% maamac.dtx -- documentation & source for maamac.tex -*-tex-*-
%%%@TeX-document-file {
%%% title = "MAAMAC -- Malvern A Macros",
%%% filename = "$texmf/doc/plain/pdcmac/maamac.dtx",
%%% version = "$Revision: 1.5 $",
%%% package = "pdcmac 1.0",
%%% date = "$Date: 1995/04/06 13:33:44 $",
%%% author = "P. Damian Cugley",
%%% email = "damian.cugley@comlab.ox.ac.uk",
%%% address = "Oxford University Computing Laboratory,"
%%% Parks Road, Oxford OX1 3QD, UK",
%%% abstract = "This document describes and is the source code for
%%% the TeX definitions file maamac.tex.
%%% Running plain TeX on this file produces both the
%%% definitions file and the printed documentation.",
%%% copyright = "Copyright (c) 1991-1995 P. Damian Cugley",
%%% copying = "This program is free software; you can redistribute
%%% it and/or modify it under the terms of the GNU
%%% General Public License as published by the Free
%%% Software Foundation; either version 2 of the License,
%%% or (at your option) any later version.",
%%% notice = "This program is distributed in the hope that it will
%%% be useful, but WITHOUT ANY WARRANTY; without even the
%%% implied warranty of MERCHANTABILITY or FITNESS FOR A
%%% PARTICULAR PURPOSE. See the GNU General Public
%%% License for more details.",
%%% notice = "You should have received a copy of the GNU General
%%% Public License along with this program; if not, write
%%% to the Free Software Foundation, Inc., 675 Mass Ave,
%%% Cambridge, MA 02139, USA.",
%%% codetable = "USASCII",
%%% dependencies = "pdccode.tex"
%%%}
%{{{ maamac
%{{{ preamble
\relax
\input pdccode
\document
\rcs$Id: maamac.dtx,v 1.5 1995/04/06 13:33:44 pdc Exp $\endrcs
\newcount\chcda \newcount\chcdc \newcount\chcdn
\def\changecodes#1#2#3#4{%
\chcda#2\relax \chcdc#3\relax \chcdn#4\relax
\loop \ifnum\chcdn>0
#1\chcda\chcdc
\advance\chcda1 \advance\chcdc1 \advance\chcdn-1
\repeat
}
\changecodes\mathcode{`a}{"7461}{26}
\codefile{maamac.tex}
%}}} preamble
%{{{ introduction
\author{P. Damian Cugley}
\title{MAAMAC---Malvern A Macros}
\advance\seccount-1
\section{Introduction}
This document describes |maamac.tex|, a collection of formatting
macros for plain \TeX\ documents using fonts with the Malvern~A
encoding. For example, it defines commands for making composite
letters and special symbols.
\subsec{About this document}
The definitions file and this printed documentation are both in
|maamac.dtx|, a `documented \TeX\ macros' file\footnote*{The
\LaTeX~2e distribution uses files with the `|dtx|' suffix for
similar purposes.} which, when processed by plain \TeX,
generates a fresh copy of |maamac.tex| in the current directory
in addition to the |dvi| file. This ensures that the printed
documentation and the code it describes are identical. The
|dtx| file uses the macros in |pdccode.tex|.
\subsec{What is in MAAMAC?}
The definitions in |maamac.tex| fall into three sections:
\smallskip
\item{\S1} Building {\it composite letters} like `\'e' from {\it
marks} like `\'{}' and {\it base letters} like `e'. This gives
new definitions for commands like |\'| and |\"|.
\item{\S2} Symbols and special letters: replacements for existing
commands like |\P| (\P) and |\ae| (\ae), and new commands like
|\registered| and |\pounds|.
\item{\S3} Switching between alphabets. The Malvern~A encoding includes
separate small capital and medium capital alphabets in addition
to the usual capitals and lower case. This section defines
commands for transliterating a list of tokens from one alphabet
to another.
\smallskip\noindent
The first two sections make changes to commands described in
{\it\TeX book} Chapter~9 (and a few maths symbols from
Chapter~16) and defined in Appendix~B.
%}}} introduction
%{{{ File identification
\subsec{File identification}
Nowadays, macro files start with some comments identifying the
file.
\code
\|\% maamac.tex \fileversion~\filedate~-- Malvern A Macros
|
|%%%@TeX-definition-file {
|%%% filename = "$texmf/tex/plain/pdcmac/maamac.tex",
\|\%\%\% version~~~~~~~~= "\fileversion",
\|\%\%\% date~~~~~~~~~~~= "\filedate",
|%%% package = "pdcmac 1.0",
|%%% author = "P. Damian Cugley",
|%%% email = "damian.cugley@comlab.ox.ac.uk",
|%%% address = "Oxford University Computing Laboratory,
|%%% Parks Road, Oxford OX1 3QD, UK",
|%%% codetable = "USASCII",
|%%% keywords = "TeX, plain TeX, macros",
|%%% supported = "Maybe",
|%%% abstract = "Macros for using the Malvern-A font coding
|%%% scheme in plain TeX documents.
|%%% This file was generated by running
\|\%\%\%~~~~~~~~~~~~~~~~~~~plain TeX on \jobname.dtx",
|%%% copyright = "Copyright (c) 1991-1995 P. Damian Cugley",
|%%% copying = "DO NOT DISTRIBUTE THIS FILE.
\|\%\%\%~~~~~~~~~~~~~~~~~~~Distribute \jobname.dtx only as part of the
|%%% package it came in.",
|%%% dependencies = "",
|%%% }
|
\|\\message\{\fileversion~<pdc \filedate>\}
|
\endcode
%}}}
%{{{ building composite letters
\section{Building composite letters}
THis section redefines commands listed at the start of Chapter~9
of the {\it\TeX book}: |\'|, |\"|, etc. The main difference is
that these now use `medium capitals' and differently-drawn marks
to make composite capital letters.
I~also define \<chardef tokens> called |\acuteglyph|,
|\twodotsglyph| which may be used to print these marks in a
document: `the acute mark ``\'{}''\dots' might be typeset with
`|the acute mark ``\acuteglyph''|\dots'. Also, if you don't
like the way my macros look, you can say `|\accent\acuteglyph
E|' rather than having to look up its glyph number.
%{{{ utility macros
\subsec{Utility macros}
First some nicknames for scratch registers.
\code
|\dimendef\dimena0 \dimendef\dimenb2
|\toksdef\toksa0
|\countdef\counta255
|\ifx\countb\UNDEFINED \csname newcount\endcsname\countb \fi
|\ifx\countc\UNDEFINED \csname newcount\endcsname\countc \fi
\endcode
A macro to find the code in \TeX's internal encoding for a
character. The character might be represented by (1)~a
character with category code~11 or~12, (2)~a token defined with
|\chardef|, or (3)~the token |\char| followed by a \<number>.
\code
|\def\code#1{%
| \ifcat#1a%
| `#1
| \else\ifcat#1+%
| `#1
| \else\ifx\char#1
\endcode
\hbox{\hskip6\fontdimen6\tentt
\% If the first token is |\char| the rest is the
\<number> we want.}
\code
| \else
| #1%
| \fi\fi\fi
|}
\endcode
So as to make conditional macros that can be skipped over if
they appear nested inside another conditional, we make include
the token |\then| amongst the arguments, making this the `if-'
token to match the |\else| and |\fi| following it. This way we
write, for example
$$\hbox{|\ifcap|\<character> |\then| \<true text>
|\else| \<false text> |\fi|}$$
\code
|\let\then=\iffalse
\smallbreak
|\def\ifcap#1\then{\expandafter\ifcapnum\code{#1}\then}
|\def\ifcapnum#1\then{\ifnum\lccode#1=#1}
\smallbreak
|\def\iflower#1\then{\expandafter\iflowernum\code{#1}\then}
|\def\iflowernum#1\then{\ifnum\uccode#1=#1}
\endcode
A macro to extract the slant-per-point parameter of a font in
toe form of a \TeX\ \<factor>.
\code
|\def\fontslant#1{\expandafter\stripPT\the\fontdimen1#1}
\smallbreak
|\begingroup \uccode`2=`p \uccode`3=`t \uppercase{\toksa{\endgroup
| \def\\#123{#1}
|}} \the\toksa
|\let\stripPT=\\
\endcode
%}}} utility macros
%{{{ macros for placing marks
\subsec{Macros for placing marks}
This simple macro places the text |#3| at displaced
$(\hbox{|#1|},\hbox{|#2|})$ from current position:
\code
|\def\putxy#1#2#3{\leavevmode\rlap{\hskip#1\raise#2\hbox{#3}}}
\endcode
Now we have a generic mark-placement macro. It is similar to
the |\accent| primitive, but takes two extra parameters:
$$
\hbox{|\putmarkxy|\arg{$xf$}\arg{$dy$}\arg{$mark$}\arg{$base$}}
$$
The horizontal position $xf$ of the optical centre of the base
letter (ignoring slant) is expressed as a \<factor> (i.e.,
decimal fraction of the width of the base glyph). The vertical
displacement $dy$ is expressed as a fragment of \TeX\ code that
will calculate $dy$. The code is expanded after setting box~0
to contain the base glyph, and must put the result in |\dimena|.
For example, we could define
$$
\hbox{|\def\fakeaccent{\putmarkxy{0.5}{\dimena\ht0
\advance\dimena-1ex }}|}
$$
This produces an approximation to the primitive |\accent|
command.
Given the $x$-factor $xf$, a base glyph with width $wd$, a font
with slant $s$, the horizontal displacement $dx$ is given by
$$
dx = xf \times wd + s \times dy\quad.
$$
In this macro it is stored in |\dimenb|.
\code
|\def\putmarkxy#1#2#3#4{{%
| \setbox0=\hbox{#4}#2\relax
| \dimenb#1\wd0 \advance\dimenb\fontslant\font\dimena
| \leavevmode
| \putxy{\dimenb}{\dimena}{\hbox to 0pt{\hss\char#3\hss}}%
| \box0
|}}
\smallbreak
|\def\putmarky{\putmarkxy{0.5}}
\endcode
%}}} macros for placing marks
%{{{ type I marks
\subsec{Type~I Marks}
A Type~I mark is one which does not have any scope for being
`squashed' to go over capital letters---for example, the
two-dots mark `\"{}' and macron `\={}'.
Here's a macro for placing a Type~I mark given by |#1| (a
\<number>) over base glyph |#2| (which is anything that |\code|
understands). The difference between this macro and simply
using |\accent| is (1)~if presented with a capital letter it
will use the corresponding medium capital letter and (2)~the the
mark is raised by an amount intended to centre it vertically
between the top of the base letter and the top of the
body-height---assuming it was original centered between
body-height and $x$-height.
If $H$ is the body height, $h$ the height of the base glyph,
$$ \mathcode`/="202F
\eqalign{dy &= { H + h \over 2 } - { H + 1\,{\rm
ex} \over 2}\cr
&= \textstyle { 1\over2}h - {1\over2}\,{\rm ex}\cr}
$$
\code
|\def\typeImark#1#2{{%
| \setbox2=\hbox{#2}%
| \ifdim \ht2=1ex
| \accent#1\relax#2%
| \else
| \counta\code{#2}%
| \ifnum\counta>64 \ifnum\counta<96
| \advance\counta 128
| \fi \fi
| \putmarky{\dimena0.5\ht0 \advance\dimena-0.5ex }
| {#1}{\char\counta}%
| \fi
|}}
\endcode
Finally we define the composite-letter-building commands.
\code
|\chardef\dotglyph22 \def\.{\typeImark\dotglyph}
|\chardef\twodotsglyph23 \def\"{\typeImark\twodotsglyph}
|\chardef\macronglyph150 \def\={\typeImark\macronglyph}
|\chardef\longmacronglyph151
| \def\longmacronmark{\typeImark\longmacronglyph}
\endcode
%}}} type I marks
%{{{ type II marks
\subsec{Type~II Marks}
These marks have an alternative glyph for use of capital and
medium capital letters. As with |\typeImark|, |#1| is a
\<number> and |#2| is anything understood by |\code|.
\code
|\def\typeIImark#1#2{{%
| \counta#1\relax \countb\code{#2}%
| \ifcap{#2}\then
| \ifnum \counta<128 \advance\counta 128 \fi
| \ifnum \countb<128 \advance\countb 128 \fi
| \fi
| \accent\counta \char\countb
|}}
\smallbreak
|\chardef\acuteglyph8 \def\'{\typeIImark\acuteglyph}
|\chardef\graveglyph9 \def\`{\typeIImark\graveglyph}
|\chardef\circumglyph10 \def\^{\typeIImark\circumglyph}
|\chardef\tildeglyph11 \def\~{\typeIImark\tildeglyph}
|\chardef\ringglyph12 \def\ringmark{\typeIImark\ringglyph}
|\chardef\caronglyph13 \def\v{\typeIImark\caronglyph}
|\chardef\breveglyph14 \def\u{\typeIImark\breveglyph}
|\chardef\hungarglyph15 \def\H{\typeIImark\hungarglyph}
\endcode
I have abbreviated `circumflex' as `circum' (consistent with
Adobe's `|asciicircum|'). I~have called the hook `\v{}'
`|caron|' because that is its name as used by ISO and Adobe.
I~have called the long Hungarian umlaut `\H{}' `|hungar|' for
want of a better name (it's shorter than |hungarumlaut|)..
%}}} type II marks
%{{{ cedilla ogonek
\subsec{Cedilla and ogonek}
The cedilla and ogonek each have a variant intended for capital
letters (in this case the variant is larger not smaller).
\code
|\def\cedillalike#1#2#3{{%
| \setbox0=\hbox{#3}%
| \ifdim\ht0>1ex
| \ooalign{\hidewidth\char#2\relax\hidewidth\crcr\unhbox0 }%
| \else
| \accent#1 #3%
| \fi
|}}
\smallbreak
|\chardef\cedillaglyph6 \chardef\Cedillaglyph134
|\def\c{\cedillalike\cedillaglyph\Cedillaglyph}
\smallbreak
|\chardef\ogonekglyph7 \chardef\Ogonekglyph135
|\def\k{\cedillalike\ogonekglyph\Ogonekglyph}
\endcode
%}}} cedilla ogonek
%}}}
%{{{ symbols
\section{Symbols and special letters}
%{{{ special letters
\subsec{Special letters}
The letters defined here that are not included in Chapter~9 of
the {\it\TeX book} are the Icelandic thorn (|\TH|, |\th|) and
eth\footnote*{Also called `edh'.} (|\DH|, |\dh|), the Polish
{\it a} and {\it e} with ogonek (|\A|, |\a|, |\E|, |\e|), and
the Sami eng\footnote\dag{Also called `ing' or `ng'.} (|\NG|,
|\ng|).
We also give replacements for {\it d}, {\it t} and {\it l} with
hook (which are usually drawn with a apostrophe-like mark), and
the Esperanto letter {\it hho} (\^h).
\code
|\chardef\TH"00 \chardef\th"10 \chardef\DH"04 \chardef\dh"14
|\chardef\A "01 \chardef\a "11 \chardef\E "05 \chardef\e "15
|\chardef\NG"02 \chardef\ng"12 \chardef\L "03 \chardef\l "13
|\chardef\AE"5E \chardef\ae"7E \chardef\OE"5F \chardef\oe"7F
|\chardef\O "5C \chardef\o "7C
\smallbreak
|\chardef\vd"9C \chardef\vt"9D \chardef\h "9E \chardef\vl"9F
|\chardef\ss"18 \chardef\i "19 \chardef\j "1A
\smallbreak
|\def\aa{\ringmark a} \def\AA{\ringmark A}
\endcode
\code
\smallbreak
|\chardef\orda"98 \chardef\ordo"99 \def\No{N\ordo} \def\no{n\ordo}
|\chardef\csuperior"98 \def\Mc{M\csuperior}
\endcode
The command |\Mc| is used for forming names like
M\flushtop{\the\scriptfont\fam \b c}Donald.
%}}} special letters
%{{{ text syms
\subsec{Other symbols used in text}
There are several new symbols here. Note that |\trademark| is
what you will get if you use the ASCII double quote character
`|"|' in your manuscript.
\code
|\chardef\trademark "22
|\chardef\copyright "FB \chardef\registered"FD
|\chardef\careof "9A \chardef\Box "A0
|\chardef\cents "A2 \chardef\pounds "A3
|\chardef\currency "A4 \chardef\permille "A5
|\chardef\yen "A8 \chardef\florin "A9
|\chardef\dag "AA \chardef\ddag "AB
|\chardef\gbdecimal "AE \chardef\minus "AF
|\chardef\S "BA \chardef\P "BB
|\chardef\degrees "C0
|\chardef\lguillemet"BC \chardef\rguillemet"BE
\endcode
The |\careof| sign ($\rm ^c\!/\!_o$) is an abbreviation for
`care of' in English-language addresses (actually, Australians
traditionally write `$\rm ^c\!/\!_-$'). The raised dot
|\gbdecimal| is the British decimal point (I~usually do
|\mathcode`.=\gbdecimal| and let maths mode do all the work).
The |\degrees| sign replaces \TeX's |$^\circ$|.
The guillemets can also be obtained with the ligatures |<<| and
|>>|.
%}}} text syms
%{{{ maths syms
\subsec{Symbols used in maths}
I have stuck with the plain \TeX\ conventions as much as
possible---even though |\bullet| is more often used in text than
maths.
\code
|\mathcode`\,"602C \mathcode`\-"20AF
|\mathcode`\."002E \mathcode`\/"002F
\smallbreak
|\mathchardef\bullet "2020 \mathchardef\times "20A6
|\mathchardef\backslash"005B \mathchardef\setminus "205B
|\mathchardef\cdot "20AE \mathchardef\cdotp "60AE
\smallbreak
|\def\langle{\delimiter"40DB30A } \delcode`\<"0DB30A
|\def\rangle{\delimiter"50DD30B } \delcode`\>"0DD30B
|\def\lbrace{\delimiter"407B308 } \let\{\lbrace \mathcode`\{"407B
|\def\rbrace{\delimiter"507D309 } \let\}\rbrace \mathcode`\}"407D
\endcode
%}}} maths syms
%}}} symbols
%{{{ switching alphabets
\section{Switching between alphabets}
This is one of the strange ideas that went into the design of
the Malvern font---I wanted to have small capitals and medium
capitals treated as separate alphabets rather than as a
different style, just as I treat old-style and ranging numerals
as separate symbols. These extra alphabets are present in all
Malvern~A encodings\footnote*{Fonts with the Malvern~A encoding
but made from normal PostScript fonts will have medium capitals
replaced with full capitals, and old-style numerals replaced
with ranging numerals.}
These extra alphabets are obtained by transliterating a sequence
of letters using the |\uppercase| and |\lowercase|
commands.\footnote\dag{Alas! this task would be so much easier
and more efficient if it were possible to declare new
alphabet-switching tables styled after the |\uccode| and
|\lccode| tables which make |\uppercase| and |\lowercase| work.
Then all this nonsense could happen in \TeX's mouth, which has
some advantages. I~can imagine commands |\newchartable| (used
to create a token which can be used in the same way as
|\uccode|) and |\usechartable| (used to defined macros that work
like |\uppercase|). E\TeX\ hackers take note!} For example, to
transliterate capitals into small capitals (the |\sc| command),
we first make assignments so that, for example, |\lccode`a| is
set to the code for small-capital $\scriptstyle A$, then apply
|\lowercase| to the token list.
\subsec{Setting upper and lower case codes for special letters}
These entries in the |\uucode| and |\lccode| tables are used by
the composite-letter-building commands rather then the
alphabet-switchign commands, but they are included here to be
near the other code-hackery.
\code
|\uccode\th\TH \lccode\TH\th \uccode\TH\TH \lccode\th\th
|\uccode\a \A \lccode\A \a \uccode\A \A \lccode\a \a
|\uccode\ng\NG \lccode\NG\ng \uccode\NG\NG \lccode\ng\ng
|\uccode\l \L \lccode\L \l \uccode\L \L \lccode\l \l
|\uccode\dh\DH \lccode\DH\dh \uccode\DH\DH \lccode\dh\dh
|\uccode\e \E \lccode\E \e \uccode\E \E \lccode\e \e
|\uccode\o \O \lccode\O \o \uccode\O \O \lccode\o \o
|\uccode\ae\AE \lccode\AE\ae \uccode\AE\AE \lccode\ae\ae
|\uccode\oe\OE \lccode\OE\oe \uccode\OE\OE \lccode\oe\oe
\smallbreak
|\uccode\i`I \uccode\j`J \lccode\i\i \lccode\j\j
|\uccode\csuperior`C
\endcode
\subsec{A macro for changing character codes}
The |\changecodes| command is used to change those segments of
character tables that are contiguous and map onto a contiguous
segment in the range. It is invoked as
$$
\hbox{\it |\changecodes| command x y n}
$$
(where $\it command$ is a token like |\uccode| or |\mathcode|
and $x$, $y$, and $n$ are \<number>s), and does the equivalent
of
$$
{\it command}\;i\mathrel{\hbox{\tt=}}
j\qquad \hbox{for all $\cases{x \le i < x + n\cr
y \le j < y + n\cr}$}
$$
\code
|\def\changecodes#1#2#3#4{%
| \counta#2\relax \countb#3\relax \countc#4\relax
| \loop
| \ifnum\countc>0
| #1\counta\countb
| \advance\counta 1
| \advance\countb 1
| \advance\countc-1
| \repeat
|}
\endcode
We can use this immediately to arrange that lowercasing medium caps
makes lower case, and uppercasing small caps makes full caps:
\code
|\changecodes\uccode{"E1}{`A}{26}
|\changecodes\uccode{"C1}{"C1}{26}
|\changecodes\lccode{"E1}{"E1}{26}
|\changecodes\lccode{"C1}{`a}{26}
\endcode
We also make old-style numerals behave as lower case for ranging
numerals, and also arrange that maths mode uses ranging figures.
\code
|\changecodes\lccode{"B0}{`0}{10}
|\changecodes\uccode{`0}{"B0}{10}
\endcode
In maths mode (1)~we use text italic for letters, since CM MAth
Italic will not match the body text and (2)~we use ranging
figures.
\code
|\changecodes\mathcode{`a}{"7461}{26}
|\changecodes\mathcode{`A}{"7441}{26}
|\changecodes\mathcode{`0}{"70B0}{10}
\endcode
\subsec{Changing the alphabet for special letters got with commands}
When applying |\uppercase| or |\lowercase| to a token list,
macros will be unaffected (|\ae| will still generate `ae'). To
get around this we define macros that redefine these control
sequences. They each take as parameters hex digits used to make
character codes.
\code
|\def\capspecials#1#2{%
| \chardef\TH"#10 \chardef\A "#11 \chardef\NG"#12
| \chardef\L "#13 \chardef\DH"#14 \chardef\E "#15
| \chardef\O "#2C \chardef\AE"#2E \chardef\OE"#2F
|}
\smallbreak
|\def\lcspecials#1#2#3{%
| \chardef\th"#10 \chardef\a "#11 \chardef\ng"#12
| \chardef\l "#13 \chardef\dh"#14 \chardef\e "#15
| \chardef\o "#2C \chardef\ae"#2E \chardef\oe"#2F
| \def\ss{\char"#23 \char"#23 }\chardef\i"#39 \chardef\j"#3A
|}
\endcode
\subsec{Pulling it all together}
Now we can define the user macros that do the work. These take
one argument, a piece of text to transliterate: |\sc{OX1~3QD}|
produces a small-capital `${\rm\scriptstyle OX}{\mit 1\
3}{\rm\scriptstyle QD}$';
|\csc{ABCdef}| produces caps and small caps
`ABC{\the\scriptfont0 DEF}'.
\code
|\def\sc#1{{%
| \changecodes\lccode{`A}{"E1}{26}\capspecials9F\lowercase{#1}%
|}}
\smallbreak
|\def\mc#1{{%
| \changecodes\lccode{`A}{"C1}{26}\capspecials8D\lowercase{#1}%
|}}
\smallbreak
|\def\csc#1{{%
| \changecodes\uccode{`a}{"E1}{26}\lcspecials9FE\uppercase{#1}%
|}}
\smallbreak
|\def\allcaps#1{{%
| \lcspecials054\uppercase{#1}%
|}}
\smallbreak
|\def\alllc#1{{%
| \capspecials17\lowercase{#1}%
|}}
\endcode
%}}} switching alphabets
\endcodefile
\enddocument
\bye
%}}} maamac
% Local variables:
% fold-folded-p: t
% fill-prefix: "\t"
% End: