Draft:TeX font encodings

(La)TeX font encodings are character-to-glyph mapping schemes used by the TeX typesetting system and related software such as LaTeX. Unlike modern Unicode encodings, TeX font encodings traditionally assign characters to positions within an 8-bit font, where each slot corresponds to a glyph rather than a universal character identity. These encodings were designed primarily for high-quality typesetting rather than text interchange.

History

Early TeX systems commonly used 7-bit fonts that contain at most 128 glyph slots. Because of this limitation, four distinct font encodings were developed for different purposes: one for text and three for various math symbols (in modern TeX language these are called OT1, OML, OMS, and OMX, but back then they did not have distinct names).

Because 7 bits is way too small to fit an entire multilingual alphabet, the alphabet was simplified into base letters and separate diacritics which had their own codepoints. Donald Knuth, the creator of TeX was initially satisfied but later, Knuth realized that 128 different characters for the text input were not enough to accommodate foreign languages (his solution for diacritics turned out to be inefficient and problematic as it caused problems with kerning, hyphenation, and text copying). To fix that, when version 3.0 of TeX was released, it gained the ability to work with 8-bit inputs, allowing 256 different characters in the text input.

Since support for 8-bit encodings was added, the natural question was: What should the ideal multilingual 8-bit encoding look like? To address this issue, a team at the TUG Annual General Meeting held in Cork, Ireland, defined a 256-glyph font encoding, which includes accented letters and non-ASCII characters essential for representing the majority of Western European languages (and a few Eastern European ones) without needing commands such as \accent that create characters from diacritic parts. This “Cork” encoding has been implemented in several fonts created with Metafont, as well as in various virtual-font mappings of different font series. Since the meeting, considerable effort has gone into designing encodings for numerous other text fonts compatible with TeX, with the Cork encoding (T1) shaping the creation of many such encodings utilized in various fonts globally.

Differences

Traditional TeX font encodings differ from general-purpose text encodings such as ASCII or Unicode because they were designed for typesetting specific fonts rather than representing plain text. Here are the most important differences:

• In normal encodings, characters 0x20 and 0xA0 are often whitespace characters, and 0x00-0x1F and 0x7F-0x9F are control characters. In TeX, the role of control characters and whitespace characters is taken over by TeX's formatting itself, meaning that TeX does not need these characters. This means that in TeX font encodings, all slots are printable, meaningful characters. 0x20 isn't space, 0x00 isn't control character NUL, and 0xA0 is not a non-breaking space. This discrepancy has caused problems when early PDF readers were developed. For example, the character 0x00 of TeX encodings (OT1 Greek Gamma) was known to break old PDF readers because the PDF reader, unfamiliar with TeX encodings, confused it with the ASCII code 0x00.
• In other text encodings, such as Unicode or Code page 437, the set of all characters that encoding supports is all that can be typed. In TeX encodings, some characters are instead simulated by macros from other characters, meaning it is possible to "type" characters that don't exist by combining other characters.

128⁺ glyph encodings (text)

OT1

Main article: OT1 encoding

OT1 (aka TeX text) is a 7-bit TeX encoding developed by Donald E. Knuth.

OT1

	0	1	2	3	4	5	6	7	8	9	A	B	C	D	E	F
0x	Γ	Δ	Θ	Λ	Ξ	Π	Σ	Υ	Φ	Ψ	Ω	ff	fi	fl	ffi	ffl
1x	ı	ȷ	`	´	ˇ	˘	ˉ	˚	¸	ß	æ	œ	ø	Æ	Œ	Ø
2x	̷	!	”	#	$	%	&	’	(	)	*	+	,	-	.	/
3x	0	1	2	3	4	5	6	7	8	9	:	;	¡	=	¿	?
4x	@	A	B	C	D	E	F	G	H	I	J	K	L	M	N	O
5x	P	Q	R	S	T	U	V	W	X	Y	Z	[	“	]	ˆ	˙
6x	‘	a	b	c	d	e	f	g	h	i	j	k	l	m	n	o
7x	p	q	r	s	t	u	v	w	x	y	z	–	—	˝	˜	¨

Notes

• The letters A, B, E, H, I, K, M, N, O, P, T, X, Z are used both as the Latin letters and their Greek homoglyphs. The remaining Greek letters are found in positions 0x00-0x0A.
• The diacritics (0x12-0x18, 0x20, 0x5E-0x5F, 0x7D-0x7F) are used as combining characters to compose different letters with diacritics. 0x20 is used to create the letter Ł , but can be reused for other letters with slanted strokes (which are not even in Unicode but may be needed for nonstandard use).

OT2

OT2 (aka UW cyrillic encoding) is a 7-bit TeX encoding developed for typesetting text in the Cyrillic script. It predates the later T2A, T2B, and T2C encodings.

OT2

	0	1	2	3	4	5	6	7	8	9	A	B	C	D	E	F
0x	Њ	Љ	Џ	Э	І	Є	Ђ	Ћ	њ	љ	џ	э	і	є	ђ	ћ
1x	Ю	Ж	Й	Ё	Ѵ	Ѳ	Ѕ	Я	ю	ж	й	ё	ѵ	ѳ	ѕ	я
2x	¨	!	”	Ѣ	˘	%	´	’	(	)	*	ѣ	,	-	.	/
3x	0	1	2	3	4	5	6	7	8	9	:	;	«	ı	»	?
4x	˘	А	Б	Ц	Д	Е	Ф	Г	Х	И	Ј	К	Л	М	Н	О
5x	П	Ч	Р	С	Т	У	В	Щ	Ш	Ы	З	[	“	]	Ь	Ъ
6x	‘	а	б	ц	д	е	ф	г	х	и	ј	к	л	м	н	о
7x	п	ч	р	с	т	у	в	щ	ш	ы	з	–	—	№	ь	ъ

OT3

OT3 (aka UW IPA encoding) is a 7-bit TeX font encoding intended for IPA. It was never really used with LaTeX 2ε following the release of the TIPA package which offers much better support for IPA and unlike OT3, supports other linguistic notations such as Germanic and Indo-European transcriptions, as well as a large portion of extIPA. No ot3enc.def file was ever produced. This encoding is supported by the wsuipa package.

OT3

	0	1	2	3	4	5	6	7	8	9	A	B	C	D	E	F
0x	ɐ	ɑ	α	ɒ	ʌ	ƀ	b	bⳆ	ɓ	β	ȼ	ɕ	ʗ	đ	d	dⳆ
1x	ɗ	ɖ	ʤ	ð	ᴅ	ə	ɚ	ɘ	ɛ	ɜ	ɝ	ɞ	ɡ	ɠ	ɢ	γ
2x	ɣ	ɤ	ƕ	ħ	ɦ	ɧ	ɥ	ɨ	ᵻ	ɩ	ɪ	ᵻ	ɟ	ɫ	ƚ	ɬ
3x	ɭ	ɮ	ꟛ	ƛ	ɱ	ɯ	ɰ	ɲ	ŋ	ɳ	ɴ	ʘ	ɵ	ɔ	ꞷ	ɷ
4x	ꝏ	ᵽ	þ	ɸ	ɾ	ɼ	ɽ	ɹ	ɻ	ɺ	ʀ	ʁ	ʂ	ʃ	ʆ	σ
5x	ʈ	ʧ	ʇ	θ	ʉ	ꞹ	ʊ	ᴜ	ᵾ	ʋ	ʍ	χ	ʎ	ʏ	ʑ	ʐ
6x	ʒ	ʓ	ʔ	ʕ	ʖ	ɂ	꟏	̪	ˈ	ˌ	̩	̚	˔	˕	̘	̙
7x	ˑ	ː	̤	ʽ	˄	˅	˂	˃	˚	˳	̜	̴	˷	ˬ	˛	̯

OT4

OT4 (also known as Polish TeX encoding) is an 8-bit TeX font encoding designed for the Polish language. The lower 128 bytes are identical to OT1. The OT4 encoding was created because OT1 was missing the ogonek diacritic, even though the Ł was included.

OT1

	0	1	2	3	4	5	6	7	8	9	A	B	C	D	E	F
0x	Γ	Δ	Θ	Λ	Ξ	Π	Σ	Υ	Φ	Ψ	Ω	ff	fi	fl	ffi	ffl
1x	ı	ȷ	`	´	ˇ	˘	ˉ	˚	¸	ß	æ	œ	ø	Æ	Œ	Ø
2x	̷	!	”	#	$	%	&	’	(	)	*	+	,	-	.	/
3x	0	1	2	3	4	5	6	7	8	9	:	;	¡	=	¿	?
4x	@	A	B	C	D	E	F	G	H	I	J	K	L	M	N	O
5x	P	Q	R	S	T	U	V	W	X	Y	Z	[	“	]	ˆ	˙
6x	‘	a	b	c	d	e	f	g	h	i	j	k	l	m	n	o
7x	p	q	r	s	t	u	v	w	x	y	z	–	—	˝	˜	¨
8x		Ą	Ć				Ę				Ł	Ń
9x		Ś								Ź		Ż
Ax		ą	ć				ę				ł	ń			«	»
Bx		ś								ź		ż
Cx
Dx				Ó
Ex
Fx				ó												„

Notes

• OT4 was created to simplify Polish typesetting in TeX before T1 was available.

OT5

The name OT5 is not currently allocated.

OT6

OT6 is a TeX font encoding for the Armenian alphabet. It was created by Sergueï Dachian, and was allocated to permit use of his Armenian fonts in a standard LaTeX environment.

OT6

	0	1	2	3	4	5	6	7	8	9	A	B	C	D	E	F
0x		՟	ՙ	Ձ	Ղ	Ճ	Է	Ը	Թ	Ժ	Շ	Չ	Ռ	Ծ	Փ	Օ
1x		֏	֍	ձ	ղ	ճ	է	ը	թ	ժ	շ	չ	ռ	ծ	փ	օ
2x	և	՜	”	#	$	%	&	՚	*	+	,	-	.	/
3x	0	1	2	3	4	5	6	7	8	9	։	;	«	=	»	՞
4x	@	Ա	Բ	Ց	Դ	Ե	Ֆ	Գ	Հ	Ի	Ջ	Կ	Լ	Մ	Ն	Ո
5x	Պ	Ք	Ր	Ս	Տ	ՈՒ	Վ	Ւ	Խ	Յ	Զ	[	“	]	{	}
6x	՝	ա	բ	ց	դ	ե	ֆ	գ	հ	ի	ջ	կ	լ	մ	ն	ո
7x	պ	ք	ր	ս	տ	ու	վ	ւ	խ	յ	զ	֊	՛	—	!	?

256 glyph encodings (text)

T1

Main article: Cork encoding

The T1 encoding (aka Cork encoding) is a character encoding used for encoding glyphs in fonts. It is named after the city of Cork in Ireland, where during a TeX Users Group (TUG) conference in 1990 a new encoding was introduced for LaTeX. It contains 256 characters supporting most west- and east-European languages with the Latin alphabet. In 8-bit TeX engines the font encoding has to match the encoding of hyphenation patterns where this encoding is most commonly used. . It is now very common.

Cork encoding

	0	1	2	3	4	5	6	7	8	9	A	B	C	D	E	F
0x	` 0060	´ 00B4	ˆ 02C6	˜ 02DC	¨ 00A8	˝ 02DD	˚ 02DA	ˇ 02C7	˘ 02D8	¯ 00AF	˙ 02D9	¸ 00B8	˛ 02DB	‚ 201A	‹ 2039	› 203A
1x	“ 201C	” 201D	„ 201E	« 00AB	» 00BB	– 2013	— 2014	ZWSP 200B	₀ 2080	ı 0131	ȷ 0237	ff FB00	fi FB01	fl FB02	ffi FB03	ffl FB04
2x	␣ 2423	!	"	#	$	%	&	’ 2019	(	)	*	+	,	-	.	/
3x	0	1	2	3	4	5	6	7	8	9	:	;	<	=	>	?
4x	@	A	B	C	D	E	F	G	H	I	J	K	L	M	N	O
5x	P	Q	R	S	T	U	V	W	X	Y	Z	[	\	]	^	_
6x	‘ 2018	a	b	c	d	e	f	g	h	i	j	k	l	m	n	o
7x	p	q	r	s	t	u	v	w	x	y	z	{	|	}	~	SHY
8x	Ă 0102	Ą 0104	Ć 0106	Č 010C	Ď 010E	Ě 011A	Ę 0118	Ğ 011E	Ĺ 0139	Ľ 013D	Ł 0141	Ń 0143	Ň 0147	Ŋ 014A	Ő 0150	Ŕ 0154
9x	Ř 0158	Ś 015A	Š 0160	Ş 015E	Ť 0164	Ţ 0162	Ű 0170	Ů 016E	Ÿ 0178	Ź 0179	Ž 017D	Ż 017B	IJ 0132	İ 0130	đ 0111	§ 00A7
Ax	ă 0103	ą 0105	ć 0107	č 010D	ď 010F	ě 011B	ę 0119	ğ 011F	ĺ 013A	ľ 013E	ł 0142	ń 0144	ň 0148	ŋ 014B	ő 0151	ŕ 0155
Bx	ř 0159	ś 015B	š 0161	ş 015F	ť 0165	ţ 0163	ű 0171	ů 016F	ÿ 00FF	ź 017A	ž 017E	ż 017C	ij 0133	¡ 00A1	¿ 00BF	£ 00A3
Cx	À	Á	Â	Ã	Ä	Å	Æ	Ç	È	É	Ê	Ë	Ì	Í	Î	Ï
Dx	Ð	Ñ	Ò	Ó	Ô	Õ	Ö	Œ 0152	Ø	Ù	Ú	Û	Ü	Ý	Þ	SS 1E9E
Ex	à	á	â	ã	ä	å	æ	ç	è	é	ê	ë	ì	í	î	ï
Fx	ð	ñ	ò	ó	ô	õ	ö	œ 0153	ø	ù	ú	û	ü	ý	þ	ß 00DF

Notes

• Hexadecimal values under the characters in the table are the Unicode character codes.
• The first 12 characters are often used as combining characters.

T2A

The T2A encoding (aka Cork encoding) is a character encoding used for encoding glyphs in fonts. It is named after the city of Cork in Ireland, where during a TeX Users Group (TUG) conference in 1990 a new encoding was introduced for LaTeX. It contains 256 characters supporting most west- and east-European languages with the Latin alphabet. In 8-bit TeX engines the font encoding has to match the encoding of hyphenation patterns where this encoding is most commonly used.

T2A encoding

	0	1	2	3	4	5	6	7	8	9	A	B	C	D	E	F
0x	` 0060	´ 00B4	ˆ 02C6	˜ 02DC	¨ 00A8	˝ 02DD	˚ 02DA	ˇ 02C7	˘ 02D8	¯ 00AF	˙ 02D9	¸ 00B8	˛ 02DB	Ӏ 04C0	〈 2329	〉 232A
1x	“ 201C	” 201D	̑ 0311	̏ 030F	˘ 02D8	– 2013	— 2014	ZWNJ 200C	₀ 2080	ı 0131	ȷ 0237	ff FB00	fi FB01	fl FB02	ffi FB03	ffl FB04
2x	␣ 2423	!	"	#	$	%	&	’ 2019	(	)	*	+	,	-	.	/
3x	0	1	2	3	4	5	6	7	8	9	:	;	<	=	>	?
4x	@	A	B	C	D	E	F	G	H	I	J	K	L	M	N	O
5x	P	Q	R	S	T	U	V	W	X	Y	Z	[	\	]	^	_
6x	‘ 2018	a	b	c	d	e	f	g	h	i	j	k	l	m	n	o
7x	p	q	r	s	t	u	v	w	x	y	z	{	|	}	~	SHY
8x	Ґ	Ғ	Ђ	Ћ	Һ	Җ	Ҙ	Љ	Ї	Қ	Ҡ	Ҝ	Ӕ	Ң	Ҥ	Ѕ
9x	Ө	Ҫ	Ў	Ү	Ұ	Ҳ	Џ	Ҹ	Ҷ	Є	Ә	Њ	Ё	№	¤	§
Ax	ґ	ғ	ђ	ћ	һ	җ	ҙ	љ	ї	қ	ҡ	ҝ	ӕ	ң	ҥ	ѕ
Bx	ө	ҫ	ў	ү	ұ	ҳ	џ	ҹ	ҷ	є	ә	њ	ё	„	«	»
Cx	А	Б	В	Г	Д	Е	Ж	З	И	Й	К	Л	М	Н	О	П
Dx	Р	С	Т	У	Ф	Х	Ц	Ч	Ш	Щ	Ъ	Ы	Ь	Э	Ю	Я
Ex	а	б	в	г	д	е	ж	з	и	й	к	л	м	н	о	п
Fx	р	с	т	у	ф	х	ц	ч	ш	щ	ъ	ы	ь	э	ю	я

References

This page will be placed in the following categories if it is moved to the article namespace.

Categories:

• TeX
• Character encoding
• Character sets
Digital typography
• Alphabets
• Free TeX software