summaryrefslogtreecommitdiff
path: root/Master/texmf-dist/doc/fonts/ly1/ly1.txt
blob: a70d0c125459cdeccccd15b9c3f3a6885011dd7c (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158


  LaTeX support for `texnansi' encoding (LY1)
  ===========================================

  ly1enc.def
  texnansi.sty
  ly1ptm.fd
  ly1phv.fd
  ly1pcr.fd
  ly1.txt

The files in this directory make known a new encoding `LY1'
to LaTeX that corresponds to the `texnansi' encoding as used
by default in the Y&Y TeX system. There is also an option
to support the `ansinew' encoding that is the default encoding
in Microsoft Windows.

The main file is ly1enc.def. This contains the declaration of
the encoding. To use `texnansi' encoded fonts within LaTeX, typically
you just need to add the line

\usepackage[LY1]{fontenc}

to your document preamble.

An alternative to the above line is to use the texnansi package

\usepackage{texnansi}

This will call the fontenc package, as above, to set up the LY1
encoding but then make further changes to the uppercase and lowercase
tables.
Note that LaTeX does not need these changes to correctly uppercase
text as its \MakeUppercase and \MakeLowercase commands only require
the uppercase and lower case codes of the letters a-z.

So for most purposes it is better to use the call to fontenc rather
than use the texnansi package. Then if you pass the file to a site
using T1 (Cork) encoded fonts the line can more easily be changed to

\usepackage[T1]{fontenc}

The texnansi package has a package option `ansinew' which modifies
the LY1 encoding to refer to that MS Windows encoding. However you
should only do this if your TeX system can not reencode fonts to
support T1 or texnansi. The `ansinew' treatment of the left quote
` as a grave accent makes it rather unsuitable for TeX use, also it
lacks several useful glyphs. However if you need to you can go:

\usepackage[ansinew]{texnansi}

The other possible reason for using the texnansi package relates
to the use of the lowercase table in the hyphenation algorithm.
This is described below, but it is unfortunately of necessity
rather technical, and you may want to stop reading at that point.

Having declared the LY1 encoding to LaTeX, LaTeX will need to know
the fonts that are available in this encoding. This distribution
includes ly1ptm.fd, ly1plv.fd and ly1pcr.fd which declare the
common Adobe Times/Helvetica/Courier `Standard PostScript' fonts.
If you install other text fonts you can make similar ly1XXX.fd files
based on this model. The LaTeX support packages for MathTime and
Lucida Bright contain suitable LY1 fd files for those fonts.
Note that these LY1 `fd' files assume that you have kept the
original font vendor's file names rather than renaming the fonts
to the Standard TeX `Karl Berry' naming scheme. The choice
of font naming scheme is however not connected to the choice of
encoding.

For each encoding LaTeX needs to know an `error font' which is its
fall back in the case that the font substitution system can find
nothing better to do. ly1enc.def specifies that the Error Font for
LY1 is Adobe Times Roman. This font is freely available from many
sources (for instance it is included with the Free Adobe Acrobat
reader) and so it seems the best choice as the error font.
If you do not have this, you may wish to edit the lines

\DeclareErrorFont{LY1}{ptm}{m}{n}{10}
\DeclareFontSubstitution{LY1}{ptm}{m}{n}

at the top of ly1enc.def to refer to some other font set.
For example if you have Y&Y TeX's `Lucida' distribution
You could change the two occurrences of {ptm} to {hlh}
Which will make LaTeX fall back on Lucida Bright if all else
fails. Note that the copyright is held on ly1enc.def.
This is the only change you should make to that file.


A Note on Lowercase Codes and Hyphenation Tables
================================================

The lowercase table is used by TeX's hyphenation algorithm,
so in principle one would need to change the lowercase table
to match the encoding used. However, it should not be changed
mid paragraph as all words, whatever font encoding they are
set in, are lowercased using the table current at the *end*
of the paragraph before being checked for hyphenation.
Because of these limitations in TeX's hyphenation algorithm,
LaTeX does not support changing the lowercase table mid document,
so if the texnansi package is used then the texnansi lowercase
table will be used for all fonts (whether or not LY1 encoded)
and conversely if it is not used, the T1 table will be used, even
for LY1 encoded fonts. The reason why this problem is not more
serious is that the majority of letters (as opposed to ligatures,
and symbols) that occur in LY1 occur in the same place in T1.

The only letters that might possibly appear in a hyphenation table
that are in different positions in T1 and LY1 are:

\oe  (low position OT1 high new)
\ss  (low position OT1 high new)
\v{s}
\v{z}
\l
\i   OT1  (although *composites* such as \'\i are in T1 position
           so this will only affect languages (Turkish?) that use
           an unaccented \i)
\"{y}


It should be noted that for those symbols that occur twice in LY1,
the `high' copy is used if it coincides with the position used
in T1 (as in the case of \ae, \o) otherwise the `low' copy will
be used which will coincide with the OT1 position (\oe, \ss).
(Unless the setup for ansinew is enabled in which case the high
position is always used.) 


Fortunately most hyphenation files for Western European languages
(I checked French, German and Danish) duplicate all the hyphenation
patterns using these letters so that the tables work with  both T1 and
OT1 (and hence LY1) encoded fonts. However if you are using a language
that makes use of say, \ss, but the hyphenation patterns just refer to
\ss as ^^ff then you will need to copy the hyphenation file and edit
all occurences of ^^ff to refer to the new position; either 
^^19 (for texnansi) or ^^df (for ansinew).

The remaining letters that are in different positions, such as s caron,
may be dealt with similarly, although they are mainly used in Eastern
European languages. The LY1 encoding does not have as many glyphs
for Eastern European languages as T1, so for those languages most likely
this encoding is not suitable. (Of course TeX can produce all the
accented letters required, but the hyphenation algorithm can not work
with such constructed composites it can only work with composite letters
that actually occur in the font.)

It would be a good thing if hyphenation tables did not refer to any
explicit character positions using the ^^ab notation, but rather always
accessed characters by control sequences such as \ss. This would allow
the remapping to be done via inputting a file such as ly1enc.def
rather than by editing the hyphenation file. However currently most of
the publicly available hyphenation tables unfortunately do `hard wire'
the T1 character codes in this way.


Copyright 1997  David Carlisle