Plan 9 from Bell Labs’s /usr/web/sources/contrib/steve/root/sys/src/cmd/tex/web2c/orig/mft.web

Copyright © 2021 Plan 9 Foundation.
Distributed under the MIT License.
Download the Plan 9 distribution.


% This program by D. E. Knuth is not copyrighted and can be used freely.
% Version 0.0 was more-or-less debugged on June 4, 1985.
% Version 0.1 improved formatting of : and added \\ (June 15, 1985).
% Version 0.2 improved formatting of good, fixed @@ bug (August 4, 1985).
% Version 0.3 fixed minor bug in change_file move (August 30, 1985).
% Version 0.4 fixed minor bug regarding empty comments (April 8, 1989).
% Version 1.0 was tuned up for the METAFONTware report (April 16, 1989).
% Version 1.1 ditto, with input handled by Hosek's idea (April 27, 1989).
% Version 2 has the new primitives of METAFONT 2.0 (October 16, 1989).

% Here is TeX material that gets inserted after \input webmac
\def\hang{\hangindent 3em\indent\ignorespaces}
\font\ninerm=cmr9
\let\mc=\ninerm % medium caps for names like SAIL
\def\PASCAL{Pascal}
\font\logo=manfnt % font used for the METAFONT logo
\def\MF{{\logo META}\-{\logo FONT}}
\def\pb{$\.|\ldots\.|$} % MF brackets (|...|)
\def\v{\.{\char'174}} % vertical (|) in typewriter font
\def\dleft{[\![} \def\dright{]\!]} % double brackets
\mathchardef\RA="3221 % right arrow
\mathchardef\BA="3224 % double arrow
\def\({} % kludge for alphabetizing certain module names
\chardef\V=`\| % vertical line in a string

\def\title{MFT}
\def\contentspagenumber{401}
\def\topofcontents{\null
  \def\titlepage{F} % include headline on the contents page
  \def\rheader{\mainfont\hfil \contentspagenumber}
  \vfill
  \centerline{\titlefont The {\ttitlefont MFT} processor}
  \vskip 15pt
  \centerline{(Version 2.0, October 1989)}
  \vfill}
\def\botofcontents{\vfill
  \centerline{\hsize 5in\baselineskip9pt
    \vbox{\ninerm\noindent
    The preparation of this report
    was supported in part by the National Science
    Foundation under grants IST-8201926, MCS-8300984, and
    CCR-8610181,
    and by the System Development Foundation. `\TeX' is a
    trademark of the American Mathematical Society.
    `{\logo hijklmnj}\kern1pt' is a trademark of Addison-Wesley
    Publishing Company.}}}
\pageno=\contentspagenumber \advance\pageno by 1

@* Introduction.
This program converts a \MF\ source file to a \TeX\ file. It was written
by D.~E. Knuth in June, 1985; a somewhat similar {\mc SAIL} program had
@^Knuth, Donald Ervin@>
been developed in January, 1980.

The general idea is to input a file called, say, \.{foo.mf} and to produce an
output file called, say, \.{foo.tex}. The latter file, when processed by \TeX,
will yield a ``prettyprinted'' representation of the input file.
@^user manual@>

Line breaks in the input are carried over into the output; moreover,
blank spaces at the beginning of a line are converted to quads of indentation
in the output. Thus, the user has full control over the indentation and line
breaks. Each line of input is translated independently of the others.

A slight change to \MF's comment convention allows further control.
Namely, `\.{\%\%}' indicates that the remainder of an input line should be
copied verbatim to the output; this interrupts the translation and forces
\.{MFT} to produce a certain result.

Furthermore, `\.{\%\%\%} $\langle\,$token$_1\,\rangle\ldots
\langle\,$token$_n\,\rangle$'
introduces a change in \.{MFT}'s formatting rules; all tokens after the first
will henceforth be translated according to the current conventions for
$\langle\,$token$_1\,\rangle$. The tokens must be symbolic (i.e., not
numeric or string tokens). For example, the input line
$$\.{\%\%\% addto fill draw filldraw}$$
says that the `\.{fill}', `\.{draw}', and `\.{filldraw}' operations of
plain \MF\ should be formatted as the primitive token `\.{addto}', i.e.,
in boldface type. (Without such reformatting commands, \.{MFT} would treat
`\.{fill}' like an ordinary tag or variable name. In fact, you need
a reformatting command even to get parentheses to act like delimiters!)

\MF\ comments, which follow a single \.\% sign, should be valid \TeX\
input.  But \MF\ material can be included in \pb\ within a comment; this
will be translated by \.{MFT} as if it were not in a comment. For example,
a phrase like `\.{make} \.{\V x2r\V} \.{zero}' will be translated into
`\.{make \$x\_\{2r\}\$ zero}'.

The rules just stated apply to lines that contain one, two, or three \.\% signs
in a row. Comments to \.{MFT} can follow `\.{\%\%\%\%}'.
Five or more \.\% signs should not be used.

Beside the normal input file, \.{MFT} also looks for a change file
(e.g., `\.{foo.ch}'), which allows substitutions to be made in the
translation. The change file follows the conventions of \.{WEB}, and
it should be null if there are no changes. (Changes usually contain
verbatim instructions to compensate for the fact that \.{MFT} cannot
format everything in an optimum way.)

There's also a third input file (e.g., `\.{plain.mft}'), which is
input before the other two. This file normally contains the `\.{\%\%\%}'
formatting commands that are necessary to tune \.{MFT} to a particular
style of \MF\ code, so it is called the style file.

The output of \.{MFT} should be accompanied by the macros in a small
package called \.{mftmac.tex}.
@.mftmac@>

Caveat: This program is not as ``bulletproof'' as the other routines
produced by Stanford's \TeX\ project. It takes care of a great deal of
tedious formatting, but it can produce strange output, because \MF\ is
an extremely general language. Users should proofread their output carefully.

@ \.{MFT} uses a few features of the local \PASCAL\ compiler that may
need to be changed in other installations:

\yskip\item{1)} Case statements have a default.
\item{2)} Input-output routines may need to be adapted for use with a particular
character set and/or for printing messages on the user's terminal.

\yskip\noindent
These features are also present in the \PASCAL\ version of \TeX, where they
are used in a similar (but more complex) way. System-dependent portions
of \.{MFT} can be identified by looking at the entries for `system
dependencies' in the index below.
@!@^system dependencies@>

The ``banner line'' defined here should be changed whenever \.{MFT}
is modified.

@d banner=='This is MFT, Version 2.0'

@ The program begins with a fairly normal header, made up of pieces that
@^system dependencies@>
will mostly be filled in later. The \.{MF} input comes from files |mf_file|,
|change_file|, and |style_file|; the \TeX\ output goes to file |tex_file|.

If it is necessary to abort the job because of a fatal error, the program
calls the `|jump_out|' procedure, which goes to the label |end_of_MFT|.

@d end_of_MFT = 9999 {go here to wrap it up}

@p @t\4@>@<Compiler directives@>@/
program MFT(@!mf_file,@!change_file,@!style_file,@!tex_file);
label end_of_MFT; {go here to finish}
const @<Constants in the outer block@>@/
type @<Types in the outer block@>@/
var @<Globals in the outer block@>@/
@<Error handling procedures@>@/
procedure initialize;
  var @<Local variables for initialization@>@/
  begin @<Set initial values@>@/
  end;

@ The \PASCAL\ compiler used to develop this system has ``compiler
directives'' that can appear in comments whose first character is a dollar sign.
In our case these directives tell the compiler to detect
@^system dependencies@>
things that are out of range.

@<Compiler directives@>=
@{@&$C+,A+,D-@} {range check, catch arithmetic overflow, no debug overhead}

@ Labels are given symbolic names by the following definitions. We insert
the label `|exit|:' just before the `\ignorespaces|end|\unskip' of a
procedure in which we have used the `|return|' statement defined below;
the label `|restart|' is occasionally used at the very beginning of a
procedure; and the label `|reswitch|' is occasionally used just prior to
a \&{case} statement in which some cases change the conditions and we wish to
branch to the newly applicable case.
Loops that are set up with the \&{loop} construction defined below are
commonly exited by going to `|done|' or to `|found|' or to `|not_found|',
and they are sometimes repeated by going to `|continue|'.

@d exit=10 {go here to leave a procedure}
@d restart=20 {go here to start a procedure again}
@d reswitch=21 {go here to start a case statement again}
@d continue=22 {go here to resume a loop}
@d done=30 {go here to exit a loop}
@d found=31 {go here when you've found it}
@d not_found=32 {go here when you've found something else}

@ Here are some macros for common programming idioms.

@d incr(#) == #:=#+1 {increase a variable by unity}
@d decr(#) == #:=#-1 {decrease a variable by unity}
@d loop == @+ while true do@+ {repeat over and over until a |goto| happens}
@d do_nothing == {empty statement}
@d return == goto exit {terminate a procedure call}
@f return == nil
@f loop == xclause

@ We assume that |case| statements may include a default case that applies
if no matching label is found. Thus, we shall use constructions like
@^system dependencies@>
$$\vbox{\halign{#\hfil\cr
|case x of|\cr
1: $\langle\,$code for $x=1\,\rangle$;\cr
3: $\langle\,$code for $x=3\,\rangle$;\cr
|othercases| $\langle\,$code for |x<>1| and |x<>3|$\,\rangle$\cr
|endcases|\cr}}$$
since most \PASCAL\ compilers have plugged this hole in the language by
incorporating some sort of default mechanism. For example, the compiler
used to develop \.{WEB} and \TeX\ allows `|others|:' as a default label,
and other \PASCAL s allow syntaxes like `\ignorespaces|else|\unskip' or
`\&{otherwise}' or `\\{otherwise}:', etc. The definitions of |othercases|
and |endcases| should be changed to agree with local conventions.
(Of course, if no default mechanism is available, the |case| statements of
this program must be extended by listing all remaining cases.)

@d othercases == others: {default for cases not listed explicitly}
@d endcases == @+end {follows the default case in an extended |case| statement}
@f othercases == else
@f endcases == end

@ The following parameters are set big enough to handle the Computer
Modern fonts, so they should be sufficient for most applications of \.{MFT}.

@<Constants...@>=
@!max_bytes=10000; {the number of bytes in tokens; must be less than 65536}
@!max_names=1000; {number of tokens}
@!hash_size=353; {should be prime}
@!buf_size=100; {maximum length of input line}
@!line_length=80; {lines of \TeX\ output have at most this many characters,
  should be less than 256}

@ A global variable called |history| will contain one of four values
at the end of every run: |spotless| means that no unusual messages were
printed; |harmless_message| means that a message of possible interest
was printed but no serious errors were detected; |error_message| means that
at least one error was found; |fatal_message| means that the program
terminated abnormally. The value of |history| does not influence the
behavior of the program; it is simply computed for the convenience
of systems that might want to use such information.

@d spotless=0 {|history| value for normal jobs}
@d harmless_message=1 {|history| value when non-serious info was printed}
@d error_message=2 {|history| value when an error was noted}
@d fatal_message=3 {|history| value when we had to stop prematurely}
@#
@d mark_harmless==@t@>@+if history=spotless then history:=harmless_message
@d mark_error==history:=error_message
@d mark_fatal==history:=fatal_message

@<Glob...@>=@!history:spotless..fatal_message; {how bad was this run?}

@ @<Set init...@>=history:=spotless;

@* The character set.
\.{MFT} works internally with ASCII codes, like all other programs
associated with \TeX\ and \MF. The present section has been lifted
almost verbatim from the \MF\ program.
@^ASCII code@>

@ Characters of text that have been converted to \MF's internal form
are said to be of type |ASCII_code|, which is a subrange of the integers.

@<Types...@>=
@!ASCII_code=0..255; {eight-bit numbers}

@ The original \PASCAL\ compiler was designed in the late 60s, when six-bit
character sets were common, so it did not make provision for lowercase
letters. Nowadays, of course, we need to deal with both capital and small
letters in a convenient way, especially in a program for font design;
so the present specification of \.{MFT} has been written under the assumption
that the \PASCAL\ compiler and run-time system permit the use of text files
with more than 64 distinguishable characters. More precisely, we assume that
the character set contains at least the letters and symbols associated
with ASCII codes @'40 through @'176. If additional characters are present,
\.{MFT} can be configured to work with them too.

Since we are dealing with more characters than were present in the first
\PASCAL\ compilers, we have to decide what to call the associated data
type. Some \PASCAL s use the original name |char| for the
characters in text files, even though there now are more than 64 such
characters, while other \PASCAL s consider |char| to be a 64-element
subrange of a larger data type that has some other name.

In order to accommodate this difference, we shall use the name |text_char|
to stand for the data type of the characters that are converted to and
from |ASCII_code| when they are input and output. We shall also assume
that |text_char| consists of the elements |chr(first_text_char)| through
|chr(last_text_char)|, inclusive. The following definitions should be
adjusted if necessary.
@^system dependencies@>

@d text_char == char {the data type of characters in text files}
@d first_text_char=0 {ordinal number of the smallest element of |text_char|}
@d last_text_char=255 {ordinal number of the largest element of |text_char|}

@<Types...@>=
@!text_file=packed file of text_char;

@ @<Local variables for init...@>=
@!i:0..255;

@ The \.{MFT} processor converts between ASCII code and
the user's external character set by means of arrays |xord| and |xchr|
that are analogous to \PASCAL's |ord| and |chr| functions.

@<Glob...@>=
@!xord: array [text_char] of ASCII_code;
  {specifies conversion of input characters}
@!xchr: array [ASCII_code] of text_char;
  {specifies conversion of output characters}

@ Since we are assuming that our \PASCAL\ system is able to read and write the
visible characters of standard ASCII (although not necessarily using the
ASCII codes to represent them), the following assignment statements initialize
most of the |xchr| array properly, without needing any system-dependent
changes. On the other hand, it is possible to implement \.{MFT} with
less complete character sets, and in such cases it will be necessary to
change something here.
@^system dependencies@>

@<Set init...@>=
xchr[@'40]:=' ';
xchr[@'41]:='!';
xchr[@'42]:='"';
xchr[@'43]:='#';
xchr[@'44]:='$';
xchr[@'45]:='%';
xchr[@'46]:='&';
xchr[@'47]:='''';@/
xchr[@'50]:='(';
xchr[@'51]:=')';
xchr[@'52]:='*';
xchr[@'53]:='+';
xchr[@'54]:=',';
xchr[@'55]:='-';
xchr[@'56]:='.';
xchr[@'57]:='/';@/
xchr[@'60]:='0';
xchr[@'61]:='1';
xchr[@'62]:='2';
xchr[@'63]:='3';
xchr[@'64]:='4';
xchr[@'65]:='5';
xchr[@'66]:='6';
xchr[@'67]:='7';@/
xchr[@'70]:='8';
xchr[@'71]:='9';
xchr[@'72]:=':';
xchr[@'73]:=';';
xchr[@'74]:='<';
xchr[@'75]:='=';
xchr[@'76]:='>';
xchr[@'77]:='?';@/
xchr[@'100]:='@@';
xchr[@'101]:='A';
xchr[@'102]:='B';
xchr[@'103]:='C';
xchr[@'104]:='D';
xchr[@'105]:='E';
xchr[@'106]:='F';
xchr[@'107]:='G';@/
xchr[@'110]:='H';
xchr[@'111]:='I';
xchr[@'112]:='J';
xchr[@'113]:='K';
xchr[@'114]:='L';
xchr[@'115]:='M';
xchr[@'116]:='N';
xchr[@'117]:='O';@/
xchr[@'120]:='P';
xchr[@'121]:='Q';
xchr[@'122]:='R';
xchr[@'123]:='S';
xchr[@'124]:='T';
xchr[@'125]:='U';
xchr[@'126]:='V';
xchr[@'127]:='W';@/
xchr[@'130]:='X';
xchr[@'131]:='Y';
xchr[@'132]:='Z';
xchr[@'133]:='[';
xchr[@'134]:='\';
xchr[@'135]:=']';
xchr[@'136]:='^';
xchr[@'137]:='_';@/
xchr[@'140]:='`';
xchr[@'141]:='a';
xchr[@'142]:='b';
xchr[@'143]:='c';
xchr[@'144]:='d';
xchr[@'145]:='e';
xchr[@'146]:='f';
xchr[@'147]:='g';@/
xchr[@'150]:='h';
xchr[@'151]:='i';
xchr[@'152]:='j';
xchr[@'153]:='k';
xchr[@'154]:='l';
xchr[@'155]:='m';
xchr[@'156]:='n';
xchr[@'157]:='o';@/
xchr[@'160]:='p';
xchr[@'161]:='q';
xchr[@'162]:='r';
xchr[@'163]:='s';
xchr[@'164]:='t';
xchr[@'165]:='u';
xchr[@'166]:='v';
xchr[@'167]:='w';@/
xchr[@'170]:='x';
xchr[@'171]:='y';
xchr[@'172]:='z';
xchr[@'173]:='{';
xchr[@'174]:='|';
xchr[@'175]:='}';
xchr[@'176]:='~';

@ The ASCII code is ``standard'' only to a certain extent, since many
computer installations have found it advantageous to have ready access
to more than 94 printing characters.  If \.{MFT} is being used
on a garden-variety \PASCAL\ for which only standard ASCII
codes will appear in the input and output files, it doesn't really matter
what codes are specified in |xchr[0..@'37]|, but the safest policy is to
blank everything out by using the code shown below.

However, other settings of |xchr| will make \.{MFT} more friendly on
computers that have an extended character set, so that users can type things
like `\.^^Z' instead of `\.{<>}', and so that \.{MFT} can echo the
page breaks found in its input.  People with extended character sets can
assign codes arbitrarily, giving an |xchr| equivalent to whatever
characters the users of \.{MFT} are allowed to have in their input files.
Appropriate changes to \.{MFT}'s |char_class| table should then be made.
(Unlike \TeX, each installation of \MF\ has a fixed assignment of category
codes, called the |char_class|.) Such changes make portability of programs
more difficult, so they should be introduced cautiously if at all.
@^character set dependencies@>
@^system dependencies@>

@<Set init...@>=
for i:=0 to @'37 do xchr[i]:=' ';
for i:=@'177 to @'377 do xchr[i]:=' ';

@ The following system-independent code makes the |xord| array contain a
suitable inverse to the information in |xchr|. Note that if |xchr[i]=xchr[j]|
where |i<j<@'177|, the value of |xord[xchr[i]]| will turn out to be
|j| or more; hence, standard ASCII code numbers will be used instead of
codes below @'40 in case there is a coincidence.

@<Set init...@>=
for i:=first_text_char to last_text_char do xord[chr(i)]:=@'177;
for i:=@'200 to @'377 do xord[xchr[i]]:=i;
for i:=1 to @'176 do xord[xchr[i]]:=i;

@* Input and output.
The I/O conventions of this program are essentially identical to those
of \.{WEAVE}.  Therefore people who need to make modifications should be
able to do so without too many headaches.

@ Terminal output is done by writing on file |term_out|, which is assumed to
consist of characters of type |text_char|:
@^system dependencies@>

@d print(#)==write(term_out,#) {`|print|' means write on the terminal}
@d print_ln(#)==write_ln(term_out,#) {`|print|' and then start new line}
@d new_line==write_ln(term_out) {start new line on the terminal}
@d print_nl(#)==  {print information starting on a new line}
  begin new_line; print(#);
  end

@<Globals...@>=
@!term_out:text_file; {the terminal as an output file}

@ Different systems have different ways of specifying that the output on a
certain file will appear on the user's terminal. Here is one way to do this
on the \PASCAL\ system that was used in \.{WEAVE}'s initial development:
@^system dependencies@>

@<Set init...@>=
rewrite(term_out,'TTY:'); {send |term_out| output to the terminal}

@ The |update_terminal| procedure is called when we want
to make sure that everything we have output to the terminal so far has
actually left the computer's internal buffers and been sent.
@^system dependencies@>

@d update_terminal == break(term_out) {empty the terminal output buffer}

@ The main input comes from |mf_file|; this input may be overridden
by changes in |change_file|. (If |change_file| is empty, there are no changes.)
Furthermore the |style_file| is input first; it is unchangeable.

@<Globals...@>=
@!mf_file:text_file; {primary input}
@!change_file:text_file; {updates}
@!style_file:text_file; {formatting bootstrap}

@ The following code opens the input files.  Since these files were listed
in the program header, we assume that the \PASCAL\ runtime system has
already checked that suitable file names have been given; therefore no
additional error checking needs to be done.
@^system dependencies@>

@p procedure open_input; {prepare to read the inputs}
begin reset(mf_file); reset(change_file); reset(style_file);
end;

@ The main output goes to |tex_file|.

@<Globals...@>=
@!tex_file: text_file;

@ The following code opens |tex_file|.
Since this file was listed in the program header, we assume that the
\PASCAL\ runtime system has checked that a suitable external file name has
been given.
@^system dependencies@>

@<Set init...@>=
rewrite(tex_file);

@ Input goes into an array called |buffer|.

@<Globals...@>=@!buffer: array[0..buf_size] of ASCII_code;

@ The |input_ln| procedure brings the next line of input from the specified
file into the |buffer| array and returns the value |true|, unless the file has
already been entirely read, in which case it returns |false|. The conventions
of \TeX\ are followed; i.e., |ASCII_code| numbers representing the next line
of the file are input into |buffer[0]|, |buffer[1]|, \dots,
|buffer[limit-1]|; trailing blanks are ignored;
and the global variable |limit| is set to the length of the
@^system dependencies@>
line. The value of |limit| must be strictly less than |buf_size|.

@p function input_ln(var f:text_file):boolean;
  {inputs a line or returns |false|}
var final_limit:0..buf_size; {|limit| without trailing blanks}
begin limit:=0; final_limit:=0;
if eof(f) then input_ln:=false
else  begin while not eoln(f) do
    begin buffer[limit]:=xord[f^]; get(f);
    incr(limit);
    if buffer[limit-1]<>" " then final_limit:=limit;
    if limit=buf_size then
      begin while not eoln(f) do get(f);
      decr(limit); {keep |buffer[buf_size]| empty}
      if final_limit>limit then final_limit:=limit;
      print_nl('! Input line too long'); loc:=0; error;
@.Input line too long@>
      end;
    end;
  read_ln(f); limit:=final_limit; input_ln:=true;
  end;
end;

@* Reporting errors to the user.
The command `|err_print('! Error message')|' will report a syntax error to
the user, by printing the error message at the beginning of a new line and
then giving an indication of where the error was spotted in the source file.
Note that no period follows the error message, since the error routine
will automatically supply a period.

The actual error indications are provided by a procedure called |error|.

@d err_print(#)==
    begin new_line; print(#); error;
    end

@<Error handling...@>=
procedure error; {prints `\..' and location of error message}
var@!k,@!l: 0..buf_size; {indices into |buffer|}
begin @<Print error location based on input buffer@>;
update_terminal; mark_error;
end;

@ The error locations can be indicated by using the global variables
|loc|, |line|, |styling|, and |changing|, which tell respectively the first
unlooked-at position in |buffer|, the current line number, and whether or not
the current line is from |style_file| or |change_file| or |mf_file|.
This routine should be modified on systems whose standard text editor
has special line-numbering conventions.
@^system dependencies@>

@<Print error location based on input buffer@>=
begin if styling then print('. (style file ')
else if changing then print('. (change file ')@+else print('. (');
print_ln('l.', line:1, ')');
if loc>=limit then l:=limit else l:=loc;
for k:=1 to l do
  print(xchr[buffer[k-1]]); {print the characters already read}
new_line;
for k:=1 to l do print(' '); {space out the next line}
for k:=l+1 to limit do print(xchr[buffer[k-1]]); {print the part not yet read}
end

@ The |jump_out| procedure just cuts across all active procedure levels
and jumps out of the program. This is the only non-local \&{goto} statement
in \.{MFT}. It is used when no recovery from a particular error has
been provided.

Some \PASCAL\ compilers do not implement non-local |goto| statements.
@^system dependencies@>
In such cases the code that appears at label |end_of_MFT| should be
copied into the |jump_out| procedure, followed by a call to a system procedure
that terminates the program.

@d fatal_error(#)==begin new_line; print(#); error; mark_fatal; jump_out;
  end

@<Error handling...@>=
procedure jump_out;
begin goto end_of_MFT;
end;

@ Sometimes the program's behavior is far different from what it should be,
and \.{MFT} prints an error message that is really for the \.{MFT}
maintenance person, not the user. In such cases the program says
|confusion('indication of where we are')|.

@d confusion(#)==fatal_error('! This can''t happen (',#,')')
@.This can't happen@>

@ An overflow stop occurs if \.{MFT}'s tables aren't large enough.

@d overflow(#)==fatal_error('! Sorry, ',#,' capacity exceeded')
@.Sorry, x capacity exceeded@>

@* Inserting the changes.
Let's turn now to the low-level routine |get_line|
that takes care of merging |change_file| into |mf_file|. The |get_line|
procedure also updates the line numbers for error messages.
(This routine was copied from \.{WEAVE}, but updated to include |styling|.)

@<Globals...@>=
@!line:integer; {the number of the current line in the current file}
@!other_line:integer; {the number of the current line in the input file that
  is not currently being read}
@!temp_line:integer; {used when interchanging |line| with |other_line|}
@!limit:0..buf_size; {the last character position occupied in the buffer}
@!loc:0..buf_size; {the next character position to be read from the buffer}
@!input_has_ended: boolean; {if |true|, there is no more input}
@!changing: boolean; {if |true|, the current line is from |change_file|}
@!styling: boolean; {if |true|, the current line is from |style_file|}

@ As we change |changing| from |true| to |false| and back again, we must
remember to swap the values of |line| and |other_line| so that the |err_print|
routine will be sure to report the correct line number.

@d change_changing==
  changing := not changing;
  temp_line:=other_line; other_line:=line; line:=temp_line
    {|line @t$\null\BA\null$@> other_line|}

@ When |changing| is |false|, the next line of |change_file| is kept in
|change_buffer[0..change_limit]|, for purposes of comparison with the next
line of |mf_file|. After the change file has been completely input, we
set |change_limit:=0|, so that no further matches will be made.

@<Globals...@>=
@!change_buffer:array[0..buf_size] of ASCII_code;
@!change_limit:0..buf_size; {the last position occupied in |change_buffer|}

@ Here's a simple function that checks if the two buffers are different.

@p function lines_dont_match:boolean;
label exit;
var k:0..buf_size; {index into the buffers}
begin lines_dont_match:=true;
if change_limit<>limit then return;
if limit>0 then
  for k:=0 to limit-1 do if change_buffer[k]<>buffer[k] then return;
lines_dont_match:=false;
exit: end;

@ Procedure |prime_the_change_buffer| sets |change_buffer| in preparation
for the next matching operation. Since blank lines in the change file are
not used for matching, we have |(change_limit=0)and not changing| if and
only if the change file is exhausted. This procedure is called only
when |changing| is true; hence error messages will be reported correctly.

@p procedure prime_the_change_buffer;
label continue, done, exit;
var k:0..buf_size; {index into the buffers}
begin change_limit:=0; {this value will be used if the change file ends}
@<Skip over comment lines in the change file; |return| if end of file@>;
@<Skip to the next nonblank line; |return| if end of file@>;
@<Move |buffer| and |limit| to |change_buffer| and |change_limit|@>;
exit: end;

@ While looking for a line that begins with \.{@@x} in the change file,
we allow lines that begin with \.{@@}, as long as they don't begin with
\.{@@y} or \.{@@z} (which would probably indicate that the change file is
fouled up).

@<Skip over comment lines in the change file...@>=
loop@+  begin incr(line);
  if not input_ln(change_file) then return;
  if limit<2 then goto continue;
  if buffer[0]<>"@@" then goto continue;
  if (buffer[1]>="X")and(buffer[1]<="Z") then
    buffer[1]:=buffer[1]+"z"-"Z"; {lowercasify}
  if buffer[1]="x" then goto done;
  if (buffer[1]="y")or(buffer[1]="z") then
    begin loc:=2; err_print('! Where is the matching @@x?');
@.Where is the match...@>
    end;
continue: end;
done:

@ Here we are looking at lines following the \.{@@x}.

@<Skip to the next nonblank line...@>=
repeat incr(line);
  if not input_ln(change_file) then
    begin err_print('! Change file ended after @@x');
@.Change file ended...@>
    return;
    end;
until limit>0;

@ @<Move |buffer| and |limit| to |change_buffer| and |change_limit|@>=
begin change_limit:=limit;
if limit>0 then for k:=0 to limit-1 do change_buffer[k]:=buffer[k];
end

@ The following procedure is used to see if the next change entry should
go into effect; it is called only when |changing| is false.
The idea is to test whether or not the current
contents of |buffer| matches the current contents of |change_buffer|.
If not, there's nothing more to do; but if so, a change is called for:
All of the text down to the \.{@@y} is supposed to match. An error
message is issued if any discrepancy is found. Then the procedure
prepares to read the next line from |change_file|.

@p procedure check_change; {switches to |change_file| if the buffers match}
label exit;
var n:integer; {the number of discrepancies found}
@!k:0..buf_size; {index into the buffers}
begin if lines_dont_match then return;
n:=0;
loop@+  begin change_changing; {now it's |true|}
  incr(line);
  if not input_ln(change_file) then
    begin err_print('! Change file ended before @@y');
@.Change file ended...@>
    change_limit:=0;  change_changing; {|false| again}
    return;
    end;
  @<If the current line starts with \.{@@y},
    report any discrepancies and |return|@>;
  @<Move |buffer| and |limit|...@>;
  change_changing; {now it's |false|}
  incr(line);
  if not input_ln(mf_file) then
    begin err_print('! MF file ended during a change');
@.MF file ended...@>
    input_has_ended:=true; return;
    end;
  if lines_dont_match then incr(n);
  end;
exit: end;

@ @<If the current line starts with \.{@@y}...@>=
if limit>1 then if buffer[0]="@@" then
  begin if (buffer[1]>="X")and(buffer[1]<="Z") then
    buffer[1]:=buffer[1]+"z"-"Z"; {lowercasify}
  if (buffer[1]="x")or(buffer[1]="z") then
    begin loc:=2; err_print('! Where is the matching @@y?');
@.Where is the match...@>
    end
  else if buffer[1]="y" then
    begin if n>0 then
      begin loc:=2; err_print('! Hmm... ',n:1,
        ' of the preceding lines failed to match');
@.Hmm... n of the preceding...@>
      end;
    return;
    end;
  end

@ Here's what we do to get the input rolling.

@<Initialize the input system@>=
begin open_input; line:=0; other_line:=0;@/
changing:=true; prime_the_change_buffer; change_changing;@/
styling:=true; limit:=0; loc:=1; buffer[0]:=" "; input_has_ended:=false;
end

@ The |get_line| procedure is called when |loc>limit|; it puts the next
line of merged input into the buffer and updates the other variables
appropriately.

@p procedure get_line; {inputs the next line}
label restart;
begin restart: if styling then
  @<Read from |style_file| and maybe turn off |styling|@>;
if not styling then
  begin if changing then
    @<Read from |change_file| and maybe turn off |changing|@>;
  if not changing then
    begin @<Read from |mf_file| and maybe turn on |changing|@>;
    if changing then goto restart;
    end;
  end;
end;

@ @<Read from |mf_file|...@>=
begin incr(line);
if not input_ln(mf_file) then input_has_ended:=true
else if limit=change_limit then
  if buffer[0]=change_buffer[0] then
    if change_limit>0 then check_change;
end

@ @<Read from |style_file|...@>=
begin incr(line);
if not input_ln(style_file) then
  begin styling:=false; line:=0;
  end;
end

@ @<Read from |change_file|...@>=
begin incr(line);
if not input_ln(change_file) then
  begin err_print('! Change file ended without @@z');
@.Change file ended...@>
  buffer[0]:="@@"; buffer[1]:="z"; limit:=2;
  end;
if limit>1 then {check if the change has ended}
  if buffer[0]="@@" then
    begin if (buffer[1]>="X")and(buffer[1]<="Z") then
      buffer[1]:=buffer[1]+"z"-"Z"; {lowercasify}
    if (buffer[1]="x")or(buffer[1]="y") then
      begin loc:=2; err_print('! Where is the matching @@z?');
@.Where is the match...@>
      end
    else if buffer[1]="z" then
      begin prime_the_change_buffer; change_changing;
      end;
    end;
end

@ At the end of the program, we will tell the user if the change file
had a line that didn't match any relevant line in |mf_file|.

@<Check that all changes have been read@>=
if change_limit<>0 then {|changing| is false}
  begin for loc:=0 to change_limit do buffer[loc]:=change_buffer[loc];
  limit:=change_limit; changing:=true; line:=other_line; loc:=change_limit;
  err_print('! Change file entry did not match');
@.Change file entry did not match@>
  end

@* Data structures.
\.{MFT} puts token names
into the large |byte_mem| array, which is packed with eight-bit integers.
Allocation is sequential, since names are never deleted.

An auxiliary array |byte_start| is used as a directory for |byte_mem|;
the |link| and |ilk| arrays give further information about names.
These auxiliary arrays consist of sixteen-bit items.

@<Types...@>=
@!eight_bits=0..255; {unsigned one-byte quantity}
@!sixteen_bits=0..65535; {unsigned two-byte quantity}

@ \.{MFT} has been designed to avoid the need for indices that are more
than sixteen bits wide, so that it can be used on most computers.

@<Globals...@>=
@!byte_mem: packed array [0..max_bytes] of ASCII_code; {characters of names}
@!byte_start: array [0..max_names] of sixteen_bits; {directory into |byte_mem|}
@!link: array [0..max_names] of sixteen_bits; {hash table links}
@!ilk: array [0..max_names] of sixteen_bits; {type codes}

@ The names of tokens are found by computing a hash address |h| and
then looking at strings of bytes signified by |hash[h]|, |link[hash[h]]|,
|link[link[hash[h]]]|, \dots, until either finding the desired name
or encountering a zero.

A `|name_pointer|' variable, which signifies a name, is an index into
|byte_start|. The actual sequence of characters in the name pointed to by
|p| appears in positions |byte_start[p]| to |byte_start[p+1]-1|, inclusive,
of |byte_mem|.

We usually have |byte_start[name_ptr]=byte_ptr|, which is
the starting position for the next name to be stored in |byte_mem|.

@d length(#)==byte_start[#+1]-byte_start[#] {the length of a name}

@<Types...@>=
@!name_pointer=0..max_names; {identifies a name}

@ @<Global...@>=
@!name_ptr:name_pointer; {first unused position in |byte_start|}
@!byte_ptr:0..max_bytes; {first unused position in |byte_mem|}

@ @<Set init...@>=
byte_start[0]:=0; byte_ptr:=0;
byte_start[1]:=0; {this makes name 0 of length zero}
name_ptr:=1;

@ The hash table described above is updated by the |lookup| procedure,
which finds a given name and returns a pointer to its index in
|byte_start|. The token is supposed to match character by character.
If it was not already present, it is inserted into the table.

Because of the way \.{MFT}'s scanning mechanism works, it is most convenient
to let |lookup| search for a token that is present in the |buffer|
array. Two other global variables specify its position in the buffer: the
first character is |buffer[id_first]|, and the last is |buffer[id_loc-1]|.

@<Glob...@>=
@!id_first:0..buf_size; {where the current token begins in the buffer}
@!id_loc:0..buf_size; {just after the current token in the buffer}
@#
@!hash:array [0..hash_size] of sixteen_bits; {heads of hash lists}

@ Initially all the hash lists are empty.

@<Local variables for init...@>=
@!h:0..hash_size; {index into hash-head array}

@ @<Set init...@>=
for h:=0 to hash_size-1 do hash[h]:=0;

@ Here now is the main procedure for finding tokens.

@p function lookup:name_pointer; {finds current token}
label found;
var i:0..buf_size; {index into |buffer|}
@!h:0..hash_size; {hash code}
@!k:0..max_bytes; {index into |byte_mem|}
@!l:0..buf_size; {length of the given token}
@!p:name_pointer; {where the token is being sought}
begin l:=id_loc-id_first; {compute the length}
@<Compute the hash code |h|@>;
@<Compute the name location |p|@>;
if p=name_ptr then @<Enter a new name into the table at position |p|@>;
lookup:=p;
end;

@ A simple hash code is used: If the sequence of
ASCII codes is $c_1c_2\ldots c_m$, its hash value will be
$$(2^{n-1}c_1+2^{n-2}c_2+\cdots+c_n)\,\bmod\,|hash_size|.$$

@<Compute the hash...@>=
h:=buffer[id_first]; i:=id_first+1;
while i<id_loc do
  begin h:=(h+h+buffer[i]) mod hash_size; incr(i);
  end

@ If the token is new, it will be placed in position |p=name_ptr|,
otherwise |p| will point to its existing location.

@<Compute the name location...@>=
p:=hash[h];
while p<>0 do
  begin if length(p)=l then
    @<Compare name |p| with current token,
      |goto found| if equal@>;
  p:=link[p];
  end;
p:=name_ptr; {the current token is new}
link[p]:=hash[h]; hash[h]:=p; {insert |p| at beginning of hash list}
found:

@ @<Compare name |p|...@>=
begin i:=id_first; k:=byte_start[p];
while (i<id_loc)and(buffer[i]=byte_mem[k]) do
  begin incr(i); incr(k);
  end;
if i=id_loc then goto found; {all characters agree}
end

@ When we begin the following segment of the program, |p=name_ptr|.

@<Enter a new name...@>=
begin if byte_ptr+l>max_bytes then overflow('byte memory');
if name_ptr+1>max_names then overflow('name');
i:=id_first; {get ready to move the token into |byte_mem|}
while i<id_loc do
  begin byte_mem[byte_ptr]:=buffer[i]; incr(byte_ptr); incr(i);
  end;
incr(name_ptr); byte_start[name_ptr]:=byte_ptr;
@<Assign the default value to |ilk[p]|@>;
end

@* Initializing the primitive tokens.
Each token read by \.{MFT} is recognized as belonging to one of the
following ``types'':

@d indentation=0 {internal code for space at beginning of a line}
@d end_of_line=1 {internal code for hypothetical token at end of a line}
@d end_of_file=2 {internal code for hypothetical token at end of the input}
@d verbatim=3 {internal code for the token `\.{\%\%}'}
@d set_format=4 {internal code for the token `\.{\%\%\%}'}
@d mft_comment=5 {internal code for the token `\.{\%\%\%\%}'}
@d min_action_type=6 {smallest code for tokens that produce ``real'' output}
@d numeric_token=6 {internal code for tokens like `\.{3.14159}'}
@d string_token=7 {internal code for tokens like `|"pie"|'}
@d min_symbolic_token=8 {smallest internal code for a symbolic token}
@d op=8 {internal code for tokens like `\.{sqrt}'}
@d command=9 {internal code for tokens like `\.{addto}'}
@d endit=10 {internal code for tokens like `\.{fi}'}
@d binary=11 {internal code for tokens like `\.{and}'}
@d abinary=12 {internal code for tokens like `\.{+}'}
@d bbinary=13 {internal code for tokens like `\.{step}'}
@d ampersand=14 {internal code for the token `\.{\char`\&}'}
@d pyth_sub=15 {internal code for the token `\.{+-+}'}
@d as_is=16 {internal code for tokens like `\.{]}'}
@d bold=17 {internal code for tokens like `\.{nullpen}'}
@d type_name=18 {internal code for tokens like `\.{numeric}'}
@d path_join=19 {internal code for the token `\.{..}'}
@d colon=20 {internal code for the token `\.:'}
@d semicolon=21 {internal code for the token `\.;'}
@d backslash=22 {internal code for the token `\.{\\}'}
@d double_back=23 {internal code for the token `\.{\\\\}'}
@d less_or_equal=24 {internal code for the token `\.{<=}'}
@d greater_or_equal=25 {internal code for the token `\.{>=}'}
@d not_equal=26 {internal code for the token `\.{<>}'}
@d sharp=27 {internal code for the token `\.{\char`\#}'}
@d comment=28 {internal code for the token `\.{\char`\%}'}
@d recomment=29 {internal code used to resume a comment after `\pb'}
@d min_suffix=30 {smallest code for symbolic tokens in suffixes}
@d internal=30 {internal code for tokens like `\.{pausing}'}
@d input_command=31 {internal code for tokens like `\.{input}'}
@d special_tag=32 {internal code for tags that take at most one subscript}
@d tag=33 {internal code for nonprimitive tokens}

@<Assign the default value to |ilk[p]|@>=ilk[p]:=tag

@ We have to get \MF's primitives into the hash table, and the
simplest way to do this is to insert them every time \.{MFT} is run.

A few macros permit us to do the initialization with a compact program.
We use the fact that the longest primitive is \.{intersectiontimes},
which is 17 letters long.

@d spr17(#)==buffer[17]:=#;cur_tok:=lookup;ilk[cur_tok]:=
@d spr16(#)==buffer[16]:=#;spr17
@d spr15(#)==buffer[15]:=#;spr16
@d spr14(#)==buffer[14]:=#;spr15
@d spr13(#)==buffer[13]:=#;spr14
@d spr12(#)==buffer[12]:=#;spr13
@d spr11(#)==buffer[11]:=#;spr12
@d spr10(#)==buffer[10]:=#;spr11
@d spr9(#)==buffer[9]:=#;spr10
@d spr8(#)==buffer[8]:=#;spr9
@d spr7(#)==buffer[7]:=#;spr8
@d spr6(#)==buffer[6]:=#;spr7
@d spr5(#)==buffer[5]:=#;spr6
@d spr4(#)==buffer[4]:=#;spr5
@d spr3(#)==buffer[3]:=#;spr4
@d spr2(#)==buffer[2]:=#;spr3
@d spr1(#)==buffer[1]:=#;spr2
@d pr1==id_first:=17; spr17
@d pr2==id_first:=16; spr16
@d pr3==id_first:=15; spr15
@d pr4==id_first:=14; spr14
@d pr5==id_first:=13; spr13
@d pr6==id_first:=12; spr12
@d pr7==id_first:=11; spr11
@d pr8==id_first:=10; spr10
@d pr9==id_first:=9; spr9
@d pr10==id_first:=8; spr8
@d pr11==id_first:=7; spr7
@d pr12==id_first:=6; spr6
@d pr13==id_first:=5; spr5
@d pr14==id_first:=4; spr4
@d pr15==id_first:=3; spr3
@d pr16==id_first:=2; spr2
@d pr17==id_first:=1; spr1

@ The intended use of the macros above might not be immediately obvious,
but the riddle is answered by the following:

@<Store all the primitives@>=
id_loc:=18;@/
pr2(".")(".")(path_join);@/
pr1("[")(as_is);@/
pr1("]")(as_is);@/
pr1("}")(as_is);@/
pr1("{")(as_is);@/
pr1(":")(colon);@/
pr2(":")(":")(colon);@/
pr3("|")("|")(":")(colon);@/
pr2(":")("=")(as_is);@/
pr1(",")(as_is);@/
pr1(";")(semicolon);@/
pr1("\")(backslash);@/
pr2("\")("\")(double_back);@/
pr5("a")("d")("d")("t")("o")(command);@/
pr2("a")("t")(bbinary);@/
pr7("a")("t")("l")("e")("a")("s")("t")(op);@/
pr10("b")("e")("g")("i")("n")("g")("r")("o")("u")("p")(command);
pr8("c")("o")("n")("t")("r")("o")("l")("s")(op);@/
pr4("c")("u")("l")("l")(command);@/
pr4("c")("u")("r")("l")(op);@/
pr10("d")("e")("l")("i")("m")("i")("t")("e")("r")("s")(command);@/
pr7("d")("i")("s")("p")("l")("a")("y")(command);@/
pr8("e")("n")("d")("g")("r")("o")("u")("p")(endit);@/
pr8("e")("v")("e")("r")("y")("j")("o")("b")(command);@/
pr6("e")("x")("i")("t")("i")("f")(command);@/
pr11("e")("x")("p")("a")("n")("d")("a")("f")("t")("e")("r")(command);@/
pr4("f")("r")("o")("m")(bbinary);@/
pr8("i")("n")("w")("i")("n")("d")("o")("w")(bbinary);@/
pr7("i")("n")("t")("e")("r")("i")("m")(command);@/
pr3("l")("e")("t")(command);@/
pr11("n")("e")("w")("i")("n")("t")("e")("r")("n")("a")("l")(command);@/
pr2("o")("f")(command);@/
pr10("o")("p")("e")("n")("w")("i")("n")("d")("o")("w")(command);@/
pr10("r")("a")("n")("d")("o")("m")("s")("e")("e")("d")(command);@/
pr4("s")("a")("v")("e")(command);@/
pr10("s")("c")("a")("n")("t")("o")("k")("e")("n")("s")(command);@/
pr7("s")("h")("i")("p")("o")("u")("t")(command);@/
pr4("s")("t")("e")("p")(bbinary);@/
pr3("s")("t")("r")(command);@/
pr7("t")("e")("n")("s")("i")("o")("n")(op);@/
pr2("t")("o")(bbinary);@/
pr5("u")("n")("t")("i")("l")(bbinary);@/
pr3("d")("e")("f")(command);@/
pr6("v")("a")("r")("d")("e")("f")(command);@/

@ (There are so many primitives, it's necessary to break this long
initialization code up into pieces so as not to overflow \.{WEAVE}'s capacity.)

@<Store all the primitives@>=
pr10("p")("r")("i")("m")("a")("r")("y")("d")("e")("f")(command);@/
pr12("s")("e")("c")("o")("n")("d")("a")("r")("y")("d")("e")("f")(command);@/
pr11("t")("e")("r")("t")("i")("a")("r")("y")("d")("e")("f")(command);@/
pr6("e")("n")("d")("d")("e")("f")(endit);@/
pr3("f")("o")("r")(command);@/
pr11("f")("o")("r")("s")("u")("f")("f")("i")("x")("e")("s")(command);@/
pr7("f")("o")("r")("e")("v")("e")("r")(command);@/
pr6("e")("n")("d")("f")("o")("r")(endit);@/
pr5("q")("u")("o")("t")("e")(command);@/
pr4("e")("x")("p")("r")(command);@/
pr6("s")("u")("f")("f")("i")("x")(command);@/
pr4("t")("e")("x")("t")(command);@/
pr7("p")("r")("i")("m")("a")("r")("y")(command);@/
pr9("s")("e")("c")("o")("n")("d")("a")("r")("y")(command);@/
pr8("t")("e")("r")("t")("i")("a")("r")("y")(command);@/
pr5("i")("n")("p")("u")("t")(input_command);@/
pr8("e")("n")("d")("i")("n")("p")("u")("t")(bold);@/
pr2("i")("f")(command);@/
pr2("f")("i")(endit);@/
pr4("e")("l")("s")("e")(command);@/
pr6("e")("l")("s")("e")("i")("f")(command);@/
pr4("t")("r")("u")("e")(bold);@/
pr5("f")("a")("l")("s")("e")(bold);@/
pr11("n")("u")("l")("l")("p")("i")("c")("t")("u")("r")("e")(bold);@/
pr7("n")("u")("l")("l")("p")("e")("n")(bold);@/
pr7("j")("o")("b")("n")("a")("m")("e")(bold);@/
pr10("r")("e")("a")("d")("s")("t")("r")("i")("n")("g")(bold);@/
pr9("p")("e")("n")("c")("i")("r")("c")("l")("e")(bold);@/
pr4("g")("o")("o")("d")(special_tag);@/
pr2("=")(":")(as_is);@/
pr3("=")(":")("|")(as_is);@/
pr4("=")(":")("|")(">")(as_is);@/
pr3("|")("=")(":")(as_is);@/
pr4("|")("=")(":")(">")(as_is);@/
pr4("|")("=")(":")("|")(as_is);@/
pr5("|")("=")(":")("|")(">")(as_is);@/
pr6("|")("=")(":")("|")(">")(">")(as_is);@/
pr4("k")("e")("r")("n")(binary);
pr6("s")("k")("i")("p")("t")("o")(command);@/

@ (Does anybody out there remember the commercials that went \.{LS-MFT}?)

@<Store all the prim...@>=
pr13("n")("o")("r")("m")("a")("l")("d")("e")("v")("i")("a")("t")("e")(op);@/
pr3("o")("d")("d")(op);@/
pr5("k")("n")("o")("w")("n")(op);@/
pr7("u")("n")("k")("n")("o")("w")("n")(op);@/
pr3("n")("o")("t")(op);@/
pr7("d")("e")("c")("i")("m")("a")("l")(op);@/
pr7("r")("e")("v")("e")("r")("s")("e")(op);@/
pr8("m")("a")("k")("e")("p")("a")("t")("h")(op);@/
pr7("m")("a")("k")("e")("p")("e")("n")(op);@/
pr11("t")("o")("t")("a")("l")("w")("e")("i")("g")("h")("t")(op);@/
pr3("o")("c")("t")(op);@/
pr3("h")("e")("x")(op);@/
pr5("A")("S")("C")("I")("I")(op);@/
pr4("c")("h")("a")("r")(op);@/
pr6("l")("e")("n")("g")("t")("h")(op);@/
pr13("t")("u")("r")("n")("i")("n")("g")("n")("u")("m")("b")("e")("r")(op);@/
pr5("x")("p")("a")("r")("t")(op);@/
pr5("y")("p")("a")("r")("t")(op);@/
pr6("x")("x")("p")("a")("r")("t")(op);@/
pr6("x")("y")("p")("a")("r")("t")(op);@/
pr6("y")("x")("p")("a")("r")("t")(op);@/
pr6("y")("y")("p")("a")("r")("t")(op);@/
pr4("s")("q")("r")("t")(op);@/
pr4("m")("e")("x")("p")(op);@/
pr4("m")("l")("o")("g")(op);@/
pr4("s")("i")("n")("d")(op);@/
pr4("c")("o")("s")("d")(op);@/
pr5("f")("l")("o")("o")("r")(op);@/
pr14("u")("n")("i")("f")("o")("r")("m")("d")("e")("v")("i")("a")("t")("e")(op);
  @/
pr10("c")("h")("a")("r")("e")("x")("i")("s")("t")("s")(op);@/
pr5("a")("n")("g")("l")("e")(op);@/
pr5("c")("y")("c")("l")("e")(op);@/

@ (If you think this \.{WEB} code is ugly, you should see the Pascal code
it produces.)

@<Store all the primitives@>=
pr13("t")("r")("a")("c")("i")("n")("g")
 ("t")("i")("t")("l")("e")("s")(internal);@/
pr16("t")("r")("a")("c")("i")("n")("g")
 ("e")("q")("u")("a")("t")("i")("o")("n")("s")(internal);@/
pr15("t")("r")("a")("c")("i")("n")("g")
 ("c")("a")("p")("s")("u")("l")("e")("s")(internal);@/
pr14("t")("r")("a")("c")("i")("n")("g")
 ("c")("h")("o")("i")("c")("e")("s")(internal);@/
pr12("t")("r")("a")("c")("i")("n")("g")
 ("s")("p")("e")("c")("s")(internal);@/
pr11("t")("r")("a")("c")("i")("n")("g")
 ("p")("e")("n")("s")(internal);@/
pr15("t")("r")("a")("c")("i")("n")("g")
 ("c")("o")("m")("m")("a")("n")("d")("s")(internal);@/
pr13("t")("r")("a")("c")("i")("n")("g")
 ("m")("a")("c")("r")("o")("s")(internal);@/
pr12("t")("r")("a")("c")("i")("n")("g")
 ("e")("d")("g")("e")("s")(internal);@/
pr13("t")("r")("a")("c")("i")("n")("g")
 ("o")("u")("t")("p")("u")("t")(internal);@/
pr12("t")("r")("a")("c")("i")("n")("g")
 ("s")("t")("a")("t")("s")(internal);@/
pr13("t")("r")("a")("c")("i")("n")("g")
 ("o")("n")("l")("i")("n")("e")(internal);@/

@ @<Store all the primitives@>=
pr4("y")("e")("a")("r")(internal);@/
pr5("m")("o")("n")("t")("h")(internal);@/
pr3("d")("a")("y")(internal);@/
pr4("t")("i")("m")("e")(internal);@/
pr8("c")("h")("a")("r")("c")("o")("d")("e")(internal);@/
pr7("c")("h")("a")("r")("f")("a")("m")(internal);@/
pr6("c")("h")("a")("r")("w")("d")(internal);@/
pr6("c")("h")("a")("r")("h")("t")(internal);@/
pr6("c")("h")("a")("r")("d")("p")(internal);@/
pr6("c")("h")("a")("r")("i")("c")(internal);@/
pr6("c")("h")("a")("r")("d")("x")(internal);@/
pr6("c")("h")("a")("r")("d")("y")(internal);@/
pr10("d")("e")("s")("i")("g")("n")("s")("i")("z")("e")(internal);@/
pr4("h")("p")("p")("p")(internal);@/
pr4("v")("p")("p")("p")(internal);@/
pr7("x")("o")("f")("f")("s")("e")("t")(internal);@/
pr7("y")("o")("f")("f")("s")("e")("t")(internal);@/
pr7("p")("a")("u")("s")("i")("n")("g")(internal);@/
pr12("s")("h")("o")("w")
 ("s")("t")("o")("p")("p")("i")("n")("g")(internal);@/
pr10("f")("o")("n")("t")("m")("a")("k")("i")("n")("g")(internal);@/
pr8("p")("r")("o")("o")("f")("i")("n")("g")(internal);@/
pr9("s")("m")("o")("o")("t")("h")("i")("n")("g")(internal);@/
pr12("a")("u")("t")("o")("r")("o")("u")("n")("d")("i")("n")("g")(internal);@/
pr11("g")("r")("a")("n")("u")("l")("a")("r")("i")("t")("y")(internal);@/
pr6("f")("i")("l")("l")("i")("n")(internal);@/
pr12("t")("u")("r")("n")("i")("n")("g")("c")("h")("e")("c")("k")(internal);@/
pr12("w")("a")("r")("n")("i")("n")("g")("c")("h")("e")("c")("k")(internal);@/
pr12("b")("o")("u")("n")("d")("a")("r")("y")("c")("h")("a")("r")(internal);@/

@ Still more.

@<Store all the prim...@>=
pr1("+")(abinary);@/
pr1("-")(abinary);@/
pr1("*")(abinary);@/
pr1("/")(as_is);@/
pr2("+")("+")(binary);@/
pr3("+")("-")("+")(pyth_sub);@/
pr3("a")("n")("d")(binary);@/
pr2("o")("r")(binary);@/
pr1("<")(as_is);@/
pr2("<")("=")(less_or_equal);@/
pr1(">")(as_is);@/
pr2(">")("=")(greater_or_equal);@/
pr1("=")(as_is);@/
pr2("<")(">")(not_equal);@/
pr9("s")("u")("b")("s")("t")("r")("i")("n")("g")(command);@/
pr7("s")("u")("b")("p")("a")("t")("h")(command);@/
pr13("d")("i")("r")("e")("c")("t")("i")("o")("n")@|
 ("t")("i")("m")("e")(command);@/
pr5("p")("o")("i")("n")("t")(command);@/
pr10("p")("r")("e")("c")("o")("n")("t")("r")("o")("l")(command);@/
pr11("p")("o")("s")("t")("c")("o")("n")("t")("r")("o")("l")(command);@/
pr9("p")("e")("n")("o")("f")("f")("s")("e")("t")(command);@/
pr1("&")(ampersand);@/
pr7("r")("o")("t")("a")("t")("e")("d")(binary);@/
pr7("s")("l")("a")("n")("t")("e")("d")(binary);@/
pr6("s")("c")("a")("l")("e")("d")(binary);@/
pr7("s")("h")("i")("f")("t")("e")("d")(binary);@/
pr11("t")("r")("a")("n")("s")("f")("o")("r")("m")("e")("d")(binary);@/
pr7("x")("s")("c")("a")("l")("e")("d")(binary);@/
pr7("y")("s")("c")("a")("l")("e")("d")(binary);@/
pr7("z")("s")("c")("a")("l")("e")("d")(binary);@/
pr17("i")("n")("t")("e")("r")("s")("e")("c")("t")("i")("o")("n")@|
 ("t")("i")("m")("e")("s")(binary);@/
pr7("n")("u")("m")("e")("r")("i")("c")(type_name);@/
pr6("s")("t")("r")("i")("n")("g")(type_name);@/
pr7("b")("o")("o")("l")("e")("a")("n")(type_name);@/
pr4("p")("a")("t")("h")(type_name);@/
pr3("p")("e")("n")(type_name);@/
pr7("p")("i")("c")("t")("u")("r")("e")(type_name);@/
pr9("t")("r")("a")("n")("s")("f")("o")("r")("m")(type_name);@/
pr4("p")("a")("i")("r")(type_name);@/

@ At last we are done with the tedious initialization of primitives.

@<Store all the prim...@>=
pr3("e")("n")("d")(endit);@/
pr4("d")("u")("m")("p")(endit);@/
pr9("b")("a")("t")("c")("h")("m")("o")("d")("e")(bold);
pr11("n")("o")("n")("s")("t")("o")("p")("m")("o")("d")("e")(bold);
pr10("s")("c")("r")("o")("l")("l")("m")("o")("d")("e")(bold);
pr13("e")("r")("r")("o")("r")("s")("t")("o")("p")@|
 ("m")("o")("d")("e")(bold);
pr5("i")("n")("n")("e")("r")(command);@/
pr5("o")("u")("t")("e")("r")(command);@/
pr9("s")("h")("o")("w")("t")("o")("k")("e")("n")(command);@/
pr9("s")("h")("o")("w")("s")("t")("a")("t")("s")(bold);@/
pr4("s")("h")("o")("w")(command);@/
pr12("s")("h")("o")("w")("v")("a")("r")("i")("a")("b")("l")("e")(command);@/
pr16("s")("h")("o")("w")@|
 ("d")("e")("p")("e")("n")("d")("e")("n")("c")("i")("e")("s")(bold);@/
pr7("c")("o")("n")("t")("o")("u")("r")(command);@/
pr10("d")("o")("u")("b")("l")("e")("p")("a")("t")("h")(command);@/
pr4("a")("l")("s")("o")(command);@/
pr7("w")("i")("t")("h")("p")("e")("n")(command);@/
pr10("w")("i")("t")("h")("w")("e")("i")("g")("h")("t")(command);@/
pr8("d")("r")("o")("p")("p")("i")("n")("g")(command);@/
pr7("k")("e")("e")("p")("i")("n")("g")(command);@/
pr7("m")("e")("s")("s")("a")("g")("e")(command);@/
pr10("e")("r")("r")("m")("e")("s")("s")("a")("g")("e")(command);@/
pr7("e")("r")("r")("h")("e")("l")("p")(command);@/
pr8("c")("h")("a")("r")("l")("i")("s")("t")(command);@/
pr8("l")("i")("g")("t")("a")("b")("l")("e")(command);@/
pr10("e")("x")("t")("e")("n")("s")("i")("b")("l")("e")(command);@/
pr10("h")("e")("a")("d")("e")("r")("b")("y")("t")("e")(command);@/
pr9("f")("o")("n")("t")("d")("i")("m")("e")("n")(command);@/
pr7("s")("p")("e")("c")("i")("a")("l")(command);@/
pr10("n")("u")("m")("s")("p")("e")("c")("i")("a")("l")(command);@/
pr1("%")(comment);@/
pr2("%")("%")(verbatim);@/
pr3("%")("%")("%")(set_format);@/
pr4("%")("%")("%")("%")(mft_comment);@/
pr1("#")(sharp);@/

@ We also want to store a few other strings of characters that are
used in \.{MFT}'s translation to \TeX\ code.

@d ttr1(#)==byte_mem[byte_ptr-1]:=#; cur_tok:=name_ptr;
  incr(name_ptr); byte_start[name_ptr]:=byte_ptr
@d ttr2(#)==byte_mem[byte_ptr-2]:=#; ttr1
@d ttr3(#)==byte_mem[byte_ptr-3]:=#; ttr2
@d ttr4(#)==byte_mem[byte_ptr-4]:=#; ttr3
@d ttr5(#)==byte_mem[byte_ptr-5]:=#; ttr4
@d tr1==incr(byte_ptr); ttr1
@d tr2==byte_ptr:=byte_ptr+2; ttr2
@d tr3==byte_ptr:=byte_ptr+3; ttr3
@d tr4==byte_ptr:=byte_ptr+4; ttr4
@d tr5==byte_ptr:=byte_ptr+5; ttr5

@<Glob...@>=
@!translation:array[ASCII_code] of name_pointer;
@!i:ASCII_code; {index into |translation|}

@ @<Store all the translations@>=
for i:=0 to 255 do translation[i]:=0;
tr2("\")("$"); translation["$"]:=cur_tok;@/
tr2("\")("#"); translation["#"]:=cur_tok;@/
tr2("\")("&"); translation["&"]:=cur_tok;@/
tr2("\")("{"); translation["{"]:=cur_tok;@/
tr2("\")("}"); translation["}"]:=cur_tok;@/
tr2("\")("_"); translation["_"]:=cur_tok;@/
tr2("\")("%"); translation["%"]:=cur_tok;@/
tr4("\")("B")("S")(" "); translation["\"]:=cur_tok;@/
tr4("\")("H")("A")(" "); translation["^"]:=cur_tok;@/
tr4("\")("T")("I")(" "); translation["~"]:=cur_tok;@/
tr5("\")("a")("s")("t")(" "); translation["*"]:=cur_tok;@/
tr4("\")("A")("M")(" "); tr_amp:=cur_tok;@/
@.\\AM, etc@>
tr4("\")("B")("L")(" "); tr_skip:=cur_tok;@/
tr4("\")("S")("H")(" "); tr_sharp:=cur_tok;@/
tr4("\")("P")("S")(" "); tr_ps:=cur_tok;@/
tr4("\")("l")("e")(" "); tr_le:=cur_tok;@/
tr4("\")("g")("e")(" "); tr_ge:=cur_tok;@/
tr4("\")("n")("e")(" "); tr_ne:=cur_tok;@/
tr5("\")("q")("u")("a")("d"); tr_quad:=cur_tok;@/

@ @<Glob...@>=
@!tr_le,@!tr_ge,@!tr_ne,@!tr_amp,@!tr_sharp,@!tr_skip,@!tr_ps,
 @!tr_quad:name_pointer; {special translations}

@* Inputting the next token.
\.{MFT}'s lexical scanning routine is called |get_next|. This procedure
inputs the next token of \MF\ input and puts its encoded meaning into
two global variables, |cur_type| and |cur_tok|.

@<Glob...@>=
@!cur_type:eight_bits; {type of token just scanned}
@!cur_tok:integer; {hash table or buffer location}
@!prev_type:eight_bits; {previous value of |cur_type|}
@!prev_tok:integer; {previous value of |cur_tok|}

@ @<Set init...@>=
cur_type:=end_of_line; cur_tok:=0;

@ Two global state variables affect the behavior of |get_next|: A space
will be considered significant when |start_of_line| is |true|,
and the buffer will be considered devoid of information when |empty_buffer|
is |true|.

@<Glob...@>=
@!start_of_line:boolean; {has the current line had nothing but spaces so far?}
@!empty_buffer:boolean; {is it time to input a new line?}

@ The 256 |ASCII_code| characters are grouped into classes by means of
the |char_class| table. Individual class numbers have no semantic
or syntactic significance, expect in a few instances defined here.
There's also |max_class|, which can be used as a basis for additional
class numbers in nonstandard extensions of \MF.

@d digit_class=0 {the class number of \.{0123456789}}
@d period_class=1 {the class number of `\..'}
@d space_class=2 {the class number of spaces and nonstandard characters}
@d percent_class=3 {the class number of `\.\%'}
@d string_class=4 {the class number of `\."'}
@d right_paren_class=8 {the class number of `\.)'}
@d isolated_classes==5,6,7,8 {characters that make length-one tokens only}
@d letter_class=9 {letters and the underline character}
@d left_bracket_class=17 {`\.['}
@d right_bracket_class=18 {`\.]'}
@d invalid_class=20 {bad character in the input}
@d end_line_class=21 {end of an input line (\.{MFT} only)}
@d max_class=21 {the largest class number}

@<Glob...@>=
@!char_class:array[ASCII_code] of 0..max_class; {the class numbers}

@ If changes are made to accommodate non-ASCII character sets, they should be
essentially the same in \.{MFT} as in \MF. However, \.{MFT} has an additional
class number, the |end_line_class|, which is used only for the special
character |carriage_return| that is placed at the end of the input buffer.
@^character set dependencies@>
@^system dependencies@>

@d carriage_return=@'15 {special code placed in |buffer[limit]|}

@<Set init...@>=
for i:="0" to "9" do char_class[i]:=digit_class;
char_class["."]:=period_class;
char_class[" "]:=space_class;
char_class["%"]:=percent_class;
char_class[""""]:=string_class;@/
char_class[","]:=5;
char_class[";"]:=6;
char_class["("]:=7;
char_class[")"]:=right_paren_class;
for i:="A" to "Z" do char_class[i]:=letter_class;
for i:="a" to "z" do char_class[i]:=letter_class;
char_class["_"]:=letter_class;@/
char_class["<"]:=10;
char_class["="]:=10;
char_class[">"]:=10;
char_class[":"]:=10;
char_class["|"]:=10;@/
char_class["`"]:=11;
char_class["'"]:=11;@/
char_class["+"]:=12;
char_class["-"]:=12;@/
char_class["/"]:=13;
char_class["*"]:=13;
char_class["\"]:=13;@/
char_class["!"]:=14;
char_class["?"]:=14;@/
char_class["#"]:=15;
char_class["&"]:=15;
char_class["@@"]:=15;
char_class["$"]:=15;@/
char_class["^"]:=16;
char_class["~"]:=16;@/
char_class["["]:=left_bracket_class;
char_class["]"]:=right_bracket_class;@/
char_class["{"]:=19;
char_class["}"]:=19;@/
for i:=0 to " "-1 do char_class[i]:=invalid_class;
char_class[carriage_return]:=end_line_class;@/
for i:=127 to 255 do char_class[i]:=invalid_class;

@ And now we're ready to take the plunge into |get_next| itself.

@d switch=25 {a label in |get_next|}
@d pass_digits=85 {another}
@d pass_fraction=86 {and still another, although |goto| is considered harmful}

@p procedure get_next; {sets |cur_type| and |cur_tok| to next token}
label switch,pass_digits,pass_fraction,done,found,exit;
var @!c:ASCII_code; {the current character in the buffer}
@!class:ASCII_code; {its class number}
begin prev_type:=cur_type; prev_tok:=cur_tok;
if empty_buffer then
  @<Bring in a new line of input; |return| if the file has ended@>;
switch: c:=buffer[loc]; id_first:=loc; incr(loc); class:=char_class[c];
@<Branch on the |class|, scan the token; |return| directly if the
  token is special, or |goto found| if it needs to be looked up@>;
found:id_loc:=loc; cur_tok:=lookup; cur_type:=ilk[cur_tok];
exit:end;

@ @d emit(#)==@t@>@+begin cur_type:=#; cur_tok:=id_first; return;@+end

@<Branch on the |class|...@>=
case class of
digit_class:goto pass_digits;
period_class:begin class:=char_class[buffer[loc]];
  if class>period_class then goto switch {ignore isolated `\..'}
  else if class<period_class then goto pass_fraction; {|class=digit_class|}
  end;
space_class:if start_of_line then emit(indentation)
  else goto switch;
end_line_class: emit(end_of_line);
string_class:@<Get a string token and |return|@>;
isolated_classes: goto found;
invalid_class:@<Decry the invalid character and |goto switch|@>;
othercases do_nothing {letters, etc.}
endcases;@/
while char_class[buffer[loc]]=class do incr(loc);
goto found;
pass_digits: while char_class[buffer[loc]]=digit_class do incr(loc);
if buffer[loc]<>"." then goto done;
if char_class[buffer[loc+1]]<>digit_class then goto done;
incr(loc);
pass_fraction:repeat incr(loc);
until char_class[buffer[loc]]<>digit_class;
done:emit(numeric_token)

@ @<Get a string token and |return|@>=
loop@+begin if buffer[loc]="""" then
    begin incr(loc); emit(string_token);
    end;
  if loc=limit then @<Decry the missing string delimiter and |goto switch|@>;
  incr(loc);
  end

@ @<Decry the missing string delimiter and |goto switch|@>=
begin err_print('! Incomplete string will be ignored'); goto switch;
@.Incomplete string...@>
end

@ @<Decry the invalid character and |goto switch|@>=
begin err_print('! Invalid character will be ignored'); goto switch;
@.Invalid character...@>
end

@ @<Bring in a new line of input; |return| if the file has ended@>=
begin get_line;
if input_has_ended then emit(end_of_file);
buffer[limit]:=carriage_return; loc:=0; start_of_line:=true;
empty_buffer:=false;
end

@* Low-level output routines.
The \TeX\ output is supposed to appear in lines at most |line_length|
characters long, so we place it into an output buffer. During the output
process, |out_line| will hold the current line number of the line about to
be output.

@<Glo...@>=
@!out_buf:array[0..line_length] of ASCII_code; {assembled characters}
@!out_ptr:0..line_length; {number of characters in |out_buf|}
@!out_line: integer; {coordinates of next line to be output}

@ The |flush_buffer| routine empties the buffer up to a given breakpoint,
and moves any remaining characters to the beginning of the next line.
If the |per_cent| parameter is |true|, a |"%"| is appended to the line
that is being output; in this case the breakpoint |b| should be strictly
less than |line_length|. If the |per_cent| parameter is |false|,
trailing blanks are suppressed.
The characters emptied from the buffer form a new line of output.

@p procedure flush_buffer(@!b:eight_bits;@!per_cent:boolean);
  {outputs |out_buf[1..b]|, where |b<=out_ptr|}
label done;
var j,@!k:0..line_length;
begin j:=b;
if not per_cent then {remove trailing blanks}
  loop@+  begin if j=0 then goto done;
    if out_buf[j]<>" " then goto done;
    decr(j);
    end;
done: for k:=1 to j do write(tex_file,xchr[out_buf[k]]);
if per_cent then write(tex_file,xchr["%"]);
write_ln(tex_file); incr(out_line);
if b<out_ptr then for k:=b+1 to out_ptr do out_buf[k-b]:=out_buf[k];
out_ptr:=out_ptr-b;
end;

@ \.{MFT} calls |flush_buffer(out_ptr,false)| before it has input
anything. We initialize the output variables
so that the first line of the output file will be `\.{\\input mftmac}'.
@.\\input mftmac@>
@.mftmac@>

@<Set init...@>=
out_ptr:=1; out_buf[1]:=" "; out_line:=1; write(tex_file,'\input mftmac');

@ When we wish to append the character |c| to the output buffer, we write
`$|out|(c)$'; this will cause the buffer to be emptied if it was already
full. Similarly, `$|out2|(c_1)(c_2)$' appends a pair of characters.
A line break will occur at a space or after a single-nonletter
\TeX\ control sequence.

@d oot(#)==@;@/
  if out_ptr=line_length then break_out;
  incr(out_ptr); out_buf[out_ptr]:=#;
@d oot1(#)==oot(#)@+end
@d oot2(#)==oot(#)@,oot1
@d oot3(#)==oot(#)@,oot2
@d oot4(#)==oot(#)@,oot3
@d oot5(#)==oot(#)@,oot4
@d out==@+begin oot1
@d out2==@+begin oot2
@d out3==@+begin oot3
@d out4==@+begin oot4
@d out5==@+begin oot5

@ The |break_out| routine is called just before the output buffer is about
to overflow. To make this routine a little faster, we initialize position
0 of the output buffer to `\.\\'; this character isn't really output.

@<Set init...@>=
out_buf[0]:="\";

@ A long line is broken at a blank space or just before a backslash that isn't
preceded by another backslash. In the latter case, a |"%"| is output at
the break. (This policy has a known bug, in the rare situation that the
backslash was in a string constant that's being output ``verbatim.'')

@p procedure break_out; {finds a way to break the output line}
label exit;
var k:0..line_length; {index into |out_buf|}
@!d:ASCII_code; {character from the buffer}
begin k:=out_ptr;
loop@+  begin if k=0 then
    @<Print warning message, break the line, |return|@>;
  d:=out_buf[k];
  if d=" " then
    begin flush_buffer(k,false); return;
    end;
  if (d="\")and(out_buf[k-1]<>"\") then {in this case |k>1|}
    begin flush_buffer(k-1,true); return;
    end;
  decr(k);
  end;
exit:end;

@ We get to this module only in unusual cases that the entire output line
consists of a string of backslashes followed by a string of nonblank
non-backslashes. In such cases it is almost always safe to break the
line by putting a |"%"| just before the last character.

@<Print warning message...@>=
begin print_nl('! Line had to be broken (output l.',out_line:1);
@.Line had to be broken@>
print_ln('):');
for k:=1 to out_ptr-1 do print(xchr[out_buf[k]]);
new_line; mark_harmless;
flush_buffer(out_ptr-1,true); return;
end

@ To output a string of bytes from |byte_mem|, we call |out_str|.

@p procedure out_str(@!p:name_pointer); {outputs a string}
var @!k:0..max_bytes; {index into |byte_mem|}
begin for k:=byte_start[p] to byte_start[p+1]-1 do out(byte_mem[k]);
end;

@ The |out_name| subroutine is used to output a symbolic token.
Unusual characters are translated into forms that won't screw up.

@p procedure out_name(@!p:name_pointer); {outputs a name}
var @!k:0..max_bytes; {index into |byte_mem|}
@!t:name_pointer; {translation of character being output, if any}
begin for k:=byte_start[p] to byte_start[p+1]-1 do
  begin t:=translation[byte_mem[k]];
        if t=0 then out(byte_mem[k])
  else out_str(t);
  end;
end;

@ We often want to output a name after calling a numeric macro
(e.g., `\.{\\1\{foo\}}').

@p procedure out_mac_and_name(@!n:ASCII_code; @!p:name_pointer);
begin out("\"); out(n);
if length(p)=1 then out_name(p)
else  begin out("{"); out_name(p); out("}");
  end;
end;

@ Here's a routine that simply copies from the input buffer to the output
buffer.

@p procedure copy(@!first_loc:integer); {output |buffer[first_loc..loc-1]|}
var @!k:0..buf_size; {|buffer| location being copied}
begin for k:=first_loc to loc-1 do out(buffer[k]);
end;

@* Translation.
The main work of \.{MFT} is accomplished by a routine that translates
the tokens, one by one, with a limited amount of lookahead/lookbehind.
Automata theorists might loosely call this a ``finite state transducer,''
because the flow of control is comparatively simple.

@p procedure do_the_translation;
label restart,reswitch,done,exit;
var @!k:0..buf_size; {looks ahead in the buffer}
@!t:integer; {type that spreads to new tokens}
begin restart:if out_ptr>0 then flush_buffer(out_ptr,false);
empty_buffer:=true;
loop@+  begin get_next;
  if start_of_line then @<Do special actions at the start of a line@>;
  reswitch:case cur_type of
  numeric_token:@<Translate a numeric token or a fraction@>;
  string_token:@<Translate a string token@>;
  indentation:out_str(tr_quad);
  end_of_line,mft_comment:@<Wind up a line of translation and |goto restart|,
    or finish a \pb\ segment and |goto reswitch|@>;
  end_of_file:return;
@t\4@>  @<Cases that translate primitive tokens@>@;
  comment,recomment:@<Translate a comment and |goto restart|,
    unless there's a \pb\ segment@>;
  verbatim:@<Copy the rest of the current input line to the output,
    then |goto restart|@>;
  set_format:@<Change the translation format of tokens,
    and |goto restart| or |reswitch|@>;
  internal,special_tag,tag:@<Translate a tag and possible subscript@>;
  end;  {all cases have been listed}
  end;
exit:end;

@ @<Do special actions at the start of a line@>=
if cur_type>=min_action_type then
  begin out("$"); start_of_line:=false;
  case cur_type of
  endit:out2("\")("!");
@.\\!@>
  binary,abinary,bbinary,ampersand,pyth_sub:out2("{")("}");
@.\{\}@>
  othercases do_nothing
  endcases;
  end
else if cur_type=end_of_line then
  begin out_str(tr_skip); goto restart;
  end
else if cur_type=mft_comment then goto restart

@ Let's start with some of the easier translations, so that the harder
ones will also be easy when we get to them. A string like |"cat"|
comes out `\.{\\7"cat"}'.

@<Translate a string token@>=
begin out2("\")("7"); copy(cur_tok);
@.\\7@>
end

@ Similarly, the translation of `\.{sqrt}' is `\.{\\1\{sqrt\}}'.

@<Cases that translate primitive tokens@>=
op: out_mac_and_name("1",cur_tok);
@.\\1@>
command: out_mac_and_name("2",cur_tok);
@.\\2@>
type_name: if prev_type=command then out_mac_and_name("1",cur_tok)
  else out_mac_and_name("2",cur_tok);
endit: out_mac_and_name("3",cur_tok);
@.\\3@>
bbinary: out_mac_and_name("4",cur_tok);
@.\\4@>
bold: out_mac_and_name("5",cur_tok);
@.\\5@>
binary: out_mac_and_name("6",cur_tok);
@.\\6@>
path_join: out_mac_and_name("8",cur_tok);
@.\\8@>
colon: out_mac_and_name("?",cur_tok);
@.\\?@>

@ Here are a few more easy cases.

@<Cases that translate primitive tokens@>=
as_is,sharp,abinary: out_name(cur_tok);
double_back: out2("\")(";");
@.\\;@>
semicolon: begin out_name(cur_tok); get_next;
  if cur_type<>end_of_line then if cur_type<>endit then out2("\")(" ");
@.\\\char32@>
  goto reswitch;
  end;

@ Some of the primitives have a fixed output (independent of |cur_tok|):

@<Cases that translate primitive tokens@>=
backslash:out_str(translation["\"]);
pyth_sub:out_str(tr_ps);
less_or_equal:out_str(tr_le);
greater_or_equal:out_str(tr_ge);
not_equal:out_str(tr_ne);
ampersand:out_str(tr_amp);

@ The remaining primitive is slightly special.

@<Cases that translate primitive tokens@>=
input_command: begin out_mac_and_name("2",cur_tok);
  out5("\")("h")("b")("o")("x");
  @<Scan the file name and output it in \.{typewriter type}@>;
  end;

@ File names have different formats on different computers, so we don't scan
them with |get_next|. Here we use
a rule that probably covers most cases satisfactorily: We ignore leading
blanks, then consider the file name to consist of all subsequent characters
up to the first blank, semicolon, comment, or end-of-line.
(A |carriage_return| appears at the end of the line.)

@<Scan the file name and output it in \.{typewriter type}@>=
while buffer[loc]=" " do incr(loc);
out5("{")("\")("t")("t")(" ");
while (buffer[loc]<>" ")and(buffer[loc]<>"%")and(buffer[loc]<>";")
  and(loc<limit) do
  begin out(buffer[loc]); incr(loc);
  end;
out("}")

@ @<Translate a numeric token or a fraction@>=
if buffer[loc]="/" then
  if char_class[buffer[loc+1]]=digit_class then {it's a fraction}
    begin out5("\")("f")("r")("a")("c"); copy(cur_tok); get_next;
@.\\frac@>
    out2("/")("{"); get_next; copy(cur_tok); out("}");
    end
  else copy(cur_tok)
else copy(cur_tok)

@ @<Translate a tag and possible subscript@>=
begin if length(cur_tok)=1 then out_name(cur_tok)
else out_mac_and_name("\",cur_tok);
@.\\\\@>
get_next;
if byte_mem[byte_start[prev_tok]]="'" then goto reswitch;
case prev_type of
internal:begin if (cur_type=numeric_token)or(cur_type>=min_suffix) then
    out2("\")(",");
@.\\,@>
  goto reswitch;
  end;
special_tag:if cur_type<min_suffix then goto reswitch
  else  begin out("."); cur_type:=internal; goto reswitch;
@..@>
    end;
tag:begin if cur_type=tag then if byte_mem[byte_start[cur_tok]]="'" then
    goto reswitch; {a sequence of primes goes on the main line}
  if (cur_type=numeric_token)or(cur_type>=min_suffix) then
    @<Translate a subscript@>
  else if cur_type=sharp then out_str(tr_sharp)
  else goto reswitch;
  end;
end; {there are no other cases}
end

@ @<Translate a subscript@>=
begin out2("_")("{");
loop@+  begin if cur_type>=min_suffix then out_name(cur_tok)
  else copy(cur_tok);
  if prev_type=special_tag then
    begin get_next; goto done;
    end;
  get_next;
  if cur_type<min_suffix then if cur_type<>numeric_token then goto done;
  if cur_type=prev_type then
    if cur_type=numeric_token then out2("\")(",")
@.\\,@>
    else if char_class[byte_mem[byte_start[cur_tok]]]=@|
     char_class[byte_mem[byte_start[prev_tok]]] then
      if byte_mem[byte_start[prev_tok]]<>"." then out(".")
      else out2("\")(",");
  end;
done: out("}"); goto reswitch;
end

@ The tricky thing about comments is that they might contain \pb.
We scan ahead for this, and replace the second `\.{\char'174}'
by a |carriage_return|.

@<Translate a comment and |goto restart|...@>=
begin if cur_type=comment then out2("\")("9");
@.\\9@>
id_first:=loc;
while (loc<limit)and(buffer[loc]<>"|") do incr(loc);
copy(id_first);
if loc<limit then
  begin start_of_line:=true; incr(loc); k:=loc;
  while (k<limit)and(buffer[k]<>"|") do incr(k);
  buffer[k]:=carriage_return;
  end
else  begin if out_buf[out_ptr]="\" then out(" ");
  out4("\")("p")("a")("r"); goto restart;
@.\\par@>
  end;
end

@ @<Copy the rest of the current input line to the output...@>=
begin id_first:=loc; loc:=limit; copy(id_first);
if out_ptr=0 then
  begin out_ptr:=1; out_buf[1]:=" ";
  end;
goto restart;
end

@ @<Wind up a line of translation...@>=
begin out("$");
if (loc<limit)and(cur_type=end_of_line) then
  begin cur_type:=recomment; goto reswitch;
  end
else  begin out4("\")("p")("a")("r"); goto restart;
@.\\par@>
  end;
end

@ @<Change the translation format...@>=
begin start_of_line:=false; get_next; t:=cur_type;
while cur_type>=min_symbolic_token do
  begin get_next;
  if cur_type>=min_symbolic_token then ilk[cur_tok]:=t;
  end;
if cur_type<>end_of_line then if cur_type<>mft_comment then
  begin err_print('! Only symbolic tokens should appear after %%%');
@.Only symbolic tokens...@>
  goto reswitch;
  end;
empty_buffer:=true; goto restart;
end

@* The main program.
Let's put it all together now: \.{MFT} starts and ends here.
@^system dependencies@>

@p begin initialize; {beginning of the main program}
print_ln(banner); {print a ``banner line''}
@<Store all the primitives@>;
@<Store all the translations@>;
@<Initialize the input...@>;
do_the_translation;
@<Check that all changes have been read@>;
end_of_MFT:{here files should be closed if the operating system requires it}
@<Print the job |history|@>;
end.

@ Some implementations may wish to pass the |history| value to the
operating system so that it can be used to govern whether or not other
programs are started. Here we simply report the history to the user.
@^system dependencies@>

@<Print the job |history|@>=
case history of
spotless: print_nl('(No errors were found.)');
harmless_message: print_nl('(Did you see the warning message above?)');
error_message: print_nl('(Pardon me, but I think I spotted something wrong.)');
fatal_message: print_nl('(That was a fatal error, my friend.)');
end {there are no other cases}

@* System-dependent changes.
This module should be replaced, if necessary, by changes to the program
that are necessary to make \.{MFT} work at a particular installation.
It is usually best to design your change file so that all changes to
previous modules preserve the module numbering; then everybody's version
will be consistent with the printed program. More extensive changes,
which introduce new modules, can be inserted here; then only the index
itself will get a new module number.
@^system dependencies@>

@* Index.

Bell Labs OSI certified Powered by Plan 9

(Return to Plan 9 Home Page)

Copyright © 2021 Plan 9 Foundation. All Rights Reserved.
Comments to webmaster@9p.io.