OBD:Text encoding

Originally created in English, Oni has been translated into the following seven languages: French, Italian, Spanish, German, Russian, Japanese and Chinese.

(An overview of the known language versions can be found HERE, whereas localized content is detailed HERE.)

Depending on the language version, vanilla Oni uses one of the following five encodings to render text:

• The original US version uses a trimmed-down Mac OS Roman code page that is effectively limited to US-ASCII (96 code points).
• European localizations (UK English, French, Italian, Spanish, German) use a custom version of Mac OS Roman (192 code points).
• The Russian localization uses a full implementation of the Windows-1251 (Cyrillic) code page (224 code points).
• The Chinese localization uses the EUC-CN implementation of GB 2312 (8,836 code points).
• The Japanese localization uses 1,357 code points mostly conforming to the Shift JIS implementation of JIS X 0208.

Properties of the fonts that are eventually used to render the text (via the encoding) are briefly described throughout the page.

(A more thorough overview of the glyphs can be found HERE.)

Encodings

US English

Below is the code page implemented by TSFFTahoma in the US English version of Oni. It is based on Mac OS Roman ("MacRoman" for short), but with two differences:

• Of the 223 printable glyphs provided by MacRoman, 42 are not implemented in TSFFTahoma (shown as grey-on-black).
• Control point 0x7F (a typically non-printable "delete" character) has a visible box-like glyph (◻) in this implementation.

	...0	...1	...2	...3	...4	...5	...6	...7	...8	...9	...A	...B	...C	...D	...E	...F
0x2...	SP	!	"	#	$	%	&	'	(	)	*	+	,	-	.	/
0x3...	0	1	2	3	4	5	6	7	8	9	:	;	<	=	>	?
0x4...	@	A	B	C	D	E	F	G	H	I	J	K	L	M	N	O
0x5...	P	Q	R	S	T	U	V	W	X	Y	Z	[	\	]	^	_
0x6...	`	a	b	c	d	e	f	g	h	i	j	k	l	m	n	o
0x7...	p	q	r	s	t	u	v	w	x	y	z	{	|	}	~	◻
0x8...	Ä	Å	Ç	É	Ñ	Ö	Ü	á	à	â	ä	ã	å	ç	é	è
0x9...	ê	ë	í	ì	î	ï	ñ	ó	ò	ô	ö	õ	ú	ù	û	ü
0xA...	†	°	￠	£	§	•	¶	ß	®	©	™	´	¨	≠	Æ	Ø
0xB...	∞	±	≤	≥	Ұ	µ	∂	∑	∏	π	∫	ª	º	Ω	æ	ø
0xC...	¿	¡	¬	√	ƒ	≈	∆	«	»	…	NB SP	À	Ã	Õ	Œ	œ
0xD...	–	—	‟	”	‛	’	÷	◊	ÿ	Ÿ	⁄	¤	‹	›	fi	fl
0xE...	‡	·	‚	„	‰	Â	Ê	Á	Ë	È	Í	Î	Ï	Ì	Ó	Ô
0xF...		Ò	Ú	Û	Ù	ı	ˆ	˜	¯	̆	̇	̊	̧	̋	̨	̌

Minor notes

• The MacRoman layout was apparently "borrowed" before 1998, when Mac OS 8.5 came out and the international currency sign a.k.a. scarab (¤), at 0xDB, was replaced with the euro symbol (€).
• The actual font (see HERE) has some unusual typographical features, such as a single-stroke Yen/Yuan symbol (Ұ) and a vertical-stroke cent symbol (￠).

Major notes

• Some of the removed glyphs (most importantly ß, ù and û, but also Ê, Ú and ú) occur in common European languages. This made the US TSFFTahoma unsuitable for EFIGS localizations, requiring the creation of a new version (see below).
• The US engine actually cannot interpret any code points beyond the US-ASCII range (first 6 rows, white background), notably failing on "…" (see "Ellipsis issue" below). This is because of a provision for Asian encoding systems (EUC-CN and Shift JIS), which use two-byte sequences starting with a high-bit byte.

---

European

The code page used by the five Western European versions (UK English, French, German, Spanish and Italian) is slightly different from the trimmed-down Mac OS Roman.

• It tends to the needs of European localizations by adding back the following characters:
  German ß; French Ê and û; French/Italian ù; Spanish/Italian Ú and ú (relatively rare).

N.B. The characters Æ and ÿ are not reinstated, despite their (very rare) occurrence in French script.

• Awkwardly enough, the six characters are not restored in their original positions (grey-on-black), but take the place of math symbols.
  Four more "math" positions are inexplicably filled with three duplicate characters (œ, ¡ and ª) and a truly enigmatic ʖ̇ , which doesn't seem to occur in any known language and has no dedicated code point in Unicode.

N.B. The broken italic font variants (see HERE) do not fully implement the 10 new glyphs and use a regular question mark instead of the ʖ̇.

	...0	...1	...2	...3	...4	...5	...6	...7	...8	...9	...A	...B	...C	...D	...E	...F
0x2...	SP	!	"	#	$	%	&	'	(	)	*	+	,	-	.	/
0x3...	0	1	2	3	4	5	6	7	8	9	:	;	<	=	>	?
0x4...	@	A	B	C	D	E	F	G	H	I	J	K	L	M	N	O
0x5...	P	Q	R	S	T	U	V	W	X	Y	Z	[	\	]	^	_
0x6...	`	a	b	c	d	e	f	g	h	i	j	k	l	m	n	o
0x7...	p	q	r	s	t	u	v	w	x	y	z	{	|	}	~	◻
0x8...	Ä		Ç	É	Ñ	Ö	Ü	á	à	â	ä	ã	å	ç	é	è
0x9...	ê	ë	í	ì	î	ï	ñ	ó	ò	ô	ö		ú	ù	û	ü
0xA...	†		￠	£	§	•		ß	®	©	™	´	¨			Ø
0xB...		±			Ұ	µ	Ê	Ú	ù	ú	û	ª	ß	œ	æ	ø
0xC...	¿	¡	¬	¡	ƒ	ʖ̇	ª	«	»	…		À		Õ	Œ	œ
0xD...	–	—	‟	”	‛	’	÷			Ÿ		¤	‹	›
0xE...	‡		‚	„	‰	Â	Ê	Á	Ë	È	Í	Î	Ï	Ì	Ó	Ô
0xF...		Ò	Ú	Û	Ù		ˆ	˜	¯

Coincidentally, with the 10 new glyphs, the European code page has exactly 96 glyphs in the US-ASCII half and 96 in the extension half (blue).

N.B. Unlike the US version, all five Western European versions (including UK English) are able to render the full extended ASCII set.

---

Cyrillic

In the Russian version of Oni, TSFFTahoma implements the Windows-1251 (Cyrillic) code page, with some deviations.

• The character 0x98, normally non-printable, is implemented as a visible box glyph (☐), slightly larger than 0x7F.
• The character 0x81, normally a "Ѓ" glyph, is replaced with a thin space of inconsistent size (2px wide for all fonts, 3px for 13pt regular and 16pt regular).
• The character 0xA0, normally a non-breaking space, is a space of not-so-consistent size (anywhere from single to triple width, depending on the font).
• The character 0xAD, normally a soft hyphen, is a visible hyphen (similar to the hyphen-minus, 0x2D) for 7pt fonts, and an inconsistently sized space for other fonts.
  (Oni's engine could in theory reserve a special treatment for soft hyphens and non-breaking spaces, specified in TSFLRoman, but in practice there is no such functionality.)

	...0	...1	...2	...3	...4	...5	...6	...7	...8	...9	...A	...B	...C	...D	...E	...F
0x2...	SP	!	"	#	$	%	&	'	(	)	*	+	,	-	.	/
0x3...	0	1	2	3	4	5	6	7	8	9	:	;	<	=	>	?
0x4...	@	A	B	C	D	E	F	G	H	I	J	K	L	M	N	O
0x5...	P	Q	R	S	T	U	V	W	X	Y	Z	[	\	]	^	_
0x6...	`	a	b	c	d	e	f	g	h	i	j	k	l	m	n	o
0x7...	p	q	r	s	t	u	v	w	x	y	z	{	|	}	~	◻
0x8...	Ђ	S   P	‚	ѓ	„	…	†	‡	€	‰	Љ	‹	Њ	Ќ	Ћ	Џ
0x9...	ђ	‘	’	“	”	•	–	—	☐	™	љ	›	њ	ќ	ћ	џ
0xA...	NB SP	Ў	ў	Ј	¤	Ґ	¦	§	Ё	©	Є	«	¬	–	®	Ї
0xB...	°	±	І	і	ґ	µ	¶	·	ё	№	є	»	ј	Ѕ	ѕ	ї
0xC...	А	Б	В	Г	Д	Е	Ж	З	И	Й	К	Л	М	Н	О	П
0xD...	Р	С	Т	У	Ф	Х	Ц	Ч	Ш	Щ	Ъ	Ы	Ь	Э	Ю	Я
0xE...	а	б	в	г	д	е	ж	з	и	й	к	л	м	н	о	п
0xF...	р	с	т	у	ф	х	ц	ч	ш	щ	ъ	ы	ь	э	ю	я

Italic fonts

The Russian version only provides an implementation of Windows-1251 for regular and bold fonts. The five italic fonts (7pt, 9pt, 10pt, 12pt and 14pt) have exactly the same data (pixels and glyph descriptors) as for the European iteration of Mac OS Roman. This makes sense because italic fonts are inherently broken (see HERE) and thus not used by any text in vanilla Oni.

Bold 14 font

Somewhat surprisingly, the size-14 TSFT in the Russian version of TSFFTahoma does not have a complete Windows-1251 code page either. Instead it is limited to the US-ASCII character set (including the "printable delete" box at code point 0x7F), i.e., the upper section of the above table (white background). This causes no issue in vanilla Oni, but only because there is no text that uses bold 14.

Incomplete transparency

A unique "feature" of the Russian/Cyrillic TSFFTahoma is that all the characters in the extended ASCII range (0x80-0xFF) have a slightly opaque background (about 3% opacity) in the regular (non-bold) font variant. This isn't visible ingame, but only because the engine (re)posterizes all the glyphs into 4-bit grayscale when rendering (so that only opacities above 6% are visible).

Glyph alignment and spacing

Last but not least, some fonts in the Russian TSFFTahoma have inconsistent vertical alignment, the most blatant example being 12 bold: some glyphs are one pixel shorter or taller than the full line height (ascender+descender), without a properly compensated vertical glyph offset; others simply have pixels that are not properly aligned within a glyph's rectangle. Besides, many glyphs have excessive padding to the left and/or right of a character, which affects readability.
N.B. There are other examples of poor alignment, e.g. for 12 bold, the character 0x9C (њ) has its right side cut off and is thus unusable (luckily it doesn't occur in Russian script).

---

Chinese

The Chinese version of Oni is unique in how the main game code resides in Oni.dat, a renamed copy of the original Oni.exe from the US version that is executed indirectly by a wrapper app called oni.exe, alongside a custom text engine, xfhsm_oni.dll. The latter DLL intercepts any text about to be displayed by "Oni.dat", first reducing it to a set of two-byte control sequences, and then (if all goes well) to a set of custom glyphs, with pixel data coming from an external font file, xf_font.dat.

Unlike for the original US engine or the Japanese one, xfhsm_oni.dll does not expect any single-byte characters in the input, does not interpret US-ASCII strings in any meaningful way and never resorts to level0_Final's TSFFTahoma for text display. The pixel data comes exclusively from xf_font.dat and the expected control sequences are exclusively two-byte code points (this includes string termination; instead of a single null char xfhsm_oni.dll expects a string to end with a pair of null chars).

Unlike for other versions of Oni, the Chinese font does not have a table listing the valid code points along with their "glyph descriptors" (i.e., instructions on how to extract a glyph from the raw pixel data). Instead all the glyphs are stored as fixed-size bitmaps (16x16 pixels each) and there are exactly 94x94=8,836 glyphs, filling up a standard GB 2312 plane (qūwèi), indexed through a compact numbering scheme known as EUC-CN: each of the 94x94 code points is indexed by a pair of bytes that are both in the 0xA1-0xFE range. Code points that are not assigned under GB 2312 (e.g., rows 10-15 and 90-94) simply have blank pixel data in the corresponding regions of xf_font.dat.

Two glyph sizes are available: 16x16 glyphs are stored in the first half of xf_font.dat, and 12x12 glyphs in the second half. Each 12x12 glyph is stored in the top left corner of a 16x16 bitmap, so the row/glyph alignment is the same in both cases: 2 bytes per pixel row and 32 bytes per glyph. The pixel packing is 1-bit black-and-white (i.e., without antialiasing), much more space-efficient than the 8-bit grayscale storage used in Oni's TSFT. Another gain comes from not having any glyph descriptors (TSGAs), and from having only two fonts instead of Oni's typical 15.

All the GB 2312 glyphs listed HERE and HERE are implemented, except for the euro sign and the ten glyphs from Vertical Forms.

Unlike for other versions of Oni, an invalid code point does not interrupt the interpretation/rendering of a text string by xfhsm_oni.dll and can lead to a wide range of unexpected behavior: at best, a blank or otherwise unintended glyph will be displayed; at worst the rendered text will be garbled (memory corruption most likely), or the game may simply crash.

---

Japanese

Japanese Oni uses a custom two-byte encoding that is mostly consistent with Shift JIS but with some of the control sequences rearranged in seemingly non-standard ways. Like Chinese Oni, the glyph data is stored in new, external files; in this case they are .fnt files stored in GameDataFolder. Three font sizes are available, with pixel sizes 11x11 (JPN_SMALL.fnt), 12x12 (JPN_MIDDLE.fnt) and 14x14 (JPN_BIG.fnt). The 14x14 font has a bold-face variant (JPN_BOLD.fnt). All four fonts are fixed-width, i.e. all glyphs have a square bounding box.

Unlike the Chinese version, the TSFFTahoma contained in the Japanese game data is not limited to the ASCII code page. There are a total of 154 double-byte code points (Romaji, punctuation, kana and kanji) across 19 code pages (TSGA) each corresponding to a different "lead byte" (0x81, 0x82, 0x83, 0x88, 0x89, 0x8A, 0x8B, 0x8C, 0x8D, 0x8E, 0x8F, 0x90, 0x91, 0x92, 0x93, 0x95, 0x96, 0x97 and 0x98).

          

 0x81 (8 glyphs, 1 anomaly) - punctuation

N.B. 0x8130 is not a legal Shift JIS sequence. The standard code for a prolonged sound mark is 0x815B.

Glyph	Shift JIS	Unicode	Designation
ー	0x8130	U+30FC	prolonged sound mark
2SP	0x8140	U+3000	ideographic space
、	0x8141	U+3001	ideographic comma
。	0x8142	U+3002	ideographic full stop
・	0x8145	U+30FB	katakana middle point
：	0x8146	U+003A	colon
！	0x8149	U+0021	exclamation mark
％	0x8193	U+0025	percent sign

          

 0x82 (42 glyphs) - numbers, letters, hiragana

N.B. There is no clear reason why numerals are limited to 2 and 6, and Roman letters are limited to A, C, D, F, S, T, W - or why these glyphs are needed at all, seeing as US-ASCII is still available. But for what it's worth, the 9 redundant glyphs come from a serifed font, whereas the US-ASCII font is sans serif.

N.B. It is also not clear why (in this font/encoding) the TU hiragana has a "lowercase" version while many other hiragana are missing.

Glyph	Shift JIS	Unicode	Designation
2	0x8251	U+0032	digit 2
6	0x8255	U+0036	digit 6
A	0x8260	U+0041	letter A
C	0x8262	U+0043	letter C
D	0x8263	U+0044	letter D
F	0x8265	U+0046	letter F
S	0x8272	U+0053	letter S
T	0x8273	U+0054	letter T
W	0x8276	U+0057	letter W
あ	0x82A0	U+3042	hiragana A
い	0x82A2	U+3044	hiragana I
う	0x82A4	U+3046	hiragana U
え	0x82A6	U+3048	hiragana E
か	0x82A9	U+304B	hiragana KA
が	0x82AA	U+304C	hiragana GA
き	0x82AB	U+304D	hiragana KI
く	0x82AD	U+304F	hiragana KU
こ	0x82B1	U+3053	hiragana KO
さ	0x82B3	U+3055	hiragana SA
し	0x82B5	U+3057	hiragana SI
じ	0x82B6	U+3058	hiragana ZI
す	0x82B7	U+3059	hiragana SU
た	0x82BD	U+305F	hiragana TA
だ	0x82BE	U+3060	hiragana DA
ち	0x82BF	U+3061	hiragana TI
っ	0x82C1	U+3063	hiragana tu
つ	0x82C2	U+3064	hiragana TU
て	0x82C4	U+3066	hiragana TE
と	0x82C6	U+3068	hiragana TO
な	0x82C8	U+306A	hiragana NA
に	0x82C9	U+306B	hiragana NI
の	0x82CC	U+306E	hiragana NO
は	0x82CD	U+306F	hiragana HA
ま	0x82DC	U+307E	hiragana MA
め	0x82DF	U+3081	hiragana ME
ゃ	0x82E1	U+307E	hiragana MA
よ	0x82E6	U+3081	hiragana ME
る	0x82E9	U+308B	hiragana RU
れ	0x82EA	U+308C	hiragana RE
わ	0x82ED	U+308F	hiragana WA
を	0x82F0	U+3092	hiragana WO
ん	0x82F1	U+3093	hiragana N

          

 0x83 (41 glyphs, 2 anomalies) - katakana

N.B. 0x8332 is not a legal Shift JIS sequence. The standard code for the BO katakana ボ is 0x837B.

N.B. 0x8333 is not a legal Shift JIS sequence. The standard code for the MA katakana マ is 0x837D.

N.B. It is not clear why (in this font/encoding) the I and O katakana have "lowercase" versions while many other katakana are missing. Also, the TU, YA, YU and YO katakana have only a lowercase version.

Glyph	Shift JIS	Unicode	Designation
ボ	0x8332	U+30DC	katakana BO
マ	0x8333	U+30DE	katakana MA
ィ	0x8342	U+30A3	katakana ı
イ	0x8343	U+30A4	katakana I
ウ	0x8345	U+30A6	katakana U
ォ	0x8348	U+30A9	katakana o
オ	0x8349	U+30AA	katakana O
キ	0x834C	U+30AD	katakana KI
ク	0x834E	U+30AF	katakana KU
グ	0x834F	U+30B0	katakana GU
ケ	0x8350	U+30B1	katakana KE
ゲ	0x8351	U+30B2	katakana GE
コ	0x8352	U+30B3	katakana KO
サ	0x8354	U+30B5	katakana SA
シ	0x8356	U+30B7	katakana SI
ジ	0x8357	U+30B8	katakana ZI
ス	0x8358	U+30B9	katakana SU
セ	0x835A	U+30BB	katakana SE
タ	0x835E	U+30BF	katakana TA
ッ	0x8362	U+30C3	katakana tu
テ	0x8365	U+30C6	katakana TE
ト	0x8367	U+30C8	katakana TO
ド	0x8368	U+30C9	katakana DO
ナ	0x8369	U+30CA	katakana NA
ニ	0x836A	U+30CB	katakana NI
ノ	0x836D	U+30CE	katakana NO
フ	0x8374	U+30D5	katakana HU
ブ	0x8375	U+30D6	katakana BU
プ	0x8376	U+30D7	katakana PU
ベ	0x8378	U+30D9	katakana BE
ポ	0x837C	U+30DD	katakana PO
ム	0x8380	U+30E0	katakana MU
ャ	0x8383	U+30E3	katakana ya
ュ	0x8385	U+30E5	katakana yu
ョ	0x8387	U+30E7	katakana yo
ラ	0x8389	U+30E9	katakana RA
リ	0x838A	U+30EA	katakana RI
ル	0x838B	U+30EB	katakana RU
レ	0x838C	U+30EC	katakana RE
ロ	0x838D	U+30ED	katakana RO
ン	0x8393	U+30F3	katakana N

          

 0x88 (4 glyphs), 0x89 (7 glyphs), 0x8A (3 glyphs), 0x8B (1 glyph) - kanji

Glyph	Shift JIS	Unicode	Designation
暗	0x88C3	U+6697	kanji AN
易	0x88D5	U+6613	kanji EKI
移	0x88DA	U+79FB	kanji I
印	0x88F3	U+5370	kanji IN
押	0x899F	U+62BC	kanji Ō
黄	0x89A9	U+2EE9 U+9EC4	kanji KI
下	0x89BA	U+4E0B	kanji SHITA
可	0x89C2	U+53EF	kanji KA
画	0x89E6	U+753B U+FAA3	kanji GA
解	0x89F0	U+89E3	kanji KAI
回	0x89F1	U+56DE	kanji KAI2
各	0x8A65	U+5404	kanji ONOONO
官	0x8AAF	U+5B98	kanji KAN
基	0x8AEE	U+57FA	kanji MOTO
許	0x8B96	U+8A31	kanji MOTO2

          

 0x8C (3 glyphs), 0x8D (4 glyphs), 0x8E (8 glyphs), 0x8F (3 glyphs) - kanji

Glyph	Shift JIS	Unicode	Designation
経	0x8C6F	U+7D4C	kanji KYŌ
庫	0x8CC9	U+5EAB	kanji KO
向	0x8CFC	U+5411	kanji MU
行	0x8D73	U+2F8F U+884C U+FA08	kanji GYŌ
高	0x8D82	U+2FBC U+9AD8	kanji TAKA
合	0x8D87	U+5408	kanji GŌ
作	0x8DEC	U+4F5C	kanji SAKU
使	0x8E67	U+4F7F	kanji SHI
司	0x8E69	U+53F8	kanji TSUKASA
始	0x8E6E	U+59CB	kanji SHI
私	0x8E84	U+79C1	kanji WATASHI
試	0x8E8E	U+8A66	kanji SHI2
字	0x8E9A	U+5B57	kanji JI
時	0x8E9E	U+6642	kanji TOKI
斜	0x8ECE	U+659C	kanji SHA
終	0x8F49	U+7D42	kanji TSUI
所	0x8F8A	U+6240	kanji SHO
場	0x8FEA	U+5834	kanji BA

          

 0x90 (3 glyphs), 0x91 (7 glyphs), 0x92 (2 glyphs), 0x93 (5 glyphs) - kanji

Glyph	Shift JIS	Unicode	Designation
色	0x9046	U+2F8A U+8272	kanji IRO
新	0x9056	U+65B0	kanji SHIN
神	0x905F	U+795E U+FA19	kanji KAMI
前	0x914F	U+524D	kanji MAE
倉	0x9171	U+5009	kanji KURA
窓	0x918B	U+7A93	kanji MADO
像	0x919C	U+50CF U+2F80B	kanji ZŌ
続	0x91B1	U+7D9A	kanji ZOKU
体	0x91CC	U+4F53	kanji TAI
替	0x91D6	U+66FF	kanji TEI
中	0x9286	U+4E2D	kanji CHU
低	0x92E1	U+4F4E	kanji HIKU
度	0x9378	U+5EA6	kanji TABI
動	0x93AE	U+52D5	katakana DO
同	0x93AF	U+540C	kanji DO2
難	0x93EF	U+96E3 U+FA68 U+FAC7	kanji NAN
入	0x93FC	U+2F0A U+5165	kanji JU

          

 0x95 (3 glyphs), 0x96 (5 glyphs, 1 anomaly), 0x97 (4 glyphs), 0x98 (1 glyph) - kanji

N.B. 0x9632 is not a legal Shift JIS sequence. The standard code for the MOTO kanji 本 is 0x967B.

Glyph	Shift JIS	Unicode	Designation
閉	0x95C2	U+9589	kanji HEI
変	0x95CF	U+5909	kanji HEN
歩	0x95E0	U+6B69	kanji HO
本	0x9632	U+672C	kanji MOTO
幕	0x968B	U+5E55	kanji MAKU
明	0x96BE	U+660E	kanji MEI
面	0x96CA	U+2FAF U+9762	kanji MEN
目	0x96DA	U+2F6C U+76EE	kanji MOKU
用	0x9770	U+2F64 U+7528	kanji YŌ
立	0x97A7	U+2F74 U+7ACB	kanji RITSU
了	0x97B9	U+4E86 U+F9BA	kanji RYŌ
令	0x97DF	U+4EE4 U+F9A8	kanji REI
路	0x9848	U+8DEF	kanji JI

As for the first code page of the Japanese TSFFTahoma, it implements only the 0x20-0x7F range of characters, i.e., is limited to US-ASCII. This is consistent with the simplified logic used by the Japanese engine, where any high-bit byte (in the 0x80-0xFF range) is treated as the start of a two-byte sequence. (In actual Shift JIS some high-bit bytes are interpreted as half-width kana, a feature that isn't supported by Oni's engine.)

It must be noted that, as compared to the separate .fnt files, the Japanese TSFFTahoma provides a very rudimentary implementation of JIS X 0208 (only coding for 154 double-byte glyphs, whereas the .fnt files implement 1,357) and is essentially useless/unusable.

• The Japanese engine requires all four .fnt files to be present (bails out if any of them are missing) and uses them for all of the vanilla text strings, which only contain double-byte control codes. Thus, under normal conditions, TSFFTahoma remains completely unused in the Japanese version, and would only be used for (artificially added) US-ASCII input.
• If the US engine is used on the Japanese game data, then the .fnt files are ignored (obviously), and the incomplete TSFFTahoma is used to render the Japanese text strings as well as the few English strings supplied by the EXE. Due to the limited character set, many strings end up broken.

It appears that the Japanese localization team initially tried to put Oni's code page system to use, and to fill in all the required JIS glyphs into TSFT and TSGA. As the number of kanji increased, supposedly, the TSFT grew prohibitively large due to the use of 8-bit grayscale storage for the pixel data, and the size taken up by the sparsely populated TSGA also increased out of proportion with the rest of the game data. At some point the engine switched to separate .fnt files, and somehow no one bothered to clean up the incomplete code pages in TSFFTahoma.

At the time of writing, the code points and pixel data in the Japanese .fnt files have not been thoroughly analyzed and compared with JIS X 0208. We know that 1,357 glyphs are implemented, across 27 "lead bytes" (roughly 50 kuten rows). This is much smaller than the full kuten plane, and makes sense in terms of space efficiency. We also know that some code points are non-standard (rearranged) as compared to regular Shift JIS, although we do not yet know if this rearrangement is consistent with any common variation of Shift JIS. As long as Japanese game data contains text strings that match the game's encoding, non-standard code points are not a problem (but should be kept in mind).

Text anomalies

Ellipsis issue

Unlike other Western versions (UK English, French, German, Italian, Spanish, Russian), the US engine treats high-bit characters as part of a two-byte control sequence (a provision for Asian encodings), and therefore fails to render any character from the extended ASCII range. This happens twice in English Oni, because the ellipsis character (…), encoded as 0xC9, was accidentally used in These Two text consoles in place of three consecutive dots (probably auto-substituted by a text editor). The result is that the two lines using a "…" are cut off at the offending character.

Invalid EUC-CN input

Unlike the Japanese version, where non-standard Shift JIS sequences are explicitly allowed in the .fnt files, the Chinese version does not have a code table and relies on a standard EUC-CN encoding, with exactly 8,836 code points (94x94). A proper EUC-CN control sequence consists of two bytes that are both in the range 0xA1-0xFE and anything else is technically illegal (single US-ASCII characters could occur in theory, but are not handled properly by the custom text engine, xfhsm_oni.dll).

The text strings in the Chinese version mostly conform to the EUC-CN scheme. A notable exception is the (A1,A0) sequence, which occurs in a few subtitles and is rendered with a blank glyph (i.e., a space between valid glyphs, undistinguishable from an ordinary ideographic space), apparently due to some kind of wraparound. At the time of writing it is not known what was meant by the (A1,A0) sequence, as it doesn't seem to be a valid control sequence under any common extension of EUC-CN.

Another illegal sequence is (0xA3,0x89), which occurs only in the SUBTmessages entry xdash1 (five identical glyphs at the end of the string) and is somehow rendered as a ㈢, which would normally be encoded with (A2,E7). Such an improbable substitution is likely unintentional, and it is not known what the intended glyph was.

At the time of writing the apparent wraparound behavior has not been investigated thoroughly, but it is established that some illegal code points are not recovered to a valid glyph at all, and instead result in garbled text or a crash. Possibly invalid EUC-CN input is what is causing most Chapters of the Chinese Oni version to crash on modern Windows systems (through varying degrees of memory corruption), although this has not been investigated thoroughly either.

Over-tall text

Although not strictly speaking a font issue, some of Oni's text fails to render because it doesn't fit vertically into a fixed-size frame (such as a text console). This is known to happen for These Two consoles in the English version, and possibly for other screens in other language versions.

Over-long text

Although Chinese text strings typically have a much smaller number of glyphs than English originals, this is not always the case. The Chinese glyphs are also much wider on average, with each glyph taking up 16x16 pixels, and so there are situations where the rendered Chinese line is much wider than the English original, no longer fitting on one line as intended by the context.

This is only known to cause a problem for the "resolution" item in the Options menu (a WMM_ generated at runtime). The actual dropdown list is wide enough to accommodate even the longest resolution strings, but the currently selected resolution appears in a small window that is only 150 pixels wide, too narrow even for the shortest resolution string "640×480×16位" (which needs 176 pixels). As a result the active resolution is always displayed on two lines, no longer fitting into the frame vertically and thus unreadable.

Chinese SUBT issues

The Chinese (Windows) version of Oni is unique in that no game content was actually localized except for text. Because of the relative simplicity of the task, the Chinese team did not build a new set of game data files, and merely modified the original .dat and .raw from the US version. WMDD, WMM_ and IGSt instances were patched inside each level's .dat, whereas the two SUBT files were patched in level0_Final.raw. In the case of an IGSt, text is stored in a fixed-size array (384 bytes), which has more than enough space for any translated text. WMDD and WMM_ also have fixed-size arrays (256 and 64 bytes, respectively) with at least some spare space. SUBT files, however, have a much more compact storage.

The text strings of a SUBT file (stored in level0_Final.raw and indexed from the .dat part of the SUBT) are typically packed right next to each other, separated only by a single null char. Chinese text typically uses fewer glyphs, but each glyph is taking up two bytes instead of one, including punctuation and the trailing null. Thus for short sentences or interjections it is possible for a Chinese translation to completely fill up the space used by the original string and even extend into the next entry.

None of the Chinese translations in SUBTmessages or SUBTsubtitles are actually longer than the original English text, and it is only the extra null byte that intrudes on the next entry's handle on several occasions. The affected handle essentially becomes a null string, and the corresponding subtitle is never found and displayed.

In SUBTmessages this happens only once (the message corresponding to "xf1" overwrites the first character of "xreload", so Konoko is never prompted to reload her gun in the last training room). In SUBTsubtitles there are as many as 29 anomalies, summed up in the following table.

List of corrupt subtitle handles in the Chinese version Culprit handle Original of culprit text (null char ° included) Victim handle Original of victim text (null char ° included) 01_01_11 Kerr:  Good luck Konoko.° 01_02_01 Konoko:  Chung...° 01_03_07 Griffin:  All right Konoko. I'm giving you a shot at this.° 01_03_08 Konoko:  Thank you Sir. I won't let you down.° 02_05_04 Muro:  Let me know when things start to get messy.° 02_05_05 Barabas:  Count on it. When I get through with them they're...° 02_06_02 Griffin:  Explain.° 02_06_03 Konoko:  This whole place - it's a trap. They planted that datapad.° 02_09_03 Griffin:  So she's still stable?° 02_09_04 Kerr:  So far as we can tell - yes, but prolonged stress could be dangerous.° 03_10_01 Barabas:  Let's get it on!° 03_10_02 Barabas:  You're strong, but this isn't over. Not by a long shot!° 03_11_01 Barabas:  She's with them.° 03_11_02 Muro:  Who is with them?° 04_17_03 Muro:  I can't allow that.° 04_17_04 Civilian:  You are with the TCTF, right? I wish there was something I could do to help.° 07_22_01 Konoko:  Showtime...° 07_23_01 Syndicate Henchman:  We've disabled the power substations. Static defenses are down. All units converge.° 07_26_15 Konoko:  Thanks.° 07_26_16 Konoko:  Things aren't looking good.° 07_26_17 Cop:  No, we haven't secured a single area -- not even our armory.° 07_26_18 Civilian:  I told them we'd just make things worse if we tried to use these guns. You should be the one using this!° 08_27_03 Konoko:  This is personal.° 08_27_04 Griffin:  Which is precisely why you should have nothing to do with it.° 09_31_02 Shinatama:  You are not who you think you are.° 09_31_03 Konoko:  What?° 09_31_03 Konoko:  What?° 09_31_04 Shinatama:  You are Mai Hasegawa. Your father...° 09_31_24 Konoko:  No!° 09_31_25 Shinatama:  Detonation in twenty, nineteen, eighteen...° 11_40_07 Mukade:  We shall see...° c11_41_01konoko Konoko:  My father's file. Griffin encouraged me not to look too deeply into my past. It seems like there's a lot he didn't want me to know... I could feel the Ninja and I knew he could feel me. Why? What am I becoming? Are we the same?° 12_46_02 Konoko:  What?° 12_46_03 Konoko:  Mai Hasegawa. That's your name, isn't it?° 12_46_06 Konoko:  Leave me alone...° 12_46_07 Konoko:  Can't do that. You are a Rogue Agent. I'm under orders to bring you in.° 13_65_05 Kerr:  This may sting a bit...° 13_65_06 Konoko:  What are you doing?° 13_65_20 Konoko:  Griffin? But why?° 13_65_21 Kerr:  Your father and I were criminals, funded by the Syndicate. We couldn't get backing from any legitimate source. They left us alone for the most part. We didn't think they were interested in our work. We were wrong.° 13_65_25 Kerr:  Muro.° 13_65_26 Konoko:  You've got to be kidding.° 13_65_36 Konoko:  What?° 13_65_37 Kerr:  You are changing...into a more powerful, resilient version of yourself. But whatever your final form, it is an expression of your true nature...° 13_66_03 Konoko:  The crane controls...° 13_66_04 Konoko:  I just hope the Chrysalis can keep me alive...° 14_52_02 Konoko:  Gotcha.° 14_52_03 Griffin:  This is a bad idea. Put the gun down.° 14_52_06 Konoko:  For you? Badly?° 14_52_07 Griffin:  That's your call. You can pull the trigger or you can walk away. It's up to you.° 00_01_09 Shinatama:  Super!° c00_01_10Shinatama Shinatama:  Great!° c00_01_10Shinatama Shinatama:  Great!° 00_01_11 Shinatama:  O.K.!° civmale3_trigger Civilian:  Hi there!° c00_01_100shinatama Shinatama: Kill me Konoko...please!° c00_01_101shinatama Shinatama: I'm sorry...so sorry! ° c00_01_102shinatama Shinatama: Forgive me...°

The systematic nature of this anomaly suggests that the Chinese team were careful not to exceed the string length of the original, and merely overlooked the extra null char (and of course didn't check the ingame rendition of the subtitles all that thoroughly).