Debian Codegraph @

texmf-dist/source/fonts/universa/universa.dtx

raw
1 % \iffalse
2 %<*(package|oml|oms|ot1|t1|u)>
3 \def\unifileversion{v2.1}
4 \def\unifiledate{2019/08/26}
5 %</(package|oml|oms|ot1|t1|u)>
6 % \fi
7 % \iffalse
8 %<*driver>
9 \documentclass{ltxdoc}
10 \usepackage{unidoc,latexsym}
11 \OnlyDescription %For short description of font and package.
12 % \AlsoImplementation %For full description of font and package.
13 \title{The \universal{} Font\\ \normalsize{Version \unifileversion}}
14 \author{Christian Holm\thanks{E--mail: \texttt{<[email protected]>}}}
15 \date{\unifiledate}
16 \begin{document}
17 \maketitle
18 \begin{abstract}
19 This is my implementation of Herbert Bayer's ``universal'' font in
20 \MF{} for \TeX{} and \LaTeX{}. Extensive support for \LaTeX{} is
21 supplied. This font is in no way intended to be a correct, not to
22 mention a comlpete implementation of Herbert Bayer's original
23 design. This document describes how to use the font with \LaTeX{},
24 and also the source code for the characters of the font.
25 \end{abstract}
26 \tableofcontents
27 \DocInput{universa.dtx}
28 \appendix
29
30 \section{Solution to the \Lcs{bauhausforms} problem}
31 \label{app:solu}
32
33 \subsection{The Problem}
34 This I owe to Joseph Collins \texttt{<[email protected]>}.
35
36 During the preperation of version~1.0 of this font, I was loosing
37 sleep over what I chose to call `The \Lcs{bauhausforms}
38 problem', after the problems the symbol \textuni{\bauforms} gave
39 me.
40
41 As can be seen from the symbol, the idea is to make a figure out of
42 a circle, a square, and a triangle. From these three figures you can
43 ofcourse make infinitly many figures, even though the sizes are
44 limited. The particular combination of the three figures I was
45 looking, first seemed simple\footnote{And after having seen Mr.\
46 Collins solution, it did again.}, but turned out to be
47 difficult. What I wanted to do was (see also
48 figure~\ref{fig:bauforms}):
49 \begin{quote}
50 Take a circle of some radius (e.g., 1). Inside this circle draw a
51 isosceles triangle $\bigtriangleup ABC$, with all vertices on the
52 circle, The singular vertice $A$ placed on the horizontal line
53 traveling left from the center of the circle.
54
55 Now draw a square $\Box PQRS$ inside of the circle, having two
56 vertices on the circle, and two on the triangle.
57
58 The wanted triangle and square are such that, the opposing side of
59 the singular vertice $|BC|$ should divede the square into to equal
60 oblongs, i.e., intersect $|PQ|$ midway between $P$ and $Q$.
61 \end{quote}
62
63 \begin{figure}[htbp]
64 \centering
65 \beginillu
66 \setlength{\unitlength}{.05\textwidth}
67 \begin{picture}(10,10)
68 \thinlines
69 \put(-1,5){\line(1,0){12}}
70 \put(5,-1){\line(0,1){12}}
71 \thicklines
72 \put(5,5){\circle{10}}
73 \put(0,5){\line(100,74){6.521}}
74 \put(0,5){\line(100,-74){6.521}}
75 \put(6.4330,0.2097){\line(0,1){9.5805}}
76 \put(3.6873,7.7457){\line(1,0){5.4914}}
77 \put(3.6873,2.2543){\line(1,0){5.4914}}
78 \put(3.6873,2.2543){\line(0,1){5.4914}}
79 \put(9.1787,2.2543){\line(0,1){5.4914}}
80 \put(0,5){\mbox{$A$}}
81 \put(6.4330,0.2097){\mbox{$B$}}
82 \put(6.4330,5){\mbox{$B^\prime$}}
83 \put(6.4330,9.7903){\mbox{$C$}}
84 \put(3.6873,2.2543){\mbox{$P$}}
85 \put(3.6873,5){\mbox{$P^\prime$}}
86 \put(9.1787,2.2543){\mbox{$Q$}}
87 \put(9.1787,5){\mbox{$Q^\prime$}}
88 \put(9.1787,7.7457){\mbox{$R$}}
89 \put(3.6873,7.7457){\mbox{$S$}}
90 \put(5,5){\mbox{$O$}}
91 \put(0,5){\arc{1}{0}{0.6401}}
92 \put(0,4.5){\mbox{$\varphi$}}
93 \end{picture}
94 \illuother
95 \begin{tabular}{%
96 @{\vline\extracolsep{1em}\space}%
97 p{.8\textwidth}%
98 @{\space\vline}}
99 \hline
100 Since you are processing this document without the
101 \textsf{eepic} package, you will not see the illustration
102 otherwise here. To compensate, I will give you a short
103 description on how to draw the figure yourself.
104
105 Take out some paper, a pencil, compasses, and a ruler. Now
106 draw a circle of some radius $r$, with its center in origo of
107 a Cartisean coordinate system. In the next subsection, you
108 will find the coordinates for points $A,B,C,O,P,Q,R,$ and $S$
109 expressed in terms of $x,y,$ and $h$. You will also find the
110 correct values for $x,y,h$ in that section. Multiply those
111 values with $r$, and plot the points above in the coordinate
112 system. Draw the triangle $\bigtriangleup ABC$, and the square
113 $\Box PQRS$, and --- voila, you have the illustration.\\
114 \hline
115 \end{tabular}
116 \endillu
117 \caption{The \Lcs{bauforms} problem.}\label{fig:bauforms}
118 \end{figure}
119
120 This shouldn't be to difficult, should it. Well I didn't think so,
121 but after many late evenings with pen, paper, ruler, compasses, and
122 heavy use of trigonometric relations, I found out that the half
123 $\varphi$ of the singular vertice should obey:
124 \begin{eqnarray}
125 0 &=& \sin \left( \cos^{-1} \left( \frac{\cos2\varphi}{\sqrt{2}}
126 \right) \right) - \frac{\cos2\varphi}{\sqrt{2}}\nonumber\\
127 & & - 2\sqrt{2}
128 \frac{\cos\varphi^2\sin\varphi}{\cos\varphi+\sin\varphi}
129 \nonumber
130 \end{eqnarray}
131 Now I dare you to find the exact solution to that.
132
133 Using numerical methodes (Newton's method), was ofcourse no
134 problem, and gave satesfactory resualt. If the expression on the
135 right above is labelled $f$, $f^\prime$ is:
136 \begin{eqnarray}
137 f^\prime &=& \frac{\sin4\varphi}{2\sqrt{\frac{3-\cos4\varphi}{4}}}
138 + \sqrt{2}\sin2\varphi\nonumber\\
139 & & - \frac{2\sqrt{2}\cos\varphi}{\cos\varphi+\sin\varphi^2}
140 \left( \left( \cos\varphi+\sin\varphi \right)
141 \left( \cos^2\varphi-2\sin^22\varphi \right)\right. \nonumber\\
142 & & - \quad \left.\left( \cos\varphi-\sin\varphi \right)
143 \cos^2\varphi\sin\varphi \right) \nonumber
144 \end{eqnarray}
145 Using these expressions for $f$ and $f^\prime$ in a \textsf{Fortran}
146 program, I reached resaults close to what Mr.\ Collins found.
147
148 \subsection{Joseph Collins' Solution}
149
150 In Mr.\ Collins notation, the points on figure~\ref{fig:bauforms}
151 has the following coordinates:
152 \begin{eqnarray*}
153 A = (0,-1) & B^\prime = (x,0) & C = (x,y)\\
154 O = (0,0) & P^\prime = (x-h,0) & Q^\prime = (x+h,0)\\
155 R = (x+h,h) & S = (x-h,h) & \\
156 \end{eqnarray*}
157 Below is what Mr.\ Collins wrote me --- thank you very much.
158
159 On the unit circle
160 \begin{equation}\label{eq:uc}
161 x^2 + y^2 = 1
162 \end{equation}
163 we have the vertices of a triangle at $(-1,0)$, $(x,y)$, and
164 $(x,-y)$. A square has four vertices $(x \pm h, \pm h)$, where the
165 two points $(x-h, \pm h)$ lie on the triangle (constraint A) and the
166 two points $(x+h, \pm h)$ lie on the circle (constraint B). Thus,
167 the vertical side of the triangle bisects the square. From
168 constraint A, upon consideration of similar triangles, we have
169 $$\frac{y}{1+x} = \frac{h}{1+x-h}, \quad \mbox{so that}\quad
170 h = \frac{(1+x) y }{1+x+y}.$$
171 By (\ref{eq:uc}), this is
172 \begin{equation} \label{eq:b1}
173 h = \frac{(1+x)\sqrt{1-x^2}}{1+x+\sqrt{1-x^2}} .
174 \end{equation}
175 From constraint B, we get
176 \begin{equation} \label{eq:b2}
177 (x+h)^2 + h^2 = 1 .
178 \end{equation}
179 Any simultaneous solution of (\ref{eq:b1}) and (\ref{eq:b2}) is also a
180 solution of
181 \begin{equation}\label{eq:eqn}
182 8x^3 - 4x^2 - 3x + 1 = 0,
183 \end{equation}
184 the relevant solution being
185 \begin{equation} \label{eq:sol}
186 x = \frac16 + \sqrt{\frac{11}{18}} \sin
187 \left[ \frac{\pi}{6}
188 - \frac13 \arctan\left(\frac{3\sqrt{237}}{23}\right) \right].
189 \end{equation}
190 Equation (\ref{eq:eqn}) and solution (\ref{eq:sol}) courtesy of
191 Mathematica. We get $y$ and $h$ from (\ref{eq:uc}) and
192 (\ref{eq:b1}), respectively. The angle at $(-1,0)$ is $\varphi = 2
193 \arctan\bigl(y/(1+x)\bigr)$.
194
195 Approximate values are
196 \begin{eqnarray*}
197 x &\simeq& 0.2865914\\
198 y &\simeq& 0.9580529\\
199 h &\simeq& 0.5491394 \\
200 \varphi &\simeq& 1.280129 \ ( \simeq 73.346^o)
201 \end{eqnarray*}
202
203 \section{Copyrights --- GNU General Public Lisence}
204 \label{app:ggpl}
205
206 This \MF{} implementation of the ``universal'' font and the \LaTeX{}
207 support package ``uni'' copyright \copyright{} 1998 Christian Holm.
208
209 This program is free software; you can redistribute it and/or modify
210 it under the terms of the GNU General Public License as published by
211 the Free Software Foundation; either version 2 of the License, or
212 (at your option) any later version.
213
214 This program is distributed in the hope that it will be useful, but
215 \textsc{without any warranty}; without even the implied warranty
216 of \textsc{merchantability} or \textsc{fitness for a particular
217 purpose}. See the GNU General Public License for more details.
218
219 You should have received a copy of the GNU General Public License
220 along with this program; if not, write to
221 \begin{quote}
222 The Free Software Foundation, Inc.\\
223 675 Mass Ave\\
224 Cambridge\\
225 MA 02139\\
226 USA
227 \end{quote}
228
229 See the file \texttt{copyright} in the distribution for the complete
230 GNU General Public License.
231
232 \section{Wishlist}
233
234 Below is a list of things I would like to do with the font and
235 package. If anyone has any suggestions, ready--made code, or new
236 ideas, please let me know.
237
238 If you would like to take on one or more of the tasks presented
239 below, please do so, but send me a note so that I may coordinate
240 with my own efforts, and perhaps have a constructive dicourse.
241
242 I should however instruct you to read the \emph{complete}
243 documentation of the package and font, since this may give some
244 reasons why I have chosen a particular approach.
245
246 \begin{itemize}
247 \item Make the font an 8--bit encoded (256 characters) font,
248 conforming somewhat to the \texttt{T1} encoding of the Cork fonts.
249 \item Making a package (perhaps \textsf{mfdoc}), to make
250 documentation of \MF{} sources, just like \textsf{doc} is fore
251 \LaTeX{} sources. This is ofcourse a independent project, and I
252 will proberly not work on it.
253 \end{itemize}
254 \clearpage
255 \section{Font Charts}
256 \label{app:charts}
257
258 Below are some charts of the \universal{} font in different series
259 and shapes (medium upright, slanted, small-caps, and strict, as well
260 as bold upright, slanted, small caps, and strict), all in size \texttt{8pt}.
261
262 \FontChart{fulmr8.mf}{medium, upright}
263 \FontChart{fulmo8.mf}{medium, oblique}
264 \FontChart{fulmc8.mf}{medium, small-caps}
265 \FontChart{fulmst8.mf}{medium, strict}
266 \FontChart{fulbr8.mf}{bold, upright}
267 \FontChart{fulbo8.mf}{bold, oblique}
268 \FontChart{fulbc8.mf}{bold, small-caps}
269 \FontChart{fulbst8.mf}{bold, strict}
270
271 \clearpage
272 \end{document}
273 %</driver>
274 %<*copyright>
275 GNU GENERAL PUBLIC LICENSE
276 Version 2, June 1991
277
278 Copyright (C) 1989, 1991 Free Software Foundation, Inc.
279 675 Mass Ave, Cambridge, MA 02139, USA
280 Everyone is permitted to copy and distribute verbatim copies
281 of this license document, but changing it is not allowed.
282
283 Preamble
284
285 The licenses for most software are designed to take away your
286 freedom to share and change it. By contrast, the GNU General Public
287 License is intended to guarantee your freedom to share and change free
288 software--to make sure the software is free for all its users. This
289 General Public License applies to most of the Free Software
290 Foundation's software and to any other program whose authors commit to
291 using it. (Some other Free Software Foundation software is covered by
292 the GNU Library General Public License instead.) You can apply it to
293 your programs, too.
294
295 When we speak of free software, we are referring to freedom, not
296 price. Our General Public Licenses are designed to make sure that you
297 have the freedom to distribute copies of free software (and charge for
298 this service if you wish), that you receive source code or can get it
299 if you want it, that you can change the software or use pieces of it
300 in new free programs; and that you know you can do these things.
301
302 To protect your rights, we need to make restrictions that forbid
303 anyone to deny you these rights or to ask you to surrender the rights.
304 These restrictions translate to certain responsibilities for you if you
305 distribute copies of the software, or if you modify it.
306
307 For example, if you distribute copies of such a program, whether
308 gratis or for a fee, you must give the recipients all the rights that
309 you have. You must make sure that they, too, receive or can get the
310 source code. And you must show them these terms so they know their
311 rights.
312
313 We protect your rights with two steps: (1) copyright the software, and
314 (2) offer you this license which gives you legal permission to copy,
315 distribute and/or modify the software.
316
317 Also, for each author's protection and ours, we want to make certain
318 that everyone understands that there is no warranty for this free
319 software. If the software is modified by someone else and passed on, we
320 want its recipients to know that what they have is not the original, so
321 that any problems introduced by others will not reflect on the original
322 authors' reputations.
323
324 Finally, any free program is threatened constantly by software
325 patents. We wish to avoid the danger that redistributors of a free
326 program will individually obtain patent licenses, in effect making the
327 program proprietary. To prevent this, we have made it clear that any
328 patent must be licensed for everyone's free use or not licensed at all.
329
330 The precise terms and conditions for copying, distribution and
331 modification follow.
332
333 GNU GENERAL PUBLIC LICENSE
334 TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION
335
336 0. This License applies to any program or other work which contains
337 a notice placed by the copyright holder saying it may be distributed
338 under the terms of this General Public License. The "Program", below,
339 refers to any such program or work, and a "work based on the Program"
340 means either the Program or any derivative work under copyright law:
341 that is to say, a work containing the Program or a portion of it,
342 either verbatim or with modifications and/or translated into another
343 language. (Hereinafter, translation is included without limitation in
344 the term "modification".) Each licensee is addressed as "you".
345
346 Activities other than copying, distribution and modification are not
347 covered by this License; they are outside its scope. The act of
348 running the Program is not restricted, and the output from the Program
349 is covered only if its contents constitute a work based on the
350 Program (independent of having been made by running the Program).
351 Whether that is true depends on what the Program does.
352
353 1. You may copy and distribute verbatim copies of the Program's
354 source code as you receive it, in any medium, provided that you
355 conspicuously and appropriately publish on each copy an appropriate
356 copyright notice and disclaimer of warranty; keep intact all the
357 notices that refer to this License and to the absence of any warranty;
358 and give any other recipients of the Program a copy of this License
359 along with the Program.
360
361 You may charge a fee for the physical act of transferring a copy, and
362 you may at your option offer warranty protection in exchange for a fee.
363
364 2. You may modify your copy or copies of the Program or any portion
365 of it, thus forming a work based on the Program, and copy and
366 distribute such modifications or work under the terms of Section 1
367 above, provided that you also meet all of these conditions:
368
369 a) You must cause the modified files to carry prominent notices
370 stating that you changed the files and the date of any change.
371
372 b) You must cause any work that you distribute or publish, that in
373 whole or in part contains or is derived from the Program or any
374 part thereof, to be licensed as a whole at no charge to all third
375 parties under the terms of this License.
376
377 c) If the modified program normally reads commands interactively
378 when run, you must cause it, when started running for such
379 interactive use in the most ordinary way, to print or display an
380 announcement including an appropriate copyright notice and a
381 notice that there is no warranty (or else, saying that you provide
382 a warranty) and that users may redistribute the program under
383 these conditions, and telling the user how to view a copy of this
384 License. (Exception: if the Program itself is interactive but
385 does not normally print such an announcement, your work based on
386 the Program is not required to print an announcement.)
387
388 These requirements apply to the modified work as a whole. If
389 identifiable sections of that work are not derived from the Program,
390 and can be reasonably considered independent and separate works in
391 themselves, then this License, and its terms, do not apply to those
392 sections when you distribute them as separate works. But when you
393 distribute the same sections as part of a whole which is a work based
394 on the Program, the distribution of the whole must be on the terms of
395 this License, whose permissions for other licensees extend to the
396 entire whole, and thus to each and every part regardless of who wrote it.
397
398 Thus, it is not the intent of this section to claim rights or contest
399 your rights to work written entirely by you; rather, the intent is to
400 exercise the right to control the distribution of derivative or
401 collective works based on the Program.
402
403 In addition, mere aggregation of another work not based on the Program
404 with the Program (or with a work based on the Program) on a volume of
405 a storage or distribution medium does not bring the other work under
406 the scope of this License.
407
408 3. You may copy and distribute the Program (or a work based on it,
409 under Section 2) in object code or executable form under the terms of
410 Sections 1 and 2 above provided that you also do one of the following:
411
412 a) Accompany it with the complete corresponding machine-readable
413 source code, which must be distributed under the terms of Sections
414 1 and 2 above on a medium customarily used for software interchange; or,
415
416 b) Accompany it with a written offer, valid for at least three
417 years, to give any third party, for a charge no more than your
418 cost of physically performing source distribution, a complete
419 machine-readable copy of the corresponding source code, to be
420 distributed under the terms of Sections 1 and 2 above on a medium
421 customarily used for software interchange; or,
422
423 c) Accompany it with the information you received as to the offer
424 to distribute corresponding source code. (This alternative is
425 allowed only for noncommercial distribution and only if you
426 received the program in object code or executable form with such
427 an offer, in accord with Subsection b above.)
428
429 The source code for a work means the preferred form of the work for
430 making modifications to it. For an executable work, complete source
431 code means all the source code for all modules it contains, plus any
432 associated interface definition files, plus the scripts used to
433 control compilation and installation of the executable. However, as a
434 special exception, the source code distributed need not include
435 anything that is normally distributed (in either source or binary
436 form) with the major components (compiler, kernel, and so on) of the
437 operating system on which the executable runs, unless that component
438 itself accompanies the executable.
439
440 If distribution of executable or object code is made by offering
441 access to copy from a designated place, then offering equivalent
442 access to copy the source code from the same place counts as
443 distribution of the source code, even though third parties are not
444 compelled to copy the source along with the object code.
445
446 4. You may not copy, modify, sublicense, or distribute the Program
447 except as expressly provided under this License. Any attempt
448 otherwise to copy, modify, sublicense or distribute the Program is
449 void, and will automatically terminate your rights under this License.
450 However, parties who have received copies, or rights, from you under
451 this License will not have their licenses terminated so long as such
452 parties remain in full compliance.
453
454 5. You are not required to accept this License, since you have not
455 signed it. However, nothing else grants you permission to modify or
456 distribute the Program or its derivative works. These actions are
457 prohibited by law if you do not accept this License. Therefore, by
458 modifying or distributing the Program (or any work based on the
459 Program), you indicate your acceptance of this License to do so, and
460 all its terms and conditions for copying, distributing or modifying
461 the Program or works based on it.
462
463 6. Each time you redistribute the Program (or any work based on the
464 Program), the recipient automatically receives a license from the
465 original licensor to copy, distribute or modify the Program subject to
466 these terms and conditions. You may not impose any further
467 restrictions on the recipients' exercise of the rights granted herein.
468 You are not responsible for enforcing compliance by third parties to
469 this License.
470
471 7. If, as a consequence of a court judgment or allegation of patent
472 infringement or for any other reason (not limited to patent issues),
473 conditions are imposed on you (whether by court order, agreement or
474 otherwise) that contradict the conditions of this License, they do not
475 excuse you from the conditions of this License. If you cannot
476 distribute so as to satisfy simultaneously your obligations under this
477 License and any other pertinent obligations, then as a consequence you
478 may not distribute the Program at all. For example, if a patent
479 license would not permit royalty-free redistribution of the Program by
480 all those who receive copies directly or indirectly through you, then
481 the only way you could satisfy both it and this License would be to
482 refrain entirely from distribution of the Program.
483
484 If any portion of this section is held invalid or unenforceable under
485 any particular circumstance, the balance of the section is intended to
486 apply and the section as a whole is intended to apply in other
487 circumstances.
488
489 It is not the purpose of this section to induce you to infringe any
490 patents or other property right claims or to contest validity of any
491 such claims; this section has the sole purpose of protecting the
492 integrity of the free software distribution system, which is
493 implemented by public license practices. Many people have made
494 generous contributions to the wide range of software distributed
495 through that system in reliance on consistent application of that
496 system; it is up to the author/donor to decide if he or she is willing
497 to distribute software through any other system and a licensee cannot
498 impose that choice.
499
500 This section is intended to make thoroughly clear what is believed to
501 be a consequence of the rest of this License.
502
503 8. If the distribution and/or use of the Program is restricted in
504 certain countries either by patents or by copyrighted interfaces, the
505 original copyright holder who places the Program under this License
506 may add an explicit geographical distribution limitation excluding
507 those countries, so that distribution is permitted only in or among
508 countries not thus excluded. In such case, this License incorporates
509 the limitation as if written in the body of this License.
510
511 9. The Free Software Foundation may publish revised and/or new versions
512 of the General Public License from time to time. Such new versions will
513 be similar in spirit to the present version, but may differ in detail to
514 address new problems or concerns.
515
516 Each version is given a distinguishing version number. If the Program
517 specifies a version number of this License which applies to it and "any
518 later version", you have the option of following the terms and conditions
519 either of that version or of any later version published by the Free
520 Software Foundation. If the Program does not specify a version number of
521 this License, you may choose any version ever published by the Free Software
522 Foundation.
523
524 10. If you wish to incorporate parts of the Program into other free
525 programs whose distribution conditions are different, write to the author
526 to ask for permission. For software which is copyrighted by the Free
527 Software Foundation, write to the Free Software Foundation; we sometimes
528 make exceptions for this. Our decision will be guided by the two goals
529 of preserving the free status of all derivatives of our free software and
530 of promoting the sharing and reuse of software generally.
531
532 NO WARRANTY
533
534 11. BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY
535 FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN
536 OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES
537 PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED
538 OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
539 MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS
540 TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE
541 PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING,
542 REPAIR OR CORRECTION.
543
544 12. IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING
545 WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY AND/OR
546 REDISTRIBUTE THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES,
547 INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING
548 OUT OF THE USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED
549 TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY
550 YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER
551 PROGRAMS), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE
552 POSSIBILITY OF SUCH DAMAGES.
553
554 END OF TERMS AND CONDITIONS
555
556 How to Apply These Terms to Your New Programs
557
558 If you develop a new program, and you want it to be of the greatest
559 possible use to the public, the best way to achieve this is to make it
560 free software which everyone can redistribute and change under these terms.
561
562 To do so, attach the following notices to the program. It is safest
563 to attach them to the start of each source file to most effectively
564 convey the exclusion of warranty; and each file should have at least
565 the "copyright" line and a pointer to where the full notice is found.
566
567 <one line to give the program's name and a brief idea of what it does.>
568 Copyright (C) 19yy <name of author>
569
570 This program is free software; you can redistribute it and/or modify
571 it under the terms of the GNU General Public License as published by
572 the Free Software Foundation; either version 2 of the License, or
573 (at your option) any later version.
574
575 This program is distributed in the hope that it will be useful,
576 but WITHOUT ANY WARRANTY; without even the implied warranty of
577 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
578 GNU General Public License for more details.
579
580 You should have received a copy of the GNU General Public License
581 along with this program; if not, write to the Free Software
582 Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
583
584 Also add information on how to contact you by electronic and paper mail.
585
586 If the program is interactive, make it output a short notice like this
587 when it starts in an interactive mode:
588
589 Gnomovision version 69, Copyright (C) 19yy name of author
590 Gnomovision comes with ABSOLUTELY NO WARRANTY; for details type `show w'.
591 This is free software, and you are welcome to redistribute it
592 under certain conditions; type `show c' for details.
593
594 The hypothetical commands `show w' and `show c' should show the appropriate
595 parts of the General Public License. Of course, the commands you use may
596 be called something other than `show w' and `show c'; they could even be
597 mouse-clicks or menu items--whatever suits your program.
598
599 You should also get your employer (if you work as a programmer) or your
600 school, if any, to sign a "copyright disclaimer" for the program, if
601 necessary. Here is a sample; alter the names:
602
603 Yoyodyne, Inc., hereby disclaims all copyright interest in the program
604 `Gnomovision' (which makes passes at compilers) written by James Hacker.
605
606 <signature of Ty Coon>, 1 April 1989
607 Ty Coon, President of Vice
608
609 This General Public License does not permit incorporating your program into
610 proprietary programs. If your program is a subroutine library, you may
611 consider it more useful to permit linking proprietary applications with the
612 library. If this is what you want to do, use the GNU Library General
613 Public License instead of this License.
614
615 %</copyright>
616 %\fi
617 % \def\fileversion{v2.1}
618 % \def\filedate{2019/08/26}
619 % \section{Introduction}
620 % \changes{v1.0}{1997/17/24}{Creation of the \universal{} font}
621 %
622 % This package contains the \MF{} source and driver files for the
623 % ``universal'' font, designed by Herbert Bayer, a teacher at the
624 % Bauhaus school in Weimar\index{Bayer, Herbert}, plus a \LaTeX{}
625 % package to utilize this font, along with a number of font definition
626 % files, as required by the New Font Selection Scheme (\NFSS{}).
627 %
628 % \subsubsection*{About the Documantation}
629 %
630 % The full documentation of this font is rather large, more then 80
631 % pages actually. Most of it is the programs for the characters for
632 % this font, so if you don't know the \MF{} language, or don't care
633 % how the charcters are created, you should insert
634 % \Lcs{OnlyDescription} into the preamble of \texttt{universa.dtx}.
635 %
636 % Notice, that \MF{} macros are \emph{not} indexed in this
637 % documentation. This is because it would take a \emph{major} rewrite
638 % of the \textsf{doc} package to do so, and I really didn't want to do
639 % that.
640 %
641 % Some of the macros of \textsf{doc} \emph{has} been redefined, and if
642 % you want to see which, or how I generally did the documentation,
643 % please refer to appendix~\ref{app:unidoc}.
644 %
645 % If you in some way are unsatisfied with some of the characters of
646 % the font, do please read the documentation of that character. There
647 % may be some notes on why the character looks the way it does. Some
648 % of the comments, however are a bit silly and should be skipped at
649 % high speed (pretend there is a conditional that says \texttt{if not
650 % silly ... fi}).
651 %
652 % \subsubsection*{This Version}
653 % \changes{v2.0}{98/08/01}{Bug corrections, new font
654 % shapes, better font programs, new/improved macros for \LaTeX{}
655 % support, better conformation to \NFSS{}, using \textsf{fontname}
656 % filenaming scheme, etc.}
657 %
658 % This new version of the \universal{} font, provides a number
659 % of new features and improvements, both to the font itself and to the
660 % \LaTeX{} support macros. Also, a number of corrections has been made
661 % to the font programs.
662 %
663 % The reason why I jumped one whole version number from 1.0 to 2.0, is
664 % that I have taken a whole new approach to the font programs and
665 % shapes of the \universal{} font, plus I decided to put some
666 % more effort into the \LaTeX{} support.
667 %
668 % \subsubsection*{Thanks and Other Stuff}
669 %
670 % To those of you who have had the (mixed) pleasure of using version
671 % 1.0 of this font, one of the most notaticable changes are to the
672 % symbol \textuni{\bauforms}. In the old documenation, I complained
673 % that I couldn't find the exact solution to the problem this symbol
674 % possed. I also encourged people to send me any solution they may
675 % have had --- and guess what --- somebody did! Therefore I would like
676 % to thanck Joseph Collins for providing me with the solution. If you
677 % like Joseph Collins and I like Mathematical puzzles take a look in
678 % Appendix~\ref{app:solu} to learn more about this problem and it's
679 % solution.
680 %
681 % I also direct your attention to Appendix~\ref{app:ggpl} for the
682 % copyright notice on the \universal{} package (it's the
683 % \textsl{Gnu General Public Lisence} to those of you who know it).
684 %
685 % If you in any way have gotten tempted to design your own font, or to
686 % implement some font into \MF{}, I feel obligated to bring you a
687 % warning, taken from the \MF{}book by Donald E.\ Knuth:
688 % \begin{quote}
689 % \uni
690 % \textsc{Warning:} Type design can be hazardous to your other
691 % interests. Once you get hooked, you will develop intense feelings
692 % about letterforms; the medium will intrude the message that you
693 % read. And you will perpetually be thincking of improvements to the
694 % fonts that you see everywhere, especially those of your own
695 % design.
696 % \end{quote}
697 %
698 % \section{The Font and it's History}
699 %
700 % \subsubsection*{Bauhaus}
701 % The Bauhaus school in germany originally located at Dessau, was a
702 % school for any kind of design, ranging from potery to furniture,
703 % from painting to --- what was considered the prime form of design
704 % --- architecture. Many famous designers came from, or taught at
705 % Bauhaus, for example Mies van der Rhoe\index{Rhoe, Mies van der},
706 % Herbert Bayer\index{Bayer, Herbert}, Kandinsky\index{Kandinsky},
707 % Walter Gropius\index{Gropius, Walter} and Gerrit
708 % Rietveld\index{Rietveld, Gerrit}. The style ``die stiil''
709 % \index{die stiil} was explored here, and painters like
710 % Mondrian\index{Mondrian} made large contributions to what today is
711 % known as
712 % ``the Bauhaus style''\index{Bauhaus style=\textuni{Bauhaus style}}.
713 %
714 % The basic idea of the Bauhaus school, was to design items, which
715 % along with it's aesthethic value, also had a high degree of
716 % functionallity. Houses were meant to be suited for all kinds of
717 % living, while still keeping the beauty that make people glad to see
718 % thier house. Chairs should be comfortable for thier use, as well be
719 % able to fit--in in a normal house. All this should be done at a
720 % price that made it possible for everybody to own
721 % designer--furniture, houses, etc.
722 %
723 % The political idea of Bauhaus, was that of a socialist one. Houses
724 % are for the people --- they have to live in them, and that living
725 % should be a good as possible. Therefore the Bauhaus school saw it
726 % has its task to provide functional, beautiful everyday items that
727 % anybody could affort. Paintings and tapestery shouldn't hang on
728 % museums or art galleries, but in peoples homes, where they would
729 % inrich thier everyday life.
730 %
731 % \subsubsection*{Herbert Bayer and the ``universal'' Font}
732 %
733 % At Bauhaus typography was also studied, not just how written text
734 % should be typeset, and how printed characters should look like, but
735 % also what the essence of writting is, in it's pratical and
736 % design--wise sense. This led Herbert Bayer\index{Bayer, Herbert} to
737 % formulate some principles of writting:
738 %
739 % \begin{quote}
740 % {\uni experiment with simplified way of writing:
741 % \begin{enumerate}
742 % \item this way of writing is recommended by all typographic
743 % designers as the future way of writing.
744 % \item by writing in minuscules our writing looses nothing,
745 % but is easier to read, considerably more economical.
746 % \item why must you for one sound have two tokens, e.g.\
747 % A and a? why two alphabets for one word, why this double
748 % set of signs, when the half is enough.
749 % \end{enumerate}
750 % \begin{flushright}
751 % herbert bayer 1925
752 % \end{flushright}}
753 % \end{quote}
754 %
755 % On this principles, Herbert Bayer designed a font, which should have
756 % no majuscles (upper case letters), easy to print, and easy to read
757 % since it didn't have any unusefull decoration, but communicated the
758 % bare meaning of characters through the simplest forms needed to
759 % reconigise a character. This font he called ``universal''.
760 %
761 % This font contained abolutly no majuscles, since Bayer believed them
762 % to be superflous, as it is clear form the quote above.
763 %
764 % At the time Herbert Bayer formulated these principles and designed
765 % the ``universal'' font, most printers used Gothic letters, which is
766 % allmost overly decorated, so his font ofcourse made
767 % contraversy.
768 %
769 % Later on, in the 1930'ies, the Bauhaus school drew the
770 % attention of \emph{Gestapo}\footnote{Gehemligche Stats Polizi} of
771 % Nazi\footnote{Nationalsocializmus} Germany. The school was finally
772 % closed in 1936 by Gestapo, because they believed they were promoters
773 % of Jewish and Communist culture and propaganda.
774 %
775 % In the aftermatch of the closing of the school, most of it's ideas
776 % were shundend by other designers, and the Bauhaus way of thinking
777 % died out. This is properly the reason why Herbert Bayers
778 % ``universal'' font is so little known today.
779 %
780 % However, the ``universal'' font still stands as one of the most
781 % compelling developments in font designing. It represents an approach
782 % to designing where the functionallity is as vital as the expression,
783 % and as such I believ it to be one of the most important fonts in the
784 % world today.
785 %
786 % \section{This \MF{} Implementation}
787 %
788 % This implementation of Herbert Bayers ``universal'' font, is not
789 % supposed to look \emph{exactly} like the original design. Ofcourse I
790 % have tried to the best of my ability to mimic his design as far as I
791 % could. However, it is not an easy matter to find a complete, not to
792 % mention exact, sample or account of Bayers design.
793 %
794 % This implementation is bassed on the samples I \emph{could} find,
795 % and other implementations of the ``universal'' font. Many of these
796 % other implementations do differ from the original samples, and
797 % include characters I couldn't find in any of the original. So
798 % whenever I found disparities, I mainly leened on the original
799 % samples and my understanding of the original design.
800 %
801 % \subsection{The \MF{} version versus the Original}
802 %
803 % As mentioned above, Bayer never did design any majuscles for this
804 % font, but nonetheless, I have included them into this
805 % implementation. This I did, because I think most people will have a
806 % hard time writting in minuscles (lowercase letters) alone. Ofcorse,
807 % if you agree with Bayer, you should simple not use them.
808 %
809 % There are also some other differences, mostly due to the fact, that
810 % I never found a complete sample of the original font. The major
811 % differences between the originalk font and this implementation is
812 % summerized below:
813 %
814 % \begin{description}
815 % \item[Majuscles:] Majuscles are present, even though they weren't
816 % in the original design.
817 % \item[Digigts:] These are based on other implementations, and my
818 % general conception of the original design.
819 % \item[Punctuations:] As above.
820 % \item[Accents:] As above.
821 % \item[Symbols:] As above.
822 % \item[Bauhaus Symbols:] I have added some various symbols I have
823 % found in connection with Bauhaus to the font. The reason is I
824 % find them beautiful, and I had some space to fill.
825 % \item[Numerous Shapes and Weights:] I don't think Bayer ever
826 % design slanted characters, a bold face version of the font, and
827 % he could never have design a small caps version of the
828 % font. However, these are present in this implementation. I
829 % included these features, because I believ them to be of general
830 % utility, and it makes the font conform more to the Computer
831 % Modern Roman font, and \NFSS{}.
832 % \end{description}
833 %
834 % \subsection{Features of the font}
835 %
836 % Rather then using \texttt{cmbase.mf}\footnote{Computer Modern Roman
837 % base file}, and then redefine some macros, I chose to make a new
838 % base file myself, i.e.,
839 % \texttt{unibase.mf}\index{unibase.mf=\texttt{unibase.mf}}. This file
840 % contains a number of macros\footnote{A better name for `base' would
841 % be `library', and then the file would be \texttt{libuni.mf}, but to
842 % conform to CM, I used `base'.} I have used in the character
843 % programs.
844 %
845 % The macros of \texttt{unibase.mf} actually reflects my conception of
846 % the font. There are three basic drawing macroes:
847 % \begin{description}
848 % \item[\texttt{unicir}] which draws a circle,
849 % \item[\texttt{uniarc}] which draws a segment (arc) of a circle, and
850 % \item[\texttt{unilne}] which draws a straight line.
851 % \end{description}
852 % I believe, that Bayer intended the font to be made of these two
853 % basic shapes: the arc, and the line. Also, to keep things simple,
854 % and therefore easy to print, all shapes should be of the same
855 % thickness, i.e.,as if drawn with a pen of equal thickness. I have
856 % made one diviation from this, however. All majuscles are drawn with
857 % a thicker `pen', which makes the output nicer, I think.
858 %
859 % Incidently, this made the programs of the characters much simpler,
860 % and shorter.
861 %
862 % \subsubsection{Series, Shapes, Sizes, and Special Characters}
863 %
864 % \Newestchanges{v2.0}{98/08/01}{Added `small caps' shape to the font.}
865 % \changes{v2.0}{98/08/01}{Added non--standard `strict' shape to the
866 % font.}
867 % \changes{v2.0}{98/08/01}{Added `slanted' shape to the font.}
868 % \changes{v2.0}{98/08/01}{Default series is changed to `medium'
869 % to the font, `bold face' support included.}
870 %
871 % Below is a sample of each series/shape combination avaliable in this
872 % implementation of the font, along with the \LaTeX{} commands that
873 % drive them:
874 %
875 % \begin{description}
876 % \item[\rm Medium upright (\Lcs{textuni}):]
877 % \textuni{The dazed brown fox quickly gave 1234-567890 jumps!}
878 % \item[\rm Medium slanted (\Lcs{textunisl}):]
879 % \textunisl{The dazed brown fox quickly gave 1234-567890 jumps!}
880 % \item[\rm Medium small caps (\Lcs{textunisc}):]
881 % \textunisc{The dazed brown fox quickly gave 1234-567890 jumps!}
882 % \item[\rm Medium strict (\Lcs{textunist}):]
883 % \textunist{The dazed brown fox quickly gave 1234-567890 jumps!}
884 % \item[\rm Bold face upright (\Lcs{textunibf}):]
885 % \textunibf{The dazed brown fox quickly gave 1234-567890 jumps!}
886 % \item[\rm Bold face slanted (\Lcs{textunibsl}):]
887 % \textunibsl{The dazed brown fox quickly gave 1234-567890 jumps!}
888 % \item[\rm Bold face small caps (\Lcs{textunibsc}):]
889 % \textunibsc{The dazed brown fox quickly gave 1234-567890 jumps!}
890 % \item[\rm Bold face strict (\Lcs{textunibst}):]
891 % \textunibst{The dazed brown fox quickly gave 1234-567890 jumps!}
892 % \end{description}
893 %
894 % Everyone of these shapes are avaliable in size 8, 9, 10, 12, 17
895 % \texttt{pt}, and \MF{} can ofcourse create others.
896 %
897 % \Newestchanges{v2.0}{98/08/01}{Some of the \textunibf{bauhaus} symbols
898 % that where avaliable in v1.0, has been taken out.}
899 % Also a number of non--standard characters are avaliable in the
900 % font. Below is a table of these characters along with the \LaTeX{}
901 % commands that drive them.
902 %
903 % Notice that \textunibf{``} and \textunibf{''} is present in the
904 % table. This is because these characters are not directly defined,
905 % but is supplied as \emph{ligatures}. This can be done, because they
906 % are simple doubles of \textunibf{`} and \textunibf{'}.
907 %
908 % A quick look on the table will also reveal some characters that
909 % generally isn't present in the standard \texttt{OT1}
910 % encoding\footnote{The \texttt{OT1} encoding is the 7 bit encoding of
911 % the Computer Modern fonts by Donald E.\ Knuth. 7 bit means it
912 % contain 128 ($=2^7$) characters.}, but generally present in the
913 % \texttt{T1} encoding\footnote{The \texttt{T1} encoding is an
914 % encoding especially designed for the (western) European
915 % languages. It was founded by the \TeX{} User's Group, on a seminar
916 % in Cork, and is the basis of the \texttt{dc} fonts. \texttt{T1} is
917 % an 8 bit encoding, which means it has 256 ($=2^8$) characters.}. I
918 % have done this, both to provide an (almost) complete font for the
919 % European languages, but also because I anticipate the \universal{}
920 % font some time in the future will shift, or a least be avaliable, in
921 % the \texttt{T1} encoding\footnote{\texttt{T1} is generally
922 % considered \emph{the} encoding of the future, and in the long term,
923 % it is most likely the encoding of \LaTeX3.}.
924 % \begin{table}[htbp]
925 % \centering
926 % \begin{tabular}{%
927 % @{\vline\extracolsep{1em}\space}ll%
928 % @{\extracolsep{1em}}ll%
929 % @{\space\vline\extracolsep{1em}}}\hline
930 % \textuni{\bausquare}&\Lcs{bausquare}&
931 % \textuni{\baucircle}&\Lcs{baucircle}\\
932 % \textuni{\bautriangle}&\Lcs{bautriangle}&
933 % \textuni{\bauhead}&\Lcs{bauhead}\\
934 % \textuni{\bauforms}&\Lcs{bauforms}&
935 % \textuni{\dh}&\Lcs{dh}\\
936 % \textuni{\dj}&\Lcs{dj}&
937 % \textuni{\ng}&\Lcs{ng}\\
938 % \textuni{\th}&\Lcs{th}&
939 % \textuni{\varQ}&\Lcs{varQ}\\
940 % \textuni{\DH}&\Lcs{DH}&
941 % \textuni{\DJ}&\Lcs{DJ}\\
942 % \textuni{\NG}&\Lcs{NG}&
943 % \textuni{\TH}&\Lcs{TH}\\
944 % \textuni{\textsection}&\Lcs{textsection} or \Lcs{S}&
945 % \textuni{\textbraceleft}&\Lcs{textbraceleft} or \Lcs{\char"7B}\\
946 % \textuni{\textbraceright}&\Lcs{textbracerigth} or \Lcs{\char"7D}&
947 % \textuni{\textbar}&\Lcs{textbar}\\
948 % \textuni{\guilsinglleft}&\Lcs{guilsinglleft}&
949 % \textuni{\guilsinglright}&\Lcs{guilsinglright}\\
950 % \textuni{\guillemoleft}&\Lcs{guillemoleft}&
951 % \textuni{\guillemoright}&\Lcs{guillemoright}\\
952 % \textuni{\quotesinglbase}&\Lcs{quotesinglbase}&
953 % \textuni{\quotedblbase}&\Lcs{quotedblbase}\\
954 % \textuni{\textquotedblleft}&\Lcs{textquotedblleft}&
955 % \textuni{\textquotedblright}&\Lcs{textquotedblright}\\
956 % \textuni{\textogonek}&\Lcs{textogonek}&
957 % \textuni{\textunderscore}&\Lcs{textunderscore}\\
958 % \hline
959 % \end{tabular}
960 % \caption{Non--standard characters in the {universal} font}
961 % \label{tab:characters}
962 % \end{table}
963 % Please note, that in the small caps shaped fonts, \Lcs{dh} does not
964 % give {\small\textuni{\DH}}, but a \textunibsc{\dh}, that is a small
965 % caps shaped version of \Lcs{varQ}. Also there is no command
966 % \Lcs{varq} defined.
967 %
968 % In appendix~\ref{app:charts} is some charts showing the font in
969 % different series and shapes.
970 %
971 % \subsubsection{File Names for the \MF{} files}
972 %
973 % \paragraph{Base File and Source Files}
974 % The base file and the files containg the code for the characters of
975 % the \universal{} font, all starts with \texttt{uni}, to reflect the
976 % connection of the files. The next five possible letters reflects
977 % what kind of code is contained within the file, e.g., the base file
978 % ends in \texttt{base}, the file containing the code for the
979 % minuscles (lower case letters) end in \texttt{lower}, and so forth.
980 %
981 % \paragraph{Font Driver Files}
982 % The font driver filenames has been chosen to conform to the
983 % \texttt{fontname} scheme, because this scheme is used by most
984 % \TeX{}, \LaTeX{}, and \MF{} systems (anyway those that use
985 % \texttt{kpathsea}, which is the most).
986 %
987 % The filename all contain the three characters \texttt{ful}, where
988 % the \texttt{f} stands for \textit{public} and \texttt{ul} for
989 % \textit{universal}.
990 %
991 % Next comes a letter which is one of \texttt{m} (\textit{medium}) or
992 % \texttt{b} (\textit{bold}), which represents the series of the
993 % font.
994 %
995 % Then comes one or two letters, which are \texttt{r}
996 % (\textit{upright} or roman), \texttt{o} (\textit{slanted}, or
997 % oblique), \texttt{c} (\textit{small caps}), or \texttt{st}
998 % (\textit{strict}), which represent the shape of the font.
999 %
1000 % Finally the filename ends with the designsize in points.
1001 %
1002 % Thus the complete syntax for the font driver file names is:
1003 % \begin{table}[htbp]
1004 % \centering
1005 % \begin{tabular}{rcl}
1006 % \meta{filename}&:=&
1007 % \meta{supplier}\meta{face}\meta{series}\meta{shape}\meta{size}\texttt{.mf}\\
1008 % \meta{supplier}&:=&\texttt{f}\\
1009 % \meta{face}&:=&\texttt{ul}\\
1010 % \meta{series}&:=&\texttt{m \char"7C\ b}\\
1011 % \meta{shape}&:=&\texttt{r \char"7C o \char"7C c \char"7C st}\\
1012 % \meta{size}&:=&\texttt{\char"7C\ 8 \char"7C\ 9 \char"7C\ 10
1013 % \char"7C\ 12 \char"7C\ 17}
1014 % \end{tabular}
1015 % \end{table}
1016 % The \texttt{fontname} scheme actually says to put \meta{encoding}
1017 % information after the \meta{shape}, but since this is \texttt{8r}
1018 % for \TeX{} Text, it would make filenames longer then 8 characters in
1019 % the cases of \meta{size} of 10, 12, and 17, so this information is
1020 % left out (which is permisable in \texttt{fontname}, but
1021 % unfortunate).
1022 %
1023 % This way of naming the font driver files will, if you use
1024 % \texttt{kpathsea}, put the \texttt{ful*.*pk} files in\newline
1025 % \indent\meta{pk--base--dir}\texttt{/public/universa/}\newline
1026 % and the \texttt{ful*.tfm} files in\newline
1027 % \indent\meta{tfm--base--dir}\texttt{/public/universa/}\newline
1028 % which I think is the intuitivly correct place to put them. This also
1029 % means, that the \texttt{*.mf} files provided with this package,
1030 % should be placed in\newline
1031 % \indent\meta{mf--source--base--dir}\texttt{/public/universa/}\newline
1032 % again very intuitive.
1033 %
1034 % Below is a table of the usual directory names under
1035 % \textsf{Unix}--like and \textsf{MSDOS}--like (including Windows95)
1036 % systems.\meta{mode} is \textsf{dvips}'s name for your printer.
1037 % \begin{table}[htbp]
1038 % \begin{tabular}{%
1039 % @{\vline\extracolsep{1em}\space}l%
1040 % @{\space\vline\extracolsep{1em}\space}l%
1041 % @{\space\vline\extracolsep{1em}\space}l%
1042 % @{\space\vline\extracolsep{1em}}}
1043 % \hline
1044 % Varibale&\textsf{Unix}--like systems&\textsf{MSDOS}--like systems\\
1045 % \hline
1046 % \meta{tfm--base--dir}&
1047 % \texttt{/var/spool/texmf/pk/}\meta{mode}\texttt{/}&
1048 % \texttt{C:\Lcs{}FONTS\Lcs{}PK\Lcs{}}\meta{mode}\Lcs{}\\
1049 % \meta{tfm--base--dir}&
1050 % \texttt{/var/spool/texmf/tfm/}&
1051 % \texttt{C:\Lcs{}FONTS\Lcs{}TFM\Lcs{}}\\
1052 % \meta{tfm--base--dir}&
1053 % \texttt{/usr/local/lib/texmf/fonts/source/}&
1054 % \texttt{C:\Lcs{}TEX\Lcs{}MFINPUTS\Lcs{}}\\
1055 % \hline
1056 % \end{tabular}
1057 % \caption{Common directory names.}
1058 % \end{table}
1059 %
1060 % \subsection{The \LaTeX\ and \NFSS\ Support}
1061 %
1062 % To use the \universal{} font with \LaTeXe{}\footnote{I have made
1063 % \emph{no} attempt to provide support for \LaTeX2.09, since this
1064 % format is obsolete, and those who \emph{do} use it, will proberly
1065 % never bother to look at CTAN for new fonts anyway.}, you should load
1066 % the package \textsf{uni} with the command
1067 % \begin{quote}
1068 % \Lcs{usepackage}\oarg{options}\marg{uni}
1069 % \end{quote}
1070 % in your preamble (i.e., after \Lcs{documentclass} and before
1071 % \Lcs{begin\{document\}}). \meta{options} can be any of the options
1072 % described below, but no other.
1073 %
1074 % \subsubsection{Options to \textsf{uni}}
1075 %
1076 % \marginpar{\raggedright\texttt{strict}}
1077 % \index{strict=\texttt{strict}\space(Option)}
1078 % \Newestchanges{v2.0}{98/08/01}{Added option \texttt{strict}.}
1079 % The \texttt{srict} option is intended to fassilitate typesetiing of
1080 % the \universal{} font in a \emph{strict} \textunibf{bauhaus}
1081 % fashion, that is \emph{only} in minuscules.
1082 %
1083 % In this font, only the series may be variated, that is, there is a
1084 % bold series strict shaped font of any size, and a medium series
1085 % strict shaped font of any size, in the \universal{} family.
1086 %
1087 % This option can be used in conjuntion with options \texttt{medium}
1088 % and \texttt{bold}. Please notice, that it doesn't make any sense to
1089 % ask for a small caps or slanted shaped font, while using this option.
1090 %
1091 % Notice that only \Lcs{textuni} and \Lcs{uni} is defined if option
1092 % \texttt{strict} was given to \textsf{uni} package.
1093 %
1094 % \marginpar{\raggedright\texttt{default}}
1095 % \index{default=\texttt{default}\space(Option)}
1096 % \Newestchanges{v2.0}{98/08/01}{Added option \texttt{default}.}
1097 % If you give the \texttt{default} option to the \textsf{uni} package,
1098 % the default font of the document will be \textunirm{universal}.
1099 %
1100 % If you also used the option \texttt{bold} the default font will be
1101 % the \universal{} font in \textbf{bold} series. Otherwise it will be
1102 % in medium series.
1103 %
1104 % With this option, \Lcs{textit}, \Lcs{it}, and \Lcs{itshape} shifts
1105 % to \textunisl{universal slanted} font, i.e., there is no
1106 % \textit{italic} font avaliable.
1107 %
1108 % \DescribeMacro{\textcmr}
1109 % \DescribeMacro{\cmr}
1110 % To make it possible to change back to Donald E.\ Knuth's Computer
1111 % Modern Roman font, even when the \texttt{default} option is given,
1112 % we define macros \Lcs{textcmr} and \Lcs{cmr}, which switches the
1113 % \Lcs{fontfamily} to \texttt{cmr} locally and gobally respecivly.
1114 %
1115 % You should use this option with some care, since the \universal{}
1116 % font isn't very suited for longer texts, but rather for short
1117 % letters, quotes, and other pieces of text where the graphical
1118 % apperance is important.
1119 %
1120 % \marginpar{\raggedright\texttt{medium}}
1121 % \index{medium=\texttt{medium}\space(Option)}
1122 % \Newestchanges{v2.0}{98/08/01}{Added option \texttt{medium}.}
1123 % When this option is given, command \Lcs{textuni} switches to
1124 % \textunirm{medium series universal font}, as do \Lcs{uni}. The other
1125 % font selection commands behave as always (see below). This is the
1126 % default option to \textsf{uni}, i.e., not normally needed.
1127 %
1128 % \marginpar{\raggedright\texttt{bold}}
1129 % \index{bold=\texttt{bold}\space(Option)}
1130 % \Newestchanges{v2.0}{98/08/01}{Added options \texttt{bold}.}
1131 % If this option is given, commands \Lcs{textuni} and \Lcs{uni}
1132 % switches to \textunibf{bold series universal font}. Other font
1133 % selection commands behave as always (see below). Notice that it
1134 % makes no sense to give both option \texttt{medium} and option
1135 % \texttt{bold} to the \textsf{uni} package.
1136 %
1137 % \subsubsection{Font Selection Commands}
1138 %
1139 % \DescribeMacro{\textuni}\DescribeMacro{\uni}
1140 % These two commands only change the current font family to
1141 % \texttt{uni} \emph{and nothing else}. That means, that if you say
1142 % for example
1143 % \begin{verbatim}
1144 % {\sl Hello \textuni{world}}
1145 % \end{verbatim}
1146 % you get \emph{both} `Hello' and `world' in slanted shape, and the
1147 % output would be
1148 % \begin{quote}
1149 % {\sl Hello \textuni{world}}
1150 % \end{quote}
1151 % To put it in another way: Font encoding, shape, size, and
1152 % baselineskip is preserved under \Lcs{textuni} and \Lcs{uni}, while
1153 % font family is not.
1154 %
1155 % \textbf{Execptions:} If you gave the \texttt{bold} option to the
1156 % \textsf{uni} package, then this command will always give you a bold
1157 % series font. If you gave the \texttt{strict} option, then this
1158 % command will always give you a strict shaped font.
1159 %
1160 % These commands can be used in conjuntion with \LaTeX{} commands
1161 % \Lcs{textbf}, \Lcs{textsl}, and even \Lcs{sc}, or \Lcs{rm} to give
1162 % different series and shapes.
1163 %
1164 % \DescribeMacro{\textunirm}\DescribeMacro{\unirm}
1165 % If you in the previous example intended to shift to
1166 % \textunirm{medium upright universal} font you could instead have
1167 % used \Lcs{textunirm}, since this command \emph{does not} preserve
1168 % font shape, i.e., the shape is always changed to upright, regardless
1169 % of the prevouis shape. So if you said
1170 % \begin{verbatim}
1171 % {\sl Hello \textunirm{world}}
1172 % \end{verbatim}
1173 % you would get
1174 % \begin{quote}
1175 % {\sl Hello \textunirm{world}}
1176 % \end{quote}
1177 %
1178 % \DescribeMacro{\textunisl}\DescribeMacro{\unisl}
1179 % \DescribeMacro{\textunisc}\DescribeMacro{\unisc}
1180 % \DescribeMacro{\textunist}\DescribeMacro{\unist}
1181 % \label{expl:cmds}
1182 % Actually \Lcs{textunirm} and \Lcs{unirm} isn't the only commands
1183 % that aggresively changes most of the font
1184 % parameters. \Lcs{textunisl}, \Lcs{textunisc}, and \Lcs{textunist},
1185 % \emph{always} gives you \textunisl{medium slanted},
1186 % \textunisc{medium small caps}, and \textunist{medium strict}
1187 % respecitivly no matter what the values of \Lcs{f@shape} was before.
1188 %
1189 % \DescribeMacro{\textunibf}\DescribeMacro{\unibf}
1190 % \DescribeMacro{\textunibsl}\DescribeMacro{\unibsl}
1191 % \DescribeMacro{\textunibsc}\DescribeMacro{\unibsc}
1192 % \DescribeMacro{\textunibst}\DescribeMacro{\unibst}
1193 % In the same category is \Lcs{textunibf}, \Lcs{textunibsl},
1194 % \Lcs{textunibsc}, and \Lcs{textunist} which always changes the font
1195 % series to \textbf{bold}, along with change in shape (upright,
1196 % slanted, small caps, and strict in that order).
1197 %
1198 % All of the `aggresive' commands, \emph{do not} however change the
1199 % \emph{size} and \emph{baselineskip} of the font. This should be done
1200 % by using \LaTeX{} commands such as \Lcs{small}, \Lcs{Large},
1201 % \Lcs{fontsize}\marg{size}\marg{lineskip},
1202 % etc.
1203 %
1204 % Notice that the `aggresive' commands always changes to the
1205 % appropiate font series. That is, even if you gave the \texttt{bold}
1206 % option to \textsf{uni}, \Lcs{textunirm} will still give you
1207 % \textunirm{medium upright universal} font. This particular instance
1208 % illustrates the use of the aggresive commands quite well I think.
1209 %
1210 % To summerize: The `aggresive' commands \emph{doesn't} preserve
1211 % fammily, series and shapes, but \emph{does} preserve encoding, size,
1212 % and baslineskip.
1213 %
1214 % \textbf{Warning:} The font shifting commands \Lcs{textuni...} and
1215 % \Lcs{uni...} in this section is \emph{not} defined if you gave the
1216 % \texttt{strict} option to the \textsf{uni} package.
1217 %
1218 % \DescribeMacro{\unifamily}
1219 % \Newestchanges{v2.0}{98/08/01}{Defined \Lcs{unifamily} to
1220 % fasilitate easier and stronger font loading.}
1221 % One can also use the rather primitive command \Lcs{unifamily} in
1222 % conjunction with \Lcs{selectfont} as described in \ftnguide{}, to
1223 % change the font family to \universal{} if absolute control is
1224 % preferd.
1225 %
1226 % \Lcs{unifamily} is used by all the other font switching commands, so
1227 % if you redefine it, or \Lcs{unifamilydefault}, you could get strange
1228 % resault.
1229 %
1230 % \DescribeMacro{\unifamilydefault}
1231 % This command normally expands to \texttt{uni}, which is the `family'
1232 % name of the \universal{} package. If you redefine this command to be
1233 % something else, e.g., \texttt{cmr}, \Lcs{unifamily} will load
1234 % another font.
1235 %
1236 % \DescribeMacro{\uniseries}
1237 % \Newestchanges{v2.0}{98/08/01}{Defined \Lcs{uniseries} to
1238 % fasilitate easier and stronger font loading.}
1239 % If the \texttt{bold} option to \textsf{uni} is used this command
1240 % will select the default series of the \universal{} font, defined in
1241 % \Lcs{uniseriesdefault}, which ofcourse defaults to \texttt{b}, i.e.,
1242 % bold series. if you redefine \Lcs{uniseriesdefault} to be
1243 % \texttt{m}, then \Lcs{uniseries} will select medium series
1244 % fonts.
1245 %
1246 % If \texttt{bold} option wasn't given, then this expands to nothing,
1247 % as do \Lcs{uniseriesdefault}.
1248 %
1249 % This command is used by \Lcs{textuni} and \Lcs{uni}.
1250 %
1251 % \DescribeMacro{\unishape}
1252 % \Newestchanges{v2.0}{98/08/01}{Defined \Lcs{unishape} to
1253 % fasilitate easier and stronger font loading.}
1254 % If the \texttt{strict} option to \textsf{uni} is used this command
1255 % will select the default shape of the \universal{} font, defined in
1256 % \Lcs{unishapedefault}, which ofcourse defaults to \texttt{st}, i.e.,
1257 % strict shape. If you redefine \Lcs{unishapedefault} to be
1258 % \texttt{n}, then \Lcs{uniseries} will select upright shaped fonts.
1259 %
1260 % If \texttt{strict} option wasn't given, then this expands to nothing,
1261 % as do \Lcs{unishapedefault}.
1262 %
1263 % This command is used by \Lcs{textuni} and \Lcs{uni}.
1264 %
1265 %
1266 % \DescribeMacro{\stshape}
1267 % \Newestchanges{v2.0}{98/08/01}{Defined \Lcs{stshape} to make package
1268 % conform to \LaTeXe{} better.}
1269 % This command switches to \emph{strict} shape, i.e., all majuscules
1270 % will be typeset as minuscles. This makes it possible in a simple way
1271 % to typeset text in the way Herbert Bayer thought it should, as
1272 % evident from the citation above.
1273 %
1274 % This command uses the command \Lcs{stdefault}, which defaults to
1275 % \texttt{st}. If you redefine this to something alse, e.g.,
1276 % \texttt{sl} you will get a \textsl{slanted} font.
1277 %
1278 % This command is used by \Lcs{textunist}, \Lcs{textunibst},
1279 % \Lcs{unibst} and \Lcs{unist}, and is defined no matter waht options
1280 % you gave to the \textsf{uni} package.
1281 %
1282 % \textbf{Warning:} Since \emph{strict} is a non--standard shape, this
1283 % command should not be used outside the \universal{} font, since this
1284 % may give you unexpected resaults.
1285 %
1286 % \subsubsection{Special Character Commands}
1287 %
1288 % For the individual commands that makes various special characters,
1289 % please consult table~\ref{tab:characters} above.
1290 %
1291 % \subsubsection{Other Commands}
1292 %
1293 % \DescribeMacro{\k}
1294 % The macro \Lcs{k} used in the \universal{} font gives the accent
1295 % ogenek, that is a reversed cidelia accent. It takes one argument,
1296 % which shuold be a single letter, under which it puts the accent. For
1297 % example, you could say \Lcs{textuni\{}\Lcs{k\{a\}\}} and get
1298 % \textuni{\k{a}}.
1299 %
1300 % \DescribeMacro{\DeclareUniChar}
1301 % \Newestchanges{v2.0}{98/08/01}{Configuration of \textsf{uni} package
1302 % provided via the commands \Lcs{DeclareUniChar} and
1303 % \Lcs{DeclareUniCommand}.}
1304 % Now you can configure the special character commands of the font,
1305 % via the commands \Lcs{DeclareUniChar} and
1306 % \Lcs{DeclareUniCommand}. The commands defined via these commands
1307 % will only work in accordence with it's definition inside the
1308 % \universal{} font, and if defined elsewhere according to it's
1309 % definition there, else it will give an error message.
1310 %
1311 % \Lcs{DeclareUniChar} is used to define a command sequence
1312 % representing a single character in the \universal{} font, much like
1313 % \Lcs{DeclareTextSymbol}, or in \TeX{} \Lcs{chardef}, though the
1314 % control sequence will produce an error message outside the
1315 % \universal{} font, and properle unexpected resaults outside the
1316 % \texttt{OT1} encoding\footnote{This should not be a problem.}.
1317 %
1318 % \Lcs{DeclareUniCommand} is used to define control sequences inside
1319 % the \universal{} font, representing many characters or doing complex
1320 % manuvers on characters and stuff. The optional argument to
1321 % \Lcs{DeclareUniCommand} can be used to say how many arguments the
1322 % control sequence should have, just like \Lcs{newcommand}. However,
1323 % it is not possible to give a default first argument.
1324 %
1325 % The below defintion uses the \textsf{color} package to typeset a
1326 % square, circle, and triangle in different colours\footnote{Since the
1327 % documentation should be avaliable to all, I can not provide you with
1328 % the outcome of this example, since it needs the \textsf{color}
1329 % package wich may not be avaliable on all sites. I suggest you try
1330 % it out, or something similar, if you can.}.
1331 % \begin{verbatim}
1332 % \DeclareUniCommand{\mybauforms}{%
1333 % \lower.5ex\hbox{\color{blue}\bautriangle}%
1334 % \kern-.5em\raise.5ex\hbox{\color{read}\baucircle}%
1335 % \kern-.5em\lower.5ex\hbox{\color{yellow}\bausquare}}
1336 % \end{verbatim}
1337 % Yet another example, using arguments could be
1338 % \begin{verbatim}
1339 % \DeclareUniCommand{\mybaulogo}[1]{\bauhead\ {\Large #1}}
1340 % \end{verbatim}
1341 % so you could say \verb+\mybaulogo{Christian Holm}+ and get:
1342 % \begin{quote}
1343 % {\uni \bauhead {\Large Christian Holm}}
1344 % \end{quote}
1345 % and I bet you can come up with some even more useful and complex
1346 % commands.
1347 %
1348 % The syntax of \Lcs{DeclareUniChar} and \Lcs{DeclareUniCommand} is
1349 % \begin{quote}
1350 % \Lcs{DeclareUniChar}\marg{cmd}\marg{slot}\newline
1351 % \Lcs{DeclareUniCommand}\marg{cmd}\oarg{arg}\marg{definition}
1352 % \end{quote}
1353 % where \meta{cmd} is the user command defined, \meta{slot} is the
1354 % number of the character in the font, \meta{arg} is the number of
1355 % arguments and \meta{definition} is what \meta{cmd} does.
1356 %
1357 % \subsubsection{File Names for the \LaTeX{} files}
1358 %
1359 % All the \LaTeX{} files contain the three letters \texttt{uni}, to
1360 % reflect the connectedness of the files. The font definition files
1361 % all start with the letter code appropiate for the encodeing.
1362 %
1363 % To follow the scheme of the \MF{} files, it would be appropiate to
1364 % place all \LaTeX{} files in\newline
1365 % \indent\meta{tex--base--dir}\footnote{On \textsf{Unix}--like systems
1366 % \meta{tex--base--dir} is usually something like
1367 % \texttt{/usr/local/lib/texmf/}, and on \textsf{MSDOS}--like systems
1368 % something like \texttt{C:\Lcs{}TEX}\Lcs{}
1369 % .}\texttt{/tex/latex/universa/}
1370 %
1371 % \subsection{The Major differences between version 1.0 and 2.0}
1372 %
1373 % \paragraph{First of:} \emph{a lot} of bugs and errors has been
1374 % corrected. In version~1.0, I had made the (stupid) mistake of
1375 % calling the macro \textuni{mode\_setup} before I defined the
1376 % unsharped units. Ofcourse a quick look in the \MFbook{} showed be
1377 % just how stupid this is. This made the font very vunerable to mode
1378 % specifications, which ofcourse isn't the idea.
1379 %
1380 % \paragraph{Secondly:} I chose a completly new approach to the
1381 % character programs, which resaultet in \texttt{unibase.mf}. The idea
1382 % is to define a few macros, and then utilize those in the character
1383 % programs, so that these programs can be kept simple, efficient, and
1384 % intuitive.
1385 %
1386 % A quick look at \texttt{unibase.mf} will also reveal that I chose a
1387 % new way of adjusting the characters. This means that the macros
1388 % \texttt{bauhaus...} present in version~1.0 no longer is needed, and
1389 % since they only tended to ubscure things rather then simplify them,
1390 % I went back to the plain \MF{} macro \texttt{beginchar}, which is
1391 % much stronger.
1392 %
1393 % All in all, \texttt{unibase.mf} provides a much stronger and uniform
1394 % frame work for character design, then did the old
1395 % \texttt{universal.mf}.
1396 %
1397 % \paragraph{Thirdly:} The file names have been keept inside
1398 % \textsc{msdos} conventions, that is first name of maximum 8
1399 % charcaters, and last name of maximum 3 characters. This does mean,
1400 % however, that some file names are not intutive, but I have tried to
1401 % make them as much as I could.
1402 %
1403 % Also, every file associated with this font, execpt the font
1404 % definition files (\texttt{*uni.fd}), and font driver files
1405 % (\texttt{ful*.mf}) begins with the three letters \texttt{uni}, to
1406 % emphasize the connection.
1407 %
1408 % \paragraph{Fouthly:} Some new font shapes are avaliable, as
1409 % explained above. I found out, during the design of the characters,
1410 % that new the base file \texttt{unibase.mf} kept showing new
1411 % potentiality, and the extension of the font to include more shapes
1412 % was very easy inside the frame of this base, so I thought ``What the
1413 % heck!''
1414 %
1415 % \paragraph{Fifthly:} Some of the \textuni{bauhaus} symbols avaliable
1416 % in version~1.0, has been taken out, and some new, more general
1417 % characters have been added. Most of the absent symbols where not
1418 % really of general use, so I decided to take them out, since I was
1419 % never really satisfied with those anyway. This also made the font
1420 % contain exactly 128 characters, just like a normal Computer Modern
1421 % Roman font.
1422 %
1423 % \paragraph{Sixthly:} I improved the \LaTeX{} and \NFSS{} support
1424 % considerably. The changes are legion, but let me sum up the most
1425 % important here.
1426 % \begin{enumerate}
1427 % \item Stronger font selection commands.
1428 % \item Command names that should be more intuitive.
1429 % \item Preperations for \texttt{T1} encoding.
1430 % \item Conformation to \LaTeXe{} style, and therefore a better
1431 % chance to conform with the future \LaTeX3 format.
1432 % \item More and better options.
1433 % \end{enumerate}
1434 %
1435 % \StopEventually{\AtEndDocument{\PrintIndex\PrintChanges}}
1436 %
1437 %\iffalse
1438 %<*package>
1439 %\fi
1440 % \section{\LaTeX\ Support Files}
1441 %
1442 % \subsection{The style file --- \texttt{uni.sty}}
1443 % \file{uni.sty}
1444 %
1445 % In version~2.0, the commands in this file has been redefined using
1446 % the macros recommended in \ftnguide{} and \clsguide{}. This should
1447 % make the commands and macros more portable, and secure. Further, it
1448 % should make it upward--compatiable with future releases of other
1449 % packages and in the end \LaTeX3.
1450 %
1451 % \subsubsection{Intilalizing}
1452 %
1453 % First we need to identify the package, its version and release date,
1454 % etc.
1455 %
1456 % \begin{macrocode}
1457 \ProvidesPackage{uni}[\unifiledate\space\unifileversion\space universal
1458 package.]
1459 % \end{macrocode}
1460 %
1461 % \begin{macro}{\if@uni}
1462 % \begin{macro}{\@unifalse}
1463 % \begin{macro}{\@unitrue}
1464 % \begin{macro}{\if@strict}
1465 % \begin{macro}{\@strictfalse}
1466 % \begin{macro}{\@stricttrue}
1467 % \begin{macro}{\ifdef@ult}
1468 % \begin{macro}{\def@ultfalse}
1469 % \begin{macro}{\def@ulttrue}
1470 % \begin{macro}{\ifm@dium}
1471 % \begin{macro}{\m@diumfalse}
1472 % \begin{macro}{\m@diumtrue}
1473 % Then we setup some new \Lcs{if} commands, to help in different
1474 % sitations, depending on options passed to the package etc.
1475 % \begin{macrocode}
1476 \newif\if@uni\@unifalse
1477 \newif\ifstri@t\stri@tfalse
1478 \newif\ifdef@ult\def@ultfalse
1479 \newif\ifm@dium\m@diumfalse
1480 % \end{macrocode}
1481 % \end{macro}
1482 % \end{macro}
1483 % \end{macro}
1484 % \end{macro}
1485 % \end{macro}
1486 % \end{macro}
1487 % \end{macro}
1488 % \end{macro}
1489 % \end{macro}
1490 % \end{macro}
1491 % \end{macro}
1492 % \end{macro}
1493 %
1494 % \subsubsection{Options}
1495 %
1496 % We define some options that can be passed to the package.
1497 % Option \texttt{strict} is intended to make it possible to do
1498 % \emph{strict} Herbert Bayer typesetting, that is only in
1499 % minuscles. Option \texttt{default} will make the \emph{default} font
1500 % \universal{} of the entire document, while option
1501 % \texttt{medium} and \texttt{bold} decides wether the default font
1502 % used by \Lcs{textuni} is normal or bold series. Finally option
1503 % \texttt{medium} is declared default.
1504 % \begin{macrocode}
1505 \DeclareOption{strict}{\stri@ttrue}
1506 \DeclareOption{default}{\def@ulttrue}
1507 \DeclareOption{medium}{\m@diumtrue}
1508 \DeclareOption{bold}{\m@diumfalse}
1509 \ExecuteOptions{medium}
1510 \ProcessOptions\relax
1511
1512 % \end{macrocode}
1513 %
1514 % \subsubsection{Special Characters}
1515 %
1516 % \begin{macro}{\uni@init}
1517 % \begin{macro}{\DeclareUniChar}
1518 % \Newestchanges{v2.1}{2018/08/26}{Prevent expansion of as-of-yet not
1519 % defined arguments. Thanks to Scott Pakin.}
1520 % \changes{v2.0}{98/08/01}{Added for faster and more stable
1521 % loading of special charactes.}
1522 % \begin{macro}{\DeclareUniCommand}
1523 % \changes{v2.0}{98/08/01}{Added for faster and more stable
1524 % decleration of \universal{} commands.}
1525 % \begin{macro}{\Declare@Uni}
1526 % \begin{macro}{\DeclareUni@xarg}
1527 % \begin{macro}{\DeclareUni@narg}
1528 % \begin{macro}{\UniError}
1529 % To facilitate the special character positions of the \universal{}
1530 % font, we define the internal commands \Lcs{uni@init}, and the user
1531 % commands \Lcs{DeclareUniChar} and \Lcs{DeclareUniCommand}.
1532 %
1533 % \Lcs{DeclareUniChar} and \Lcs{DeclareUniCommand} is very much like
1534 % the \LaTeX{} commands \Lcs{DeclareTextSymbol} and
1535 % \Lcs{DeclareTextCommand}, execpt we have to take into account, that
1536 % the commands may very well be defined in \texttt{ot1enc.def} or the
1537 % like. Since some of the characters provided with this font, is
1538 % usually part of an 8 bit font, I can't just use the character
1539 % numbers straight of (this is a 7 bit font), and I have some very
1540 % non--standard characters as well.
1541 %
1542 % Though the two commands differ a bit, the main mechanism is the
1543 % same, so I will explain them together, and note the
1544 % differences\footnote{The syntax of the macroes are explanied on
1545 % page~\pageref{expl:cmds}.} .
1546 % \begin{enumerate}
1547 % \item When either of the commands is used, we first call the macro
1548 % \Lcs{Declare@uni}, which sets up 2 or 3 commands.
1549 % \item Inside \Lcs{Declare@Uni}, we define \Lcs{temp@a} to be
1550 % \Lcs{uni@}\meta{cmd-name}, where \meta{cmd-name} is \meta{cmd}
1551 % stripped of the escape character (\Lcs{}).
1552 % \item Then we check wether \Lcs{cmd} is defined.
1553 % \begin{itemize}
1554 % \item If \meta{cmd} \emph{isn't} defined, then we define
1555 % \Lcs{temp@c} to give \Lcs{uni@}\meta{cmd-name} (which isn't
1556 % defined yet) inside the \universal{} font, otherwise an error
1557 % message (\Lcs{UniError}).
1558 % \item If \meta{cmd} \emph{is} defined, we define \Lcs{temp@b} to
1559 % be \Lcs{no@uni@}\meta{cmd-name}, \Lcs{temp@c} to give
1560 % \Lcs{uni@}\meta{cmd-name} inside the \universal{} font, and
1561 % \Lcs{no@uni@}\meta{cmd-name} (which isn't defined yet)
1562 % elsewhere. Then we define \Lcs{no@uni@}\meta{cmd-name} to be
1563 % what \meta{cmd} currently is.
1564 % \end{itemize}
1565 % This finishes \Lcs{Declare@Uni} off.
1566 % \item The next step depends a bit on which of the macros that is
1567 % utiliezed. What is coomon is that \Lcs{uni@}\meta{cmd-name} (hold
1568 % in \Lcs{temp@a}) is defined to be the last argument of the command.
1569 % \begin{itemize}
1570 % \item In \Lcs{DeclareUniChar}, we define \Lcs{uni@}\meta{cmd-name}
1571 % via the \TeX{} command \Lcs{chardef} to be \meta{slot}, which
1572 % should be a number (decimal, octal, or hexal) between 0 and
1573 % 127. Using the \TeX{} primitive will make \LaTeX{} think of
1574 % \meta{cmd} as a single character.
1575 % \item If \Lcs{DeclareUniCommand} was given an optional argument,
1576 % we use the command \Lcs{DeclareUni@xarg} to define
1577 % \Lcs{uni@}\meta{cmd-name} --- which uses \Lcs{newcommand} with
1578 % \emph{one} optinal argument --- to be
1579 % \meta{definition}. Otherwise we use \Lcs{DeclareUni@narg} ---
1580 % which uses \Lcs{newcommand} with \emph{no} optinal arguments ---
1581 % to do the trick.
1582 % Using the \LaTeX{} complex \Lcs{newcommand}, makes \meta{cmd} a
1583 % strong command.
1584 % \end{itemize}
1585 % \item Finally we define \meta{cmd} to be \Lcs{temp@d}, which was the
1586 % conditional command defined in step 2.
1587 % \end{enumerate}
1588 %
1589 % Notice the use of \Lcs{aftergroup}. This means that \meta{cmd-name}
1590 % shouldn't be too long, i.e., $>300$ characters, which isn't a real
1591 % limitation, but now I said it. This macro, together with the group
1592 % in \Lcs{temp@c} is what made it possible to give arguments to
1593 % \meta{cmd} defined by \Lcs{DeclareUniCommand}.
1594 %
1595 % This way of doing things has a number of benefiets. 1) First of I
1596 % makes it very simple to redefine the behaviour of the \universal{}
1597 % speciffic character commands. All you have to do is to use
1598 % \Lcs{DeclareUniChar} or \Lcs{DeclareUniChar} in your preamble (and
1599 % \emph{only} there). 2) Since \TeX{} are faster at evaluating
1600 % \Lcs{if...}'s, switching to \universal{} commands should take a
1601 % minimum amount of time. 3) From a portability point of view, this
1602 % way of shifting meanings of commands such as \Lcs{textquotedblleft},
1603 % according to context, doesn't screw up older documents where the
1604 % \textsf{uni} package is added to.
1605 %
1606 % There are some disadvantage. 1) Most noticably there is \emph{no}
1607 % conformation to \emph{any} existing encoding, which however is a
1608 % problem of the in the font, not the \textsf{uni} package. 2)
1609 % The heavy use of \Lcs{def}, \Lcs{edef}, and \Lcs{let}, is not
1610 % exactly \LaTeX3 policy, but is needed here. 3) The definition of the
1611 % commands may slow \TeX{} down a bit, but should be bearable.
1612 % \begin{macrocode}
1613 \newcommand{\uni@init}{\@unitrue}
1614 \DeclareRobustCommand{\Declare@Uni}[1]{%
1615 \edef\temp@{\expandafter\@gobble\string#1}
1616 \edef\temp@a{\csname uni@\temp@\endcsname}
1617 \@ifundefined{\temp@}{%
1618 \edef\temp@c{{%
1619 \noexpand\if@uni\noexpand\aftergroup\temp@a\noexpand%
1620 \else\noexpand\UniError{\noexpand#1}\noexpand\fi}}}{%
1621 \edef\temp@b{\csname no@uni@\temp@\endcsname}
1622 \edef\temp@c{{%
1623 \noexpand\if@uni\noexpand\aftergroup\temp@a\noexpand%
1624 \else\noexpand\aftergroup\temp@b\noexpand\fi}}
1625 \expandafter\let\temp@b#1}}
1626 \DeclareRobustCommand{\DeclareUniChar}[2]{%
1627 \Declare@Uni{#1}
1628 \expandafter\chardef\temp@a=#2
1629 \let#1\temp@c}
1630 \def\DeclareUniCommand#1{%
1631 \@ifnextchar[{\DeclareUni@xarg#1}{\DeclareUni@narg#1}}
1632 \def\DeclareUni@xarg#1[#2]#3{%
1633 \Declare@Uni{#1}
1634 \expandafter\DeclareRobustCommand\temp@a[#2]{#3}
1635 \let#1\temp@c}
1636 \def\DeclareUni@narg#1#2{%
1637 \Declare@Uni{#1}
1638 \expandafter\DeclareRobustCommand\temp@a{#2}
1639 \let#1\temp@c}
1640 \@onlypreamble\DeclareUniChar\relax
1641 \@onlypreamble\DeclareUniCommand\relax
1642 \DeclareRobustCommand{\UniError}[1]{%
1643 \PackageError{uni}{%
1644 Command \string#1 not defined outside universal font.}{%
1645 Correct your source file, so that \string#1 doesn't appear^^J%
1646 ouside of universal font, then run LaTeX again.^^J%
1647 Prepare your self for some gastly output.}}
1648 % \expandafter\show\csname no@uni@\temp@\endcsname
1649 % \expandafter\show\csname uni@\temp@\endcsname
1650 % \expandafter\show\csname \temp@\endcsname
1651
1652 % \end{macrocode}
1653 % \end{macro}
1654 % \end{macro}
1655 % \end{macro}
1656 % \end{macro}
1657 % \end{macro}
1658 % \end{macro}
1659 % \end{macro}
1660 %
1661 % \begin{macro}{\bausquare}
1662 % \Newestchanges{v2.0}{98/08/01}{\universal{} specific
1663 % characters declared by seperate macro.}
1664 % \begin{macro}{\baucircle}
1665 % \begin{macro}{\bautriangle}
1666 % \begin{macro}{\bauhead}
1667 % \begin{macro}{\bauforms}
1668 % \begin{macro}{\dh}
1669 % \begin{macro}{\ng}
1670 % \begin{macro}{\th}
1671 % \begin{macro}{\varQ}
1672 % \begin{macro}{\NG}
1673 % \begin{macro}{\TH}
1674 % \begin{macro}{\textogonek}
1675 % \begin{macro}{\textbraceleft}
1676 % \begin{macro}{\textbraceright}
1677 % \begin{macro}{\textsection}
1678 % \begin{macro}{\textbar}
1679 % \begin{macro}{\guilsinglleft}
1680 % \begin{macro}{\guilsinglright}
1681 % \begin{macro}{\textquotedblleft}
1682 % \begin{macro}{\textquotedblright}
1683 % \begin{macro}{\quotesinglbase}
1684 % \begin{macro}{\quotedblbase}
1685 % \begin{macro}{\guillemoleft}
1686 % \begin{macro}{\guillemoright}
1687 % \begin{macro}{\k}
1688 % \begin{macro}{\DH}
1689 % \begin{macro}{\DJ}
1690 % \begin{macro}{\dj}
1691 % \begin{macro}{\textunderscore}
1692 % This command takes care of commands for non--standard placement, and
1693 % symbols in the \universal{} font, while using standard \LaTeXe{}
1694 % encodings, such as \texttt{OT1}.
1695 %
1696 % For instance, the symbols \textunibf{\}}, \textunibf{\{} isn't
1697 % present in Computer Modern fonts, but is accesiable through the
1698 % command \Lcs{}\texttt{\{}, and \Lcs{}\texttt{\}} in the \universal{}
1699 % font. On the other hand \char"00{} isn't present in this font, but
1700 % is replaced by \textunibf{\char"00}.
1701 %
1702 % Some of the commands avaliable are well--known to the \texttt{T1}
1703 % encodeing, and are provided with the same commmand names as in that
1704 % encoding.
1705 %
1706 % The decleration of these commands use my (slick) macros
1707 % \Lcs{DcelareUniChar} and \Lcs{DcelareUniCommand} --- very neat I
1708 % think.
1709 %
1710 % \begin{macrocode}
1711 \DeclareUniChar{\bausquare}{"00}
1712 \DeclareUniChar{\baucircle}{"01}
1713 \DeclareUniChar{\bautriangle}{"02}
1714 \DeclareUniChar{\bauhead}{"03}
1715 \DeclareUniChar{\bauforms}{"04}
1716 \DeclareUniChar{\dh}{"05}
1717 \DeclareUniChar{\ng}{"06}
1718 \DeclareUniChar{\th}{"07}
1719 \DeclareUniChar{\varQ}{"08}
1720 \DeclareUniChar{\NG}{"09}
1721 \DeclareUniChar{\TH}{"0A}
1722 \DeclareUniChar{\textogonek}{"0B}
1723 \DeclareUniChar{\textbraceleft}{"0C}
1724 \DeclareUniChar{\textbraceright}{"0D}
1725 \DeclareUniChar{\textsection}{"0E}
1726 \DeclareUniChar{\textbar}{"0F}
1727 \DeclareUniChar{\guilsinglleft}{"22}
1728 \DeclareUniChar{\guilsinglright}{"5C}
1729 \DeclareUniChar{\quotesinglbase}{"2C}
1730 \DeclareUniCommand{\quotedblbase}{\char"2C\kern-.3em\char"2C}
1731 \DeclareUniCommand{\textquotedblleft}{``}
1732 \DeclareUniCommand{\textquotedblright}{''}
1733 \DeclareUniCommand{\guillemoleft}{\char"22\char"22}
1734 \DeclareUniCommand{\guillemoright}{\char"5C\char"5C}
1735 \DeclareUniCommand{\DH}{\raise.5ex\hbox{\char"2D}\kern-.5em D}
1736 \DeclareUniCommand{\DJ}{\raise.5ex\hbox{\char"2D}\kern-.5em D}
1737 \DeclareUniCommand{\dj}{\raise.75ex\hbox{\char"2D}\kern-1em d}
1738 \DeclareUniCommand{\textunderscore}{\lower.5ex\hbox{\char"7B}}
1739 \DeclareUniCommand{\k}[1]{%
1740 \leavevmode\setbox\z@\hbox{#1}\ifdim\ht\z@=1ex\accent11 #1%
1741 \else{\ooalign{\unhbox\z@\crcr\hidewidth\char11\hidewidth}}\fi}
1742
1743 % \end{macrocode}
1744 % \end{macro}
1745 % \end{macro}
1746 % \end{macro}
1747 % \end{macro}
1748 % \end{macro}
1749 % \end{macro}
1750 % \end{macro}
1751 % \end{macro}
1752 % \end{macro}
1753 % \end{macro}
1754 % \end{macro}
1755 % \end{macro}
1756 % \end{macro}
1757 % \end{macro}
1758 % \end{macro}
1759 % \end{macro}
1760 % \end{macro}
1761 % \end{macro}
1762 % \end{macro}
1763 % \end{macro}
1764 % \end{macro}
1765 % \end{macro}
1766 % \end{macro}
1767 % \end{macro}
1768 % \end{macro}
1769 % \end{macro}
1770 % \end{macro}
1771 % \end{macro}
1772 % \end{macro}
1773 %
1774 % \subsubsection{Font Selection}
1775 %
1776 % \begin{macro}{\unifamily}
1777 % \Newestchanges{v2.0}{98/08/01}{Defined \Lcs{unifamily} to
1778 % fasilitate easier and stronger font loading.}
1779 % The real command. This switches to \universal{} font, and can
1780 % be used by authors in conjunction with \LaTeXe{} primitive
1781 % \Lcs{selectfont}. \Lcs{unifamily} is used below in \Lcs{textuni...},
1782 % and \Lcs{uni...}.
1783 %
1784 % \begin{macro}{\unifamilydefault}
1785 % \changes{v2.0}{98/08/01}{Defined to make package conform to
1786 % \LaTeXe{} better.}
1787 % Here \Lcs{unifamilydefault} is by default defined to be
1788 % \texttt{uni}, but could ofcourse expand to something
1789 % different. However, the expnasion should be a font family (like
1790 % \texttt{cmr}) or the like.
1791 %
1792 % \begin{macrocode}
1793 \DeclareRobustCommand\unifamily{%
1794 \not@math@alphabet\unifamily\relax%
1795 \fontfamily\unifamilydefault\selectfont}
1796 \newcommand{\unifamilydefault}{uni}
1797
1798 % \end{macrocode}
1799 % \end{macro}
1800 % \end{macro}
1801 %
1802 % \begin{macro}{\uniseries}
1803 % \Newestchanges{v2.0}{98/08/01}{Defined \Lcs{uniseries} to
1804 % fasilitate easier and stronger font loading.}
1805 % If the \texttt{bold} option is used with the \textsf{uni} package,
1806 % then this command switches to whatever series is defined via
1807 % \Lcs{uniseriesdefault} (default is \textbf{bold} series), otherwise
1808 % it does nothing. Since this command is used in \Lcs{textuni} and
1809 % \Lcs{uni}, this way will not allow anyother series if \texttt{bold}
1810 % option is given, expect \textbf{bold}. However, if \texttt{bold}
1811 % isn't given, then \emph{any} series is possible.
1812 %
1813 % \begin{macro}{\uniseriesdefault}
1814 % \changes{v2.0}{98/08/01}{Defined to make package conform to
1815 % \LaTeXe{} better.}
1816 % Here \Lcs{uniseriesdefault} is defined to be \texttt{b}
1817 % (\textbf{medium}) if the \texttt{bold} option \emph{was} given to
1818 % \textsf{uni} package. Otherwise it expans to nothing
1819 % (\Lcs{relax}). This macro is used by \Lcs{uniseries}.
1820 % \begin{macrocode}
1821 \ifm@dium
1822 \let\uniseries\relax
1823 \let\uniseriesdefault\relax
1824 \else
1825 \DeclareRobustCommand\uniseries{%
1826 \not@math@alphabet\uniseries\relax%
1827 \fontseries\uniseriesdefault\selectfont}
1828 \newcommand{\uniseriesdefault}{b}
1829 \fi
1830
1831 % \end{macrocode}
1832 % \end{macro}
1833 % \end{macro}
1834 %
1835 % \begin{macro}{\unishape}
1836 % \Newestchanges{v2.0}{98/08/01}{Defined \Lcs{unishape} to
1837 % fasilitate easier and stronger font loading.}
1838 % If the \texttt{strict} option is used with the \textsf{uni} package,
1839 % then this command switches to whatever shape is defined via
1840 % \Lcs{unishapedefault} (default is \emph{strict} shape), otherwise
1841 % it does nothing. Since this command is used in \Lcs{textuni} and
1842 % \Lcs{uni}, this way will not allow anyother shape if \texttt{strict}
1843 % option is given, expect \emph{strict}. However, if \texttt{strict}
1844 % isn't given, then \emph{any} series is possible.
1845 %
1846 % \begin{macro}{\unishapedefault}
1847 % \changes{v2.0}{98/08/01}{Defined to make package conform to
1848 % \LaTeXe{} better.}
1849 % Here \Lcs{unishapedefault} is defined to be \texttt{st}
1850 % (\emph{strict}) if the \texttt{strict} option \emph{was} given to
1851 % \textsf{uni} package. Otherwise it expans to nothing
1852 % (\Lcs{relax}). This macro is used by \Lcs{unishape}.
1853 % \begin{macrocode}
1854 \ifstri@t
1855 \DeclareRobustCommand\unishape{%
1856 \not@math@alphabet\unishape\relax%
1857 \fontseries\unishapedefault\selectfont}
1858 \newcommand{\unishapedefault}{st}
1859 \else
1860 \let\unishape\relax
1861 \let\unishapedefault\relax
1862 \fi
1863 % \end{macrocode}
1864 % \end{macro}
1865 % \end{macro}
1866 %
1867 % \begin{macro}{\stshape}
1868 % \changes{v2.0}{98/08/01}{Defined \Lcs{stshape} to make package
1869 % conform to \LaTeXe{} better.}
1870 % This command selects the shape holded by \Lcs{stdefault} (defaults
1871 % to \texttt{st} --- strict) in any family or series. This is used by
1872 % commands \Lcs{textunist}, \Lcs{textunibst}, \Lcs{unist}, and
1873 % \Lcs{unibst}.
1874 %
1875 % \begin{macro}{\stdefault}
1876 % \changes{v2.0}{98/08/01}{Defined to make package conform to
1877 % \LaTeXe{} better.}
1878 % The shape selected by \Lcs{stshape}.
1879 % \begin{macrocode}
1880 \DeclareRobustCommand\stshape{%
1881 \not@math@alphabet\stshape\relax%
1882 \fontshape\stdefault\selectfont}
1883 \newcommand{\stdefault}{st}
1884
1885 % \end{macrocode}
1886 % \end{macro}
1887 % \end{macro}
1888 %
1889 % \paragraph{Local Font Selection}
1890 % \begin{macro}{\textuni}
1891 % \Newestchanges{v2.0}{98/08/01}{\Lcs{textbayer} changed to
1892 % \Lcs{textuni}, and defined using \LaTeXe{} scheme.}
1893 % \begin{macro}{\textunirm}
1894 % \begin{macro}{\textunibf}
1895 % \begin{macro}{\textunisl}
1896 % \begin{macro}{\textunisc}
1897 % \begin{macro}{\textunibsl}
1898 % \begin{macro}{\textunibsc}
1899 % These are the macros that switches to \universal{} font,
1900 % locally, i.e., for a few words, in various series and shapes. The
1901 % names are pretty selfexplanetory, I think. These macros will
1902 % \emph{always} change the font series and shape accroding to it's
1903 % definiton, nothing more, and nothing less, exepct \Lcs{textuni},
1904 % which only changes the cuurent family to \texttt{uni}. Notice it
1905 % simply uses the \Lcs{unifamily} command to do the trick.
1906 %
1907 % The second argument to \Lcs{DeclareTextFontCommand} should not
1908 % normally contain commands that typeset, or commands not relevant to
1909 % the selection of fonts, but to make up for the non--default
1910 % characeters and character placements, we \emph{do} include
1911 % \Lcs{uni@init} in the argument.
1912 %
1913 % Notice that only \Lcs{textuni} is defined if option \texttt{strict}
1914 % was given to \textsf{uni} package.
1915 % \begin{macrocode}
1916 \DeclareTextFontCommand{\textuni}{%
1917 \uni@init\unifamily\uniseries\unishape}
1918 \ifstri@t
1919 \else
1920 \DeclareTextFontCommand{\textunirm}{%
1921 \uni@init\unifamily\mdseries\upshape}
1922 \DeclareTextFontCommand{\textunibf}{%
1923 \uni@init\unifamily\bfseries\upshape}
1924 \DeclareTextFontCommand{\textunisl}{%
1925 \uni@init\unifamily\mdseries\slshape}
1926 \DeclareTextFontCommand{\textunisc}{%
1927 \uni@init\unifamily\mdseries\scshape}
1928 \DeclareTextFontCommand{\textunist}{%
1929 \uni@init\unifamily\mdseries\stshape}
1930 \DeclareTextFontCommand{\textunibsl}{%
1931 \uni@init\unifamily\bfseries\slshape}
1932 \DeclareTextFontCommand{\textunibsc}{%
1933 \uni@init\unifamily\bfseries\scshape}
1934 \DeclareTextFontCommand{\textunibst}{%
1935 \uni@init\unifamily\bfseries\stshape}
1936 \fi
1937
1938 % \end{macrocode}
1939 % \end{macro}
1940 % \end{macro}
1941 % \end{macro}
1942 % \end{macro}
1943 % \end{macro}
1944 % \end{macro}
1945 % \end{macro}
1946 %
1947 % \paragraph{Global Font Selection}
1948 % \begin{macro}{\uni}
1949 % \Newestchanges{v2.0}{98/08/01}{\Lcs{bayer} changed to \Lcs{uni}}
1950 % \begin{macro}{\unibf}
1951 % \begin{macro}{\unibf}
1952 % \begin{macro}{\unisl}
1953 % \begin{macro}{\unisc}
1954 % \begin{macro}{\unibsl}
1955 % \begin{macro}{\unibsc}
1956 % These macros works as \Lcs{textuni...} above, execpt the take effect
1957 % from the point used, to the end of the current group. Again font
1958 % selection is \emph{exactly} as given in the command name, execpt for
1959 % \Lcs{uni} which only changes the family.
1960 %
1961 % Notice that only \Lcs{uni} is defined if option \texttt{strict}
1962 % was given to \textsf{uni} package.
1963 % \begin{macrocode}
1964 \DeclareOldFontCommand{\uni}{%
1965 \uni@init\unifamily\uniseries\unishape}{}
1966 \ifstri@t
1967 \else
1968 \DeclareOldFontCommand{\unirm}{%
1969 \uni@init\unifamily\mdseries\upshape}{}
1970 \DeclareOldFontCommand{\unibf}{%
1971 \uni@init\unifamily\bfseries\upshape}{}
1972 \DeclareOldFontCommand{\unisl}{%
1973 \uni@init\unifamily\mdseries\slshape}{}
1974 \DeclareOldFontCommand{\unisc}{%
1975 \uni@init\unifamily\mdseries\scshape}{}
1976 \DeclareOldFontCommand{\unist}{%
1977 \uni@init\unifamily\mdseries\stshape}{}
1978 \DeclareOldFontCommand{\unibsl}{%
1979 \uni@init\unifamily\bfseries\slshape}{}
1980 \DeclareOldFontCommand{\unibsc}{%
1981 \uni@init\unifamily\bfseries\scshape}{}
1982 \DeclareOldFontCommand{\unibst}{%
1983 \uni@init\unifamily\bfseries\stshape}{}
1984 \fi
1985
1986 % \end{macrocode}
1987 % \end{macro}
1988 % \end{macro}
1989 % \end{macro}
1990 % \end{macro}
1991 % \end{macro}
1992 % \end{macro}
1993 % \end{macro}
1994 %
1995 % \paragraph{\texttt{default} Option Font}
1996 %
1997 % \begin{macro}{\ifdef@ult}
1998 % \begin{macro}{\def@ulttrue}
1999 % \begin{macro}{\def@ultfalse}
2000 % \begin{macro}{\ifm@dium}
2001 % \begin{macro}{\m@diumtrue}
2002 % \begin{macro}{\m@diumfalse}
2003 % Now if you gave the \texttt{default} option to the package,
2004 % \Lcs{ifdef@ult} evaluates to true, and so we setup the default font
2005 % to be \universal{}, in the medium or bold version, depending
2006 % on wether you gave the \texttt{medium} option or not.
2007 % \begin{macrocode}
2008 \ifdef@ult
2009 \renewcommand{\familydefault}{uni}
2010 \ifm@dium\renewcommand{\seriesdefault}{m}
2011 \else\renewcommand{\seriesdefault}{b}\fi
2012 \renewcommand{\itdefault}{sl}
2013 % \end{macrocode}
2014 % \end{macro}
2015 % \end{macro}
2016 % \end{macro}
2017 % \end{macro}
2018 % \end{macro}
2019 % \end{macro}
2020 %
2021 % \begin{macro}{\cmrfamily}
2022 % \begin{macro}{\cmrdefault}
2023 % \begin{macro}{\textcmr}
2024 % \begin{macro}{\cmr}
2025 % We define \Lcs{cmrfamily} and \Lcs{cmrdefault} to change back to
2026 % Computer Modern Roman font, if so wanted. The macros \Lcs{textcmr}
2027 % and \Lcs{cmr} are the logical extensions of \Lcs{cmrfamily}.
2028 % \begin{macrocode}
2029 \DeclareRobustCommand\cmrfamily{%
2030 \not@math@alphabet\cmrfamily\relax%
2031 \fontencoding\cmrenc\fontfamily\cmrdefault\selectfont}
2032 \newcommand{\cmrdefault}{cmr}
2033 \newcommand{\cmrenc}{OT1}
2034 \DeclareTextFontCommand{\textcmr}{\cmrfamily}
2035 \DeclareOldFontCommand{\cmr}{\cmrfamily}{}
2036 \fi
2037
2038 % \end{macrocode}
2039 % \end{macro}
2040 % \end{macro}
2041 % \end{macro}
2042 % \end{macro}
2043 %\iffalse
2044 %</package>
2045 %\fi
2046 % \subsection{The Font Definition Files --- \texttt{*uni.fd}}
2047 %
2048 % These files is needed in the New Font Selection Scheme, used by
2049 % \LaTeXe{}, but really isn't nessecary for pure \TeX{} or \LaTeX2.09
2050 % users, but since those are treathned races, I did put in the extra
2051 % effort and made the files.
2052 %
2053 % What they really do, is to speciffy what font driver file should be
2054 % loaded when the user swichtes font. Notice, alot of the font shapes
2055 % etc.\ avaliable in the Computer Modern Roman scheme isn't avaliable
2056 % in the \universal{} font, so we substitute with whatever is
2057 % closest, something from the \universal{} or Computer Modern
2058 % Roman scheme.
2059 %
2060 % The reason why the file names may look a bit wierd, is to make the
2061 % font conform to the standard set by \texttt{fontname} used by
2062 % \texttt{kpathsea} in most \TeX{}, \LaTeX{}, and \MF{} systems.
2063 %
2064 % \paragraph*{Math Encoding}\file{oml.fd}
2065 % First comes the the font definition file for \emph{math}, but since
2066 % no math characters is defined in the \universal{} font, we
2067 % substitute with the relevant fonts from Computer Modern Roman.
2068 %\iffalse
2069 %<*oml>
2070 %\fi
2071 % \begin{macrocode}
2072 \ProvidesFile{omluni.fd}
2073 [\unifiledate\space\unifileversion\space Non Standard LaTeX font definitions]
2074 \DeclareFontFamily{OML}{uni}{\skewchar\font127 }
2075 \DeclareFontShape{OML}{uni}{m}{n}{<-> ssub * cmm/m/it}{}
2076 \DeclareFontShape{OML}{uni}{m}{it}{<-> ssub * cmm/m/it}{}
2077 \DeclareFontShape{OML}{uni}{m}{sl}{<-> ssub * cmm/m/it}{}
2078 \DeclareFontShape{OML}{uni}{m}{sc}{<-> ssub * cmm/m/it}{}
2079 \DeclareFontShape{OML}{uni}{bx}{n}{<-> ssub * cmm/b/it}{}
2080 \DeclareFontShape{OML}{uni}{bx}{it}{<-> ssub * cmm/b/it}{}
2081 \DeclareFontShape{OML}{uni}{bx}{sl}{<-> ssub * cmm/b/it}{}
2082 \DeclareFontShape{OML}{uni}{bx}{sc}{<-> ssub * cmm/b/it}{}
2083
2084 % \end{macrocode}
2085 %\iffalse
2086 %</oml>
2087 %\fi
2088 %
2089 % \paragraph*{Symbols Encoding}\file{omsuni.fd}
2090 % Next is the definitions for \emph{symbols} fonts, but as above,
2091 % there is no seperate symbol font defined for the \universal{}
2092 % font, so we substitute for default.
2093 %\iffalse
2094 %<*oms>
2095 %\fi
2096 % \begin{macrocode}
2097 \ProvidesFile{omsuni.fd}
2098 [\unifiledate\space\unifileversion\space Non Standard LaTeX font definitions]
2099 \DeclareFontFamily{OMS}{uni}{\skewchar\font48 }
2100 \DeclareFontShape{OMS}{uni}{m}{n}{<-> ssub * cmsy/m/n}{}
2101 \DeclareFontShape{OMS}{uni}{m}{it}{<-> ssub * cmsy/m/n}{}
2102 \DeclareFontShape{OMS}{uni}{m}{sl}{<-> ssub * cmsy/m/n}{}
2103 \DeclareFontShape{OMS}{uni}{m}{sc}{<-> ssub * cmsy/m/n}{}
2104 \DeclareFontShape{OMS}{uni}{bx}{n}{<-> ssub * cmsy/b/n}{}
2105 \DeclareFontShape{OMS}{uni}{bx}{it}{<-> ssub * cmsy/b/n}{}
2106 \DeclareFontShape{OMS}{uni}{bx}{sl}{<-> ssub * cmsy/b/n}{}
2107 \DeclareFontShape{OMS}{uni}{bx}{sc}{<-> ssub * cmsy/b/n}{}
2108
2109 % \end{macrocode}
2110 %\iffalse
2111 %</oms>
2112 %\fi
2113 %
2114 % \paragraph*{Normal Encoding}\file{ot1uni.fd}
2115 % Now the definitions for the normal font. This \emph{is} ofcourse
2116 % defined, anything else would be ludicruis. It is very straight
2117 % forward most of the way. However, notice the substitutions at the
2118 % end of the file.
2119 %
2120 %\iffalse
2121 %<*ot1>
2122 %\fi
2123 % \begin{macrocode}
2124 \ProvidesFile{ot1uni.fd}%
2125 [\unifiledate\space\unifileversion\space Non Standard LaTeX font definitions]
2126 \DeclareFontFamily{OT1}{uni}{\hyphenchar\font45 }
2127
2128 % \end{macrocode}
2129 % Next comes the specifications of what to load in \textunirm{normal
2130 % upright shape}
2131 % \begin{macrocode}
2132 \DeclareFontShape{OT1}{uni}{m}{n}{
2133 <5><6><7><8>fulmr8
2134 <9>fulmr9
2135 <10><10.95>fulmr10
2136 <12><14.4>fulmr12
2137 <17.28><20.74><24.88>fulmr17
2138 }{}
2139
2140 % \end{macrocode}
2141 % Next comes the specifications of what to load in \textunisl{normal
2142 % slanted shape}
2143 % \begin{macrocode}
2144 \DeclareFontShape{OT1}{uni}{m}{sl}{
2145 <5><6><7><8>fulmo8
2146 <9>fulmo9
2147 <10><10.95>fulmo10
2148 <12><14.4>fulmo12
2149 <17.28><20.74><24.88>fulmo17
2150 }{}
2151
2152 % \end{macrocode}
2153 % Now for \textunisc{small caps medium} definitions.
2154 % \begin{macrocode}
2155 \DeclareFontShape{OT1}{uni}{m}{sc}{
2156 <5><6><7><8>fulmc8
2157 <9>fulmc9
2158 <10><10.95>fulmc10
2159 <12><14.4>fulmc12
2160 <17.28><20.74><24.88>fulmc17
2161 }{}
2162
2163 % \end{macrocode}
2164 % Now for \textunist{strict medium} definitions.
2165 % \begin{macrocode}
2166 \DeclareFontShape{OT1}{uni}{m}{st}{
2167 <5><6><7><8>fulmst8
2168 <9>fulmst9
2169 <10><10.95>fulmst10
2170 <12><14.4>fulmst12
2171 <17.28><20.74><24.88>fulmst17
2172 }{}
2173
2174 % \end{macrocode}
2175 % Next comes the specifications of what to load in \textunibf{bold
2176 % face upright shape}
2177 % \begin{macrocode}
2178 \DeclareFontShape{OT1}{uni}{b}{n}{
2179 <5><6><7><8>fulbr8
2180 <9>fulbr9
2181 <10><10.95>fulbr10
2182 <12><14.4>fulbr12
2183 <17.28><20.74><24.88>fulbr17
2184 }{}
2185
2186 % \end{macrocode}
2187 % Next comes the specifications of what to load in
2188 % \textunibsl{bold face slanted shape}
2189 % \begin{macrocode}
2190 \DeclareFontShape{OT1}{uni}{b}{sl}{
2191 <5><6><7><8>fulbo8
2192 <9>fulbo9
2193 <10><10.95>fulbo10
2194 <12><14.4>fulbo12
2195 <17.28><20.74><24.88>fulbo17
2196 }{}
2197
2198 % \end{macrocode}
2199 % And \textunibsc{small caps bold face} definitions.
2200 % \begin{macrocode}
2201 \DeclareFontShape{OT1}{uni}{b}{sc}{
2202 <5><6><7><8>fulbc8
2203 <9>fulbc9
2204 <10><10.95>fulbc10
2205 <12><14.4>fulbc12
2206 <17.28><20.74><24.88>fulbc17
2207 }{}
2208
2209 % \end{macrocode}
2210 % And \textunibst{strict bold face} definitions.
2211 % \begin{macrocode}
2212 \DeclareFontShape{OT1}{uni}{b}{st}{
2213 <5><6><7><8>fulbst8
2214 <9>fulbst9
2215 <10><10.95>fulbst10
2216 <12><14.4>fulbst12
2217 <17.28><20.74><24.88>fulbst17
2218 }{}
2219
2220 % \end{macrocode}
2221 %
2222 % Now for the substitutions. This is straight forward, that
2223 % is, upright slanted is substituted by upright; bold face condensed,
2224 % and bold face extra is substituted by boldface; and italic by
2225 % slanted\footnote{As mentioned earlier, one of the characteristics
2226 % of this dont, is that it has no serifs, so it would be strange
2227 % to include italics in this font.}.
2228 % \begin{macrocode}
2229 \DeclareFontShape{OT1}{uni}{m}{it}{<->ssub*uni/m/sl}{}
2230 \DeclareFontShape{OT1}{uni}{m}{ui}{<->ssub*uni/m/n}{}
2231 \DeclareFontShape{OT1}{uni}{b}{it}{<->ssub*uni/b/sl}{}
2232 \DeclareFontShape{OT1}{uni}{bx}{n}{<->ssub*uni/b/n}{}
2233 \DeclareFontShape{OT1}{uni}{bx}{sl}{<->ssub*uni/b/sl}{}
2234 \DeclareFontShape{OT1}{uni}{bx}{sc}{<->ssub*uni/b/sc}{}
2235 \DeclareFontShape{OT1}{uni}{bx}{st}{<->ssub*uni/b/st}{}
2236 \DeclareFontShape{OT1}{uni}{bx}{it}{<->ssub*uni/b/sl}{}
2237 \DeclareFontShape{OT1}{uni}{bc}{ui}{<->ssub*uni/b/n}{}
2238 \DeclareFontShape{OT1}{uni}{sbc}{n}{<->ssub*uni/b/n}{}
2239
2240 % \end{macrocode}
2241 %\iffalse
2242 %</ot1>
2243 %\fi
2244 %
2245 % \paragraph*{T1 encoding}\file{t1uni.fd}
2246 % Since the T1 encoding relies on the font scheme of \texttt{exbase}
2247 % (EC fonts), and \texttt{dxbase} (DC fonts), by J\"org Knappen and
2248 % Nobert Schwartz, which is a whole new way of making fonts, that
2249 % requires some special messaures that I couldn't incorporate into the
2250 % \universal{} font without a whole lot of work, I chose not to
2251 % make support for the T1 scheme. Therefore this file issues a warning
2252 % and then exists the whole \LaTeX{} run. This may seem a bit drastic,
2253 % but it ensures that you don't get screwed up resuats.
2254 %\iffalse
2255 %<*t1>
2256 %\fi
2257 % \begin{macrocode}
2258 \ProvidesFile{t1uni.fd}
2259 [\unifiledate\space\unifileversion\space Non Standard LaTeX font definitions]
2260 \typeout{WARNING: universal font not avaliable in t1 definition.}
2261 \typeout{WARNING: exiting, correct your source.}
2262 \stop
2263
2264 % \end{macrocode}
2265 %\iffalse
2266 %</t1>
2267 %\fi
2268 %
2269 % \paragraph*{U Encoding}\file{uuni.fd}
2270 % This is similar to the normal encoding above, execpt encoding is U.
2271 %\iffalse
2272 %<*u>
2273 %\fi
2274 % \begin{macrocode}
2275 \ProvidesFile{uuni.fd}
2276 [\unifiledate\space\unifileversion\space Non Standard LaTeX font definitions]
2277 \DeclareFontFamiliy{U}{uni}{\hyphenchar\font45}
2278 \DeclareFontShape{U}{uni}{m}{n}{
2279 <5><6><7><8>fulmr8
2280 <9>fulmr9
2281 <10><10.95>fulmr10
2282 <12><14.4>fulmr12
2283 <17.28><20.74><24.88>fulmr17
2284 }{}
2285 \DeclareFontShape{U}{uni}{m}{sl}{
2286 <5><6><7><8>fulmo8
2287 <9>fulmo9
2288 <10><10.95>fulmo10
2289 <12><14.4>fulmo12
2290 <17.28><20.74><24.88>fulmo17
2291 }{}
2292 \DeclareFontShape{U}{uni}{m}{sc}{
2293 <5><6><7><8>fulmc8
2294 <9>fulmc9
2295 <10><10.95>fulmc10
2296 <12><14.4>fulmc12
2297 <17.28><20.74><24.88>fulmc17
2298 }{}
2299 \DeclareFontShape{U}{uni}{m}{st}{
2300 <5><6><7><8>fulmst8
2301 <9>fulmst9
2302 <10><10.95>fulmst10
2303 <12><14.4>fulmst12
2304 <17.28><20.74><24.88>fulmst17
2305 }{}
2306 \DeclareFontShape{U}{uni}{b}{n}{
2307 <5><6><7><8>fulbr8
2308 <9>fulbr9
2309 <10><10.95>fulbr10
2310 <12><14.4>fulbf12
2311 <17.28><20.74><24.88>fulbr17
2312 }{}
2313 \DeclareFontShape{U}{uni}{b}{sl}{
2314 <5><6><7><8>fulbo8
2315 <9>fulbo9
2316 <10><10.95>fulbo10
2317 <12><14.4>fulbo12
2318 <17.28><20.74><24.88>fulbo17
2319 }{}
2320 \DeclareFontShape{U}{uni}{b}{sc}{
2321 <5><6><7><8>fulbc8
2322 <9>fulbc9
2323 <10><10.95>fulbc10
2324 <12><14.4>fulbc12
2325 <17.28><20.74><24.88>fulbc17
2326 }{}
2327 \DeclareFontShape{U}{uni}{b}{st}{
2328 <5><6><7><8>fulbst8
2329 <9>fulbst9
2330 <10><10.95>fulbst10
2331 <12><14.4>fulbst12
2332 <17.28><20.74><24.88>fulbst17
2333 }{}
2334 \DeclareFontShape{U}{uni}{m}{it}{<->ssub*uni/m/sl}{}
2335 \DeclareFontShape{U}{uni}{m}{ui}{<->ssub*uni/m/n}{}
2336 \DeclareFontShape{U}{uni}{b}{it}{<->ssub*uni/b/sl}{}
2337 \DeclareFontShape{U}{uni}{bx}{n}{<->ssub*uni/b/n}{}
2338 \DeclareFontShape{U}{uni}{bx}{sl}{<->ssub*uni/b/sl}{}
2339 \DeclareFontShape{U}{uni}{bx}{sc}{<->ssub*uni/b/sc}{}
2340 \DeclareFontShape{U}{uni}{bx}{st}{<->ssub*uni/b/st}{}
2341 \DeclareFontShape{U}{uni}{bx}{it}{<->ssub*uni/b/sl}{}
2342 \DeclareFontShape{U}{uni}{bc}{ui}{<->ssub*uni/b/n}{}
2343 \DeclareFontShape{U}{uni}{sbc}{n}{<->ssub*uni/b/n}{}
2344
2345 % \end{macrocode}
2346 %\iffalse
2347 %</u>
2348 %\fi
2349
2350 %\iffalse
2351 %<*unibase>
2352 %\fi
2353 %
2354 % \section{The Font Macros and Programs}
2355 %
2356 % \subsection{Base file --- \texttt{unibase.mf}}
2357 % \file{unibase.mf}
2358 % \Newestchanges{v2.0}{98/08/01}{Base file \texttt{unibase.mf} swa the
2359 % light of day, providing many driver, font control, and drawing
2360 % macros.}
2361 %
2362 % This file defines all the macroes used in the source programs
2363 % for the font. I could have chosen to use \texttt{cmbase}, but
2364 % rather then defining a lot of variables that isn't used, I chose
2365 % to make a base file myself. Also, some af the drawing macros
2366 % presented here still needed to be defined. This file could be
2367 % made into a binary file, but I think that would be a bit to
2368 % much.
2369 %
2370 % First we define the variable \texttt{unibase}, to check later if
2371 % this file has been loaded.
2372 % \begin{macrocode}
2373 unibase := 1 ;
2374 string univer ;
2375 univer = "2.0" ;
2376 if proofing > 1 :
2377 message "Base file for font `universal', version " & univer ;
2378 fi
2379 % \end{macrocode}
2380 % Secondly, we define the variables \texttt{medium} and \texttt{bold}
2381 % to control the font serie to be produced. If \texttt{medium} is set
2382 % to \textsl{true} in the driver file, then a medium series
2383 % (\Lcs{mdseries}) font will be produced, while \texttt{bold} ofcourse
2384 % generates a boldface font. The medium serie may very well be
2385 % more suitiable for long texts then the default font. In $10pt$
2386 % the default thickness of the pen is $\frac{5}{6}pt\approx 0.8pt$
2387 % wide for the minuscles, and a bit wider for the majuscules. If
2388 % \texttt{medium} is \texttt{true} then it is $\frac{5}{9}pt\approx
2389 % 0.6pt$ wide.
2390 %
2391 % Booleans \texttt{slant} and \texttt{caps} is used to control the
2392 % shape of the font. If both is \textsl{false} a normal (upright)
2393 % shaped font (\Lcs{upshape}) font is generated. If \texttt{slant} or
2394 % \texttt{caps} is true, either a slanted (\Lcs{slshape}) or small
2395 % caps (\Lcs{scshape}) shaped font is generated. The values of these
2396 % booleans should be set in the driver file, \emph{before} any call to
2397 % \texttt{start\_font}.
2398 % \begin{macrocode}
2399 boolean medium , slant , bold , caps , strict ;
2400 medium := false ; slant := false ; bold := false ;
2401 caps := false ; strict := false ;
2402 % \end{macrocode}
2403 % And then the font is identified as \texttt{"universal"}.
2404 % \begin{macrocode}
2405 font_identifier:="universal";
2406
2407 % \end{macrocode}
2408 %
2409 % \subsubsection{Font macros}
2410 %
2411 % These macroes have to do with the general initialization and
2412 % production of the font. The macros needed later on is
2413 % \texttt{start\_font}, and \texttt{end\_font}.
2414 %
2415 % The general idea is, that as little as possible should be
2416 % defined in the driver file, so that the font will have a uniform
2417 % look in all series, shapes, and sizes. All that is needed in the
2418 % driver file is to define some of the above booleans to true, call
2419 % the macro \texttt{start\_font} giving it the size (in \texttt{pt\#})
2420 % of the font as argument, include of source files, and call the macro
2421 % \texttt{end\_font}.
2422 %
2423 % \begin{fontmacro}{start\_font}
2424 % This is the macro for setting up the needed dimensions and
2425 % variables. First we take into acount the \textit{mode} used, by
2426 % calling \texttt{mode\_setup}, then we define the correct,
2427 % \textit{sharped} units, bassed on the argument to the macro, wich
2428 % should be the size of the font in \texttt{pt\#}.
2429 % \begin{macrocode}
2430 def start_font(expr size) =
2431 u# := size/18 ;
2432 mode_setup;
2433 x_h# := 8u# ; x_w# := 8u# ; m_w# := 11u# ; l_h# := 13u# ;
2434 p_d# := 5u# ; l_s# := 2u# ; r_s# := 2u# ; A_w# := 10u# ;
2435 M_w# := 14u# ; c_d# := 3u#/2 ; C_d# := 2u# ;
2436 % \end{macrocode}
2437 % We test to see, if any of the above booleans is set to
2438 % \textsl{true}, and if so, takes appropiate action. If
2439 % \texttt{medium} is set, the pens gets thinner, thereby producing
2440 % a lighter font. If \texttt{bold} the pens gets thicker, and so a
2441 % heavier font is produced. If \texttt{slant} is set, then the
2442 % inclination (\texttt{slantratio}) of the font is assigned.
2443 %
2444 % No other dimensions then the pen thickness should be differ among
2445 % the \emph{medium} and \emph{bold} font. However, two dimension
2446 % \texttt{c\_d} and \texttt{C\_d} is always set to the same value as
2447 % the pens would be in the current size, if the font to be made was
2448 % \emph{bold}.
2449 %
2450 % Also notice, that two different pens thicknesses is defined
2451 % \texttt{p\_t} and \texttt{P\_t}, so that the majuscules may come out
2452 % a bit thicker then the minuscles. Ofcourse this adds to the
2453 % complexity of the drawing macroes defined below, but in a very
2454 % straight forward way, so I dont't think it will be a problem.
2455 %
2456 % Then we set \texttt{currenttransform} to take into account the
2457 % varibale \texttt{slantratio}.
2458 % \begin{macrocode}
2459 if (medium and bold) :
2460 message "Font can't be both medium and bold, making medium font." ;
2461 medium := true ;
2462 fi;
2463 if medium : p_t# := 2u#/3 ; P_t# := u# ; P_T# := u# ; fi ;
2464 if bold : P_T# := 3u#/2 ; p_t# := 3u#/2 ; P_t# := 2u# ; fi ;
2465 a_h# := x_h#+p_t#/2 ;
2466 if (slant and (caps or strict)) : double_shape := 1 ; fi ;
2467 if (caps and (slant or strict)) : double_shape := 1 ; fi ;
2468 if (strict and (caps or slant)) : double_shape := 1 ; fi ;
2469 if known double_shape :
2470 message "Font cannot be two shapes, doing upright shape." ;
2471 slant := false ; caps := false ; strict := false ;
2472 fi ;
2473 if slant : slantratio := 1/6 ; else: slantratio := 0 ; fi ;
2474 currenttransform := identity slanted slantratio ;
2475 % \end{macrocode}
2476 % We need to set some paramteres for the Font Metric File
2477 % (\texttt{*.tfm}), so that \TeX{} knows how to typeset the characters
2478 % of the font. This is pretty straight, as desribed in the \MFbook{}.
2479 % \begin{macrocode}
2480 font_x_height 8u# ; font_quad 12u#+2u# ;
2481 font_normal_space 8u# ; font_normal_stretch 3u# ;
2482 font_normal_shrink 2u# ; font_extra_space 2u# ;
2483 font_coding_scheme:="TeX text" ; font_size 18u# ;
2484 % \end{macrocode}
2485 % Next we assigns some variables we need later on, defines the
2486 % \emph{un\/}sharped (mode dependent) units, declares some variables
2487 % for later use, and call the macro \texttt{do\_pens} to save the pens
2488 % needed by the font.
2489 % \begin{macrocode}
2490 spanish_shriek:=hex"3C" ; spanish_query:=hex"3E";
2491 define_pixels(u,x_h,x_w,m_w,l_h,p_d,A_w,M_w,c_d,C_d,a_h);
2492 define_blacker_pixels(p_t,P_t,P_T);
2493 picture e_pic, a_pic ;
2494 do_pens ;
2495 % \end{macrocode}
2496 % The definitions of \texttt{extra\_beginchar} and
2497 % \texttt{extra\_endchar} is responsible for the extra stem width
2498 % of the font. Since this font is pretty uniform in width, we
2499 % can easily do this globally. I got the trick from \texttt{cmbase},
2500 % where you instead use \texttt{adjust\_fit} after \texttt{beginchar}.
2501 % However, for the above reasons, and because it lessens the source
2502 % programs considerably, I chose to do it this way.
2503 %
2504 % The inclusion of the assignement
2505 % \begin{verbatim}
2506 % charcode := charcode + code_offset ;
2507 % \end{verbatim}
2508 % is for the small caps font, so that characters may be placed at
2509 % non--standard positions in the font. When we are not making a small
2510 % caps font, \texttt{code\_offset} should ofcourse be set to 0 (zero).
2511 % \begin{macrocode}
2512 extra_beginchar := extra_beginchar&"fit_font;" ;
2513 extra_endchar := extra_endchar&
2514 "end_fit_font;extra_proof_rules;charcode:=charcode+code_offset;" ;
2515 code_offset := 0 ;
2516 enddef;
2517
2518 % \end{macrocode}
2519 % \end{fontmacro}
2520 %
2521 % \begin{fontmacro}{do\_pens}
2522 % This macro is used to set up the pens needed in this font. It is
2523 % called every time one of the dimensions \texttt{p\_t\#} or
2524 % \texttt{P\_t\#} is changed, since those are the \texttt{p/P}en
2525 % \texttt{t/T}hicknesses.
2526 %
2527 % Apart from saving a lot of time, as explained in the \MFbook{}, this
2528 % also garanties, that all pens are transformed correctly when the
2529 % font shape is \textsl{slanted}.
2530 % \begin{macrocode}
2531 def do_pens =
2532 clear_pen_memory ;
2533 pickup pencircle scaled P_T slanted slantratio ;
2534 UniPencir := savepen ;
2535 pickup pencircle scaled p_t slanted slantratio ;
2536 unipencir := savepen ;
2537 pickup pencircle scaled P_t slanted slantratio ;
2538 Unipencir := savepen ;
2539 pickup pensquare scaled P_T slanted slantratio ;
2540 UniPensqu := savepen ;
2541 pickup pensquare scaled p_t slanted slantratio ;
2542 unipensqu := savepen ;
2543 pickup pensquare scaled P_t slanted slantratio ;
2544 Unipensqu := savepen ;
2545 enddef ;
2546
2547 % \end{macrocode}
2548 % \end{fontmacro}
2549 %
2550 % \begin{fontmacro}{fit\_font}
2551 % This is the macro that changes the width of the font. It is
2552 % derived from \texttt{cmbase}s \texttt{normal\_adjust\_fit}, though
2553 % used in a slightly different way here. As explained in the \MFbook,
2554 % this is a very nice macro, since it is essesially an add--on
2555 % feature. We still specify the width of character in
2556 % \texttt{beginchar}, and \texttt{w} is still this dimension. However,
2557 % extra space is added to the character at both the left and right,
2558 % so that characters don't get too close.
2559 % \begin{macrocode}
2560 def fit_font =
2561 l := -hround(l_s#*hppp) ; interim xoffset := -l ;
2562 charwd := charwd+l_s#+r_s# ; r := l+hround(charwd*hppp) ;
2563 w := r-hround(r_s#*hppp) ;
2564 italcorr charht*slantratio ;
2565 enddef;
2566
2567 % \end{macrocode}
2568 % \end{fontmacro}
2569 % \begin{fontmacro}{end\_fit\_font}
2570 % Again, this is taken from \texttt{cmbase}, so that extra space is
2571 % added on the sides of the character.
2572 % \begin{macrocode}
2573 def end_fit_font =
2574 w := r+l ;
2575 enddef;
2576
2577 % \end{macrocode}
2578 % \end{fontmacro}
2579 %
2580 % \begin{fontmacro}{extra\_proof\_rules}
2581 % When doing proofing, it is nice to have some extra rules on the
2582 % proofs, so that errors and mistakes are easiler seen. This macro
2583 % does just that. A verticatal rule is produced at the left and
2584 % right edge, and through the middle of the character. Horizontal
2585 % rules are produced at one quarter, half, and three quarters of the
2586 % character height. Also, if the charcter extence below the baseline,
2587 % a rule is produced at one half the depth.
2588 % \begin{macrocode}
2589 def extra_proof_rules =
2590 if proofing > 1 :
2591 for xa = l,r,w/2: proofrule((xa,-d), (xa,h)) ; endfor ;
2592 for ya = 0,h/4,h/2,3h/4,h: proofrule((l,ya), (r,ya)) ; endfor ;
2593 if d > 0 :
2594 proofrule((l,-d/2), (r,-d/2)) ;
2595 proofrule((l,-d), (r,-d)) ;
2596 fi
2597 fi
2598 enddef ;
2599
2600 % \end{macrocode}
2601 % \end{fontmacro}
2602 % \begin{fontmacro}{end\_font}
2603 % \texttt{end\_font} is used at the end of the driver file, to set
2604 % ligatures, and end of the file. There is nothing fancy here. Do,
2605 % however, notice no `f' ligatures is defined, as in the Computer
2606 % Modern Roman. This is because they are not meant to be there,
2607 % because `ff' shouldn't come out as `\textunibf{f\kern-.25ex f}',
2608 % which looks bad, but rather as `\textunibf{ff}'.
2609 %
2610 % Also notice, that we haven't defined any characters \textunibf{``}
2611 % or \textunibf{''}, since we make these up of two \textunibf{`} or
2612 % \textunibf{'}. This gives us two more slots we can put other
2613 % characters into.
2614 %
2615 % Note, we preface the definition by \texttt{inner bye} so that we can
2616 % use this macro inside other macros (Kudo's to \TeX{}Live
2617 % developers).
2618 %
2619 % \begin{macrocode}
2620 inner bye;
2621 def end_font =
2622 ligtable "u" : "h" kern u# ;
2623 ligtable "f" : "f" kern 0 , "l" kern 0 , "i" kern 0 ;
2624 ligtable "o" : "l" kern u# ;
2625 ligtable "?" : "`" =: spanish_query ;
2626 ligtable "!" : "`" =: spanish_shriek ;
2627 ligtable "`" : "`" kern -l_s#-r_s#+p_t# ;
2628 ligtable "'" : "'" kern -l_s#-r_s#+p_t# ,
2629 "?" kern 2u#, "!" kern 2u# ;
2630 ligtable "-" : "-" =: hex"7B" ;
2631 ligtable hex"7B" : "-" =: hex"7C" ;
2632 ligtable hex"22" : hex"22" kern -l_s#-r_s# ;
2633 ligtable hex"5C" : hex"5C" kern -l_s#-r_s# ;
2634 ligtable hex"20" :
2635 "l" kern -3c_d#*cosd(15)-p_t#*sind(15)-l_s#-r_s# ,
2636 "L" kern -3c_d#*cosd(15)-p_t#*sind(15)-l_s#-r_s# ;
2637 bye
2638 enddef ;
2639
2640 % \end{macrocode}
2641 % \end{fontmacro}
2642 % \begin{fontmacro}{do\_caps}
2643 % This next macro is for the \textunisc{small caps} version of the
2644 % font. Most of it is taken from \texttt{csc.mf} of the Computer
2645 % Modern Font, though not as many dimensions have been redefined in
2646 % this macro. The idea is to utilize this macro whenever you want to
2647 % make a small caps version of majuscles. This macro is used in both
2648 % \texttt{unilow.mf} and \texttt{unilig.mf}.
2649 %
2650 % The argument to \texttt{do\_caps} is the number the new characters
2651 % should be offset from it's original position in the font.
2652 %
2653 % The last \texttt{do\_pens} is needed, since the programs uses saved
2654 % pens, and we have just redefined the dimensions for the pens.
2655 % \begin{macrocode}
2656 def do_caps(expr offset) =
2657 sc_shrink := x_h#/l_h# ;
2658 P_t# := sc_shrink*P_t# ; l_h# := sc_shrink*l_h# ;
2659 A_w# := sc_shrink*A_w# ; M_w# := sc_shrink*M_w# ;
2660 C_d# := sc_shrink*C_d# ;
2661 code_offset := offset ;
2662 define_pixels(l_h,A_w,M_w,C_d) ;
2663 define_blacker_pixels(P_t) ;
2664 do_pens ;
2665 enddef ;
2666
2667 % \end{macrocode}
2668 % \end{fontmacro}
2669 % \begin{fontmacro}{end\_do\_caps}
2670 % When ever we need to get back to the default setting of dimensions,
2671 % after the use of \texttt{do\_caps}, we call on this macro to do the
2672 % job. The idea is to multiply by \texttt{no\_shrink} which is the
2673 % reciprocal of \texttt{sc\_shrink}.
2674 % \begin{macrocode}
2675 def end_do_caps =
2676 no_shrink := 1/sc_shrink ;
2677 P_t# := no_shrink*P_t# ; l_h# := no_shrink*l_h# ;
2678 A_w# := no_shrink*A_w# ; M_w# := no_shrink*M_w# ;
2679 C_d# := no_shrink*C_d# ;
2680 code_offset := 0 ;
2681 define_pixels(l_h,A_w,M_w,C_d) ;
2682 define_blacker_pixels(P_t) ;
2683 do_pens ;
2684 enddef ;
2685
2686 % \end{macrocode}
2687 % \end{fontmacro}
2688 %
2689 % \subsubsection{Control macroes}
2690 %
2691 % These macroes are designed to ease the control of individual
2692 % characters.
2693 %
2694 % \begin{fontmacro}{no\_adjust}
2695 % If a character is not supposed to have extra space on either side
2696 % you can say \texttt{no\_adjust} just before
2697 % \begin{quote}
2698 % \texttt{extra\_beginchar := extra\_beginchar\&"}
2699 % \meta{your begin char stuff}\texttt{";}\\
2700 % \texttt{extra\_endchar := extra\_endchar\&"}
2701 % \meta{your stuff endchar stuff}\texttt{";}\\
2702 % \texttt{beginchar(}\meta{usual arguments}\texttt{);}
2703 % \end{quote}
2704 % where the two string assigments are optional.
2705 % \begin{macrocode}
2706 def no_adjust=
2707 old_begin_adjust := extra_beginchar ;
2708 old_end_adjust := extra_endchar ;
2709 extra_beginchar := " " ;
2710 extra_endchar :=
2711 "extra_proof_rules;charcode:=charcode+code_offset" ;
2712 enddef;
2713
2714 % \end{macrocode}
2715 % \end{fontmacro}
2716 %
2717 % \begin{fontmacro}{restore\_adjust}
2718 % Use this command to restore automaticly adding extra space around
2719 % the characters. Must be used between a \texttt{endchar} and
2720 % \texttt{beginchar}.
2721 % \begin{macrocode}
2722 def restore_adjust=
2723 extra_beginchar := old_begin_adjust ;
2724 extra_endchar := old_end_adjust ;
2725 enddef;
2726 string old_begin_adjust, old_end_adjust ;
2727
2728 % \end{macrocode}
2729 % \end{fontmacro}
2730 %
2731 % \subsubsection{Drawing macroes}
2732 %
2733 % I have made this macros for drawing, to ease up the writing of
2734 % source programs. They all simulate a pen of breadth \textit{thick}
2735 % always moving perpendecular to the current path. This is needed
2736 % in the \universal{} font, since all pen breadths are supposed
2737 % to be equal anywhere (at least within a paricular `group'), in any
2738 % direction.
2739 %
2740 % All of the macros heavely use \texttt{penpos}, so the use
2741 % is pretty reliant on the appropiate use of this macro. This
2742 % put some restrictions on the arguments, since \texttt{penpos}
2743 % deals with angles of all sorts.
2744 %
2745 % In the macroes, we use vectors with number $m$, as assigened in the
2746 % last argument of the macros. $m$ should be assigend so that the
2747 % vectors $m,m+1,\ldots,m+4$ does not conflict with already assigned
2748 % vectors, or vectors to be assigend in the current picture.
2749 %
2750 % Though it may seem a somewhat \textit{ad hoc} solution, it is the
2751 % best way I could think of. Ofcourse I could check to see what
2752 % vectors are defined, and then use some others, but that would
2753 % complicate the code unessecary, and since the idea is to hide the
2754 % actual vector operations in the macros, I descided to use the
2755 % course.
2756 %
2757 % (No labels are produced on the proofsheets by this macroes,
2758 % since they use very high numbers, which is rather destracting.)
2759 %
2760 % \begin{fontmacro}{unicir}
2761 % Draws a circle of outer radius \texttt{rad}, and centered at
2762 % \texttt{center}, which shoud be a pair. Notice, no effort is
2763 % made to keep the circle within the character, so the use of
2764 % this macro, should be governed with some caution.
2765 %
2766 % The second argument is the \emph{thickness} of the pen drawing the
2767 % segment.
2768 %
2769 % The last argument \texttt{m} is to insure the uniqueness of
2770 % all pairs. That is, if you need more then one circle in a letter,
2771 % then \texttt{m} should take different values for each circle, so that
2772 % $m_{i}-m_{i+1}>4$. Also \texttt{m} should not take any value
2773 % corrosponing to vectors already defined. This is most easily done if
2774 % \texttt{m} is asigned a large value, like say a multipla of 10, or
2775 % even 100.
2776 % \begin{macrocode}
2777 def unicir(expr rad, thick, center, m)=
2778 pickup pensquare scaled thick;
2779 top z[m] = (0,rad) + center ; penpos[m](thick,90) ;
2780 lft z[m+1] = (-rad,0) + center ; penpos[m+1](thick,180) ;
2781 bot z[m+2] = (0,-rad) + center ; penpos[m+2](thick,270) ;
2782 rt z[m+3] = (rad,0) + center ; penpos[m+3](thick,0) ;
2783 penstroke z[m]e{left}..z[m+1]e{down}..
2784 z[m+2]e{right}..z[m+3]e{up}..cycle ;
2785 penlabels(range m thru m+3);
2786 enddef;
2787
2788 % \end{macrocode}
2789 % \end{fontmacro}
2790 %
2791 % \begin{fontmacro}{uniarc}
2792 % Produce a segment of a circle, with radius \texttt{rad}, and
2793 % centered at \texttt{center}.
2794 %
2795 % The size of the segment is determind by the arguments
2796 % \texttt{ang\_a} and \texttt{ang\_b}, consider as angles from a
2797 % horizontal linesegment from \texttt{center} traveling right. That
2798 % means you should specify the angles in at counter--clockwise way,
2799 % and between -360 and 360. \texttt{ang\_b} \emph{must} be bigger then
2800 % \texttt{ang\_a}, otherwise the output is totally unpredictiable.
2801 %
2802 % The second argument is the \emph{thickness} of the pen drawing the
2803 % segment, Again the last argument \texttt{m} is to insure the
2804 % uniqueness of the vectors. The comments above on the argument also
2805 % applies here.
2806 % \begin{macrocode}
2807 def uniarc(expr rad, thick, center, ang_a, ang_b, m)=
2808 pickup pensquare scaled thick ;
2809 path arc.l, arc.r, arc.e ; n:=m ; ang_c := abs(ang_b-ang_a) ;
2810 % \end{macrocode}
2811 % First we check to see, if the user somehow requested an arc of more
2812 % then $360^\circ$. If so, we just call \texttt{unicir}, since this is
2813 % properly intended. If not, we go on to define relevant vectors.
2814 % \begin{macrocode}
2815 if ang_c >= 360 :
2816 unicir(rad,center,n) ;
2817 else:
2818 % \end{macrocode}
2819 % We acctualy define the vectors as if we started in the arc on the
2820 % positiv $x$--axis, considering \texttt{center} as origo, an then
2821 % latter on turn the arc the appropiate angle. Notice the heavy use of
2822 % \texttt{penpos} and \texttt{[ ]} to evaluate variable \texttt{m}.
2823 % \begin{macrocode}
2824 top z[n] = (0,rad) + center ; penpos[n](thick,90) ;
2825 arc.r := z[n]r{left} ;
2826 arc.l := z[n]l{left} ;
2827 % \end{macrocode}
2828 % The next \texttt{if...fi} loop ensures we don't define a
2829 % whole lot of useless vectors. The \MF{} primitiv path macro
2830 % \texttt{...} does a very good job at approching a segment of a
2831 % circle even if only two points are defined in an interval of
2832 % $90^\circ$, if we at the same time utilizes the direction mechanisme
2833 % of \MF{}.
2834 % \begin{macrocode}
2835 if ang_c > 90 :
2836 n := n+1 ; lft z[n] = (-rad,0) + center ; penpos[n](thick,180) ;
2837 arc.r := arc.r&z[n-1]r..z[n]r{down} ;
2838 arc.l := arc.l&z[n-1]l..z[n]l{down} ;
2839 if ang_c > 180 :
2840 n := n+1 ; bot z[n] = (0,-rad) + center ; penpos[n](thick,270) ;
2841 arc.r := arc.r&z[n-1]r..z[n]r{right} ;
2842 arc.l := arc.l&z[n-1]l..z[n]l{right} ;
2843 if ang_c > 270 :
2844 n := n+1 ; rt z[n] = (rad,0) + center ; penpos[n](thick,0) ;
2845 arc.r := arc.r&z[n-1]r..z[n]r{up} ;
2846 arc.l := arc.l&z[n-1]l..z[n]l{up} ;
2847 fi ;
2848 fi ;
2849 fi ;
2850 % \end{macrocode}
2851 % After the \texttt{if} loop, we define the final point, make the
2852 % complete path, and finishes off by drawing the arc, turned the
2853 % appropiate angle.
2854 % \begin{macrocode}
2855 n:=n+1 ; z[n] = z[m] rotatedabout(center,ang_c) ;
2856 penpos[n](thick,90+ang_c) ;
2857 arc.r := arc.r&z[n-1]r..z[n]r{dir(180+ang_c)} ;
2858 arc.l := arc.l&z[n-1]l..z[n]l{dir(180+ang_c)} ;
2859 arc.e := arc.l--reverse arc.r--cycle;
2860 fill (arc.e rotatedabout(center,-90+ang_a));
2861 fi;
2862 penlabels(range m thru n);
2863 enddef;
2864
2865 % \end{macrocode}
2866 % Notice the use of two paths, and reversal of one. This is
2867 % essentially what the \texttt{plain.mf} macro \texttt{penstroke}
2868 % does. However, that macro couldn't be used here, since we are
2869 % building a path on the fly, so every step is needed to be fully
2870 % known, and \texttt{penstroke} makes new variables to do the job.
2871 %
2872 % Also, we use the integer \texttt{n} for surfixes, since this
2873 % limits the creation of vectors, and eases the path building.
2874 % \end{fontmacro}
2875 %
2876 % \begin{fontmacro}{uniend}
2877 % This macro is somewhat similar to \texttt{uniarc} above, but has a
2878 % different use. Where \texttt{uniarc} is intended for arcs with a
2879 % radius bigger then the pen thickness, this macro is intended for
2880 % arcs with a radius of exactly same size as the pen thickness, hence
2881 % this macro produces a \emph{filled} segment of a circle with a
2882 % radius of the current pen thickness.
2883 %
2884 % This macro, like \texttt{uniarc}, takes 5 arguments: the radius of
2885 % the arc; the center of the arc, which should be of type
2886 % \texttt{pair}; the starting and ending angle; plus the first number
2887 % of the vectors needed to draw this figure. The considerations above
2888 % in \texttt{uniarc} conserning the angles also apply here.
2889 %
2890 % Notice once again the use of \texttt{if..fi} loops to define the
2891 % relevant vectors and paths.
2892 % \begin{macrocode}
2893 def uniend(expr rad, center, ang_a, ang_b, m)=
2894 path arc ; n:=m ; ang_c := abs(ang_b-ang_a) ;
2895 if ang_c >= 360 :
2896 Rad := 2*rad ;
2897 pickup pencircle scaled Rad ; z[n]=center ; drawdot z[n] ;
2898 else:
2899 z[n] = (0,rad) + center ; arc := center{up}--z[n]{left} ;
2900 if ang_c > 90 :
2901 n := n+1 ; z[n] = (-rad,0) + center ;
2902 arc := arc&z[n-1]..z[n]{down} ;
2903 if ang_c > 180 :
2904 n := n+1 ; z[n] = (0,-rad) + center ;
2905 arc := arc&z[n-1]..z[n]{right} ;
2906 if ang_c > 270 :
2907 n := n+1 ; z[n] = (rad,0) + center ;
2908 arc := arc&z[n-1]..z[n]{up} ;
2909 fi ;
2910 fi ;
2911 fi ;
2912 n := n+1 ; z[n] = z[m] rotatedabout(center,ang_c) ;
2913 arc := arc&z[n-1]..z[n]{dir(180+ang_c)}--cycle ;
2914 fill (arc rotatedabout(center,-90+ang_a));
2915 fi;
2916 penlabels(range m thru n);
2917 enddef;
2918
2919 % \end{macrocode}
2920 % Notice, once more we are building a path on the fly, so some caution
2921 % was used here. Again the \texttt{plain.mf} macro \texttt{penstroke}
2922 % doesn't do the job here. This is because only a part of a stroke is
2923 % actually wanted. You might think then, that the \texttt{plain.mf}
2924 % macro \texttt{cutoff} then would do the job, but in my experience
2925 % that macro has some serious flaws\footnote{A quick use of the Unix
2926 % command \texttt{grep} on most of the Computer Modern Fonts reveals
2927 % that \texttt{cutoff} is \emph{never} used, which I take to
2928 % indicate its weakness.}.
2929 % \end{fontmacro}
2930 %
2931 % \begin{fontmacro}{unilne}
2932 % This macro is intended for drawing non--vertical or horizontal
2933 % lines, since this is easily done with \texttt{draw}. It takes
2934 % six arguments, the thickness of the pen drawing the line, the
2935 % starting point of the line (\texttt{start}), the ending point of
2936 % line (\texttt{stop}), which should be higher and more to the right
2937 % then the starting point, and the cutoff--style
2938 % (\texttt{endcutstart, endcutstop}).
2939
2940 % If \texttt{endcut}$x$ is positiv, then, the line is cut off
2941 % horizontally at the relevant point $x$.
2942 %
2943 % If \texttt{endcut}$x$ is negativ, then, the line is cut off
2944 % vertically at $x$.
2945 %
2946 % If it is zero, then the line is cut off perpendicular to the slant
2947 % of the line at $x$.
2948 %
2949 % Also, if \texttt{endcut}$x$ is zero, the starting point is the
2950 % point corrosponding to the lowest height and leftmost width, and the
2951 % ending point corrosponds to the heighest height and rightmost width
2952 % of the line. Again \texttt{m} insures unique vectors.
2953 % \begin{macrocode}
2954 def unilne(expr thick, start, stop, endcutstart, endcutstop, m)=
2955 pickup pensquare scaled thick;
2956 if endcutstart = endcutstop :
2957 if endcutstart = 0 :
2958 lft bot z[m] = start ; rt top z[m+1] = stop ;
2959 ang_c := angle(z[m+1]-z[m]) ;
2960 penpos[m](thick,-90+ang_c) ;
2961 penpos[m+1](thick,-90+ang_c) ;
2962 elseif endcutstart > 0 :
2963 lft z[m] = start ; rt z[m+1] = stop ;
2964 ang_c := angle(z[m+1]-z[m]) ;
2965 penpos[m](thick/sind(ang_c),0) ;
2966 penpos[m+1](thick/sind(ang_c),0) ;
2967 elseif endcutstart < 0:
2968 bot z[m] = start ; top z[m+1] = stop ;
2969 ang_c := angle(z[m+1]-z[m]) ;
2970 penpos[m](thick/cosd(ang_c),90) ;
2971 penpos[m+1](thick/cosd(ang_c),90) ;
2972 fi;
2973 penstroke z[m]e--z[m+1]e ;
2974 else :
2975 message "Different endcuts on unilne not implemented yet.";
2976 fi;
2977 n := m + 1 ;
2978 penlabels(range m thru n);
2979 enddef;
2980
2981 % \end{macrocode}
2982 % \end{fontmacro}
2983 %
2984 %\iffalse
2985 %</unibase>
2986 %\fi
2987 %
2988 %\iffalse
2989 %<*low>
2990 %\fi
2991 % \subsection{The Minuscles (lowercase letters) --- \texttt{unilow.mf}}
2992 % \file{unilow.mf}
2993 %
2994 % \Newestchanges{v2.0}{98/08/01}{All of \texttt{unilow.mf} is defined
2995 % within a conditional, to ease the generation of the small caps
2996 % shaped fonts.}
2997 % This whole file is loaded only conditionally to make the generation
2998 % of small caps font easier.
2999 % \begin{macrocode}
3000 if not caps :
3001
3002 % \end{macrocode}
3003 % \begin{fontcode}{a}
3004 % The letter \textunibf{a} is interresting in the sense, that the bottom
3005 % part of the curve, is \emph{not} a segement of an arc. The exact
3006 % curvature of this part, is controversial; some font designers have
3007 % chosen to make it a segement of an arc, making all of the curve, a
3008 % circle; while others have chosen to make it a straight line. I
3009 % decided to take the middleroad, since it appeals more to me
3010 % aestheticly, even though it deviates from the principle of building
3011 % all characters from the two simple forms: the cirle and the line.
3012 %
3013 % The \textuni{a} is also saved in the picture variable
3014 % \texttt{a\_pic}, for use in \texttt{unilig.mf}.
3015 % \begin{macrocode}
3016 beginchar("a",x_w#,x_h#,0); "The letter a";
3017 uniarc(w/2,p_t,(w/2,h/2),0,290,10);
3018 pickup unipensqu ;
3019 z1 = ((cosd(290)*(w-p_t)+w)/2,(sind(290)*(w-p_t)+w)/2);
3020 z2 = (w-p_t/2,2h/5) ; penpos1(p_t,-70); penpos2(p_t,-45);
3021 fill z1l{dir(20)}..z2l{dir(45)}--
3022 z2r{dir(225)}..z1r{dir(200)}--cycle;
3023 top rt z3 = (w,.5h) ; bot rt z4 = (w,0) ; draw z3{down}--z4 ;
3024 penlabels(1,2,3,4);
3025 a_pic := currentpicture ;
3026 endchar;
3027
3028 % \end{macrocode}
3029 % \end{fontcode}
3030 % \begin{fontcode}{b}
3031 % There is not much to say about this letter, only that it is a
3032 % \textunibf{d} reflected through the vertical center axis.
3033 % \begin{macrocode}
3034 beginchar("b",x_w#,l_h#,0); "The letter b";
3035 unicir(w/2,p_t,(w/2,w/2),10);
3036 pickup unipensqu ;
3037 lft bot z1 = (0,w/2) ; lft top z2 = (0,h) ; draw z1{up}--z2 ;
3038 penlabels(1,2);
3039 endchar;
3040
3041 % \end{macrocode}
3042 % \end{fontcode}
3043 % \begin{fontcode}{c}
3044 % This is a rather short program, but is interresting, since some care
3045 % has to be taken, when assigning the width. Since a \textunibf{c} is not
3046 % a full cirle, but only a segment opening towards the right, it would
3047 % look wrong, if the letter was as wide, as it is tall.
3048 % \begin{macrocode}
3049 beginchar("c",x_w#*(1+cosd(30))/2,x_h#,0); "The letter c";
3050 uniarc(h/2,p_t,(h/2,h/2),30,330,10);
3051 endchar;
3052
3053 % \end{macrocode}
3054 % \end{fontcode}
3055 % \begin{fontcode}{d}
3056 % A flipped \textunibf{b} --- that's it!
3057 % \begin{macrocode}
3058 beginchar("d",x_w#,l_h#,0); "The letter d";
3059 unicir(w/2,p_t,(w/2,w/2),10);
3060 pickup unipensqu ;
3061 rt bot z1 = (w,w/2) ; rt top z2 = (w,h) ; draw z1{up}--z2 ;
3062 penlabels(1,2);
3063 endchar;
3064
3065 % \end{macrocode}
3066 % \end{fontcode}
3067 % \begin{fontcode}{e}
3068 % Notice the middle bar. The arc can not go all the way down to the
3069 % bar, because then the \textunibf{e} would end in a arc, on the right
3070 % side. We want a flat end, so an extra dot has to be made.
3071 %
3072 % The \textuni{e} is also saved in the picture variable
3073 % \texttt{e\_pic}, for use in \texttt{unilig.mf}.
3074 % \begin{macrocode}
3075 beginchar("e",x_w#,x_h#,0); "The letter e";
3076 uniarc(w/2,p_t,(w/2,h/2),0,330,10);
3077 temp := p_t/2 ;
3078 pickup pensquare xscaled p_t yscaled temp slanted slantratio ;
3079 top rt z1 = (w,h/2) ; drawdot z1 ;
3080 pickup unipensqu ;
3081 z2 = (p_t,h/2) ; z3 = (w-p_t,h/2) ; draw z2--z3 ;
3082 penlabels(1,2,3);
3083 e_pic := currentpicture ;
3084 endchar;
3085
3086 % \end{macrocode}
3087 % \end{fontcode}
3088 % \begin{fontcode}{f}
3089 % The horizontal placement of the bar is somewhat controversial. I
3090 % think it looks best if it is placed, so that the top of the bar, is
3091 % at the same height, as the top of an \textunibf{x}, that is $1ex$.
3092 % \begin{macrocode}
3093 beginchar("f",4p_t#,l_h#,0); "The letter f";
3094 uniarc(3/4w,p_t,(3/4w,h-3/4w),90,180,10);
3095 pickup unipensqu ;
3096 top rt z1 = (w,h) ; drawdot z1 ;
3097 lft top z2 = (0,h-3/4w) ; bot lft z3 = (0,0) ; draw z2--z3 ;
3098 top lft z4 = (0,x_h) ; top rt z5 = (w,x_h) ; draw z4--z5 ;
3099 penlabels(1,2,3,4,5);
3100 endchar;
3101
3102 % \end{macrocode}
3103 % \end{fontcode}
3104 % \begin{fontcode}{g}
3105 % The \textunibf{g} is a very good measure of how low ascenders should
3106 % go. Some might want to add a square on top of the ascender, so that
3107 % the letter ended in flat to the right, but I found it displeacing,
3108 % and therefore left it out.
3109 % \begin{macrocode}
3110 beginchar("g",x_w#,x_h#,p_d#); "The letter g" ;
3111 uniarc(w/2,p_t,(w/2,h/2),30,360,10) ;
3112 pickup unipensqu ;
3113 top rt z1 = (w,h/2) ; bot rt z2 = (w,-d) ; draw z1--z2;
3114 penlabels(1,2);
3115 endchar;
3116
3117 % \end{macrocode}
3118 % \end{fontcode}
3119 % \begin{fontcode}{h}
3120 % It can be argued, that this letter is too wide. However, I decieded
3121 % to weight the concept of homongeious circles higher in this case.
3122 % \begin{macrocode}
3123 beginchar("h",x_w#,l_h#,0); "The letter h" ;
3124 uniarc(w/2,p_t,(w/2,w/2),0,180,10) ;
3125 pickup unipensqu ;
3126 top lft z1 = (0,h) ; bot lft z2 = (0,0) ; draw z1--z2 ;
3127 top rt z3 = (w,x_h-w/2) ; bot rt z4 = (w,0) ; draw z3--z4 ;
3128 penlabels(1,2,3,4) ;
3129 endchar;
3130
3131 % \end{macrocode}
3132 % \end{fontcode}
3133 % \begin{fontcode}{i}
3134 % Quite a simple letter, though it can be argued wether the dot should
3135 % be square, or as here circular.
3136 % \begin{macrocode}
3137 beginchar("i",p_t#,l_h#,0) ; "The letter i" ;
3138 pickup unipensqu ;
3139 top lft z1 = (0,x_h) ; bot lft z2 = (0,0) ; draw z1--z2 ;
3140 pickup unipencir ;
3141 top lft z3 = (0,h) ; drawdot z3 ;
3142 penlabels(1,2,3) ;
3143 endchar;
3144
3145 % \end{macrocode}
3146 % \end{fontcode}
3147 % \begin{fontcode}{j}
3148 % No left stem is added at the bottom, since it isn't needed!
3149 % \begin{macrocode}
3150 beginchar("j",p_t#,l_h#,p_d#); "The letter j" ;
3151 pickup unipensqu ;
3152 top lft z1 = (0,x_h) ; bot lft z2 = (0,-d) ; draw z1--z2 ;
3153 pickup unipencir ;
3154 top lft z3 = (0,h) ; drawdot z3 ;
3155 penlabels(1,2,3) ;
3156 endchar;
3157
3158 % \end{macrocode}
3159 % \end{fontcode}
3160 % \begin{fontcode}{k}
3161 % Again, this letter is narrower then most of the letters. This is
3162 % because, the to arcs isn't full halfarcs, but rather quarterarcs.
3163 % \begin{macrocode}
3164 beginchar("k",(x_h#+3p_t#)/2,x_h#,0) ; "The letter k" ;
3165 uniarc((h+p_t)/2,p_t,(p_t,h),270,360,10) ;
3166 uniarc((h+p_t)/2,p_t,(p_t,0),0,90,20) ;
3167 pickup unipensqu ;
3168 top lft z1 = (0,h) ; bot lft z2 = (0,0) ; draw z1--z2 ;
3169 penlabels(1,2);
3170 endchar;
3171
3172 % \end{macrocode}
3173 % \end{fontcode}
3174 % \begin{fontcode}{l}
3175 % What is a line, but a line! Well right now it is an `l'. You may
3176 % wonder why no stems are added, to destingish an `l' from a vertical
3177 % bar, as in math mode. Well, how often will you put a math vertical
3178 % bar next to an text `l'.
3179 % \begin{macrocode}
3180 beginchar("l",p_t#,l_h#,0) ; "The letter l" ;
3181 pickup unipensqu ;
3182 top lft z1 = (0,h) ; bot lft z2 = (0,0) ; draw z1--z2;
3183 penlabels(1,2);
3184 endchar;
3185
3186 % \end{macrocode}
3187 % \end{fontcode}
3188 % \begin{fontcode}{m}
3189 % This proved more tricky then I thought. Originally, I wanted the arc
3190 % of the \textunibf{n} to be as high as those of the \textunibf{m}. This
3191 % meant, that either the \textunibf{n} should be very narrow, or the
3192 % \textunibf{m} very wide; either way it wouldn't look good. So after a
3193 % few experiments, I decided for the current form.
3194 % \begin{macrocode}
3195 beginchar("m",m_w#,x_h#,0) ; "The letter m" ;
3196 temp := .25*(w+p_t);
3197 uniarc(temp,p_t,(temp,h-temp),0,180,10) ;
3198 uniarc(temp,p_t,(w-temp,h-temp),0,180,20) ;
3199 pickup unipensqu ;
3200 top lft z1 = (0,h-temp) ; bot lft z2 = (0,0) ; draw z1--z2 ;
3201 top z3 = (w/2,h-temp) ; bot z4 = (w/2,0) ; draw z3--z4 ;
3202 top rt z5 = (w,h-temp) ; bot rt z6 = (w,0) ; draw z5--z6 ;
3203 penlabels(1,2,3,4,5,6);
3204 endchar;
3205
3206 % \end{macrocode}
3207 % \end{fontcode}
3208 % \begin{fontcode}{n}
3209 % Tricky because of the \textunibf{m} (see above).
3210 % \begin{macrocode}
3211 beginchar("n",x_w#,x_h#,0); "The letter n" ;
3212 uniarc(w/2,p_t,(w/2,h-w/2),0,180,10) ;
3213 pickup unipensqu ;
3214 top lft z1 = (0,h-w/2) ; bot lft z2 = (0,0) ; draw z1--z2 ;
3215 top rt z3 = (w,h-w/2) ; bot rt z4 = (w,0) ; draw z3--z4 ;
3216 penlabels(1,2,3,4) ;
3217 endchar;
3218
3219 % \end{macrocode}
3220 % \end{fontcode}
3221 % \begin{fontcode}{o}
3222 % Well, this is just a circle, nothing fancy. Why not make it oval,
3223 % you say. I say, listen --- Herbert Bayer just turned in his grave!
3224 % \begin{macrocode}
3225 beginchar("o",x_w#,x_h#,0) ; "The letter o";
3226 unicir(w/2,p_t,(w/2,h/2),10);
3227 endchar;
3228
3229 % \end{macrocode}
3230 % \end{fontcode}
3231 % \begin{fontcode}{p}
3232 % Try to flip the \textunibf{b} horizontally, and tell me what you get?
3233 % \begin{macrocode}
3234 beginchar("p",x_w#,x_h#,p_d#) ; "The letter p" ;
3235 unicir(w/2,p_t,(w/2,w/2),10);
3236 pickup unipensqu ;
3237 lft top z1 = (0,h/2) ; lft bot z2 = (0,-d) ; draw z1--z2 ;
3238 penlabels(1,2);
3239 endchar;
3240
3241 % \end{macrocode}
3242 % \end{fontcode}
3243 % \begin{fontcode}{q}
3244 % Why not try to destingish the \textunibf{q} more from the \textunibf{g}?
3245 % Well if you look at the two letters, I think, you will get the
3246 % answer. Also, my intuition of a \textunibf{q} is that of a reflected
3247 % \textunibf{p}, but that is just \emph{my} intuition!
3248 % \begin{macrocode}
3249 beginchar("q",x_w#,x_h#,p_d#) ; "The letter q" ;
3250 unicir(w/2,p_t,(w/2,w/2),10);
3251 pickup unipensqu ;
3252 rt top z1 = (w,h/2) ; rt bot z2 = (w,-d) ; draw z1--z2 ;
3253 penlabels(1,2);
3254 endchar;
3255
3256 % \end{macrocode}
3257 % \end{fontcode}
3258 % \begin{fontcode}{r}
3259 % Well that is really simplle, but illustrates a fine point --- why
3260 % use more strokes, when fewer is needed?
3261 % \begin{macrocode}
3262 beginchar("r",3p_t#,x_h#,0) ; "The letter r" ;
3263 uniarc(w,p_t,(w,h-w),90,180,10) ;
3264 pickup unipensqu ;
3265 top lft z1 = (0,h-w) ; bot lft z2 = (0,0) ; draw z1--z2;
3266 penlabels(1,2);
3267 endchar;
3268
3269 % \end{macrocode}
3270 % \end{fontcode}
3271 % \begin{fontcode}{s}
3272 % \Newestchanges{v2.0}{98/08/01}{Considerable simplifications.}
3273 % This is a `slow' letter, i.e.\ it takes \MF{} quite some time to
3274 % compute. The reason is simple --- a \emph{lot} of curves is drawn.
3275 % \begin{macrocode}
3276 beginchar("s",x_w#,x_h#,0) ; "The letter s" ;
3277 temp := (h+p_t)/4 ;
3278 uniarc(temp,p_t,(temp,temp),180,270,10) ;
3279 uniarc(temp,p_t,(w-temp,temp),270,450,20) ;
3280 uniarc(temp,p_t,(temp,h-temp),90,270,30) ;
3281 uniarc(temp,p_t,(w-temp,h-temp),0,90,40) ;
3282 pickup unipensqu ;
3283 bot lft z1 = (temp,0) ; bot rt z2 = (w-temp,0) ; draw z1--z2 ;
3284 lft z3 = (temp,h/2) ; rt z4 = (w-temp,h/2) ; draw z3--z4 ;
3285 top lft z5 = (temp,h) ; top rt z6 = (w-temp,h) ; draw z5--z6 ;
3286 penlabels(1,2,3,4,5,6) ;
3287 endchar;
3288
3289 % \end{macrocode}
3290 % \end{fontcode}
3291 % \begin{fontcode}{t}
3292 % As I argued before, the placement of the bar in the \textunibf{f} and
3293 % \textunibf{t} is controversial.
3294 % \begin{macrocode}
3295 beginchar("t",p_t#+2c_d#,l_h#,0); "The letter t" ;
3296 pickup unipensqu ;
3297 top z1 = (w/2,h) ; bot z2 = (w/2,0) ; draw z1--z2 ;
3298 top lft z3 = (0,x_h) ; top rt z4 = (w,x_h) ; draw z3--z4 ;
3299 penlabels(1,2,3,4);
3300 endchar;
3301
3302 % \end{macrocode}
3303 % \end{fontcode}
3304 % \begin{fontcode}{u}
3305 % Flip a letter here, and flip a letter there --- all in a days work
3306 % of a font designer!
3307 % \begin{macrocode}
3308 beginchar("u",x_w#,x_h#,0); "The letter u" ;
3309 uniarc(w/2,p_t,(w/2,w/2),180,360,10) ;
3310 pickup unipensqu ;
3311 top lft z1 = (0,h) ; bot lft z2 = (0,w/2) ; draw z1--z2 ;
3312 top rt z3 = (w,h) ; bot rt z4 = (w,w/2) ; draw z3--z4 ;
3313 penlabels(1,2,3,4) ;
3314 endchar;
3315
3316 % \end{macrocode}
3317 % \end{fontcode}
3318 % \begin{fontcode}{v}
3319 % The lines and the arc doesn't intersect in a flowing way, but the
3320 % error is so small it will not be seen in normal use. However, if
3321 % the font is maginified to, say $70pt$, it might be a problem. I
3322 % hope to get into this, some time in the future. For now, make do.
3323 % \begin{macrocode}
3324 beginchar("v",x_w#,x_h#,0); "The letter v" ;
3325 unilne(p_t,(0,h),(w/2,p_t),1,1,10);
3326 unilne(p_t,(w/2,p_t),(w,h),1,1,20);
3327 uniarc(p_t,p_t,(w/2,p_t),180,360,30);
3328 endchar;
3329
3330 % \end{macrocode}
3331 % \end{fontcode}
3332 % \begin{fontcode}{w}
3333 % Flippedy--flop.
3334 % \begin{macrocode}
3335 beginchar("w",m_w#,x_h#,0) ; "The letter w" ;
3336 temp := .25*(w+p_t);
3337 uniarc(temp,p_t,(temp,temp),180,360,10) ;
3338 uniarc(temp,p_t,(w-temp,temp),180,360,20) ;
3339 pickup unipensqu ;
3340 top lft z1 = (0,h) ; bot lft z2 = (0,temp) ; draw z1--z2 ;
3341 top z3 = (w/2,h) ; bot z4 = (w/2,temp) ; draw z3--z4 ;
3342 top rt z5 = (w,h) ; bot rt z6 = (w,temp) ; draw z5--z6 ;
3343 penlabels(1,2,3,4,5,6);
3344 endchar;
3345
3346 % \end{macrocode}
3347 % \end{fontcode}
3348 % \begin{fontcode}{x}
3349 % This is a bit tricky --- the two arces are each half the width wide,
3350 % but since they share middle part, they leave the width of the
3351 % strokes uncoverd. To remedy this problem, four dots are drawn. It
3352 % would be silly, if the \textunibf{x} wasn't as wide as it is tall,
3353 % namely \texttt{x\_w}.
3354 % \begin{macrocode}
3355 beginchar("x",x_w#,x_h#,0) ; "The letter x" ;
3356 uniarc(h/2,p_t,(p_t/2,h/2),270,90,10);
3357 uniarc(h/2,p_t,(w-p_t/2,h/2),90,270,20);
3358 temp := (w-h+p_t)/2 ;
3359 pickup pensquare xscaled temp yscaled p_t slanted slantratio ;
3360 lft top z1 = (0,h) ; drawdot z1 ;
3361 lft bot z2 = (0,0) ; drawdot z2 ;
3362 rt top z3 = (w,h) ; drawdot z3 ;
3363 rt bot z4 = (w,0) ; drawdot z4 ;
3364 penlabels(1,2,3,4);
3365 endchar;
3366
3367 % \end{macrocode}
3368 % \end{fontcode}
3369 % \begin{fontcode}{y}
3370 % Flip an reflect, or just turn, it's all the same, it will still be a
3371 % transformation of an \textunibf{h}
3372 % \begin{macrocode}
3373 beginchar("y",x_w#,x_h#,p_d#); "The letter y" ;
3374 uniarc(w/2,p_t,(w/2,w/2),180,360,10) ;
3375 pickup unipensqu ;
3376 top lft z1 = (0,h) ; bot lft z2 = (0,w/2); draw z1--z2 ;
3377 top rt z3 = (w,h) ; bot rt z4 = (w,-d) ; draw z3--z4 ;
3378 penlabels(1,2,3,4) ;
3379 endchar;
3380
3381 % \end{macrocode}
3382 % \end{fontcode}
3383 % \begin{fontcode}{z}
3384 % This is a bit tricky letter, since what I wanted was for middle line
3385 % to have an inclination such that its edges would be tangents to the
3386 % circle segments at the endpoints of the horizontal lines. I tried to
3387 % solve this problem analytically, but gave up. I also tried to use
3388 % some of \texttt{plain.mf}'s macros, but none of them gave me the
3389 % resulat I wanted. So I finally made the approximation seen below.
3390 % If you can come up with a good solution to this problem, please let
3391 % me know.
3392 % \begin{macrocode}
3393 beginchar("z",x_w#,x_h#,0) ; "The letter z" ;
3394 pickup unipensqu ;
3395 top lft z1 = (0,h) ; top rt z2 = (w-p_t,h) ; draw z1--z2 ;
3396 bot lft z3 = (p_t,0) ; bot rt z4 = (w,0) ; draw z3--z4 ;
3397 phi := angle((w,h-p_t)-(0,p_t)) ; % Uuuuupps !?!!!!!!
3398 z5 = (p_t,p_t) ; z6 = z5 + p_t*(-sind(phi),cosd(phi)) ;
3399 z7 = (w-p_t,h-p_t) ; z8 = z7 + p_t*(sind(phi),-cosd(phi)) ;
3400 uniend(p_t,z5,90+phi,270,10);
3401 uniend(p_t,z7,-90+phi,90,20);
3402 fill z5--z6--z7--z8--cycle ;
3403 penlabels(range 1 thru 8);
3404 endchar;
3405
3406 % \end{macrocode}
3407 % \end{fontcode}
3408 % As explain in the begining of the section, this file is one big
3409 % \texttt{if...else...fi} loop to mak the generation of the small caps
3410 % font easier.
3411 %
3412 % The idea is, that if \texttt{caps} is
3413 % \texttt{true}, then the main part of this file shouldn't be used,
3414 % but rather \texttt{uniupp.mf} should be loaded again, but with
3415 % different diffintions of dimensions for uppercase letters, so that
3416 % they come out as small as lowercase letters, hence --- small
3417 % caps.
3418 %
3419 % This is done by the macro \texttt{do\_caps}, which we utilies
3420 % inside a group, so that the changes made by \texttt{do\_caps} will be
3421 % undone after the group, and hence not affect the rest of the font.
3422 %
3423 % \begin{macrocode}
3424 else :
3425 do_caps(ASCII"a" - ASCII"A") ;
3426 input uniupp ;
3427 end_do_caps ;
3428 fi ;
3429
3430 % \end{macrocode}
3431 %\iffalse
3432 %</low>
3433 %\fi
3434 %\iffalse
3435 %<*upp>
3436 %\fi
3437 % \subsection{The Majuscles (uppercase letters) --- \texttt{uniupp.mf}}
3438 % \file{uniupp.mf}
3439 %
3440 % As noted earlier, Herbert Bayer never did design any majuscles,
3441 % since he considered them useless. However, majuscles are avaliable
3442 % in the current implentation and are to be used at your lesuire. Of
3443 % course, if you agree with the Bauhaus school, and think uppercase
3444 % letters are useless, you shouldn't use them. Then, perhaps, you
3445 % should also refrain from using non--standard letters shuch as German
3446 % double s, Scandinavian ligatures, Spanish schriek, and perhaps even
3447 % accents of any kind. This may seem a little drastic, but the
3448 % train--of--tought should be clear: Why use letters and symbols that
3449 % are in the best case obsolete, and in the worst useless or without
3450 % significans?
3451 %
3452 % \Newestchanges{v2.0}{98/08/01}{All of \texttt{unilow.mf} is defined
3453 % within a conditional, to ease the generation of the strict shaped
3454 % fonts.}
3455 %
3456 % To facilitate a use of the \universal{} font conforming to Bayers'
3457 % original idea, that is without majusceles this file is loaded only
3458 % under the condition that \texttt{strict} is \emph{false}.
3459 % \begin{macrocode}
3460 if not strict :
3461
3462 % \end{macrocode}
3463 % \begin{fontcode}{A}
3464 % It can be argued where the bar should be placed. I choose place it
3465 % the middle, since most of the majuscles tend to have a symmetry axis
3466 % long this line, so I thought it apropiate to follow this line of
3467 % thought. Ofcourse, I'm the only one to blane for this, since Herbert
3468 % Bayer \emph{never} designed any majuscles! (So flame me for
3469 % guessing).
3470 % \begin{macrocode}
3471 beginchar("A",A_w#,l_h#,0); "The letter A";
3472 uniarc(w/2,P_t,(w/2,h-w/2),0,180,10);
3473 pickup Unipensqu ;
3474 top lft z1 = (0,h-w/2) ; bot lft z2 = (0,0) ; draw z1--z2 ;
3475 top rt z3 = (w,h-w/2) ; bot rt z4 = (w,0) ; draw z3--z4 ;
3476 lft z5 = (0,h/2) ; rt z6 = (w,h/2) ; draw z5--z6 ;
3477 penlabels(1,2,3,4,5,6);
3478 endchar;
3479
3480 % \end{macrocode}
3481 % \end{fontcode}
3482 % \begin{fontcode}{B}
3483 % This letter is somewhat tricky, since it is not clear wether the
3484 % arces should be placed directly on the left down stoke, or there
3485 % should be some distance filled with horizontal bars. I chose the
3486 % second solution, even though it isn't the most simple, and hence in
3487 % better agrement with the concept of this font design. However, the
3488 % first solution does make the letter very narrow compared with some
3489 % of the other letters.
3490 % \begin{macrocode}
3491 beginchar("B",A_w#,l_h#,0) ; "The letter B" ;
3492 rad := 1/4(h+P_t) ;
3493 uniarc(rad,P_t,(w-rad,h-rad),-90,90,10) ;
3494 uniarc(rad,P_t,(w-rad,rad),-90,90,20) ;
3495 uniend(P_t,(P_t,h-P_t),90,180,30) ;
3496 uniend(P_t,(P_t,P_t),180,270,40) ;
3497 pickup Unipensqu ;
3498 top lft z1 = (0,h-P_t) ; bot lft z2 = (0,P_t) ; draw z1--z2 ;
3499 top lft z3 = (P_t,h) ; top rt z4 = (w-rad,h) ; draw z3--z4 ;
3500 lft z5 = (P_t,h/2) ; rt z6 = (w-rad,1/2h) ; draw z5--z6 ;
3501 bot lft z7 = (P_t,0) ; bot rt z8 = (w-rad,0) ; draw z7--z8 ;
3502 penlabels(range 1 thru 8);
3503 endchar;
3504
3505 % \end{macrocode}
3506 % \end{fontcode}
3507 % \begin{fontcode}{C}
3508 % This is really simple, since it is just an enlargement of the
3509 % corresponding minuscule, so the is really nothing interresting
3510 % here.
3511 % \begin{macrocode}
3512 beginchar("C",l_h#*(1+cosd(30))/2,l_h#,0); "The letter C";
3513 uniarc(h/2,P_t,(h/2,h/2),30,330,10);
3514 endchar;
3515
3516 % \end{macrocode}
3517 % \end{fontcode}
3518 % \begin{fontcode}{D}
3519 % As with the B, there are two ways to do this letter: One way would
3520 % be to make the arc a true half segment of a circle, the other would
3521 % be to compose it of two quarter segments, joined by a vertical
3522 % line. The second solution also entails, that two horizontal lines
3523 % should connect the two quarter circles with the left vertical
3524 % line. Again I chose the simple solution, which --- ofcourse ---
3525 % amounts to the first mentioned above.
3526 % \begin{macrocode}
3527 beginchar("D",A_w#,l_h#,0); "The letter D";
3528 uniarc(1/2h,P_t,(w-1/2h,1/2h),-90,90,10);
3529 uniend(P_t,(P_t,h-P_t),90,180,20);
3530 uniend(P_t,(P_t,P_t),180,270,30);
3531 pickup Unipensqu ;
3532 top lft z1 = (0,h-P_t) ; bot lft z2 = (0,P_t) ; draw z1--z2 ;
3533 top lft z3 = (P_t,h) ; top rt z4 = (w-1/2h,h) ; draw z3--z4 ;
3534 bot lft z5 = (P_t,0) ; bot rt z6 = (w-1/2h,0) ; draw z5--z6 ;
3535 penlabels(1,2) ;
3536 endchar;
3537
3538 % \end{macrocode}
3539 % \end{fontcode}
3540 % \begin{fontcode}{E}
3541 % It can be argued, that the middle bar should be shorter then the top
3542 % and bottom bar, but since this only complicates things, I chose not
3543 % to do so. However, I am not going to be stuborn on this, and it may
3544 % well prove to be better aestethics, if the middle bar is shortend.
3545 % \begin{macrocode}
3546 beginchar("E",A_w#,l_h#,0); "The letter E" ;
3547 uniend(P_t,(P_t,h-P_t),90,180,10);
3548 uniend(P_t,(P_t,P_t),180,270,20);
3549 pickup Unipensqu ;
3550 lft top z1 = (P_t,h) ; rt top z2 = (w,h) ; draw z1--z2 ;
3551 lft z3 = (P_t,h/2) ; rt z4 = (w,h/2) ; draw z3--z4 ;
3552 lft bot z5 = (P_t,0) ; rt bot z6 = (w,0) ; draw z5--z6 ;
3553 lft top z7 = (0,h-P_t) ; lft bot z8 = (0,P_t) ; draw z7--z8 ;
3554 penlabels(range 1 thru 8) ;
3555 endchar;
3556
3557 % \end{macrocode}
3558 % \end{fontcode}
3559 % \begin{fontcode}{F}
3560 % Ofcourse, very similar to the \textunibf{E}. The consideration
3561 % concerning the middle bar of the \textunibf{E} will ofcourse reflect
3562 % on this letter.
3563 % \begin{macrocode}
3564 beginchar("F",A_w#,l_h#,0); "The letter F" ;
3565 uniend(P_t,(P_t,h-P_t),90,180,10);
3566 pickup Unipensqu ;
3567 lft top z1 = (P_t,h) ; rt top z2 = (w,h) ; draw z1--z2 ;
3568 lft z3 = (P_t,h/2) ; rt z4 = (w,h/2) ; draw z3--z4 ;
3569 lft top z5 = (0,h-P_t) ; lft bot z6 = (0,0) ; draw z5--z6 ;
3570 penlabels(range 1 thru 8);
3571 endchar;
3572
3573 % \end{macrocode}
3574 % \end{fontcode}
3575 % \begin{fontcode}{G}
3576 % This is very similar to the minuscle \textunibf{g}, the only difference
3577 % is the size, and that the down stroke doens't extend below the
3578 % baseline.
3579 % \begin{macrocode}
3580 beginchar("G",l_h#,l_h#,0); "The letter G" ;
3581 uniarc(w/2,P_t,(w/2,h/2),30,360,10) ;
3582 pickup Unipensqu ;
3583 top rt z1 = (w,h/2) ; bot rt z2 = (w,0) ;
3584 draw z1--z2;
3585 penlabels(1,2);
3586 endchar;
3587
3588 % \end{macrocode}
3589 % \end{fontcode}
3590 % \begin{fontcode}{H}
3591 % An awfull simple letter. There can be no doubt on how this should
3592 % look.
3593 % \begin{macrocode}
3594 beginchar("H",A_w#,l_h#,0); "The letter H" ;
3595 pickup Unipensqu ;
3596 top lft z1 = (0,h) ; bot lft z2 = (0,0) ; draw z1--z2 ;
3597 top rt z3 = (w,h) ; bot rt z4 = (w,0) ; draw z3--z4 ;
3598 lft z5 = (0,h/2) ; rt z6 = (w,h/2) ; draw z5--z6 ;
3599 penlabels(1,2,3,4,5,6);
3600 endchar;
3601
3602 % \end{macrocode}
3603 % \end{fontcode}
3604 % \begin{fontcode}{I}
3605 % It doesn't get any simpler!
3606 % \begin{macrocode}
3607 beginchar("I",P_t#,l_h#,0); "The letter I" ;
3608 pickup Unipensqu ;
3609 top lft z1 = (0,h) ; bot lft z2 = (0,0) ; draw z1--z2 ;
3610 penlabels(1,2);
3611 endchar;
3612
3613 % \end{macrocode}
3614 % \end{fontcode}
3615 % \begin{fontcode}{J}
3616 % \begin{macrocode}
3617 beginchar("J",P_t#+2C_d#,l_h#,0); "The letter J";
3618 uniend(P_t,(w-P_t,P_t),270,360,10);
3619 pickup Unipensqu ;
3620 lft bot z1 = (0,0) ; rt bot z2 = (w-P_t,0) ; draw z1--z2 ;
3621 rt bot z3 = (w,P_t) ; rt top z4 = (w,h) ; draw z3--z4 ;
3622 penlabels(1,2,3,4);
3623 endchar;
3624
3625 % \end{macrocode}
3626 % \end{fontcode}
3627 % \begin{fontcode}{K}
3628 % Similar to \textunibf{k}.
3629 % \begin{macrocode}
3630 beginchar("K",(l_h#+3P_t#)/2,l_h#,0) ; "The letter K" ;
3631 uniarc((h+P_t)/2,P_t,(P_t,h),270,360,10) ;
3632 uniarc((h+P_t)/2,P_t,(P_t,0),0,90,20) ;
3633 pickup Unipensqu ;
3634 top lft z1 = (0,h) ; bot lft z2 = (0,0) ; draw z1--z2 ;
3635 penlabels(1,2);
3636 endchar;
3637
3638 % \end{macrocode}
3639 % \end{fontcode}
3640 % \begin{fontcode}{L}
3641 % This is just a reflected \textunibf{J}.
3642 % \begin{macrocode}
3643 beginchar("L",P_t#+2C_d#,l_h#,0) ; "The letter L" ;
3644 uniend(P_t,(P_t,P_t),180,270,10) ;
3645 pickup Unipensqu ;
3646 top lft z1 = (0,h) ; bot lft z2 = (0,P_t) ; draw z1--z2 ;
3647 bot lft z3 = (P_t,0) ; bot rt z4 = (w,0) ; draw z3--z4 ;
3648 penlabels(1,2,3,4) ;
3649 endchar;
3650
3651 % \end{macrocode}
3652 % \end{fontcode}
3653 % \begin{fontcode}{M}
3654 % The same kind of considerations as in \textunibf{m} went into this
3655 % letter.
3656 % \begin{macrocode}
3657 beginchar("M",M_w#,l_h#,0) ; "The letter M" ;
3658 temp := .25*(w+P_t);
3659 uniarc(temp,P_t,(temp,h-temp),0,180,10) ;
3660 uniarc(temp,P_t,(w-temp,h-temp),0,180,20) ;
3661 pickup Unipensqu ;
3662 top lft z1 = (0,h-temp) ; bot lft z2 = (0,0) ; draw z1--z2 ;
3663 top z3 = (w/2,h-temp) ; bot z4 = (w/2,0) ; draw z3--z4 ;
3664 top rt z5 = (w,h-temp) ; bot rt z6 = (w,0) ; draw z5--z6 ;
3665 penlabels(1,2,3,4,5,6);
3666 endchar;
3667
3668 % \end{macrocode}
3669 % \end{fontcode}
3670 % \begin{fontcode}{N}
3671 % Tricky because of the \textunibf{M} (see above).
3672 % \begin{macrocode}
3673 beginchar("N",A_w#,l_h#,0); "The letter N" ;
3674 uniarc(w/2,P_t,(w/2,h-w/2),0,180,10) ;
3675 pickup Unipensqu ;
3676 top lft z1 = (0,h-w/2) ; bot lft z2 = (0,0); draw z1--z2 ;
3677 top rt z3 = (w,h-w/2) ; bot rt z4 = (w,0) ; draw z3--z4 ;
3678 penlabels(1,2,3,4) ;
3679 endchar;
3680
3681 % \end{macrocode}
3682 % \end{fontcode}
3683 % \begin{fontcode}{O}
3684 % No, this is not going to be an oval, but a \emph{circle\/}!
3685 % \begin{macrocode}
3686 beginchar("O",l_h#,l_h#,0) ; "The letter O";
3687 unicir(w/2,P_t,(w/2,h/2),10);
3688 endchar;
3689
3690 % \end{macrocode}
3691 % \end{fontcode}
3692 % \begin{fontcode}{P}
3693 % No surprises.
3694 % \begin{macrocode}
3695 beginchar("P",A_w#,l_h#,0) ; "The letter P";
3696 unicir(w/2,P_t,(w/2,h-w/2),10)
3697 pickup Unipensqu ;
3698 lft top z1 = (0,h-w/2) ; lft bot z2 = (0,0) ; draw z1--z2 ;
3699 penlabels(1,2);
3700 endchar;
3701
3702 % \end{macrocode}
3703 % \end{fontcode}
3704 % \begin{fontcode}{Q}
3705 % There are to variants of this letter, one is an reflected
3706 % \textunibf{P}, and the other is a vaiant of an \textunibf{O}. The
3707 % default one, is the \textunibf{Q} variant. If you want to use the
3708 % other variant, you can acceses it directly with \Lcs{varQ}, or you
3709 % can pass the argument \texttt{varQ} to the \textsc{uni} package.
3710 % (Option \texttt{varQ} not implemented yet!)
3711 % \begin{macrocode}
3712 beginchar("Q",l_h#,l_h#,p_d#) ; "The letter Q";
3713 unicir(w/2,P_t,(w/2,w/2),10);
3714 pickup Unipensqu ;
3715 top z1 = (w/2,3P_t) ; bot z2 = (w/2,-d) ; draw z1--z2 ;
3716 penlabels(1,2);
3717 endchar;
3718
3719 % \end{macrocode}
3720 % \end{fontcode}
3721 % \begin{fontcode}{R}
3722 % I'm not enterily happy with this letter, but quite frankly, I don't
3723 % know how else to make it. Please give me your suggestion.
3724 % \begin{macrocode}
3725 beginchar("R",A_w#,l_h#,0) ; "The letter R";
3726 unicir(w/2,P_t,(w/2,h-w/2),10) ;
3727 rad := h-w+P_t ;
3728 uniarc(rad,P_t,(w-rad,0),0,90,20) ;
3729 pickup Unipensqu ;
3730 lft top z1 = (0,h-w/2) ; bot lft z2 = (0,0) ; draw z1--z2 ;
3731 temp := w/2-rad ;
3732 if temp>0 :
3733 pickup pensquare xscaled temp yscaled P_t ;
3734 lft top z3 = (1/2w,rad) ; drawdot z3 ;
3735 fi;
3736 penlabels(1,2,3) ;
3737 endchar;
3738
3739 % \end{macrocode}
3740 % \end{fontcode}
3741 % \begin{fontcode}{S}
3742 % Just like the minuscle \textunibf{s}, this is a `slow' letter, i.e.,
3743 % it takes \MF{} quite some time to compute.
3744 % \begin{macrocode}
3745 beginchar("S",A_w#,l_h#,0) ; "The letter S" ;
3746 temp := (h+P_t)/4 ;
3747 uniarc(temp,P_t,(temp,temp),180,270,10) ;
3748 uniarc(temp,P_t,(w-temp,temp),270,450,20) ;
3749 uniarc(temp,P_t,(temp,h-temp),90,270,30) ;
3750 uniarc(temp,P_t,(w-temp,h-temp),0,90,40) ;
3751 pickup Unipensqu ;
3752 bot lft z1 = (temp,0) ; bot rt z2 = (w-temp,0) ; draw z1--z2 ;
3753 lft z3 = (temp,h/2) ; rt z4 = (w-temp,h/2) ; draw z3--z4 ;
3754 top lft z5 = (temp,h) ; top rt z6 = (w-temp,h) ; draw z5--z6 ;
3755 penlabels(1,2,3,4,5,6) ;
3756 endchar;
3757
3758 % \end{macrocode}
3759 % \end{fontcode}
3760 % \begin{fontcode}{T}
3761 % Again --- simple!
3762 % \begin{macrocode}
3763 beginchar("T",A_w#,l_h#,0); "The letter T" ;
3764 pickup Unipensqu ;
3765 top z1 = (w/2,h) ; bot z2 = (w/2,0) ; draw z1--z2 ;
3766 top lft z3 = (0,h) ; top rt z4 = (w,h) ; draw z3--z4 ;
3767 penlabels(1,2,3,4);
3768 endchar;
3769
3770 % \end{macrocode}
3771 % \end{fontcode}
3772 % \begin{fontcode}{U}
3773 % \emph{Font--jepody:} The character of the \universal{} font, that,
3774 % when reflected gives \textunibf{U}.
3775 % \begin{macrocode}
3776 beginchar("U",A_w#,l_h#,0); "The letter U" ;
3777 uniarc(w/2,P_t,(w/2,w/2),180,360,10) ;
3778 pickup Unipensqu ;
3779 top lft z1 = (0,h) ; bot lft z2 = (0,w/2) ; draw z1--z2 ;
3780 top rt z3 = (w,h) ; bot rt z4 = (w,w/2) ; draw z3--z4 ;
3781 penlabels(1,2,3,4) ;
3782 endchar;
3783
3784 % \end{macrocode}
3785 % \end{fontcode}
3786 % \begin{fontcode}{V}
3787 % See also the minuscle \textunibf{v}.
3788 % \begin{macrocode}
3789 beginchar("V",A_w#,l_h#,0); "The letter V" ;
3790 unilne(P_t,(0,h),(w/2,P_t),1,1,10);
3791 unilne(P_t,(w/2,P_t),(w,h),1,1,20);
3792 uniarc(P_t,P_t,(w/2,P_t),180,360,30);
3793 endchar;
3794
3795 % \end{macrocode}
3796 % \end{fontcode}
3797 % \begin{fontcode}{W}
3798 % Flippedy--flop.
3799 % \begin{macrocode}
3800 beginchar("W",M_w#,l_h#,0) ; "The letter W" ;
3801 temp := .25*(w+P_t);
3802 uniarc(temp,P_t,(temp,temp),180,360,10) ;
3803 uniarc(temp,P_t,(w-temp,temp),180,360,20) ;
3804 pickup Unipensqu ;
3805 top lft z1 = (0,h) ; bot lft z2 = (0,temp) ; draw z1--z2 ;
3806 top z3 = (w/2,h) ; bot z4 = (w/2,temp) ; draw z3--z4 ;
3807 top rt z5 = (w,h) ; bot rt z6 = (w,temp) ; draw z5--z6 ;
3808 penlabels(1,2,3,4,5,6);
3809 endchar;
3810
3811 % \end{macrocode}
3812 % \end{fontcode}
3813 % \begin{fontcode}{X}
3814 % This is a bit tricky --- see the minuscle \textunibf{x}.
3815 % \begin{macrocode}
3816 beginchar("X",l_h#,l_h#,0) ; "The letter X" ;
3817 uniarc(h/2,P_t,(P_t/2,h/2),270,90,10);
3818 uniarc(h/2,P_t,(w-P_t/2,h/2),90,270,20);
3819 temp := (h-w+P_t)/2 ;
3820 pickup pensquare xscaled temp yscaled P_t slanted slantratio ;
3821 lft top z1 = (0,h) ; draw z1 ;
3822 lft bot z2 = (0,0) ; draw z2 ;
3823 rt top z3 = (w,h) ; draw z3 ;
3824 rt bot z4 = (w,0) ; draw z4 ;
3825 penlabels(1,2,3,4);
3826 endchar;
3827
3828 % \end{macrocode}
3829 % \end{fontcode}
3830 % \begin{fontcode}{Y}
3831 % \Newestchanges{v2.0}{98/08/01}{Character \textunibf{Y} changed.}
3832 % This actually a bit different from \textunibf{y}, but it looks right.
3833 % \begin{macrocode}
3834 beginchar("Y",A_w#,l_h#,0); "The letter Y" ;
3835 uniarc(w/2,P_t,(w/2,h-P_t),180,360,10) ;
3836 pickup Unipensqu ;
3837 top lft z1 = (0,h) ; drawdot z1 ;
3838 top rt z2 = (w,h) ; drawdot z2 ;
3839 top z3 = (w/2,h-w/2) ; bot z4 = (w/2,0) ; draw z3--z4 ;
3840 penlabels(1,2,3,4) ;
3841 endchar;
3842
3843 % \end{macrocode}
3844 % \end{fontcode}
3845 % \begin{fontcode}{Z}
3846 % Again a tricky letter. For more onm this please refer to the
3847 % minuscle \textunibf{z}.
3848 %
3849 % Notice that the angle \texttt{phi} in this majuscle is not the
3850 % same as in the minuscles \textunibf{z}, since this letter doesn't fit
3851 % in a square, but is retangular.
3852 % \begin{macrocode}
3853 beginchar("Z",A_w#,l_h#,0) ; "The letter Z" ;
3854 pickup Unipensqu ;
3855 top lft z1 = (0,h) ; top rt z2 = (w-P_t,h) ; draw z1--z2 ;
3856 bot lft z3 = (P_t,0) ; bot rt z4 = (w,0) ; draw z3--z4 ;
3857 phi := angle((w,h-P_t)-(0,P_t)) ;
3858 z5 = (P_t,P_t) ; z6 = z5 + P_t*(-sind(phi),cosd(phi)) ;
3859 z7 = (w-P_t,h-P_t) ; z8 = z7 + P_t*(sind(phi),-cosd(phi)) ;
3860 uniend(P_t,z5,90+phi,270,10);
3861 uniend(P_t,z7,-90+phi,90,20);
3862 fill z5--z6--z7--z8--cycle ;
3863 penlabels(range 1 thru 8);
3864 endchar;
3865
3866 % \end{macrocode}
3867 % \end{fontcode}
3868 % As explain in the begining of the section, this file is one big
3869 % \texttt{if...else...fi} loop to mak the generation of the strict
3870 % font easier.
3871 %
3872 % The idea is, that if \texttt{strict} is \texttt{true}, then the main
3873 % part of this file shouldn't be used, but rather \texttt{unilow.mf}
3874 % should be loaded again, but with this time the characters should be
3875 % placed where the majuscules should be.
3876 %
3877 % This is done by defining \texttt{code\_offset} to be the (negative)
3878 % difference the ASCII code for \texttt{A} and \texttt{a}. After we
3879 % have input \texttt{unilow.mf} once more, we reset
3880 % \texttt{code\_offset}.
3881 % \begin{macrocode}
3882 else :
3883 code_offset := ASCII"A" - ASCII"a" ;
3884 input unilow ;
3885 code_offset := 0 ;
3886 fi ;
3887
3888 % \end{macrocode}
3889 %\iffalse
3890 %</upp>
3891 %\fi
3892 %\iffalse
3893 %<*dig>
3894 %\fi
3895 % \subsection{The Numbers --- (\texttt{unidig.mf})}
3896 % \file{unidig.mf}
3897 %
3898 % I couldn't find any design of numbers, I could directly ascribe to
3899 % Herbert Bayer. I, however, found some other examples, and from
3900 % these, and my general conception of the font, I designed the
3901 % numerals, as they are seen below.
3902 %
3903 % \begin{fontcode}{0}
3904 % Perhaps this should have been a circle, but to destingice the zero
3905 % from the majuscle {\uni O}, I chose to strecth it.
3906 % \begin{macrocode}
3907 beginchar("0",A_w#,l_h#,0) ; "The number 0" ;
3908 uniarc(w/2,p_t,(w/2,h-w/2),0,180,10) ;
3909 uniarc(w/2,p_t,(w/2,w/2),180,360,20) ;
3910 pickup unipensqu ;
3911 top lft z1 = (0,h-w/2) ; bot lft z2 = (0,w/2) ; draw z1--z2 ;
3912 top rt z3 = (w,h-w/2) ; bot rt z4 = (w,w/2) ; draw z3--z4 ;
3913 penlabels(1,2,3,4) ;
3914 endchar;
3915
3916 % \end{macrocode}
3917 % \end{fontcode}
3918 % \begin{fontcode}{1}
3919 % Here we use the dimension \texttt{c\_d}, since the stem should be
3920 % thee same lenght in the medium font as in the bold face one.
3921 % \begin{macrocode}
3922 beginchar("1",2c_d#,l_h#,0) ; "The number 1" ;
3923 uniend(p_t,(w-p_t,h-p_t),0,90,10) ;
3924 pickup unipensqu ;
3925 lft top z1 = (0,h) ; rt top z2 = (w-p_t,h) ; draw z1--z2 ;
3926 rt top z3 = (w,h-p_t) ; rt bot z4 = (w,0) ; draw z3--z4 ;
3927 penlabels(1,2,3,4) ;
3928 endchar;
3929
3930 % \end{macrocode}
3931 % \end{fontcode}
3932 % \begin{fontcode}{2}
3933 % Again, this is a bit tricky, for the same reason as the minuscle
3934 % {\uni z}.
3935 % \begin{macrocode}
3936 beginchar("2",A_w#,l_h#,0) ; "The number 2" ;
3937 uniarc(w/2,p_t,(w/2,h-w/2),0,180,10) ;
3938 phi := angle((w,h-w/2)-(0,p_t)) ;
3939 z1 = (w-p_t,h-w/2) ; z2 = z1 + p_t*(sind(phi),-cosd(phi)) ;
3940 z3 = (p_t,p_t) ; z4 = z3 + p_t*(-sind(phi),cosd(phi)) ;
3941 fill z1--z2--z3--z4--cycle ;
3942 uniend(p_t,z1,270+phi,360,20) ;
3943 uniend(p_t,z3,90+phi,270,30) ;
3944 pickup unipensqu ;
3945 bot lft z5 = (p_t,0) ; bot rt z6 = (w,0) ; draw z5--z6 ;
3946 penlabels(range 1 thru 6) ;
3947 endchar;
3948
3949 % \end{macrocode}
3950 % \end{fontcode}
3951 % \begin{fontcode}{3}
3952 % By first defining to paths similar to the two circle segments, it is
3953 % possible to calculate the intersectionpoint, and thereby the angle
3954 % \texttt{phi}, where the to segments meet, relative to one of the
3955 % centers. This angle is then used to define the \emph{real} path, and
3956 % the rightmost point of the middle bar.
3957 % \begin{macrocode}
3958 beginchar("3",A_w#,l_h#,0) ; "The number 3" ;
3959 path arc_a, arc_b ;
3960 arc_a = (w,h-w/2)..(w/2,h-w) ; arc_b = (w,w/2)..(w/2,w) ;
3961 z3 = arc_a intersectionpoint arc_b ; phi := angle(z3-(w/2,w/2)) ;
3962 uniarc(w/2,p_t,(w/2,h-w/2),-phi,180,10) ;
3963 uniarc(w/2,p_t,(w/2,w/2),-180,phi,20) ;
3964 pickup unipensqu ;
3965 rt z1 = (w/2*(1+cosd(phi)),h/2) ; lft z2 = (w/2,h/2) ; draw z1--z2 ;
3966 penlabels(1,2) ;
3967 endchar;
3968
3969 % \end{macrocode}
3970 % \end{fontcode}
3971 % \begin{fontcode}{4}
3972 % One could argue, that the top left stroke should be vertical, but I
3973 % chose to go for a $45^\circ$ slanted line, for aestethic reasons.
3974 % \begin{macrocode}
3975 beginchar("4",A_w#,l_h#,0) ; "The number 4" ;
3976 z1 = (w-p_t,h-p_t) ; z2 = z1 + p_t*(-1/sqrt(2),1/sqrt(2)) ;
3977 z3 = (p_t,h-w+p_t) ; z4 = z3 + p_t*(-1/sqrt(2),1/sqrt(2)) ;
3978 fill z1--z2--z4--z3--cycle ;
3979 uniend(p_t,z1,0,135,10) ;
3980 uniend(p_t,z3,135,270,20) ;
3981 pickup unipensqu ;
3982 top rt z5 = (w,h-p_t) ; bot rt z6 = (w,0) ; draw z5--z6 ;
3983 bot lft z7 = (p_t,h-w) ; bot rt z8 = (w,h-w) ; draw z7--z8 ;
3984 penlabels(range 1 thru 8) ;
3985 endchar;
3986
3987 % \end{macrocode}
3988 % \end{fontcode}
3989 % \begin{fontcode}{5}
3990 % This gave me some trouble. I originally wanted the arc to have a
3991 % diameter the width of the character, but that meant, that the to
3992 % horizontal lines of the font, came so close to each other, that it
3993 % didn't look good. So I played a little with \texttt{rad}, and
3994 % decided that the current value gave the nicest resault.
3995 % \begin{macrocode}
3996 beginchar("5",A_w#,l_h#,0) ; "The number 5" ;
3997 rad := 9w/20 ;
3998 uniarc(rad,p_t,(w-rad,rad),-150,90,10) ;
3999 uniend(p_t,(p_t,h-p_t),90,180,20) ;
4000 uniend(p_t,(p_t,2rad),180,270,30) ;
4001 pickup unipensqu ;
4002 top lft z1 = (p_t,2rad) ; top rt z2 = (w-rad,2rad) ; draw z1--z2 ;
4003 bot lft z3 = (0,2rad) ; top lft z4 = (0,h-p_t) ; draw z3--z4 ;
4004 top lft z5 = (p_t,h) ; top rt z6 = (w,h) ; draw z5--z6 ;
4005 penlabels(range 1 thru 6) ;
4006 endchar;
4007
4008 % \end{macrocode}
4009 % \end{fontcode}
4010 % \begin{fontcode}{6}
4011 % This was almost to easy --- a circle, an arc, and a line connecting
4012 % them.
4013 % \begin{macrocode}
4014 beginchar("6",A_w#,l_h#,0) ; "The number 6" ;
4015 unicir(w/2,p_t,(w/2,w/2),10) ;
4016 uniarc(w/2,p_t,(w/2,h-w/2),60,180,20) ;
4017 pickup unipensqu ;
4018 top lft z1 = (0,h-w/2) ; bot lft z2 = (0,w/2) ; draw z1--z2 ;
4019 penlabels(1,2) ;
4020 endchar;
4021
4022 % \end{macrocode}
4023 % \end{fontcode}
4024 % \begin{fontcode}{7}
4025 % Apperently, this character could have the same problems as the
4026 % \textuni{2}, but it didn't --- can you see why not?
4027 % \begin{macrocode}
4028 beginchar("7",A_w#,l_h#,0) ; "The number 7" ;
4029 z1 = (w-p_t,h-p_t) ; phi := angle(z1) ;
4030 uniend(p_t,z1,-90+phi,90,10) ;
4031 z2 = z1 + p_t*(cosd(90-phi),-sind(90-phi)) ;
4032 z3 = (p_t/cosd(90-phi),0) ; z4 = (0,0) ;
4033 fill z1--z2--z3--z4--cycle ;
4034 pickup unipensqu ;
4035 top lft z5 = (0,h) ; top rt z6 = (w-p_t,h) ; draw z5--z6 ;
4036 penlabels(range 1 thru 6) ;
4037 endchar;
4038
4039 % \end{macrocode}
4040 % \end{fontcode}
4041 % \begin{fontcode}{8}
4042 % Ha ha, to circles, one on top of the other. Some may prefer if the
4043 % bottom most circle had a slightly larger radius then the top most,
4044 % but I believe that to be in conflict with the general idea of
4045 % simplicity.
4046 % \begin{macrocode}
4047 beginchar("8",1/2*(l_h#+p_t#),l_h#,0) ; "The number 8" ;
4048 unicir(w/2,p_t,(w/2,w/2),10) ;
4049 unicir(w/2,p_t,(w/2,h-w/2),20) ;
4050 endchar;
4051
4052 % \end{macrocode}
4053 % \end{fontcode}
4054 % \begin{fontcode}{9}
4055 % See above under \textuni{6}.
4056 % \begin{macrocode}
4057 beginchar("9",A_w#,l_h#,0) ; "The number 9" ;
4058 unicir(w/2,p_t,(w/2,h-w/2),10) ;
4059 uniarc(w/2,p_t,(w/2,w/2),-150,0,20) ;
4060 pickup unipensqu ;
4061 top rt z1 = (w,h-w/2) ; bot rt z2 = (w,w/2) ; draw z1--z2 ;
4062 penlabels(1,2) ;
4063 endchar;
4064
4065 % \end{macrocode}
4066 % \end{fontcode}
4067 %\iffalse
4068 %</dig>
4069 %\fi
4070 %\iffalse
4071 %<*spe>
4072 %\fi
4073 % \subsection{Specials --- \texttt{unispe.mf}}
4074 % \file{unispe.mf}
4075 %
4076 % For these characters, I couldn't find any original design,
4077 % so once again, I did some inter-- and extrapolation, plus I had some
4078 % print--outs of other implenations of Herbert Bayers universal
4079 % font. So once again, I am to blame for any aestethic judgements.
4080 %
4081 % \begin{fontcode}{\char"10}
4082 % At first I thought I didn't had to do this character, but then I
4083 % realized that you \emph{do} needed, since some words are spelled
4084 % with some sort of accent over the ``i'', other then a single dot,
4085 % e.g., \textunibf{na\"ive}.
4086 % \begin{macrocode}
4087 beginchar(hex"10",p_t#,x_h#,0) ; "The dotless letter i" ;
4088 pickup unipensqu ;
4089 bot lft z1 = (0,0) ; top lft z2 = (0,h) ; draw z1--z2;
4090 penlabels(1,2);
4091 endchar;
4092
4093 % \end{macrocode}
4094 % \end{fontcode}
4095 % \begin{fontcode}{\char"11}
4096 % Non single dot j's are not as frequent as those of i's, so for the
4097 % sake of completness, if nothing else, this letter is included.
4098 % \begin{macrocode}
4099 beginchar(hex"11",p_t#,x_h#,p_d#) ; "The dotless letter j" ;
4100 pickup unipensqu ;
4101 bot lft z1 = (0,-d) ; top lft z2 = (0,h) ; draw z1--z2 ;
4102 penlabels(1,2);
4103 endchar;
4104
4105 % \end{macrocode}
4106 % \end{fontcode}
4107 % \begin{fontcode}{\char"24}
4108 % I couldn't make up my mind wether this should just be a \textunibf{s},
4109 % with a vertical bar through, or if I should make it differntly. In
4110 % the end, I chose to make it different, since I think it looks best
4111 % in the context of the font.
4112 % \begin{macrocode}
4113 beginchar("$",(l_h#-c_d#+p_t#)/2,l_h#,c_d#) ; "The dollar sign ($)" ;
4114 uniarc(w/2,p_t,(w/2,w/2),180,450,10) ;
4115 uniarc(w/2,p_t,(w/2,h-c_d-w/2),0,270,20) ;
4116 pickup unipensqu ;
4117 top z1 = (w/2,h) ; bot z2 = (w/2,-d) ; draw z1--z2 ;
4118 penlabels(1,2) ;
4119 endchar;
4120
4121 % \end{macrocode}
4122 % \end{fontcode}
4123 % \begin{fontcode}{\char"26}
4124 % You could argue back and forth on the shape of the ampersand, but I
4125 % decided to go for a simple design.
4126 % \begin{macrocode}
4127 beginchar("&",(l_h#+p_t#)/2+c_d#,l_h#,c_d#) ; "The ampersand (&)" ;
4128 rad := (w-c_d)/2 ;
4129 unicir(rad,p_t,(rad,h-rad),10) ;
4130 uniarc(rad,p_t,(rad,rad),0,270,20) ;
4131 pickup unipensqu ;
4132 lft bot z1 = (rad,0) ; rt bot z2 = (w,0) ; draw z1--z2 ;
4133 rt top z3 = (w-c_d,rad) ; rt bot z4 = (w-c_d,-d) ; draw z3--z4 ;
4134 penlabels(1,2,3,4) ;
4135 endchar;
4136
4137 % \end{macrocode}
4138 % \end{fontcode}
4139 % \begin{fontcode}{?}
4140 % I did have some considerations concerning this. Since I had kind of
4141 % decided to make avery junction ``smooth'', this also had to
4142 % be. However, that meant, that my original plan, to make this letter
4143 % as wide as an \textunibf{A}, didn't work, so I had to make it less
4144 % wide, which in the end does look better.
4145 % \begin{macrocode}
4146 beginchar("?",l_h#-5c_d#/2,l_h#,0) ; "The question mark (?)" ;
4147 temp := angle((P_T,-w/2)) - 0.01 ;
4148 uniarc(w/2,P_T,(w/2,h-w/2),temp,180,10) ;
4149 uniend(P_T,((w+P_T)/2,h-w),90,180,20) ;
4150 uniend(P_T,((w+P_T)/2,h-w+P_T),270,360+temp,30) ;
4151 pickup UniPensqu ;
4152 top z1 = (w/2,h-w) ; bot z2 = (w/2,2P_T) ; draw z1--z2 ;
4153 pickup UniPencir ;
4154 bot z3 = (w/2,0) ; drawdot z3 ;
4155 penlabels(1,2,3) ;
4156 endchar;
4157
4158 % \end{macrocode}
4159 % \end{fontcode}
4160 % \begin{fontcode}{\char"3E}
4161 % These Spanish special letters, are just rotations, so nothing
4162 % special is going on here.
4163 % \begin{macrocode}
4164 beginchar(spanish_query,l_h#-5c_d#/2,x_h#,p_d#) ; "The Spanish query" ;
4165 temp := angle((-P_T,w/2))-0.01 ;
4166 uniarc(w/2,P_T,(w/2,-d+w/2),temp,360,10) ;
4167 uniend(P_T,((w-P_T)/2,-d+w),270,360,20) ;
4168 uniend(P_T,((w-P_T)/2,-d+w-P_T),90,temp,30) ;
4169 pickup UniPensqu ;
4170 bot z1 = (w/2,-d+w) ; top z2 = (w/2,h-2P_T) ; draw z1--z2 ;
4171 pickup UniPencir ;
4172 top z3 = (w/2,h) ; drawdot z3 ;
4173 penlabels(1,2,3) ;
4174 endchar;
4175
4176 % \end{macrocode}
4177 % \end{fontcode}
4178 % \begin{fontcode}{\char"3C}
4179 % See above.
4180 % \begin{macrocode}
4181 beginchar(spanish_shriek,P_T#,x_h#,p_d#) ; "The Spanish shriek" ;
4182 pickup UniPencir ; top z1 = (w/2,h) ; drawdot z1 ;
4183 pickup UniPensqu ; top z2 = (w/2,h-2P_T) ; bot z3 = (w/2,-d) ;
4184 draw z2--z3 ; penlabels(1,2,3) ;
4185 endchar;
4186
4187 % \end{macrocode}
4188 % \end{fontcode}
4189 %\iffalse
4190 %</spe>
4191 %\fi
4192 %\iffalse
4193 %<*pun>
4194 %\fi
4195 % \subsection{Punctuations --- \texttt{unipun.mf}}
4196 % \file{unipun.mf}
4197 %
4198 % This is the programs for general punctuation marks, ranging from
4199 % period to ast.
4200 %
4201 % \begin{fontcode}{\char"21}
4202 % Quite a simple character. Notice the use of \texttt{unipencir}.
4203 % \begin{macrocode}
4204 beginchar("!",P_T#,l_h#,0) ; "The exclamation mark (!)" ;
4205 pickup UniPensqu ;
4206 top lft z1 = (0,h) ; bot lft z2 = (0,2P_T) ; draw z1--z2 ;
4207 pickup UniPencir ;
4208 bot lft z3 = (0,0) ; drawdot z3 ;
4209 penlabels(1,2,3) ;
4210 endchar;
4211
4212 % \end{macrocode}
4213 % \end{fontcode}
4214 % \begin{fontcode}{(}
4215 % Here, and in the next character ()) we don't use pens, since we can
4216 % get a much better resault using the \MF{} path macro \texttt{...} .
4217 % \begin{macrocode}
4218 beginchar("(",x_w#/2,l_h#,p_d#) ; "The left paranthesis (()" ;
4219 z1 = (w,h) ; z2 = (p_t,(h-d)/2) ; z3 = (w,-d) ;
4220 z4 = (w-p_t,-d) ; z5 = (0,(h-d)/2) ; z6 = (w-p_t,h) ;
4221 fill z1..z2..z3--z4..z5..z6--cycle ;
4222 penlabels(range 1 thru 6) ;
4223 endchar;
4224
4225 % \end{macrocode}
4226 % \end{fontcode}
4227 % \begin{fontcode}{)}
4228 % \begin{macrocode}
4229 beginchar(")",x_w#/2,l_h#,p_d#) ; "The right paranthesis ())" ;
4230 z1 = (p_t,h) ; z2 = (w,(h-d)/2) ; z3 = (p_t,-d) ;
4231 z4 = (0,-d) ; z5 = (w-p_t,(h-d)/2) ; z6 = (0,h) ;
4232 fill z1..z2..z3--z4..z5..z6--cycle ;
4233 penlabels(range 1 thru 6) ;
4234 endchar;
4235
4236 % \end{macrocode}
4237 % \end{fontcode}
4238 %
4239 % \begin{fontcode}{\#}
4240 % I never really got satisfied with this letter, but this is the best
4241 % I could do --- who ever uses a \# anyway!
4242 % \begin{macrocode}
4243 beginchar("#",x_w#,x_h#,0) ; "The hash mark (#)" ;
4244 unilne(p_t,(0,0),(w/2,h),1,1,10) ;
4245 unilne(p_t,(w/2,0),(w,h),1,1,20) ;
4246 pickup unipensqu ;
4247 top lft z1 = (0,(h-c_d)/2) ; top rt z2 = (w,(h-c_d)/2) ;
4248 bot lft z3 = (0,(h+c_d)/2) ; bot rt z4 = (w,(h+c_d)/2) ;
4249 draw z1--z2 ; draw z3--z4 ;
4250 penlabels(1,2,3,4) ;
4251 endchar;
4252
4253 % \end{macrocode}
4254 % \end{fontcode}
4255 % \begin{fontcode}{\%}
4256 % Among other characters, this one made me define a dimension called
4257 % \texttt{c\_d}, which is to be used for all kinds of curves. The idea
4258 % is that, some sizes curves should depend on the weight of the font,
4259 % and others not. Those that \emph{shouldn't} will use \texttt{c\_d}.
4260 % \begin{macrocode}
4261 beginchar("%",x_w#,l_h#,0) ; "The percent sign (%)" ;
4262 temp := 3c_d/2 ;
4263 unicir(temp,p_t,(temp,h-temp),10) ;
4264 unicir(temp,p_t,(w-temp,temp),20) ;
4265 unilne(p_t,(0,0),(w,h),1,1,30) ;
4266 endchar;
4267
4268 % \end{macrocode}
4269 % \end{fontcode}
4270 % \begin{fontcode}{'}
4271 % Agian we have a translated comma --- the apostrophe.
4272 % \begin{macrocode}
4273 beginchar("'",2P_T#,l_h#,0) ; "The apostrophe (')" ;
4274 uniend(w/2,(w/2,h-w/2),0,180,10) ;
4275 z1 = (0,h-w/2) ; z2 = (w,h-w/2) ; z3 = (0,h-3w/2) ;
4276 fill z1--z2--z3--cycle ; penlabels(1,2,3) ;
4277 endchar;
4278
4279 % \end{macrocode}
4280 % \end{fontcode}
4281 % \begin{fontcode}{*}
4282 % This isn't really complicated execpt for the value of \texttt{temp},
4283 % which insures that the top of ast will be in the character.
4284 % \begin{macrocode}
4285 beginchar("*",x_w#,l_h#,0) ; "The ast (*)" ;
4286 path rect ;
4287 temp := h-(w*sind(60)-p_t*cosd(60))/2 ;
4288 pickup unipensqu ;
4289 lft z1 = (0,temp) ; rt z2 = (w,temp) ; draw z1--z2 ;
4290 rect = (bot lft z1)--(top lft z1)--(top rt z2)--(bot rt z2)--cycle ;
4291 fill rect rotatedabout((w/2,temp),60) ;
4292 fill rect rotatedabout((w/2,temp),120) ;
4293 penlabels(1,2) ;
4294 endchar;
4295
4296 % \end{macrocode}
4297 % \end{fontcode}
4298 % \begin{fontcode}{+}
4299 % Plus and minus, now that's easy.
4300 % \begin{macrocode}
4301 beginchar("+",x_w#,x_h#,0) ; "The plus sign (+)" ;
4302 pickup unipensqu ;
4303 lft z1 = (0,w/2) ; rt z2 = (w,w/2) ; draw z1--z2 ;
4304 bot z3 = (w/2,0) ; top z4 = (w/2,w) ; draw z3--z4 ;
4305 endchar;
4306
4307 % \end{macrocode}
4308 % \end{fontcode}
4309 % \begin{fontcode}{,}
4310 % Incedently, the apostrophs (both single and reversed), are
4311 % just translations and reflections of the comma.
4312 %
4313 % \textsl{Mathemathical problem:} Work out which algebraic group the
4314 % characters `\textunibf{,}', `\textunibf{'}', and `\textunibf{`}',
4315 % belongs to?
4316 % \begin{macrocode}
4317 beginchar(",",2P_T#,2P_T#,2P_T#) ; "The punctuation ," ;
4318 uniend(w/2,(w/2,0),0,180,10) ;
4319 z1 = (0,0) ; z2 = (w,0) ; z3 = (0,-d) ;
4320 fill z1--z2--z3--cycle ; penlabels(1,2,3) ;
4321 endchar;
4322
4323 % \end{macrocode}
4324 % \end{fontcode}
4325 % \begin{fontcode}{-}
4326 % Minus? --- isn't it the opposite of plus?
4327 % \begin{macrocode}
4328 beginchar("-",4x_w#/5,x_h#,0) ; "The hyphen (-)" ;
4329 pickup unipensqu ;
4330 lft z1 = (0,h/2) ; rt z2 = (w,h/2) ; draw z1--z2 ;
4331 penlabels(1,2);
4332 endchar;
4333
4334 % \end{macrocode}
4335 % \end{fontcode}
4336 % \begin{fontcode}{.}
4337 % Wow, a dot!
4338 % \begin{macrocode}
4339 beginchar(".",2P_T#,2P_T#,0) ; "The punctuation ." ;
4340 uniend(w/2,(w/2,h/2),0,360,10) ;
4341 % pickup UniPencir ; lft bot z1 = (0,0) ; drawdot z1 ; penlabels(1);
4342 endchar;
4343
4344 % \end{macrocode}
4345 % \end{fontcode}
4346 % \begin{fontcode}{/}
4347 % This is a nice utilization of \texttt{unilne}.
4348 % \begin{macrocode}
4349 beginchar("/",x_w#,l_h#,0) ; "The slash (/)" ;
4350 unilne(p_t,(0,0),(w,h),1,1,10) ;
4351 endchar;
4352
4353 % \end{macrocode}
4354 % \end{fontcode}
4355 % \begin{fontcode}{:}
4356 % Wauw, \emph{two} dots --- cool!
4357 % \begin{macrocode}
4358 beginchar(":",2P_T#,x_h#,0) ; "The punctuation :" ;
4359 uniend(w/2,(w/2,h-w/2),0,360,10) ;
4360 uniend(w/2,(w/2,w/2),0,360,20) ;
4361 endchar;
4362
4363 % \end{macrocode}
4364 % \end{fontcode}
4365 % \begin{fontcode}{;}
4366 % Again, this is straight forward.
4367 % \begin{macrocode}
4368 beginchar(";",2P_T#,x_h#,2P_T#) ; "The punctuation ;" ;
4369 uniend(w/2,(w/2,h-w/2),0,360,10) ;
4370 uniend(w/2,(w/2,0),0,180,20) ;
4371 z2 = (0,0) ; z3 = (w,0) ; z4 = (0,-d) ; fill z2--z3--z4--cycle ;
4372 penlabels(1,2,3,4) ;
4373 endchar;
4374
4375 % \end{macrocode}
4376 % \end{fontcode}
4377 % \begin{fontcode}{\char"3D}
4378 % These characters are rediculously simple.
4379 % \begin{macrocode}
4380 beginchar("=",x_w#,x_h#,0) ; "The equal sign (=)" ;
4381 pickup unipensqu ;
4382 top lft z1 = (0,(h-p_t)/2) ; top rt z2 = (w,(h-p_t)/2) ;
4383 bot lft z3 = (0,(h+p_t)/2) ; bot rt z4 = (w,(h+p_t)/2) ;
4384 draw z1--z2 ; draw z3--z4 ;
4385 penlabels(1,2,3,4) ;
4386 endchar;
4387
4388 % \end{macrocode}
4389 % \end{fontcode}
4390 % \begin{fontcode}{@}
4391 % Now this was a bit of a challenge. It is clear, that the left stroke
4392 % should be a straight line tarveling all the way to the baseline, but
4393 % how big should the small circle be. I decided to go for the largest
4394 % one, that didn't touch or intersected the outer arc more then one
4395 % place, i.e., at the far left.
4396 % \begin{macrocode}
4397 beginchar("@",l_h#,l_h#,0) ; "The at sign (@)" ;
4398 uniarc(w/2,p_t,(w/2,h/2),0,240,10) ;
4399 unicir(w/3,p_t,(2w/3,w/3),20) ;
4400 pickup unipensqu ;
4401 top rt z1 = (w,w/2) ; bot rt z2 = (w,0) ; draw z1--z2 ;
4402 penlabels(1,2) ;
4403 endchar;
4404
4405 % \end{macrocode}
4406 % \end{fontcode}
4407 % \begin{fontcode}{[}
4408 % Notice the soft corners. This is ofcourse accomplieshed with
4409 % \texttt{uniend}, which turned out to be a very nice macro.
4410 % \begin{macrocode}
4411 beginchar("[",3p_t#,l_h#,p_d#) ; "The left bracket ([)" ;
4412 uniend(p_t,(p_t,h-p_t),90,180,10) ;
4413 uniend(p_t,(p_t,-d+p_t),180,270,20) ;
4414 pickup unipensqu ;
4415 top lft z1 = (p_t,h) ; top rt z2 = (w,h) ; draw z1--z2 ;
4416 top lft z3 = (0,h-p_t) ; bot lft z4 = (0,-d+p_t) ; draw z3--z4 ;
4417 bot lft z5 = (p_t,-d) ; bot rt z6 = (w,-d) ; draw z5--z6 ;
4418 penlabels(range 1 thru 6) ;
4419 endchar;
4420
4421 % \end{macrocode}
4422 % \end{fontcode}
4423 % \begin{fontcode}{]}
4424 % Speaking of reflections.
4425 % \begin{macrocode}
4426 beginchar("]",3p_t#,l_h#,p_d#) ; "The right braket (])" ;
4427 uniend(p_t,(w-p_t,h-p_t),0,90,10) ;
4428 uniend(p_t,(w-p_t,-d+p_t),270,360,20) ;
4429 pickup unipensqu ;
4430 top lft z1 = (0,h) ; top rt z2 = (w-p_t,h) ; draw z1--z2 ;
4431 top rt z3 = (w,h-p_t) ; bot rt z4 = (w,-d+p_t) ; draw z3--z4 ;
4432 bot lft z5 = (0,-d) ; bot rt z6 = (w-p_t,-d) ; draw z5--z6 ;
4433 penlabels(range 1 thru 6) ;
4434 endchar;
4435
4436 % \end{macrocode}
4437 % \end{fontcode}
4438 % \begin{fontcode}{`}
4439 % Who is this `comma' character --- he keeps poping up the strangest
4440 % places.
4441 % \begin{macrocode}
4442 beginchar("`",2P_T#,l_h#,0) ; "The reversed apostroph (`)" ;
4443 uniend(w/2,(w/2,h-w),180,360,10) ;
4444 z1 = (0,h) ; z2 = (0,h-w) ; z3 = (w,h-w) ; fill z1--z2--z3--cycle ;
4445 penlabels(1,2,3) ;
4446 endchar;
4447
4448 % \end{macrocode}
4449 % \end{fontcode}
4450 % \begin{fontcode}{--}
4451 % Ah, a horizontal line --- how fascinating!
4452 % \begin{macrocode}
4453 beginchar(hex"7B",m_w#,x_h#,0) ; "The en-dash (--)" ;
4454 pickup unipensqu ;
4455 lft z1 = (0,h/2) ; rt z2 = (w,h/2) ; draw z1--z2 ;
4456 penlabels(1,2) ;
4457 endchar;
4458
4459 % \end{macrocode}
4460 % \end{fontcode}
4461 % \begin{fontcode}{---}
4462 % And another horizontal line --- and it's even longer then the
4463 % previous one.
4464 % \begin{macrocode}
4465 beginchar(hex"7C",M_w#,x_h#,0) ; "The em-dash (---)" ;
4466 pickup unipensqu ;
4467 lft z1 = (0,h/2) ; rt z2 = (w,h/2) ; draw z1--z2 ;
4468 penlabels(1,2) ;
4469 endchar;
4470
4471 % \end{macrocode}
4472 % \end{fontcode}
4473 %\iffalse
4474 %</pun>
4475 %\fi
4476 %\iffalse
4477 %<*acc>
4478 %\fi
4479 % \subsection{Accents --- \texttt{uniacc.mf}}
4480 % \file{uniacc.mf}
4481 % \begin{fontcode}{\char"12}
4482 % Even though you use a reversed apostrophe in \LaTeX{} to get this
4483 % accent, the accent shouldn't look like a apostophe, that would be
4484 % far to clumsy. So instead we \emph{rotate} a comma and translate it
4485 % --- algebraic groups again!
4486 % \begin{macrocode}
4487 beginchar(hex"12",x_w#,l_h#,0) ; "The accent `" ;
4488 uniend(P_T/2,(w/2,a_h+P_T/2),90,270,10) ;
4489 z1 = (w/2,a_h+P_T) ; z2 = (w/2,a_h) ; z3 = (w/2+P_T,a_h) ;
4490 fill z1--z2--z3--cycle ; penlabels(1,2,3) ;
4491 endchar ;
4492
4493 % \end{macrocode}
4494 % \end{fontcode}
4495 % \begin{fontcode}{\char"13}
4496 % See above.
4497 % \begin{macrocode}
4498 beginchar(hex"13",x_w#,l_h#,0) ; "The accent '" ;
4499 uniend(P_T/2,(w/2,a_h+P_T/2),-90,90,10) ;
4500 z1 = (w/2,a_h+P_T) ; z2 = (w/2,a_h) ; z3 = (w/2-P_T,a_h) ;
4501 fill z1--z2--z3--cycle ; penlabels(1,2,3) ;
4502 endchar ;
4503
4504 % \end{macrocode}
4505 % \end{fontcode}
4506 % \begin{fontcode}{\char"14}
4507 % \begin{macrocode}
4508 beginchar(hex"14",x_w#,l_h#,0) ; "The accent hachek" ;
4509 z1 = (w/2,a_h+p_t) ; uniend(p_t,z1,225,315,10) ;
4510 z2 = z1 + p_t*(-1,-1)/sqrt(2) ; z3 = z1 + p_t*(-1,1) ;
4511 z4 = z3 + (-p_t*sqrt(2),0) ; path rombe ;
4512 rombe := z1--z2--z4--z3--cycle ; fill rombe ;
4513 fill rombe reflectedabout ((w/2,0),(w/2,h)) ;
4514 penlabels(1,2,3,4) ;
4515 endchar ;
4516
4517 % \end{macrocode}
4518 % \end{fontcode}
4519 % \begin{fontcode}{\char"15}
4520 % \begin{macrocode}
4521 beginchar(hex"15",x_w#,l_h#,0) ; "The accent breve" ;
4522 rad := 3p_t/2 ; uniarc(rad,p_t,(w/2,a_h+rad),180,360,10) ;
4523 endchar ;
4524
4525 % \end{macrocode}
4526 % \end{fontcode}
4527 % \begin{fontcode}{\char"16}
4528 % \begin{macrocode}
4529 beginchar(hex"16",x_w#,a_h#+p_t#,0) ; "The accent macron (bar)" ;
4530 pickup unipensqu ;
4531 bot lft z1 = (w/2-2p_t,a_h) ; bot rt z2 = (w/2+2p_t,a_h) ;
4532 draw z1--z2 ; penlabels(1,2) ;
4533 endchar ;
4534
4535 % \end{macrocode}
4536 % \end{fontcode}
4537 % \begin{fontcode}{\char"17}
4538 % \begin{macrocode}
4539 beginchar(hex"17",A_w#+4/3*(l_h#-x_h#)*slantratio,l_h#,0) ;
4540 "The accent circle" ;
4541 rad := P_T ; unicir(rad,p_t,(w/2,h-rad),10) ;
4542 endchar ;
4543
4544 % \end{macrocode}
4545 % \end{fontcode}
4546 % \begin{fontcode}{\char"18}
4547 % \begin{macrocode}
4548 beginchar(hex"18",x_w#,x_h#,2c_d#) ; "The accent cidelia" ;
4549 uniarc(d/2,2P_T/3,(w/2+d/4,-d/2),90,270,10) ;
4550 endchar ;
4551
4552 % \end{macrocode}
4553 % \end{fontcode}
4554 % \begin{fontcode}{\char"20}
4555 % \begin{macrocode}
4556 phi := 15 ; tmp# := 3c_d# ; define_pixels(tmp) ;
4557 beginchar(hex"20",tmp#*cosd(phi)+p_t#*sind(phi),l_h#,0) ;
4558 "The accent bar over l" ;
4559 z1 = (0,(h+p_t*cosd(phi)-tmp*sind(phi))/2) ;
4560 z2 = (p_t*sind(phi),(h-p_t*cosd(phi)-tmp*sind(phi))/2) ;
4561 z3 = (w,(h-p_t*cosd(phi)+tmp*sind(phi))/2) ;
4562 z4 = (w-p_t*sind(phi),(h+p_t*cosd(phi)+tmp*sind(phi))/2) ;
4563 fill z1--z2--z3--z4--cycle ;
4564 penlabels(1,2,3,4) ;
4565 endchar ;
4566
4567 % \end{macrocode}
4568 % \end{fontcode}
4569 % \begin{fontcode}{\char"5E}
4570 % \begin{macrocode}
4571 beginchar(hex"5E",x_w#,l_h#,0) ; "The accent hat" ;
4572 z1 = (w/2,a_h+p_t) ; uniend(p_t,z1,45,136,10) ;
4573 z2 = z1 + p_t*(-1,1)/sqrt(2) ; z3 = z1 + p_t*(-1,-1) ;
4574 z4 = z3 + (-p_t*sqrt(2),0) ; path rombe ;
4575 rombe := z1--z2--z4--z3--cycle ; fill rombe ;
4576 fill rombe reflectedabout ((w/2,0),(w/2,h)) ;
4577 penlabels(1,2,3,4) ;
4578 endchar ;
4579
4580 % \end{macrocode}
4581 % \end{fontcode}
4582 % \begin{fontcode}{\char"5F}
4583 % \begin{macrocode}
4584 beginchar(hex"5F",x_w#,l_h#,0) ; "The accent dot" ;
4585 pickup UniPencir ; bot z1 = (w/2,a_h) ; drawdot z1 ; penlabels(1) ;
4586 endchar ;
4587
4588 % \end{macrocode}
4589 % \end{fontcode}
4590 % \begin{fontcode}{\char"7D}
4591 % \begin{macrocode}
4592 beginchar(hex"7D",x_w#,l_h#,0) ; "The accent ''" ;
4593 rad := P_T/2 ;
4594 uniend(rad,(w/2-rad,a_h+rad),-90,90,10) ;
4595 uniend(rad,(w/2+P_T,a_h+rad),-90,90,20) ;
4596 z1 = (w/2-rad-P_T,a_h) ; z2 = (w/2-rad,a_h) ;
4597 z3 = (w/2-rad,a_h+P_T) ; fill z1--z2--z3--cycle ;
4598 z4 = (w/2,a_h) ; z5 = (w/2+P_T,a_h) ;
4599 z6 = (w/2+P_T,a_h+P_T) ; fill z4--z5--z6--cycle ;
4600 penlabels(range 1 thru 6) ;
4601 endchar ;
4602
4603 % \end{macrocode}
4604 % \end{fontcode}
4605 % \begin{fontcode}{\char"7E}
4606 % \begin{macrocode}
4607 beginchar(hex"7E",x_w#,l_h#,0) ; "The accent ~" ;
4608 rad := (w*sqrt(2)+p_t)/4 ; temp := (2w-p_t*sqrt(2))/8 ;
4609 uniarc(rad,p_t,(w/2+temp,a_h+rad),225,315,10) ;
4610 uniarc(rad,p_t,(w/2-temp,a_h+rad-2temp),45,135,20) ;
4611 endchar ;
4612
4613 % \end{macrocode}
4614 % \end{fontcode}
4615 % \begin{fontcode}{\char"7F}
4616 % \begin{macrocode}
4617 beginchar(hex"7F",x_w#,l_h#,0) ; "The accent umlout (double dot)" ;
4618 pickup UniPencir ;
4619 bot z1 = (w/2-P_T,a_h) ; bot z2 = (w/2+P_T,a_h) ;
4620 drawdot z1 ; drawdot z2 ; penlabels(1,2) ;
4621 endchar ;
4622
4623 % \end{macrocode}
4624 % \end{fontcode}
4625 %\iffalse
4626 %</acc>
4627 %\fi
4628 %\iffalse
4629 %<*lig>
4630 %\fi
4631 % \subsection{Ligatures --- \texttt{unilig.mf}}
4632 % \file{unilig.mf}
4633 %
4634 % \begin{fontcode}{SS}
4635 % If we are making a small caps shaped font, then we need to make an
4636 % upper case version of the german double s (\ss), so we start of
4637 % testing if the \texttt{caps} flag is set.
4638 % \begin{macrocode}
4639 if caps:
4640 beginchar(hex"19",5/2x_w#,x_h#,0) ; "Ligature SS" ;
4641 temp := (h+p_t)/4 ;
4642 uniarc(temp,p_t,(temp,temp),180,270,10) ;
4643 uniarc(temp,p_t,(2w/5-temp,temp),270,450,20) ;
4644 uniarc(temp,p_t,(temp,h-temp),90,270,30) ;
4645 uniarc(temp,p_t,(2w/5-temp,h-temp),0,90,40) ;
4646 pickup unipensqu ;
4647 bot lft z1 = (temp,0) ; bot rt z2 = (2w/5-temp,0) ;
4648 lft z3 = (temp,h/2) ; rt z4 = (2w/5-temp,h/2) ;
4649 top lft z5 = (temp,h) ; top rt z6 = (2w/5-temp,h) ;
4650 draw z1--z2 ; draw z3--z4 ; draw z5--z6 ;
4651 penlabels(1,2,3,4,5,6) ;
4652 currentpicture := currentpicture + (currentpicture shifted
4653 (3/5w,0)) ;
4654 endchar ;
4655
4656 % \end{macrocode}
4657 % \end{fontcode}
4658 % \begin{fontcode}{\char"19}
4659 % If we are not making a small caps font, then the usual german double
4660 % s is made.
4661 % \begin{macrocode}
4662 else:
4663 beginchar(hex"19",(l_h#+3p_t#)/2,l_h#,p_t#) ; "Ligature ss" ;
4664 rad := (w-p_t)/2 ;
4665 uniarc(rad,p_t,(p_t+rad,h-rad),-90,180,10) ;
4666 uniarc(rad,p_t,(p_t+rad,rad),-90,90,20) ;
4667 pickup unipensqu ;
4668 top lft z1 = (p_t,h-rad) ; bot lft z2 = (p_t,-d) ; draw z1--z2 ;
4669 lft z3 = (0,h/2) ; rt z4 = (w-rad,h/2) ; draw z3--z4 ;
4670 penlabels(1,2,3,4) ;
4671 endchar ;
4672 fi ;
4673
4674 % \end{macrocode}
4675 % \end{fontcode}
4676 % The next three characters should not be made in a small caps font,
4677 % so these are all put in side the conditinal
4678 % \begin{verbatim}
4679 % if not caps :
4680 % \end{verbatim}
4681 % However, we need to put small caps versions of these letters in the
4682 % position of these charcaters, so we use a little trick later on to
4683 % do this.
4684 %
4685 % And if we are making a \emph{strict} shaped font, then we need to
4686 % put these characters into two slots each, with three slots between
4687 % the to replicas of the character. In that case, we first evaluate
4688 % the code once, then if \texttt{strict} is false, we go on in the
4689 % code, otherwise, we evaluate the code once more, \emph{but} the
4690 % characters is put into new slots 3 slots later then before, since
4691 % \texttt{code\_offset} now is 3. Finally in the end we reset
4692 % \texttt{code\_offset}.
4693 %
4694 % This trick is used also in \texttt{uniext.mf} and
4695 % \texttt{unilow.mf}.
4696 %
4697 % \begin{fontcode}{\char"1A}
4698 % This is just a combination of an \textunibf{a} and \textunibf{e},
4699 % which we earlier on saved in the picture variables \texttt{a\_pic}
4700 % and \texttt{e\_pic}.
4701 % \begin{macrocode}
4702 if not caps :
4703 for temp = 1,2 :
4704 if temp = 2 : code_offset := 3 ; fi ;
4705 beginchar(hex"1A",2x_w#-p_t#,x_h#,0) ; "Ligature ae" ;
4706 currentpicture := a_pic + ( e_pic shifted ((w-p_t)/2,0)) ;
4707 endchar ;
4708
4709 % \end{macrocode}
4710 % \end{fontcode}
4711 % \begin{fontcode}{\char"1B}
4712 % This is just a compinition of an \textunibf{o} and
4713 % \textunibf{e}. The \textunibf{e} we saved earlier on in the picture
4714 % variable \texttt{e\_pic}. The \textunibf{o} part is drawn anew.
4715 % \begin{macrocode}
4716 beginchar(hex"1B",2x_w#-p_t#,x_h#,0) ; "Ligature oe" ;
4717 unicir(h/2,p_t,(h/2,h/2),10) ;
4718 currentpicture := currentpicture + (e_pic shifted ((w-p_t)/2,0)) ;
4719 endchar ;
4720
4721 % \end{macrocode}
4722 % \end{fontcode}
4723 % \begin{fontcode}{\char"1C}
4724 % This is really a simple letter. It utilizes the to macros
4725 % \texttt{unicir} and \texttt{unilne} in a very nice way, I think.
4726 % \begin{macrocode}
4727 beginchar(hex"1C",x_w#,x_h#,0) ; "Ligature o/" ;
4728 unicir(w/2,p_t,(w/2,h/2),10) ;
4729 unilne(p_t,(0,0),(w,h),1,1,20) ;
4730 endchar ;
4731 exitif (not strict) ;
4732 if temp = 2 : code_offset := 0 ; fi ;
4733 endfor ;
4734 fi ;
4735
4736 % \end{macrocode}
4737 % \end{fontcode}
4738 % Since we don't want any majusculues in the \emph{strict} shaped
4739 % font, the next three characteres is only generated if
4740 % \texttt{strict} is \texttt{false}.
4741 %
4742 % Since we have to make two version of the majuscules below in a small
4743 % caps font, we set up an \texttt{fot} loop. The trick is this: The
4744 % loop while always execute once, so the characters will always be
4745 % made in the original size. Now, if we are making a small caps font,
4746 % then the loop will execute once more, and we call the macro
4747 % \texttt{do\_font} (\texttt{temp = 2}), so that the new characters
4748 % will have the right dimensions.
4749 % \begin{macrocode}
4750 if not strict :
4751 for temp = 1,2 :
4752 if temp = 2 : do_caps(-3) ; fi ;
4753
4754 % \end{macrocode}
4755 % \begin{fontcode}{\char"1D}
4756 % It wouln't make any sense to just do a combinition of \textunibf{A}
4757 % and \textunibf{E}, as were done in minuscules above, so we draw the
4758 % whole letter anew.
4759 % \begin{macrocode}
4760 beginchar(hex"1D",M_w#,l_h#,0) ; "Ligature AE" ;
4761 uniarc(w/2,P_t,(w/2,h-w/2),90,180,21) ;
4762 uniend(P_t,((w+P_t)/2,P_t),180,270,30) ;
4763 pickup Unipensqu ;
4764 top lft z1 = (w/2,h) ; top rt z2 = (w,h) ;
4765 top lft z3 = (0,h-w/2) ; bot lft z4 = (0,0) ;
4766 lft z5 = (0,h/2) ; rt z6 = (w,h/2) ;
4767 bot lft z7 = ((w+P_t)/2,0) ; bot rt z8 = (w,0) ;
4768 top z9 = (w/2,h) ; bot z10 = (w/2,P_t) ;
4769 draw z1--z2 ; draw z3--z4 ; draw z5--z6 ;
4770 draw z7--z8 ; draw z9--z10 ;
4771 penlabels(range 1 thru 10) ;
4772 endchar ;
4773
4774 % \end{macrocode}
4775 % \end{fontcode}
4776 % \begin{fontcode}{\char"1E}
4777 % Again, a simple combination of \textunibf{O} and \textunibf{E},
4778 % wouldn't work.
4779 % \begin{macrocode}
4780 beginchar(hex"1E",M_w#,l_h#,0) ; "Ligature OE" ;
4781 uniarc(h/2,P_t,(h/2,h/2),90,270,10) ;
4782 pickup Unipensqu ;
4783 top lft z1 = (h/2,h) ; top rt z2 = (w,h) ; draw z1--z2 ;
4784 lft z3 = (h/2,h/2) ; rt z4 = (w,h/2) ; draw z3--z4 ;
4785 bot lft z5 = (h/2,0) ; bot rt z6 = (w,0) ; draw z5--z6 ;
4786 top z7 = (h/2,h) ; bot z8 = (h/2,0) ; draw z7--z8 ;
4787 penlabels(range 1 thru 8) ;
4788 endchar ;
4789
4790 % \end{macrocode}
4791 % \end{fontcode}
4792 % \begin{fontcode}{\char"1F}
4793 % What a said above the minuscule \textunibf{\o} also goes for this
4794 % letter, since this is nothing but an enlargement of that character.
4795 % \begin{macrocode}
4796 beginchar(hex"1F",l_h#,l_h#,0) ; "Liguture O/" ;
4797 unicir(w/2,P_t,(w/2,h/2),10) ;
4798 unilne(P_t,(0,0),(w,h),1,1,20) ;
4799 endchar ;
4800
4801 % \end{macrocode}
4802 % \end{fontcode}
4803 % This is where it gets tricky. First we check to see if we are
4804 % maiking an small caps font, and if not so, we should exit the loop
4805 % right away, so that we don't screw up the minuscules we already have
4806 % made above. If, however, we are making a small caps font, we should,
4807 % after the second run, restore the dimensions to it's original state.
4808 % \begin{macrocode}
4809 exitif (not caps) ;
4810 if temp = 2 : end_do_caps ; fi ;
4811 endfor ;
4812 fi ;
4813
4814 % \end{macrocode}
4815 %\iffalse
4816 %</lig>
4817 %\fi
4818 %\iffalse
4819 %<*ext>
4820 %\fi
4821 % \subsection{Extras --- \texttt{uniext.mf}}
4822 % \file{uniext.mf}
4823 %
4824 % Please notice I included \textunibf{\char"0B}, \textunibf{\char"0C},
4825 % \textunibf{\char"0D}, and \textunibf{\char"0E}, which isn't
4826 % standard Computer Modern Roman policy. See more below.
4827 %
4828 % \Newestchanges{v2.0}{98/08/01}{Some of the \textunibf{bauhaus} symbols
4829 % that where avaliable in v1.0, has been taken out.}
4830 % The first part of this file is some symbols relevant to the
4831 % \textunibf{bauhaus} school. Please notice that some of the symbols
4832 % present in version~1.0 has been taken out\footnote{If you have used
4833 % version~1.0 of this font, and liked those characters and found them
4834 % useful, you should e--mail me \texttt{<[email protected]>} and let me
4835 % know. If I get enough letters on this I will incorporate them again
4836 % in future releases.}, since the were seldomly used, and it made the
4837 % font contain exactly 128 characters.
4838 %
4839 % These characters should \emph{not} be slanted in a slanted shaped
4840 % font, hence we redefine \texttt{currenttrans} to \texttt{identity},
4841 % just before the next character, and resets it after the last
4842 % character. Also notice that \textuni{\bauhead} and
4843 % \textuni{\bauforms} may take some time to generate.
4844 % \begin{macrocode}
4845 currenttransform := identity ;
4846
4847 % \end{macrocode}
4848 %
4849 % \begin{fontcode}{\char"00}
4850 % The next three characters are really not the complicated or useful
4851 % by themselves, but if you combine them, maybe even with the
4852 % \textsf{color} package of \LaTeX{}, you can really add some
4853 % \textunibf{bauhaus} flavour to your document.
4854 %
4855 % The square, ofcourse is obvious.
4856 % \begin{macrocode}
4857 beginchar(hex"00",l_h#,l_h#,0) ; "The filled square symbol" ;
4858 z1 = (0,0) ; z2 = (w,0) ; z3 = (w,h) ; z4 = (0,h) ;
4859 fill z1--z2--z3--z4--cycle ;
4860 penlabels(1,2,3,4) ;
4861 endchar ;
4862
4863 % \end{macrocode}
4864 % \end{fontcode}
4865 % \begin{fontcode}{\char"01}
4866 % Also four vectors are defined, put translated a bit, and another
4867 % path macro --- it is amazing how much you can do with so little.
4868 % \begin{macrocode}
4869 beginchar(hex"01",l_h#,l_h#,0) ; "The filled circle symbol" ;
4870 z1 = (w,h/2) ; z2 = (w/2,h) ; z3 = (0,h/2) ; z4 = (w/2,0) ;
4871 fill z1..z2..z3..z4..cycle ;
4872 penlabels(1,2,3,4) ;
4873 endchar ;
4874
4875 % \end{macrocode}
4876 % \end{fontcode}
4877 % \begin{fontcode}{\char"02}
4878 % Maybe every side of this trangle should be of the same length, mut
4879 % then I had to make the triangle either lower, or wider then the
4880 % square and circle. I couldn't decide, so I went for equal character
4881 % dimensions --- so I believe in equality, sue me!
4882 % \begin{macrocode}
4883 beginchar(hex"02",l_h#,l_h#,0) ; "The filled triangle symbol" ;
4884 z1 = (0,0) ; z2 = (w,0) ; z3 = (w/2,h) ;
4885 fill z1--z2--z3--cycle ;
4886 penlabels(1,2,3) ;
4887 endchar ;
4888
4889 % \end{macrocode}
4890 % \end{fontcode}
4891 % \begin{fontcode}{\char"03}
4892 % This is a typical \textuni{bauhaus} symbol. I think it is very
4893 % beautiful.
4894 %
4895 % First we draw the surrounding circle.
4896 % \begin{macrocode}
4897 beginchar(hex"03",2l_h#,2l_h#,0) ; "The bauhaus-head symbol" ;
4898 unicir(w/2,2c_d/9,(w/2,w/2),40) ;
4899 % \end{macrocode}
4900 % Then we define some horizontal variables $a_{1\ldots6}$ and some
4901 % vertical variables $b_{1\ldots8}$.
4902 % \begin{macrocode}
4903 a1 := .702h ; a2 := .561h ; a3 := .491h ; a4 := .351h ;
4904 a5 := .272h ; a6 := .140h ;
4905 b1 := .202w ; b2 := .307w ; b3 := .412w ; b4 := .447w ;
4906 b5 := .465w ; b6 := .518w ; b7 := .588w ; b8 := .623w ;
4907 % \end{macrocode}
4908 % Then we define the vectors needed for the eye, and draw the eye.
4909 % \begin{macrocode}
4910 temp := c_d/6 ; pickup pensquare scaled temp ;
4911 top lft z1 = (b1,a1) ; top rt z2 = (b2,a1) ; draw z1--z2 ;
4912 bot rt z3 = (b4,a3) ; top rt z4 = (b4,a2) ; draw z3--z4 ;
4913 z5 = (b2,a1) ; z6 = (b4,a1) ; z7 = (b4,a2) ; z8 = (b2,a2) ;
4914 fill z5--z6--z7--z8--cycle ;
4915 % \end{macrocode}
4916 % Then we define the vectors needed for the nostril, and draw it. We
4917 % still use the pen used for the edges of the eye.
4918 % \begin{macrocode}
4919 lft bot z9 = (b4,a4) ; lft bot z10 = (b7,a4) ; draw z9--z10 ;
4920 % \end{macrocode}
4921 % Next up is the chin. Again the same pen.
4922 % \begin{macrocode}
4923 lft bot z11 = (b2,a6) ; rt bot z12 = (b6,a6) ; draw z11--z12 ;
4924 % \end{macrocode}
4925 % Now it's time for the lips. They need to be thicker then the chin
4926 % and edges of the eye, so we pickup a new pen.
4927 % \begin{macrocode}
4928 temp := 3c_d/5 ; pickup pensquare scaled temp ;
4929 lft top z13 = (b5,a5) ; rt top z14 = (b6,a5) ; draw z13--z14 ;
4930 % \end{macrocode}
4931 % It's time for the lower part of the face. Here we don't need to find
4932 % strange paths or the like, so we do it all with a \texttt{fill}
4933 % command.
4934 % \begin{macrocode}
4935 z15 = (b6,a4) ; z16 = (b7,a4) ; z17 = (b7,a6) ; z18 = (b6,a6) ;
4936 fill z15--z16--z17--z18--cycle ;
4937 % \end{macrocode}
4938 % Since it is not know where the forehead will end at the top, or
4939 % where the nect will end in the bottom, we define some temporay
4940 % paths, and make up the real ones from those.
4941 %
4942 % First we need to declare the paths.
4943 % \begin{macrocode}
4944 path noselft, nosert, necklft, neckrt ;
4945 path headcircle , parttop, partbot;
4946 % \end{macrocode}
4947 % Then we build five temporary paths, one for each side of the
4948 % forehead and neck, and one for the surronding circle.
4949 % \begin{macrocode}
4950 z19 = (b7,a4) ; z20 = (b7,h) ; noselft = z19--z20 ;
4951 z21 = (b8,h) ; z22 = (b8,a4) ; nosert = z21--z22 ;
4952 z23 = (b3,a6) ; z24 = (b3,0) ; necklft = z23--z24 ;
4953 z25 = (b6,0) ; z26 = (b6,a6) ; neckrt = z25--z26 ;
4954 z27 = (w/2,h) ; z28 = (w,h/2) ; z29 = (w/2,0) ; z30 = (0,h/2) ;
4955 headcircle := z27..z28..z29..z30..cycle ;
4956 % \end{macrocode}
4957 % Now we can determine the intersection points between our temporary
4958 % paths and the circle, and define the vectors we need.
4959 %
4960 % We also find the parts of the path \texttt{headcircle} that are
4961 % relevant for our to areas we need to fill. At last we can fill the
4962 % areas.
4963 % \begin{macrocode}
4964 z31 = noselft intersectiontimes headcircle ;
4965 z32 = nosert intersectiontimes headcircle ;
4966 parttop = subpath(y31,y32) of headcircle ;
4967 z33 = neckrt intersectiontimes headcircle ;
4968 z34 = necklft intersectiontimes headcircle ;
4969 partbot = subpath(y34,y33) of headcircle ;
4970 fill z19--parttop--z22--cycle ;
4971 fill z23--partbot--z26--cycle ;
4972 penlabels(range 1 thru 34) ;
4973 endchar ;
4974
4975 % \end{macrocode}
4976 % \end{fontcode}
4977 % \begin{fontcode}{\char"04}
4978 % \Newestchanges{v2.0}{98/08/01}{Joseph Collins came up with the
4979 % analytic solution --- thanks JC.}
4980 % This is an interresting character. In version~1.0, I posed it as an
4981 % mathemathical puzzle, and engourage people to work on it, and if
4982 % they came up with a solution, pass it on to me. Well, Joseph Collins
4983 % \index{Collins, Joseph} did just that, and for this I owe him many
4984 % thanks. For more details on the problem and its solution, take a
4985 % look in appendix~\ref{app:solu}.
4986 %
4987 % First we define some pen sizes we want to use, the variable
4988 % \texttt{phi} and draw the circle around the character.
4989 % \begin{macrocode}
4990 beginchar(hex"04",2l_h#,2l_h#,0) ; "The bauhaus-forms symbol" ;
4991 phi := 36.673 ;
4992 thn_pen := 2c_d/9 ; med_pen := 3c_d/9 ; thk_pen := 4c_d/9 ;
4993 unicir(w/2,thn_pen,(w/2,w/2),20) ;
4994 % \end{macrocode}
4995 % Then we define the three vertices of the triangle --- points $A,B,C$
4996 % in the notation of appendix~\ref{app:solu} --- and draw the curved
4997 % ends there.
4998 % \begin{macrocode}
4999 z1 = (thn_pen+med_pen,h/2) ;
5000 z2 = (.6433w-med_pen,.9790h-med_pen) ;
5001 z3 = (.6433w-med_pen,.0210h+med_pen) ;
5002 uniend(med_pen,z1,phi+90,270-phi,30) ;
5003 uniend(med_pen,z2,0,phi+90,40) ;
5004 uniend(med_pen,z3,270-phi,360,50) ;
5005 % \end{macrocode}
5006 % After this, we calculate the rest of the points needed for the top
5007 % most edge of the triangle, line $|AC|$. Then we reflect it
5008 % horizontaly to get the bottom most line $|AB|$.
5009 %
5010 % Then we draw the vertical line $|BC|$ via a pen.
5011 % \begin{macrocode}
5012 z4 = z1 + med_pen*(-sind(phi),cosd(phi)) ;
5013 z5 = z2 + med_pen*(-sind(phi),cosd(phi)) ;
5014 path sl_line ; sl_line = z1--z4--z5--z2--cycle ; fill sl_line ;
5015 fill sl_line reflectedabout((0,h/2),(w,h/2)) ;
5016 pickup pensquare scaled med_pen ;
5017 lft bot z6 = z2 ; lft top z7 = z3 ; draw z6--z7 ;
5018 % \end{macrocode}
5019 % Now it's time for the four vertices $P,Q,R,S$ of the square. Again
5020 % we draw the cured ends at those four vertices. This time we use a
5021 % thicker pen.
5022 % \begin{macrocode}
5023 z8 = (.3687w+thk_pen,.2254h+thk_pen) ;
5024 z9 = (.9179w-thk_pen,.2254h+thk_pen) ;
5025 z10 = (.9179w-thk_pen,.7746h-thk_pen) ;
5026 z11 = (.3687w+thk_pen,.7746h-thk_pen) ;
5027 uniend(thk_pen,z8,180,270,60) ;
5028 uniend(thk_pen,z9,270,360,70) ;
5029 uniend(thk_pen,z10,0,90,80) ;
5030 uniend(thk_pen,z11,90,180,90) ;
5031 % \end{macrocode}
5032 % Finally we define the remaning needed vectors, and draw the edges
5033 % $|PQ|,|QR|,|RS|$ and $|SP|$ of the square.
5034 % \begin{macrocode}
5035 pickup pensquare scaled thk_pen ;
5036 lft top z12 = z8 ; rt top z13 = z9 ; draw z12--z13 ;
5037 lft bot z14 = z9 ; lft top z15 = z10 ; draw z14--z15 ;
5038 bot rt z16 = z10 ; bot lft z17 = z11 ; draw z16--z17 ;
5039 top rt z18 = z11 ; bot rt z19 = z8 ; draw z18--z19 ;
5040 penlabels(range 1 thru 19) ;
5041 endchar ;
5042
5043 % \end{macrocode}
5044 % After these five characters we reset \texttt{currenttransform} to
5045 % include the slant specified by \texttt{slantratio}, so that the rest
5046 % of the font will come out as it slanted if such a font is begin made.
5047 % \begin{macrocode}
5048 currenttransform := identity slanted slantratio ;
5049
5050 % \end{macrocode}
5051 % \end{fontcode}
5052 %
5053 % Next we provide 6 charecters not directly avaliable in the
5054 % \texttt{OT1} encoding, but avaliable in the \texttt{T1} encoding of
5055 % the Cork fonts. These are \textunibf{\NG}, \textunibf{\TH},
5056 % \textunibf{\dh}, \textunibf{\ng}, and \textunibf{\th}, plus the
5057 % accent `ogoneck' (on an \textunibf{a} it looks like
5058 % \textunibf{\k{a}}).
5059 %
5060 % We do need to check again, wether we are making a small caps shaped
5061 % font, and if so, we should not generate the minuscles
5062 % \textunibf{\dh}, \textunibf{\ng}, and \textunibf{\th}.
5063 %
5064 % However, if we are making a \emph{strict} shaped font, then we need
5065 % to make the characters twice, and we employe the trick used in
5066 % \texttt{unilig.mf} again.
5067 %
5068 % \begin{fontcode}{\char"05}
5069 % I am not completly happy with this character, but it was the best I
5070 % could think of. Please let me know if you have any kinds of
5071 % suggestions.
5072 % \begin{macrocode}
5073 if not caps :
5074 for temp = 1,2 :
5075 if temp = 2 : code_offset := 3 ; fi ;
5076 beginchar(hex"05",x_w#,l_h#,0) ; "The letter eth" ;
5077 unicir(w/2,p_t,(w/2,w/2),10) ;
5078 uniarc(w/2,p_t,(w/2,h-w/2),0,120,20) ;
5079 pickup unipensqu ;
5080 top rt z1 = (w,h-w/2) ; bot rt z2 = (w,w/2) ; draw z1--z2 ;
5081 z3 = (w/2,h-w/2+p_t/sqrt(2)) ; z4 = (w/2+p_t/sqrt(2),h-w/2) ;
5082 z5 = (w,h-p_t/sqrt(2)) ; z6 = (w-p_t/sqrt(2),h) ;
5083 fill z3--z4--z5--z6--cycle ;
5084 penlabels(range 1 thru 6) ;
5085 endchar ;
5086
5087 % \end{macrocode}
5088 % \end{fontcode}
5089 % \begin{fontcode}{\char"06}
5090 % This is very much like \textunibf{n}, but has a longer rightmost
5091 % down stroke.
5092 % \begin{macrocode}
5093 beginchar(hex"06",x_w#,x_h#,p_d#) ; "The letter eng" ;
5094 uniarc(w/2,p_t,(w/2,h-w/2),0,180,10) ;
5095 pickup unipensqu ;
5096 top lft z1 = (0,h-w/2) ; bot lft z2 = (0,0) ; draw z1--z2 ;
5097 top rt z3 = (w,h-w/2) ; bot rt z4 = (w,-d) ; draw z3--z4 ;
5098 penlabels(1,2,3,4) ;
5099 endchar ;
5100
5101 % \end{macrocode}
5102 % \end{fontcode}
5103 % \begin{fontcode}{\char"07}
5104 % \begin{macrocode}
5105 beginchar(hex"07",x_w#,l_h#,p_d#) ; "The letter thorn" ;
5106 rad := (h+d)/4 ;
5107 uniarc(rad,p_t,(w-rad,rad),-90,90,10) ;
5108 pickup unipensqu ;
5109 top lft z1 = (0,h) ; bot lft z2 = (0,-d) ; draw z1--z2 ;
5110 top lft z3 = (p_t,2rad) ; top rt z4 = (w-rad,2rad) ;
5111 bot lft z5 = (p_t,0) ; bot rt z6 = (w-rad,0) ;
5112 draw z3--z4 ; draw z5--z6 ;
5113 penlabels(range 1 thru 6) ;
5114 endchar ;
5115 exitif (not strict) ;
5116 if temp = 2 : code_offset := 0 ; fi ;
5117 endfor ;
5118 fi ;
5119
5120 % \end{macrocode}
5121 % \end{fontcode}
5122 %
5123 % Like in \texttt{unilig.mf} we now employe a trick to generate to
5124 % versions of the uppercase letters \textunibf{\NG} and
5125 % \textunibf{\TH}, that is \textunibsc{\ng} and \textunibsc{\th}.
5126 %
5127 % Since there us only two majuscules here, we might as well take
5128 % advantige of the extra slot, and make a small caps version of
5129 % \textunibf{\varQ}, so this is defined before \textunibf{\NG}, to
5130 % take the place of \textunibf{dh}.
5131 % \begin{macrocode}
5132 if not strict :
5133 for temp = 1,2 :
5134 if temp = 2 : do_caps(-3) ; fi ;
5135
5136 % \end{macrocode}
5137 % \begin{fontcode}{\char"08}
5138 % This is the variant of \textunibf{Q} based an \textunibf{P}. This
5139 % one is \emph{non}--default, you can acces this letter with
5140 % \Lcs{varQ}\footnote{I am somewhat troubled with the majuscle `Q',
5141 % because I don't know if the default one should be `\textunibf{Q}' or
5142 % `\textunibf{\varQ}'. I will keep both characters in the font, but if
5143 % enough wnats to chage the default, I will consider this
5144 % ofcourse. Tell me \emph{your} opinon!}.
5145 % \begin{macrocode}
5146 beginchar(hex"08",A_w#,l_h#,0) ; "The variant letter Q";
5147 unicir(w/2,P_t,(w/2,h-w/2),10)
5148 pickup Unipensqu ;
5149 rt top z1 = (w,h-w/2) ; rt bot z2 = (w,0) ; draw z1--z2 ;
5150 penlabels(1,2);
5151 endchar ;
5152
5153 % \end{macrocode}
5154 % \end{fontcode}
5155 % \begin{fontcode}{\char"09}
5156 % This is similar to \textunibf{N}, execpt the rightmost vertical
5157 % stroke ends in an arc.
5158 % \begin{macrocode}
5159 beginchar(hex"09",A_w#,l_h#,0) ; "The letter Eng" ;
5160 rad := 3C_d ;
5161 uniarc(w/2,P_t,(w/2,h-w/2),0,180,10) ;
5162 uniarc(rad,P_t,(w-rad,rad),-90,0,20) ;
5163 pickup Unipensqu ;
5164 top lft z1 = (0,h-w/2) ; bot lft z2 = (0,0); draw z1--z2 ;
5165 top rt z3 = (w,h-w/2) ; bot rt z4 = (w,rad) ; draw z3--z4 ;
5166 penlabels(1,2,3,4) ;
5167 endchar ;
5168
5169 % \end{macrocode}
5170 % \end{fontcode}
5171 % \begin{fontcode}{\char"0A}
5172 % \begin{macrocode}
5173 beginchar(hex"0A",A_w#,l_h#,0) ; "The letter Thorn" ;
5174 rad := h/4 ;
5175 uniarc(rad,P_t,(w-rad,h/2),-90,90,10) ;
5176 pickup Unipensqu ;
5177 top lft z1 = (0,h) ; bot lft z2 = (0,0) ; draw z1--z2 ;
5178 top lft z3 = (P_t,h/2+rad) ; top rt z4 = (w-rad,h/2+rad) ;
5179 bot lft z5 = (P_t,h/2-rad) ; bot rt z6 = (w-rad,h/2-rad) ;
5180 draw z3--z4 ; draw z5--z6 ;
5181 penlabels(range 1 thru 6) ;
5182 endchar ;
5183 exitif (not caps) ;
5184 if temp = 2 : end_do_caps ; fi ;
5185 endfor ;
5186 fi ;
5187
5188 % \end{macrocode}
5189 % \end{fontcode}
5190 % \begin{fontcode}{\char"0B}
5191 % This is actually a reversed cidelia accent.
5192 % \begin{macrocode}
5193 beginchar(hex"0B",x_w#,x_h#,2c_d#) ; "The accent ogoneck" ;
5194 uniarc(d/2,2P_T/3,(w/2-d/4,-d/2),-90,90,10) ;
5195 endchar ;
5196
5197 % \end{macrocode}
5198 % \end{fontcode}
5199 % \begin{fontcode}{\char"0C}
5200 % I chose to include the braces in the font, since you might needed
5201 % them, and the Computer Modern Roman braces, doesn't really go with
5202 % the rest of the font, so if you had to use those, it wouldn't look
5203 % very good. They are, however not in the standard position, since
5204 % that is already taken up.
5205 % \begin{macrocode}
5206 beginchar(hex"0C",3c_d#,l_h#,p_d#) ; "The left brace ({)" ;
5207 uniarc(2c_d,p_t,(3c_d,h-2c_d),90,180,10) ;
5208 uniarc(2c_d,p_t,(3c_d,-d+2c_d),180,270,20) ;
5209 pickup unipensqu ;
5210 top lft z1 = (c_d,h-2c_d) ; bot lft z2 = (c_d,-d+2c_d) ; draw z1--z2 ;
5211 lft z3 = (0,(h-d)/2) ; rt z4 = (c_d,(h-d)/2) ; draw z3--z4 ;
5212 penlabels(1,2,3,4) ;
5213 endchar ;
5214
5215 % \end{macrocode}
5216 % \end{fontcode}
5217 % \begin{fontcode}{\char"0D}
5218 % Discussion above.
5219 % \begin{macrocode}
5220 beginchar(hex"0D",3c_d#,l_h#,p_d#) ; "The right brace (})" ;
5221 uniarc(2c_d,p_t,(0,h-2c_d),0,90,10) ;
5222 uniarc(2c_d,p_t,(0,-d+2c_d),270,360,20) ;
5223 pickup unipensqu ;
5224 top rt z1 = (w-c_d,h-2c_d) ; bot rt z2 = (w-c_d,-d+2c_d) ;
5225 draw z1--z2 ;
5226 rt z3 = (w,(h-d)/2) ; lft z4 = (w-c_d,(h-d)/2) ; draw z3--z4 ;
5227 penlabels(1,2,3,4) ;
5228 endchar ;
5229
5230 % \end{macrocode}
5231 % \end{fontcode}
5232 % \begin{fontcode}{\char"0E}
5233 % The paragraph sign is usually in the \texttt{cmsy} family, avaliable
5234 % in \texttt{OT1} encoding only, but for the same reasons as the
5235 % braces above, I decided to include this charater in the font.
5236 % \begin{macrocode}
5237 beginchar(hex"0E",(l_h#+p_d#+2p_t#)/3,l_h#,p_d#) ;
5238 "The Paragraph sign" ;
5239 uniarc(w/2,p_t,(w/2,h-w/2),30,270,10) ;
5240 uniarc(w/2,p_t,(w/2,w/2-d),-150,90,20) ;
5241 unicir(w/2,p_t,(w/2,(h-d)/2),30) ;
5242 endchar ;
5243
5244 % \end{macrocode}
5245 % \end{fontcode}
5246 % \begin{fontcode}{\char"0F}
5247 % I guess there are no real surprises here --- it's just a bar!
5248 % \begin{macrocode}
5249 beginchar(hex"0F",p_t#,l_h#,p_d#) ; "The symbol bar (|)" ;
5250 pickup unipensqu ;
5251 lft top z1 = (0,h) ; lft bot z2 = (0,-d) ; draw z1--z2 ;
5252 penlabels(1,2);
5253 endchar ;
5254
5255 % \end{macrocode}
5256 % \end{fontcode}
5257 %
5258 % The next two characters is included, since I think it would be a
5259 % good idea to use \universal{} characters for German
5260 % quotation, instead of some Computer Modern Roman character, which
5261 % would be totally out of place.
5262 %
5263 % If you use the \textsf{uni} package you can say
5264 % \begin{verbatim}
5265 % {\uni
5266 % \guillemoleft Hello World\guillemoright\newline
5267 % \guilsinglleft Hello World\guilsinglright}
5268 % \end{verbatim}
5269 % and get
5270 % \begin{quote}
5271 % {\uni
5272 % \guillemoleft Hello World\guillemoright\newline
5273 % \guilsinglleft Hello World\guilsinglright}
5274 % \end{quote}
5275 %
5276 % \begin{fontcode}{\char"22}
5277 % This is quote a simple character. Do notice the nice use of
5278 % \texttt{uniend}, to produce the soft turn.
5279 % \begin{macrocode}
5280 beginchar(hex"22",2x_w#/3,x_h#,0) ; "The single left guillomot" ;
5281 z1 = (p_t,h/2) ; z2 = (w,h) ; phi := angle(z2-z1) ;
5282 uniend(p_t,(z1),90+phi,270-phi,10) ;
5283 z3 = (w-p_t/sind(phi),h) ; z4 = z1 + p_t*(-sind(phi),cosd(phi)) ;
5284 path rombe ; rombe := z1--z2--z3--z4--cycle ; fill rombe ;
5285 fill rombe reflectedabout ((0,h/2),(w,h/2)) ;
5286 penlabels(1,2,3,4) ;
5287 endchar ;
5288
5289 % \end{macrocode}
5290 % \end{fontcode}
5291 % \begin{fontcode}{\char"5C}
5292 % \begin{macrocode}
5293 beginchar(hex"5C",2x_w#/3,x_h#,0) ; "The single right guillomot";
5294 z1 = (w-p_t,h/2) ; z2 = (0,h) ; phi := 180-angle(z2-z1) ;
5295 uniend(p_t,(z1),phi-90,90-phi,10) ;
5296 z3 = (p_t/sind(phi),h) ; z4 = z1 + p_t*(sind(phi),cosd(phi)) ;
5297 path rombe ; rombe := z1--z2--z3--z4--cycle ; fill rombe ;
5298 fill rombe reflectedabout ((0,h/2),(w,h/2)) ;
5299 penlabels(1,2,3,4) ;
5300 endchar ;
5301
5302 % \end{macrocode}
5303 % \end{fontcode}
5304 %\iffalse
5305 %</ext>
5306 %\fi
5307 %\iffalse
5308 %<*font>
5309 %\fi
5310 % \section{Font driver files}\file{ful*.mf}
5311 %
5312 % This section describes the driver files
5313 % \texttt{uni}\meta{series}\meta{shape}\meta{size}\texttt{.mf}, where
5314 % \meta{series} is one of \texttt{r} (medium), or \texttt{bf}
5315 % (boldface), \meta{shape} one of upright (no letter), \texttt{sl}
5316 % (slanted), or \texttt{sc} (small caps), and \meta{size} is the
5317 % designsize in points, one of 8, 9, 10, 12, or 17.
5318 %
5319 % To make the source file shorter, I employed a little trick,
5320 % exploiting some of the features of the \textsf{docstrip}
5321 % program. That, on the other hand means that the line numbers you see
5322 % here, doesn't conform to the actual ones in the driver files.
5323 %
5324 % First we make sure the base file \texttt{unibase.mf} is loaded, but
5325 % not more then once in a run. This is needed since this file contains
5326 % all the macros and definitions for the font.
5327 %
5328 % \begin{macrocode}
5329 if unknown unibase : input unibase; fi;
5330
5331 % \end{macrocode}
5332 % Next we define some \textsl{boolean} variables, determining the
5333 % shape of the font we are to create. This are straight forward, and
5334 % should be self explanetory. Any number of variables may be set to
5335 % \texttt{true}, execpt that it makes no sense to define both
5336 % \texttt{bold} and \texttt{medium} to \texttt{true}, so this will be
5337 % considered an error, and a medium font will be made. (See alse the
5338 % explanation of the \texttt{start\_font} macro in the
5339 % \texttt{unibase.mf} section.)
5340 %
5341 % \begin{verbatim}
5342 %<+medium>medium := true ;
5343 %<+bold>bold := true ;
5344 %<+slanted>slant := true ;
5345 %<+caps>caps := true ;
5346 %<+strict>strict := true ;
5347 %<+eight>start_font(8pt#) ;
5348 %<+nine>start_font(9pt#) ;
5349 %<+ten>start_font(10pt#) ;
5350 %<+twelve>start_font(12pt#) ;
5351 %<+seventeen>start_font(17pt#) ;
5352 %\end{verbatim}
5353 %
5354 % The next part of the files just inputs the nessecary source files,
5355 % and there is nothing special here. Finally we end the font with the
5356 % macro \texttt{end\_font}.
5357 % \begin{macrocode}
5358 input uniupp ; input unilow ; input unidig ; input unispe ;
5359 input unilig ; input unipun ; input uniacc ; input uniext ;
5360
5361 end_font;
5362
5363 % \end{macrocode}
5364 %\iffalse
5365 %</font>
5366 %\fi
5367 %\iffalse
5368 %<*docpackage>
5369 %\fi
5370 % \section{Commands for the Documentation --- \texttt{unidoc.sty}}
5371 % \file{unidoc.sty}\label{app:unidoc}
5372 %
5373 % \textsc{Warning:} Most of the macros defined in this file uses
5374 % \TeX{} primitivs like \Lcs{def} and \Lcs{let}, contorary to the
5375 % recommendations in \clsguide{}. This means you shouldn't try to use
5376 % this package in conjunction with normal \LaTeXe{} documents, since
5377 % it will not nessecarily by portable, and may generate unwanted
5378 % resaults. This package is \emph{only} intended for the documentation
5379 % of the \universal{} package, \emph{and nothing else!}
5380 %
5381 % \subsection{Initializing}
5382 %
5383 % First we need to identify the package, its version and release
5384 % date, etc.
5385 % \begin{macrocode}
5386 \ProvidesPackage{unidoc}
5387 [\unifiledate\space\unifileversion\space universal documenation package.]
5388 % \end{macrocode}
5389 %
5390 % \begin{macro}{\ifeepic}
5391 % \begin{macro}{\eepictrue}
5392 % \begin{macro}{\eepicfalse}
5393 %
5394 % In case we can load the package \textsf{eepic}, we will set
5395 % \cs{eepictrue}, and make a plot of the shapes problem.
5396 %
5397 % \begin{macrocode}
5398 \newif\ifeepic\eepicfalse
5399 % \end{macrocode}
5400 % \end{macro}
5401 % \end{macro}
5402 % \end{macro}
5403 %
5404 % Then we load some pacakges needed. \textsf{doc} is ofcourse needed
5405 % to do all the actual typesetting of macros etc., while \textsf{uni}
5406 % is needed for the fonts\footnote{This means ofcourse, that you
5407 % needed to have all the \MF{} sources ready, or in other words, you
5408 % should have run \texttt{universa.ins} through \LaTeX{}.}, and
5409 % finally \textsf{mflogo} is needed for the \MF{} logo.
5410 %
5411 % \begin{macrocode}
5412 \DeclareOption*{\PassOptionsToPackage{\CurrentOption}{uni}}
5413 \ProcessOptions\relax
5414 \RequirePackage{doc}
5415 \RequirePackage{mflogo}
5416 \RequirePackage[colorlinks,%
5417 linkcolor=red,%
5418 anchorcolor=black,%
5419 citecolor=green,%
5420 filecolor=black,%
5421 menucolor=black,%
5422 urlcolor=blue,%
5423 hyperindex,%
5424 linktocpage,%
5425 pdfduplex=duplexFlipLongEdge,%
5426 a4paper,%
5427 bookmarks=true]{hyperref}
5428 \RequirePackage{uni}
5429 \IfFileExists{eepic.sty}{%
5430 \RequirePackage{eepic}\eepictrue}{%
5431 \PackageWarning{unidoc}{`eepic.sty' not found, so there will be no
5432 pictures - you are warned.}\eepicfalse}
5433
5434 % \end{macrocode}
5435 %
5436 % \subsection{Macros for the Index}
5437 %
5438 % \begin{macro}{\file}
5439 % First we save the argument --- the current output filename --- in a
5440 % command, set the codeline counter to zero (since every starts at
5441 % line 1, obviously). Then we check to see if we are outputting files
5442 % at all. If we are, we write an index entry.
5443 % \begin{macrocode}
5444 \def\file#1{
5445 \def\c@r{#1}
5446 \c@CodelineNo0
5447 \if@filesw
5448 \immediate\write\@indexfile
5449 {\string\indexentry{#1=\string\texttt{#1}%
5450 \string\space(File)}%
5451 {\thepage}}\fi}
5452
5453 % \end{macrocode}
5454 % \end{macro}
5455 % \begin{macro}{\main}
5456 % This command is used be \texttt{doc.sty}, to choose the font for the
5457 % codelines numbers in the index. It is set to \texttt{typewriter}.
5458 % \begin{macrocode}
5459 \def\main#1{\texttt{{\small\c@r}:\ #1}}
5460
5461 % \end{macrocode}
5462 % \end{macro}
5463 % \begin{macro}{\src}
5464 % To make the index reflect, that more then one file is present, and
5465 % therefore more then one entry could have the same codeline number,
5466 % we put the file name before the codeline number in the index. The
5467 % name is gotten from the \Lcs{file} command. This ofcourse takes up
5468 % quite some space in the index, so later on we speciffy the index
5469 % to only have 2 columns.
5470 % \begin{macrocode}
5471 \def\src#1#2{\texttt{{\small#1}:\ #2}}
5472
5473 % \end{macrocode}
5474 % \end{macro}
5475 % \begin{macro}{\fontcodeindex}
5476 % The next two commands are responsible for writting the index entries
5477 % of the above environments. They are similar to the \texttt{doc}
5478 % package command \Lcs{codeline@wrindex} but uses the filenames
5479 % defined by \Lcs{file} (via the \Lcs{src} command), and the
5480 % \Lcs{fontcodeindex} uses a font appropiat for the font descriped
5481 % (i.e., \Lcs{FontCodeFont})
5482 % \begin{macrocode}
5483 \def\fontcodeindex#1{%
5484 \@bsphack\begingroup\let\_=\protect\_%
5485 \special@index{%
5486 \expandafter\string#1\actualchar%
5487 {\string\FontCodeFont\space#1}%
5488 \encapchar src{\c@r}}\endgroup\@esphack}
5489
5490 % \end{macrocode}
5491 % \end{macro}
5492 % \begin{macro}{\fontmacroindex}
5493 % \begin{macrocode}
5494 \def\fontmacroindex#1{%
5495 \@bsphack\begingroup\let\_=\protect\_%
5496 \special@index{%
5497 \expandafter\string#1\actualchar%
5498 {\string\MacroFont\space#1}%
5499 \encapchar src{\c@r}}\endgroup\@esphack}
5500
5501 % \end{macrocode}
5502 % \end{macro}
5503 %
5504 % \subsection{Descriptive Environments}
5505 %
5506 % The next counters are to do some more advanced stuff on the on line
5507 % numbers in the index. Instead of just writting the beginning line of
5508 % the fontcode, we writte the complete interval of the fontcode. See
5509 % also the command \Lcs{f@ntcodeindex} and \Lcs{f@ntmacroindex}.
5510 % \begin{macrocode}
5511 \newcount\startfontcodeline
5512 \newcount\startfontmacroline
5513
5514 % \end{macrocode}
5515 % \begin{macro}{\fontcode}
5516 % \begin{macro}{\endfontcode}
5517 % \begin{macro}{\FontCodeFont}
5518 % \begin{macro}{\PrintFontName}
5519 % Next comes two environments for descriping the font programs, and
5520 % font macros. They are similar to the environment \texttt{macro} in
5521 % the \texttt{doc} package. However, some alterations is made, so that
5522 % the right font is used in the index. (See also the
5523 % \Lcs{fontcodeindex} macro below.) Also, notice the defintion of
5524 % \Lcs{saved@macroname} and \Lcs{PrintFontName}. This is done so that
5525 % control sequnces in the argument to \texttt{fontcode} environment,
5526 % which should expand to a single letter, comes out expanded, rather
5527 % then the actual control sequences, which isn't wanted here.
5528 % \begin{macrocode}
5529 \def\fontcode#1{
5530 \def\saved@fontname{#1}
5531 \let\temp@MF\MacroFont\let\MacroFont\FontCodeFont
5532 \let\temp@PMN\PrintMacroName\let\PrintMacroName\PrintFontName
5533 \let\temp@SMI\SpecialMainIndex
5534 \let\SpecialMainIndex\fontcodeindex
5535 \macro{#1}}
5536 \def\endfontcode{
5537 \endmacro
5538 \let\MacroFont\temp@MF
5539 \let\SpecialMainIndex\temp@SMI
5540 \let\PrintMacroName\temp@PMN}
5541 \let\FontCodeFont\unibf
5542 \def\PrintFontName#1{\strut\MacroFont\saved@fontname\ }
5543
5544 % \end{macrocode}
5545 % \end{macro}
5546 % \end{macro}
5547 % \end{macro}
5548 % \end{macro}
5549 %
5550 % \begin{macro}{\fontmacro}
5551 % \begin{macro}{\endfontmacro}
5552 % This is similar to \texttt{fontcode} environment, but is used to
5553 % descripe the macros of \texttt{unibase}. The reason why we employ to
5554 % environments, and not just one, is that the margin notes should come
5555 % out in \texttt{typewriter} font in this environment, while it
5556 % should be in \universal{} font above.
5557 % \begin{macrocode}
5558 \def\fontmacro#1{
5559 \let\temp@under\_
5560 \let\_\MacroUnderscore
5561 \let\temp@SMI\SpecialMainIndex
5562 \let\SpecialMainIndex\fontmacroindex
5563 \macro{#1}}
5564 \def\endfontmacro{%
5565 \endmacro\let\SpecialMainIndex\temp@SMI\let\_\temp@under}
5566 \def\MacroUnderscore{{\char"5F}}
5567
5568 % \end{macrocode}
5569 % \end{macro}
5570 % \end{macro}
5571 %
5572 % \begin{macro}{\illubegin}
5573 % \begin{macro}{\illuother}
5574 % \begin{macro}{\endillu}
5575 % \begin{macrocode}
5576 \let\beginillu\ifeepic
5577 \let\illuother\else
5578 \let\endillu\fi
5579 % \end{macrocode}
5580 % \end{macro}
5581 % \end{macro}
5582 % \end{macro}
5583 %
5584 % \subsection{Useful Commands}
5585 %
5586 % \begin{macro}{\Newestchanges}
5587 % This macro is intended to be used like the \Lcs{changes} macro of
5588 % the \textsf{doc} package. However, it should only be used on the last
5589 % most change. But rather then just making an entry in the the
5590 % glossary, it will also print a line saying when this macro was last
5591 % changed. This is something like what is done in
5592 % \texttt{ltxguide.cls}, but a bit different, since in that class not
5593 % enough is being done glossary--wise, and too much is done
5594 % margin--wise. When making an update to a macro, the new
5595 % \Lcs{changes} should be made with \Lcs{Newestchanges}, and the old
5596 % \Lcs{Newestchanges} should be renamed to \Lcs{changes}. Notice that
5597 % \texttt{doc.sty} says \Lcs{def}\Lcs{saved@macroname}\texttt{\{\}},
5598 % so we can't say
5599 % \begin{verbatim}
5600 % \ifx\saved@macroname\relax
5601 % \end{verbatim}
5602 % since \Lcs{relax} \emph{isn't} the same as an empty definition, but
5603 % \Lcs{@empty} is. In the case that \Lcs{saved@macroname} is defined
5604 % to something, then we put a bullet ($\bullet$) in the margin, and
5605 % attach a footnote to it, where the footnote text contains the proper
5606 % information. Ofcourse this could led to a lot of footnotes, but the
5607 % alternative is to redefine the \texttt{macro} environment, and I
5608 % really didn't have the stomach for that, Sorry. Notice the use of
5609 % \Lcs{@gobble}, that removes the leading `v' in the version number.
5610 % \begin{macrocode}
5611 \def\Newestchanges#1#2#3{
5612 \ifx\saved@macroname\@empty%
5613 \marginpar{\small\raggedright{\sl Change on} #2, {\sl Version}
5614 \expandafter\@gobble#1}
5615 \else%
5616 \llap{$\bullet$\footnote{{\sl Change of }
5617 {\MacroFont\expandafter\string\saved@macroname}
5618 {\sl on} #2, {\sl Version }\expandafter\@gobble#1}}
5619 \fi
5620 \changes{#1}{#2}{#3}}
5621
5622 % \end{macrocode}
5623 % \end{macro}
5624 %
5625 % \begin{macro}{\Lcs}
5626 % Finally we define a short hand for \Lcs{...}, to make the
5627 % documentaion easier.
5628 % \begin{macrocode}
5629 \def\Lcs#1{\texttt{\char"5C{}#1}}
5630
5631 % \end{macrocode}
5632 % \end{macro}
5633 %
5634 % \subsubsection{The Font Charts}
5635 %
5636 % \begin{macro}{\FontChart}
5637 % This macro is for typesetting a table of the font. It is taken from
5638 % \texttt{testfont.tex}.
5639 %
5640 % First we define some counters to use in the table. \Lcs{n} is the
5641 % counts the number of letters and symbols typeset so far, while
5642 % \Lcs{m} is counting the number of typeset characters in current
5643 % line. \Lcs{p} is a penalty counter. The dimension \Lcs{dim} is used
5644 % for setting the characters appropiate.
5645 %
5646 % We also define a new \Lcs{if}, used to see if we should go on.
5647 %
5648 % \Lcs{oct} and \Lcs{hex} are used for typesetting numbers in
5649 % octadecimal and hexadecimal notation respectivily.
5650 % \begin{macrocode}
5651 \newcount\n\newcount\m\newcount\p\newdimen\dim
5652 \newif\ifskipping
5653 \def\oct#1{\hbox{\rm\'{}\kern-.2em\it#1\/\kern.05em}}
5654 \def\hex#1{\hbox{\rm\H{}\tt#1}}
5655
5656 % \end{macrocode}
5657 % \begin{macro}{\setdigs}
5658 % This is used to set the counters \Lcs{n} and \Lcs{m} every time a
5659 % new (double) line is started.
5660 % \begin{macrocode}
5661 \def\setdigs#1"#2{\gdef\h{#2}
5662 \m=\n \divide\m by 64 \xdef\0{\the\m}%
5663 \multiply\m by-64 \advance\m by\n \divide\m by 8 \xdef\1{\the\m}}
5664
5665 % \end{macrocode}
5666 % \end{macro}
5667 % \begin{macro}{\testrow}
5668 % This is used to see how we should typeset the current row.
5669 % \begin{macrocode}
5670 \def\testrow{\setbox0=\hbox{\penalty 1\def\\{\char"\h}%
5671 \\0\\1\\2\\3\\4\\5\\6\\7\\8\\9\\A\\B\\C\\D\\E\\F%
5672 \global\p=\lastpenalty}}
5673
5674 % \end{macrocode}
5675 % \end{macro}
5676 % \begin{macro}{\oddline}
5677 % This is used to typeset topmost line of a double line. It basically
5678 % just putting a line and some saces into the chart.
5679 % \begin{macrocode}
5680 \def\oddline{
5681 \cr\noalign{\nointerlineskip}\multispan{19}\hrulefill&
5682 \setbox0=\hbox{\lower 2.3pt\hbox{\hex{\h x}}}\smash{\box0}\cr
5683 \noalign{\nointerlineskip}}
5684
5685 % \end{macrocode}
5686 % \end{macro}
5687 % \begin{macro}{\evenline}
5688 % This is used to typeset bottommost line of a double line. It
5689 % basically does the same as \Lcs{oddline}, but also checks to see if
5690 % we need to end the chart or go on.
5691 % \begin{macrocode}
5692 \def\evenline{
5693 \loop\skippingfalse
5694 \ifnum\n<256 \m=\n \divide\m 16 \chardef\next=\m
5695 \expandafter\setdigs\meaning\next \testrow
5696 \ifnum\p=1 \skippingtrue \fi\fi
5697 \ifskipping \global\advance\n 16\repeat
5698 \ifnum\n=256 \let\next=\endchart\else\let\next=\morechart\fi
5699 \next}
5700
5701 % \end{macrocode}
5702 % \end{macro}
5703 %
5704 % \begin{macro}{\morechart}
5705 % This macro is called every time we need to typeset a new double line
5706 % of the chart.
5707 % \begin{macrocode}
5708 \def\morechart{
5709 \cr\noalign{\hrule\penalty5000}
5710 \chartline \oddline \m=\1 \advance\m 1 \xdef\1{\the\m}
5711 \chartline \evenline}
5712
5713 % \end{macrocode}
5714 % \end{macro}
5715 %
5716 % \begin{macro}{\chartline}
5717 % \begin{macro}{\chartstrut}
5718 % \Lcs{chartline} is called every time we need to typeset a line of
5719 % the chart. Notice it uses the macro \Lcs{:} to do the
5720 % job. \Lcs{chartstrut} is used to get approiate spacing.
5721 % \begin{macrocode}
5722 \def\chartline{&\oct{\0\1x}&&\:&&\:&&\:&&\:&&\:&&\:&&\:&&\:&&}
5723 \def\chartstrut{\lower4.5pt\vbox to14pt{}}
5724
5725 % \end{macrocode}
5726 % \end{macro}
5727 % \end{macro}
5728 %
5729 % \begin{macro}{\:}
5730 % \begin{macro}{\reposition}
5731 % \begin{macro}{\centerlargechars}
5732 % These macros are used for typesetting the actual characters in the
5733 % chart. \Lcs{:} gives the appropiate space for a character and
5734 % encriments the \Lcs{n} counter, while \Lcs{reposition} places the
5735 % character nicely in the the chart. \Lcs{centerlargechars} does the
5736 % same for {\large large} characters.
5737 % \begin{macrocode}
5738 \def\:{\setbox0=\hbox{\char\n}%
5739 \ifdim\ht0>7.5pt\reposition
5740 \else\ifdim\dp0>2.5pt\reposition\fi\fi
5741 \box0\global\advance\n 1 }
5742 \def\reposition{
5743 \setbox0=\vbox{\kern2pt\box0}\dim=\dp0
5744 \advance\dim 2pt \dp0=\dim}
5745 \def\centerlargechars{
5746 \def\reposition{\setbox0=\hbox{$\vcenter{\kern2pt\box0\kern2pt}$}}}
5747
5748 % \end{macrocode}
5749 % \end{macro}
5750 % \end{macro}
5751 % \end{macro}
5752 % \begin{macro}{\chart}
5753 % \begin{macro}{\endchart}
5754 % These two macros is what starts and ends the chart. \Lcs{chart} sets
5755 % up the appropiate stuff, and writes the top most line in the
5756 % chart. \Lcs{endchart} is called by \Lcs{evenline} if no more
5757 % characters should be typeset, and then it finished everything off,
5758 % plus writes the last line of the chart.
5759 % \begin{macrocode}
5760 \def\chart{$$\global\n=0
5761 \halign to\hsize\bgroup
5762 \chartstrut##\tabskip0pt plus10pt&&\hfil##\hfil&\vrule##\cr
5763 \lower6.5pt\null&&&
5764 \oct0&&\oct1&&\oct2&&\oct3&&\oct4&&\oct5&&\oct6&&\oct7&\evenline}
5765 \def\endchart{
5766 \cr\noalign{\hrule}
5767 \raise11.5pt\null&&&\hex 8&&\hex 9&&\hex A&&\hex B&
5768 &\hex C&&\hex D&&\hex E&&\hex F&\cr
5769 \egroup$$\par}
5770
5771 % \end{macrocode}
5772 % \end{macro}
5773 % \end{macro}
5774 %
5775 % Finally we get to the \emph{real} macro. This macro takes one
5776 % argument, the name of the \meta{font driver file}. We put the chart
5777 % inside a \texttt{figure} environment, so that we may refer to it
5778 % latter. This is done using the label \texttt{chart:}\meta{font driver
5779 % file}. The chart is encapsulated in a group, so that the definition
5780 % of \Lcs{chart@font} doesn't get in the way elsewhere in the document.
5781 % \begin{macrocode}
5782 \def\FontChart#1#2{
5783 %% \marginpar{\small\raggedright Letters and symbols in \texttt{#1}
5784 %% font.}
5785 \begin{figure}[htbp]
5786 \begingroup
5787 \font\chart@font=#1
5788 \chart@font
5789 \chart
5790 \endgroup
5791 \caption{Letters and symbols in \texttt{#1} --- #2}
5792 \label{fig:font:#1}
5793 \end{figure}}
5794
5795 % \end{macrocode}
5796 % \end{macro}
5797 %
5798 % \subsection{Parameters and Miscillaneous Commands}
5799 %
5800 % This is used --- as explaind in \texttt{doc.dvi} --- to give a
5801 % different prologue to the index, more suitable to our needs.
5802 % \begin{macrocode}
5803 \IndexPrologue{%
5804 \section{Index}\markboth{Index}{Index}
5805 The Numbers written in \textit{italic} refer to the pages, where
5806 a macros usage is descriped, while those in \texttt{typewrite} refer
5807 to line numbers in the files, mentioned before, where the definition
5808 is, while \textsl{slanted} shows the places it is used. Normal letters
5809 refer to pages, wether it be descriptions or usage.}
5810 \GlossaryPrologue{%
5811 \section{Change History}\markboth{Change History}{Change History}
5812 Below is listed the changes made to the \universal{} font and the
5813 \LaTeX{} support package \textsf{uni}, from version~1.0 to 2.0.}
5814
5815 % \end{macrocode}
5816 % Also we set the number of columns in the index to 2, since some
5817 % space is needed for the entries
5818 % \begin{macrocode}
5819 \setcounter{IndexColumns}{2}
5820
5821 % \end{macrocode}
5822 % We don't want the index flutted with silly command sequences, so we
5823 % leave some of them out.
5824 % \begin{macrocode}
5825 \DoNotIndex{\@bsphack,\@empty,\@esphack,\@gobble,\@ifnextchar,%
5826 \@ifundefined,\@indexfile,\@onlypreamble,\iffalse,\_,\,\accent,%
5827 \aftergroup,\actualchar,\advance,\begin,\begingroup,\bgroup,%
5828 \box,\bullet,\changes,\char,\chardef,\CheckModules,\CodeLineIndex,%
5829 \cr,\crcr,\csname,\DeclareFontFamily,\DeclareFontShape,\DeclareMacro,%
5830 \dim,\divide,\dp,\DeclareOption,\def,\DoNotIndex,\edef,\egroup,\else,%
5831 \end,\endcsname,\EnableCrossrefs,\encapchar,\endgroup,\endmacro,%
5832 \expandafter,\fbox,\fi,\font,\footnote,\fontemcoding,\fontfamuliy,%
5833 \fontseries,\fontshape,\gdef,\global,\H,\h,\halign,\hbox,\hfil,%
5834 \hidewidth,\hrule,\hrulefill,\hsize,\ht,\hyphenchar,\if@filesw,%
5835 \ifdim,\IfFileExists,\ifnum,\ifskipping,\ifx,\immediate,\indexentry,%
5836 \it,\itdefault,\kern,\lastpenalty,\LaTeX,\LaTeXe,\leavevmode,\let,%
5837 \llap,\loop,\lower,\m,\macro,\MacroFont,\marginpar,\marginparwidth,%
5838 \markboth,\mdseries,\meaning,\multiply,\multispan,\n,\newcommand,%
5839 \newcount,\newdimen,\newif,\newline,\next,\noalign,\noexpand,%
5840 \nointerlineskip,\not@math@alphabet,\null,\ooalign,\p,\par,\penalty,%
5841 \ProcessOptions,\protect,\ProvidesFile,\ProvidesPackage,%
5842 \RecordChanges,\raggedright,\raise,\relax,\renewcommand,\repeat,%
5843 \RequirePackage,\rm,\saved@macroname,\scshape,\section,\selectfont,%
5844 \setcounter,\setbox,\show,\skippingfalse,\skippingture,\sl,\slshape,%
5845 \small,\smash,\space,\special@index,\SpecialMainIndex,\stop,\string,%
5846 \strut,\tabskip,\TeX,\TeXbook,\textit,\textsf,\textsl,\texttt,\the,%
5847 \thepage,\tt,\typeout,\unhbox,\universal,\upshape,\usefont,\vbox,%
5848 \vcenter,\vrule,\write,\xdef,\z@,\PackageError,\PackageWarning,
5849 \PassOptionsToPackage,\GlossaryPrologue,\',\0,\1,\bfseries,
5850 \centerlargechars,\DeclareRobustCommand,\DeclareTextFontCommand,
5851 \ExecuteOptions,\familydefault,\fontencoding,\fontfamily,\hex,\oct,
5852 \IndexPrologue,\oddline}
5853
5854 % \end{macrocode}
5855 % And then we define the behaviour of the \textsf{doc} package, by
5856 % setting a number of flags (as explained in the documentation of the
5857 % \textsf{doc} package). We also set the margin paragraph widths to be
5858 % a bit smaller then usual.
5859 % \begin{macrocode}
5860 \EnableCrossrefs
5861 \CodelineIndex
5862 \RecordChanges
5863 \CheckModules
5864 \marginparwidth=54pt
5865 % \end{macrocode}
5866 %
5867 % \begin{macro}{\NFSS}
5868 % \begin{macro}{\ftnguide}
5869 % \begin{macro}{\clsguide}
5870 % \begin{macro}{\MFbook}
5871 % \begin{macro}{\universal}
5872 %
5873 % Some utility macros used in the documentation.
5874 %
5875 % \begin{macrocode}
5876 \newcommand{\NFSS}{\textsf{NFSS}}
5877 \newcommand{\ftnguide}{\textit{\LaTeXe{} Font Selection}}
5878 \newcommand{\clsguide}{ \textit{\LaTeXe{} for Class and Packages
5879 Writers}} \newcommand{\TeXbook}{\TeX\textsf{book}}
5880 \newcommand{\MFbook}{\MF\textsf{book}}
5881 \newcommand{\universal}{\textunibf{universal}}
5882
5883 % \end{macrocode}
5884 % \end{macro}
5885 % \end{macro}
5886 % \end{macro}
5887 % \end{macro}
5888 % \end{macro}
5889 % %\iffalse
5890 %</docpackage>
5891 %\fi
5892 % \Finale
5893 %
5894 % \iffalse
5895 % Local Variables:
5896 % mode: docTeX
5897 % End:
5898 % \fi