644
index 000000000..05f9a1dd4
--- /dev/null
+++ b/tools/preproc/char_util.h
@@ -0,0 +1,65 @@
+// Copyright(c) 2016 YamaArashi
+//
+// Permission is hereby granted, free of charge, to any person obtaining a copy
+// of this software and associated documentation files (the "Software"), to deal
+// in the Software without restriction, including without limitation the rights
+// to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+// copies of the Software, and to permit persons to whom the Software is
+// furnished to do so, subject to the following conditions:
+//
+// The above copyright notice and this permission notice shall be included in
+// all copies or substantial portions of the Software.
+//
+// THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+// IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+// FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+// AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+// LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+// OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+// THE SOFTWARE.
+
+#ifndef CHAR_UTIL_H
+#define CHAR_UTIL_H
+
+#include <cstdint>
+#include <cassert>
+
+inline bool IsAscii(unsigned char c)
+{
+    return (c < 128);
+}
+
+inline bool IsAsciiAlpha(unsigned char c)
+{
+    return ((c >= 'A' && c <= 'Z') || (c >= 'a' && c <= 'z'));
+}
+
+inline bool IsAsciiDigit(unsigned char c)
+{
+    return (c >= '0' && c <= '9');
+}
+
+inline bool IsAsciiHexDigit(unsigned char c)
+{
+    return ((c >= '0' && c <= '9')
+        || (c >= 'a' && c <= 'f')
+        || (c >= 'A' && c <= 'F'));
+}
+
+inline bool IsAsciiAlphanum(unsigned char c)
+{
+    return (IsAsciiAlpha(c) || IsAsciiDigit(c));
+}
+
+inline bool IsAsciiPrintable(unsigned char c)
+{
+    return (c >= ' ' && c <= '~');
+}
+
+// Returns whether the character can be used in the identifier of a "{FOO}" constant in strings.
+inline bool IsIdentifierChar(unsigned char c)
+{
+    return IsAsciiAlphanum(c) || c == '_';
+}
+
+#endif // CHAR_UTIL_H
diff --git a/tools/preproc/charmap.cpp b/tools/preproc/charmap.cpp
new file mode 100644
index 000000000..de693eda5
--- /dev/null
+++ b/tools/preproc/charmap.cpp
@@ -0,0 +1,397 @@
+// Copyright(c) 2016 YamaArashi
+//
+// Permission is hereby granted, free of charge, to any person obtaining a copy
+// of this software and associated documentation files (the "Software"), to deal
+// in the Software without restriction, including without limitation the rights
+// to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+// copies of the Software, and to permit persons to whom the Software is
+// furnished to do so, subject to the following conditions:
+//
+// The above copyright notice and this permission notice shall be included in
+// all copies or substantial portions of the Software.
+//
+// THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+// IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+// FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+// AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+// LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+// OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+// THE SOFTWARE.
+
+#include <cstdio>
+#include <cstdint>
+#include <cstdarg>
+#include "preproc.h"
+#include "charmap.h"
+#include "char_util.h"
+#include "utf8.h"
+
+enum LhsType
+{
+    Char,
+    Escape,
+    Constant,
+    None
+};
+
+struct Lhs
+{
+    LhsType type;
+    std::string name;
+    std::int32_t code;
+};
+
+class CharmapReader
+{
+public:
+    CharmapReader(std::string filename);
+    CharmapReader(const CharmapReader&) = delete;
+    ~CharmapReader();
+    Lhs ReadLhs();
+    void ExpectEqualsSign();
+    std::string ReadSequence();
+    void ExpectEmptyRestOfLine();
+
+private:
+    char* m_buffer;
+    long m_pos;
+    long m_size;
+    long m_lineNum;
+    std::string m_filename;
+
+    void RaiseError(const char* format, ...);
+    void RemoveComments();
+    std::string ReadConstant();
+    void SkipWhitespace();
+};
+
+CharmapReader::CharmapReader(std::string filename) : m_filename(filename)
+{
+    FILE *fp = std::fopen(filename.c_str(), "rb");
+
+    if (fp == NULL)
+        FATAL_ERROR("Failed to open \"%s\" for reading.\n", filename.c_str());
+
+    std::fseek(fp, 0, SEEK_END);
+
+    m_size = std::ftell(fp);
+
+    m_buffer = new char[m_size + 1];
+
+    std::rewind(fp);
+
+    if (std::fread(m_buffer, m_size, 1, fp) != 1)
+        FATAL_ERROR("Failed to read \"%s\".\n", filename.c_str());
+
+    m_buffer[m_size] = 0;
+
+    std::fclose(fp);
+
+    m_pos = 0;
+    m_lineNum = 1;
+
+    RemoveComments();
+}
+
+CharmapReader::~CharmapReader()
+{
+    delete[] m_buffer;
+}
+
+Lhs CharmapReader::ReadLhs()
+{
+    Lhs lhs;
+
+    for (;;)
+    {
+        SkipWhitespace();
+
+        if (m_buffer[m_pos] == '\n')
+        {
+            m_pos++;
+            m_lineNum++;
+        }
+        else
+        {
+            break;
+        }
+    }
+    
+    if (m_buffer[m_pos] == '\'')
+    {
+        m_pos++;
+
+        bool isEscape = (m_buffer[m_pos] == '\\');
+
+        if (isEscape)
+        {
+            m_pos++;
+        }
+
+        unsigned char c = m_buffer[m_pos];
+
+        if (c == 0)
+        {
+            if (m_pos >= m_size)
+                RaiseError("unexpected EOF in UTF-8 character literal");
+            else
+                RaiseError("unexpected null character in UTF-8 character literal");
+        }
+
+        if (IsAscii(c) && !IsAsciiPrintable(c))
+            RaiseError("unexpected character U+%X in UTF-8 character literal", c);
+
+        UnicodeChar unicodeChar = DecodeUtf8(&m_buffer[m_pos]);
+        std::int32_t code = unicodeChar.code;
+
+        if (code == -1)
+            RaiseError("invalid encoding in UTF-8 character literal");
+
+        m_pos += unicodeChar.encodingLength;
+
+        if (m_buffer[m_pos] != '\'')
+            RaiseError("unterminated character literal");
+
+        m_pos++;
+
+        lhs.code = code;
+
+        if (isEscape)
+        {
+            if (code >= 128)
+                RaiseError("escapes using non-ASCII characters are invalid");
+
+            switch (code)
+            {
+            case '\'':
+                lhs.type = LhsType::Char;
+                break;
+            case '\\':
+                lhs.type = LhsType::Char;
+            case '"':
+                RaiseError("cannot escape double quote");
+                break;
+            default:
+                lhs.type = LhsType::Escape;
+            }
+        }
+        else
+        {
+            if (code == '\'')
+                RaiseError("empty character literal");
+
+            lhs.type = LhsType::Char;
+        }
+    }
+    else if (IsIdentifierChar(m_buffer[m_pos]))
+    {
+        lhs.type = LhsType::Constant;
+        lhs.name = ReadConstant();
+    }
+    else if (m_buffer[m_pos] == '\r')
+    {
+        RaiseError("only Unix-style LF newlines are supported");
+    }
+    else if (m_buffer[m_pos] == 0)
+    {
+        if (m_pos < m_size)
+            RaiseError("unexpected null character");
+        lhs.type = LhsType::None;
+    }
+    else
+    {
+        RaiseError("junk at start of line");
+    }
+
+    return lhs;
+}
+
+void CharmapReader::ExpectEqualsSign()
+{
+    SkipWhitespace();
+
+    if (m_buffer[m_pos] != '=')
+        RaiseError("expected equals sign");
+
+    m_pos++;
+}
+
+static unsigned int ConvertHexDigit(char c)
+{
+    unsigned int digit = 0;
+
+    if (c >= '0' && c <= '9')
+        digit = c - '0';
+    else if (c >= 'A' && c <= 'F')
+        digit = 10 + c - 'A';
+    else if (c >= 'a' && c <= 'f')
+        digit = 10 + c - 'a';
+
+    return digit;
+}
+
+std::string CharmapReader::ReadSequence()
+{
+    SkipWhitespace();
+
+    long startPos = m_pos;
+
+    unsigned int length = 0;
+
+    while (IsAsciiHexDigit(m_buffer[m_pos]) && IsAsciiHexDigit(m_buffer[m_pos + 1]))
+    {
+        m_pos += 2;
+        length++;
+
+        if (length > kMaxCharmapSequenceLength)
+            RaiseError("byte sequence too long (max is %lu bytes)", kMaxCharmapSequenceLength);
+
+        SkipWhitespace();
+    }
+
+    if (IsAsciiHexDigit(m_buffer[m_pos]))
+        RaiseError("each byte must have 2 hex digits");
+
+    if (length == 0)
+        RaiseError("expected byte sequence");
+
+    std::string sequence;
+    sequence.reserve(length);
+
+    m_pos = startPos;
+
+    for (unsigned int i = 0; i < length; i++)
+    {
+        unsigned int digit1 = ConvertHexDigit(m_buffer[m_pos]);
+        unsigned int digit2 = ConvertHexDigit(m_buffer[m_pos + 1]);
+        unsigned char byte = digit1 * 16 + digit2;
+        sequence += byte;
+
+        m_pos += 2;
+        SkipWhitespace();
+    }
+
+    return sequence;
+}
+
+void CharmapReader::ExpectEmptyRestOfLine()
+{
+    SkipWhitespace();
+
+    if (m_buffer[m_pos] == 0)
+    {
+        if (m_pos < m_size)
+            RaiseError("unexpected null character");
+    }
+    else if (m_buffer[m_pos] == '\n')
+    {
+        m_pos++;
+        m_lineNum++;
+    }
+    else if (m_buffer[m_pos] == '\r')
+    {
+        RaiseError("only Unix-style LF newlines are supported");
+    }
+    else
+    {
+        RaiseError("junk at end of line");
+    }
+}
+
+void CharmapReader::RaiseError(const char* format, ...)
+{
+    const int bufferSize = 1024;
+    char buffer[bufferSize];
+
+    std::va_list args;
+    va_start(args, format);
+    std::vsnprintf(buffer, bufferSize, format, args);
+    va_end(args);
+
+    std::fprintf(stderr, "%s:%ld: error: %s\n", m_filename.c_str(), m_lineNum, buffer);
+}
+
+void CharmapReader::RemoveComments()
+{
+    long pos = 0;
+    bool inString = false;
+
+    for (;;)
+    {
+        if (m_buffer[pos] == 0)
+            return;
+
+        if (inString)
+        {
+            if (m_buffer[pos] == '\\' && m_buffer[pos + 1] == '\'')
+            {
+                pos += 2;
+            }
+            else
+            {
+                if (m_buffer[pos] == '\'')
+                    inString = false;
+                pos++;
+            }
+        }
+        else if (m_buffer[pos] == '@')
+        {
+            while (m_buffer[pos] != '\n' && m_buffer[pos] != 0)
+                m_buffer[pos++] = ' ';
+        }
+        else
+        {
+            if (m_buffer[pos] == '\'')
+                inString = true;
+            pos++;
+        }
+    }
+}
+
+std::string CharmapReader::ReadConstant()
+{
+    long startPos = m_pos;
+
+    while (IsIdentifierChar(m_buffer[m_pos]))
+        m_pos++;
+
+    return std::string(&m_buffer[startPos], m_pos - startPos);
+}
+
+void CharmapReader::SkipWhitespace()
+{
+    while (m_buffer[m_pos] == '\t' || m_buffer[m_pos] == ' ')
+        m_pos++;
+}
+
+Charmap::Charmap(std::string filename)
+{
+    CharmapReader reader(filename);
+
+    for (;;)
+    {
+        Lhs lhs = reader.ReadLhs();
+
+        if (lhs.type == LhsType::None)
+            return;
+
+        reader.ExpectEqualsSign();
+
+        std::string sequence = reader.ReadSequence();
+
+        switch (lhs.type)
+        {
+        case LhsType::Char:
+            m_chars[lhs.code] = sequence;
+            break;
+        case LhsType::Escape:
+            m_escapes[lhs.code] = sequence;
+            break;
+        case LhsType::Constant:
+            m_constants[lhs.name] = sequence;
+            break;
+        }
+
+        reader.ExpectEmptyRestOfLine();
+    }
+}
diff --git a/tools/preproc/charmap.h b/tools/preproc/charmap.h
new file mode 100644
index 000000000..0d752ac99
--- /dev/null
+++ b/tools/preproc/charmap.h
@@ -0,0 +1,64 @@
+// Copyright(c) 2016 YamaArashi
+//
+// Permission is hereby granted, free of charge, to any person obtaining a copy
+// of this software and associated documentation files (the "Software"), to deal
+// in the Software without restriction, including without limitation the rights
+// to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+// copies of the Software, and to permit persons to whom the Software is
+// furnished to do so, subject to the following conditions:
+//
+// The above copyright notice and this permission notice shall be included in
+// all copies or substantial portions of the Software.
+//
+// THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+// IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+// FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+// AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+// LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+// OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+// THE SOFTWARE.
+
+#ifndef CHARMAP_H
+#define CHARMAP_H
+
+#include <cstdint>
+#include <string>
+#include <map>
+#include <vector>
+
+class Charmap
+{
+public:
+    Charmap(std::string filename);
+
+    std::string Char(std::int32_t code)
+    {
+        auto it = m_chars.find(code);
+
+        if (it == m_chars.end())
+            return std::string();
+
+        return it->second;
+    }
+
+    std::string Escape(unsigned char code)
+    {
+        return m_escapes[code];
+    }
+
+    std::string Constant(std::string identifier)
+    {
+        auto it = m_constants.find(identifier);
+
+        if (it == m_constants.end())
+            return std::string();
+
+        return it->second;
+    }
+private:
+    std::map<std::int32_t, std::string> m_chars;
+    std::string m_escapes[128];
+    std::map<std::string, std::string> m_constants;
+};
+
+#endif // CHARMAP_H
diff --git a/tools/preproc/preproc.cpp b/tools/preproc/preproc.cpp
new file mode 100644
index 000000000..7fc3f5c10
--- /dev/null
+++ b/tools/preproc/preproc.cpp
@@ -0,0 +1,83 @@
+// Copyright(c) 2016 YamaArashi
+//
+// Permission is hereby granted, free of charge, to any person obtaining a copy
+// of this software and associated documentation files (the "Software"), to deal
+// in the Software without restriction, including without limitation the rights
+// to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+// copies of the Software, and to permit persons to whom the Software is
+// furnished to do so, subject to the following conditions:
+//
+// The above copyright notice and this permission notice shall be included in
+// all copies or substantial portions of the Software.
+//
+// THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+// IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+// FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+// AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+// LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+// OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+// THE SOFTWARE.
+
+#include <stack>
+#include "preproc.h"
+#include "asm_file.h"
+#include "charmap.h"
+
+Charmap* g_charmap;
+
+int main(int argc, char **argv)
+{
+    if (argc != 3)
+    {
+        fprintf(stderr, "Usage: %s ASM_FILE CHARMAP_FILE", argv[0]);
+        return 1;
+    }
+
+    g_charmap = new Charmap(argv[2]);
+
+    std::stack<AsmFile> stack;
+
+    stack.push(AsmFile(argv[1]));
+
+    for (;;)
+    {
+        while (stack.top().IsAtEnd())
+        {
+            stack.pop();
+            
+            if (stack.empty())
+                return 0;
+            else
+                stack.top().OutputLocation();
+        }
+
+        Directive directive = stack.top().GetDirective();
+
+        switch (directive)
+        {
+        case Directive::Include:
+            stack.push(AsmFile(stack.top().ReadPath()));
+            stack.top().OutputLocation();
+            break;
+        case Directive::String:
+        {
+            unsigned char s[kMaxStringLength];
+            int length = stack.top().ReadString(s);
+
+            printf("\t.byte ");
+            for (int i = 0; i < length; i++)
+            {
+                printf("0x%02X", s[i]);
+
+                if (i < length - 1)
+                    printf(", ");
+            }
+            putchar('\n');
+            break;
+        }
+        case Directive::Unknown:
+            stack.top().OutputLine();
+            break;
+        }
+    }
+}
diff --git a/tools/preproc/preproc.h b/tools/preproc/preproc.h
new file mode 100644
index 000000000..c9e1a8414
--- /dev/null
+++ b/tools/preproc/preproc.h
@@ -0,0 +1,52 @@
+// Copyright(c) 2016 YamaArashi
+//
+// Permission is hereby granted, free of charge, to any person obtaining a copy
+// of this software and associated documentation files (the "Software"), to deal
+// in the Software without restriction, including without limitation the rights
+// to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+// copies of the Software, and to permit persons to whom the Software is
+// furnished to do so, subject to the following conditions:
+//
+// The above copyright notice and this permission notice shall be included in
+// all copies or substantial portions of the Software.
+//
+// THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+// IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+// FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+// AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+// LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+// OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+// THE SOFTWARE.
+
+#ifndef PREPROC_H
+#define PREPROC_H
+
+#include "charmap.h"
+
+#ifdef _MSC_VER
+
+#define FATAL_ERROR(format, ...)          \
+do                                        \
+{                                         \
+    fprintf(stderr, format, __VA_ARGS__); \
+    exit(1);                              \
+} while (0)
+
+#else
+
+#define FATAL_ERROR(format, ...)            \
+do                                          \
+{                                           \
+    fprintf(stderr, format, ##__VA_ARGS__); \
+    exit(1);                                \
+} while (0)
+
+#endif // _MSC_VER
+
+const int kMaxPath = 256;
+const int kMaxStringLength = 256;
+const unsigned long kMaxCharmapSequenceLength = 16;
+
+extern Charmap* g_charmap;
+
+#endif // PREPROC_H
diff --git a/tools/preproc/utf8.cpp b/tools/preproc/utf8.cpp
new file mode 100644
index 000000000..0aed83f4a
--- /dev/null
+++ b/tools/preproc/utf8.cpp
@@ -0,0 +1,94 @@
+// Copyright (c) 2008-2009 Bjoern Hoehrmann <bjoern@hoehrmann.de>
+// See http://bjoern.hoehrmann.de/utf-8/decoder/dfa/ for details.
+//
+// Copyright(c) 2016 YamaArashi
+//
+// Permission is hereby granted, free of charge, to any person obtaining a copy
+// of this software and associated documentation files (the "Software"), to deal
+// in the Software without restriction, including without limitation the rights
+// to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+// copies of the Software, and to permit persons to whom the Software is
+// furnished to do so, subject to the following conditions:
+//
+// The above copyright notice and this permission notice shall be included in
+// all copies or substantial portions of the Software.
+//
+// THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+// IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+// FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+// AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+// LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+// OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+// THE SOFTWARE.
+
+#include <cstdint>
+#include "utf8.h"
+
+static const unsigned char s_byteTypeTable[] =
+{
+    0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, // 00..1f
+    0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, // 20..3f
+    0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, // 40..5f
+    0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, // 60..7f
+    1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, // 80..9f
+    7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7, // a0..bf
+    8,8,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2, // c0..df
+    0xa,0x3,0x3,0x3,0x3,0x3,0x3,0x3,0x3,0x3,0x3,0x3,0x3,0x4,0x3,0x3, // e0..ef
+    0xb,0x6,0x6,0x6,0x5,0x8,0x8,0x8,0x8,0x8,0x8,0x8,0x8,0x8,0x8,0x8, // f0..ff
+};
+
+const unsigned char s0 = 0 * 12;
+const unsigned char s1 = 1 * 12;
+const unsigned char s2 = 2 * 12;
+const unsigned char s3 = 3 * 12;
+const unsigned char s4 = 4 * 12;
+const unsigned char s5 = 5 * 12;
+const unsigned char s6 = 6 * 12;
+const unsigned char s7 = 7 * 12;
+const unsigned char s8 = 8 * 12;
+
+static const unsigned char s_transitionTable[] =
+{
+    s0,s1,s2,s3,s5,s8,s7,s1,s1,s1,s4,s6, // s0
+    s1,s1,s1,s1,s1,s1,s1,s1,s1,s1,s1,s1, // s1
+    s1,s0,s1,s1,s1,s1,s1,s0,s1,s0,s1,s1, // s2
+    s1,s2,s1,s1,s1,s1,s1,s2,s1,s2,s1,s1, // s3
+    s1,s1,s1,s1,s1,s1,s1,s2,s1,s1,s1,s1, // s4
+    s1,s2,s1,s1,s1,s1,s1,s1,s1,s2,s1,s1, // s5
+    s1,s1,s1,s1,s1,s1,s1,s3,s1,s3,s1,s1, // s6
+    s1,s3,s1,s1,s1,s1,s1,s3,s1,s3,s1,s1, // s7
+    s1,s3,s1,s1,s1,s1,s1,s1,s1,s1,s1,s1, // s8
+};
+
+// Decodes UTF-8 encoded Unicode code point at "s".
+// If the encoding is valid, it returns the code point and advances "s" past the byte sequence.
+// If the encoding is not valid, it returns -1 and doesn't advance "s".
+UnicodeChar DecodeUtf8(const char* s)
+{
+    UnicodeChar unicodeChar;
+    int state = s0;
+    auto start = s;
+
+    do
+    {
+        unsigned char byte = *s++;
+        int type = s_byteTypeTable[byte];
+
+        if (state == s0)
+            unicodeChar.code = (0xFF >> type) & byte;
+        else
+            unicodeChar.code = (unicodeChar.code << 6) | (byte & 0x3F);
+
+        state = s_transitionTable[state + type];
+
+        if (state == s1)
+        {
+            unicodeChar.code = -1;
+            return unicodeChar;
+        }
+    } while (state != s0);
+
+    unicodeChar.encodingLength = s - start;
+
+    return unicodeChar;
+}
diff --git a/tools/preproc/utf8.h b/tools/preproc/utf8.h
new file mode 100644
index 000000000..259de67c5
--- /dev/null
+++ b/tools/preproc/utf8.h
@@ -0,0 +1,34 @@
+// Copyright(c) 2016 YamaArashi
+//
+// Permission is hereby granted, free of charge, to any person obtaining a copy
+// of this software and associated documentation files (the "Software"), to deal
+// in the Software without restriction, including without limitation the rights
+// to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+// copies of the Software, and to permit persons to whom the Software is
+// furnished to do so, subject to the following conditions:
+//
+// The above copyright notice and this permission notice shall be included in
+// all copies or substantial portions of the Software.
+//
+// THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+// IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+// FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+// AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+// LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+// OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+// THE SOFTWARE.
+
+#ifndef UTF8_H
+#define UTF8_H
+
+#include <cstdint>
+
+struct UnicodeChar
+{
+    std::int32_t code;
+    int encodingLength;
+};
+
+UnicodeChar DecodeUtf8(const char* s);
+
+#endif // UTF8_H