@defcodeindex fl @defcodeindex op
This document is based on `dvips.tex' by Tomas Rokicki. It is in the public domain.
The `dvips' program has a number of features that set it apart from other PostScript drivers for TeX. This rather long section describes the advantages of using dvips, and may be skipped if you are just interested in learning how to use the program. See section Installation, for details of compilation and installation.
The dvips driver generates excellent, standard PostScript, that can be included in other documents as figures or printed through a variety of spoolers. The generated PostScript requires very little printer memory, so very complex documents with a lot of fonts can easily be printed even on PostScript printers without much memory, such as the original Apple LaserWriter. The PostScript output is also compact, requiring less disk space to store and making it feasible as a transfer format.
Even those documents that are too complex to print in their entirety on a particular printer can be printed, since dvips will automatically split such documents into pieces, reclaiming the printer memory between each piece.
The dvips program supports graphics in a natural way, allowing PostScript graphics to be included and automatically scaled and positioned in a variety of ways.
Printers with resolutions other than 300 dpi are also supported, even if they have different resolutions in the horizontal and vertical directions. High resolution output is supported for typesetters, including an option that compresses the bitmapped fonts so that typesetter virtual memory is not exhausted. This option also significantly reduces the size of the PostScript file and decoding in the printer is very fast.
Missing fonts can be automatically generated if METAFONT exists on the system, or fonts can be converted from `gf' to `pk' format on demand. If a font cannot be generated, a scaled version of the same font at a different size can be used instead, although dvips will complain loudly about the poor aesthetics of the resulting output.
Users will appreciate features such as collated copies and support for `tpic', `psfig', `emtex', and `METAPOST'; system administrators will love the support for multiple printers, each with their own configuration file, and the ability to pipe the output directly to a program such as `lpr'. Support for MS-DOS, OS/2, and VMS in addition to UNIX is provided in the standard distribution, and porting to other systems is easy.
One of the most important features is the support of virtual fonts, which add an entirely new level of flexibility to TeX. Virtual fonts are used to give dvips its excellent PostScript font support, handling all the font remapping in a natural, portable, elegant, and extensible way. The dvips `afm2tfm' driver even comes with its own `afm2tfm' program that creates the necessary virtual fonts and TeX font metric files automatically from the Adobe font metric files.
Source is provided and freely distributable, so adding a site-specific feature is possible. Adding such features is made easier by the highly modular structure of the program.
There is really no reason to use another driver, and the more people use dvips, the less time will be spent fighting with PostScript and the more time will be available to create beautiful documents. So if you don't use dvips on your system, get it today.
To use dvips, simply type
dvips foo
where `foo.dvi' is the output of TeX that you want to print. If dvips has been installed correctly, the document should come out of your default printer.
If you use fonts that have not been used on your system before, they may be automatically generated; this process can take a few minutes. The next time that document is printed, these fonts will already exist, so printing will go much faster.
Many options are available; they are described in a later section. For a brief summary of available options, just type
dvips
Most TeX documents at a particular site are designed to use the standard paper size (letter size in the United States, A4 in Europe.) The dvips program defaults to these paper sizes and can be customized for the defaults at each site or on each printer.
But many documents are designed for other paper sizes. For instance, you may want to design a document that has the long edge of the paper horizontal. This can be useful when typesetting booklets, brochures, complex tables, or many other documents. This type of paper orientation is called landscape orientation (the `normal' orientation is portrait). Alternatively, a document might be designed for ledger or A3 paper.
Since the intended paper size is a document design decision, and not a decision that is made at printing time, such information should be given in the TeX file and not on the dvips command line. For this reason, dvips supports a `papersize' special. It is hoped that this special will become standard over time for TeX previewers and other printer drivers.
The format of the `papersize' special is
\special{papersize=8.5in,11in}
where the dimensions given above are for a standard letter sheet. The first dimension given is the horizontal size of the page, and the second is the vertical size. The dimensions supported are the same as for TeX; namely, in (inches), cm (centimeters), mm (millimeters), pt (points), sp (scaled points), bp (big points, the same as the default PostScript unit), pc (picas), dd (didot points), and cc (ciceros).
For a landscape document, the `papersize' comment would be given as
\special{papersize=11in,8.5in}
An alternate specification of `landscape' is to have a special of the form
\special{landscape}
This is supported for backward compatibility, but it is hoped that eventually the `papersize' comment will dominate.
Of course, using such a command only informs dvips of the desired paper size; you must still adjust the `hsize' and `vsize' in your TeX document to actually use the full page.
The `papersize' special must occur somewhere on the first page of the document.
Scaling and including PostScript graphics is a breeze--if the PostScript file is correctly formed. Even if it is not, however, the file can usually be accommodated with just a little more work. The most important feature of a good PostScript file--from the standpoint of including it in another document--is an accurate bounding box comment.
Every well-formed PostScript file has a comment describing where on the page the graphic is located, and how big that graphic is. This information is given in terms of the lower left and upper right corners of a box just enclosing the graphic, and is thus referred to as a bounding box. These coordinates are given in PostScript units (there are precisely 72 PostScript units to the inch) with respect to the lower left corner of the sheet of paper.
To see if a PostScript file has a bounding box comment, just look at the first few lines of the file. (PostScript is standard ASCII, so you can use any text editor to do this.) If within the first few dozen lines there is a line of the form
%%BoundingBox: 0 1 2 3
(with any numbers), chances are very good that the file is Encapsulated PostScript and will work easily with dvips. If the file contains instead a line like
%%BoundingBox: (atend)the file is still probably Encapsulated PostScript, but the bounding box (that dvips needs to position the graphic) is at the end of the file and should be moved to the position of the line above. This can be done with that same text editor, or with a simple Perl script.
If the document lacks a bounding box altogether, one can easily be added. Simply print the file. Now, take a ruler, and make the following measurements. All measurements should be in PostScript units, so measure it in inches and multiply by 72. Alternatively, the `bbfig' program distributed with dvips in the `contrib' directory can be used to automate this process.
From the left edge of the paper to the leftmost mark on the paper is llx, the first number. From the bottom edge of the paper to the bottommost mark on the paper is lly, the second number. From the left edge of the paper to the rightmost mark on the paper is urx, the third number. The fourth and final number, ury, is the distance from the bottom of the page to the uppermost mark on the paper.
Now, add a comment of the following form as the second line of the document. (The first line should already be a line starting with the two characters `%!'; if it is not, the file probably isn't PostScript.)
%%BoundingBox: llx lly urx ury
Or, if you don't want to modify the file, you can simply write these numbers down in a convenient place and use them when you import the graphic.
If the document does not have such a bounding box, or if the bounding box is given at the end of the document, please complain to the authors of the software package that generated the file; without such a line, including PostScript graphics can be tedious.
Now you are ready to include the graphic into a TeX file. Simply add to the top of your TeX file a line like
@flindex epsf.tex
\input epsf
(or, if your document is in LaTeX or SliTeX, add the `epsf' style option, as was done to the following line).
\documentstyle[12pt,epsf]{article}
This only needs to be done once, no matter how many figures you plan to include. Now, at the point you want to include the file, enter a line such as
\epsffile{foo.ps}
If you are using LaTeX or SliTeX, you may need to add a `\leavevmode' command immediately before the `\epsffile' command to get certain environments to work correctly. If your file did not (or does not currently) have a bounding box comment, you should supply those numbers you wrote down as in the following example:
\epsffile[100 100 500 500]{foo.ps}
(in the same order they would have been in a normal bounding box comment). Now, save your changes and run TeX and dvips; the output should have your graphic positioned at precisely the point you indicated, with the proper amount of space reserved.
The effect of the `epsffile' macro is to typeset the figure as a TeX `vbox' at the point of the page that the command is executed. By default, the graphic will have its `natural' width (namely, the width of its bounding box). The TeX box will have depth zero and a `natural' height. The graphic will be scaled by any `dvi' magnification in effect at the time.
Any PostScript graphics included by any method in this document (except `bop-hook' and its ilk) are scaled by the current `dvi' magnification. For graphics included with `epsffile' where the size is given in TeX dimensions, this scaling will produce the correct, or expected, results. For compatibility with old PostScript drivers, it is possible to turn this scaling off with the following TeX command:
\special{! /magscale false def}
Use of this command is not recommended because it will make the `epsffile' graphics the wrong size if global magnification is used in a `dvi' document, and it will cause any PostScript graphics to appear improperly scaled and out of position if a `dvi' to `dvi' program is used to scale or otherwise modify the document.
You can enlarge or reduce the figure by putting
\epsfxsize=dimen
right before the call to `epsffile'. Then the width of the TeX box will be dimen and its' height will be scaled proportionately. Alternatively you can force the vertical size to a particular size with
\epsfysize=dimen
in which case the height will be set and the width will be scaled proportionally. If you set both, the aspect ratio of the included graphic will be distorted but both size specifications will be honored.
By default, clipping is disabled for included EPSF images. This is because clipping to the bounding box dimensions often cuts off a small portion of the figure, due to slightly inaccurate bounding box arguments. The problem might be subtle; lines around the boundary of the image might be half their intended width, or the tops or bottoms of some text annotations might be sliced off. If you want to turn clipping on, just use the command
\epsfclipon
and to turn clipping back off, use
\epsfclipoff
A more general facility for sizing is available by defining the `epsfsize' macro. You can redefine this macro to do almost anything. This TeX macro is passed two parameters by `epsffile'. The first parameter is the natural horizontal size of the PostScript graphic, and the second parameter is the natural vertical size. This macro is responsible for returning the desired horizontal size of the graph (the same as assigning `epsfxsize' above).
In the definitions given below, only the body is given; it should be inserted in
\def\epsfsize#1#2{body}
Some common definitions are:
If you want TeX to report the size of the figure as a message on your terminal when it processes each figure, give the command
\epsfverbosetrue
Often in order to get a particular graphic file to work, a certain header file might need to be sent first. Sometimes this is even desirable, since the size of the header macros can dominate the size of certain PostScript graphics files. The dvips program provides support for this with the `header=' special command. For instance, to ensure that `foo.ps' gets downloaded as part of the header material, the following command should be added to the TeX file:
\special{header=foo.ps}
The dictionary stack will be at the `userdict' level when these header files are included.
For these and all other header files (including the headers required by dvips itself and any downloaded fonts), the printer VM budget is debited by some value. If the header file has, in its first 1024 bytes, a line of the form
%%VMusage: min max
then the maximum value is used. If it doesn't, then the total size of the header file in bytes is used as an approximation of the memory requirements.
For simple graphics, or just for experimentation, literal PostScript graphics can be included. Simply use a special command that starts with a double quote (`"').
For instance, the following (simple) graphic: was created by typing:
\vbox to 100bp{\vss % a bp is the same as a PostScript unit
\special{" newpath 0 0 moveto 100 100 lineto 394 0 lineto
closepath gsave 0.8 setgray fill grestore stroke}}
(You are responsible for leaving space for such literal graphics.) Literal graphics are discouraged because of their nonportability.
Similarly, you can define your own macros for use in such literal graphics through the use of literal macros. Literal macros are defined just like literal graphics, only you begin the special with an exclamation point instead of a double quote. These literal macros are included as part of the header material in a special dictionary called `SDict'. This dictionary is the first one on the PostScript dictionary stack when any PostScript graphic is included, whether by literal inclusion or through the `epsffile' macros.
There are other ways to include graphics with dvips. One is to use an existing package, such as `emtex', `psfig', `tpic', or `METAPOST', all of which `dvips' supports.
Other facilities are available for historical reasons, but their use is discouraged, in hope that some `sane' form of PostScript inclusion shall become standard. The main advantage of the `epsffile' macros is that they can be adapted for whatever form of special eventually becomes standard, and thus only minor modifications to that one file need to be made, rather than revising an entire library of TeX documents.
Most of these specials use a flexible key and value scheme:
\special{psfile=filename.ps[key=value]*}
This will download the PostScript file called `filename.ps' such that the current point will be the origin of the PostScript coordinate system. The optional key/value assignments allow you to specify transformations on the PostScript.
The possible keys are:
The dimension parameters are all given in PostScript units. The `hscale' and `vscale' are given in non-dimensioned percentage units, and the rotation value is specified in degrees. Thus
\special{psfile=foo.ps hoffset=72 hscale=90 vscale=90}
will shift the graphics produced by file `foo.ps' right by one inch and will draw it at 0.9 times normal size. Offsets are given relative to the point of the special command, and are unaffected by scaling or rotation. Rotation is counterclockwise about the origin. The order of operations is to rotate the figure, scale it, then offset it.
For compatibility with older PostScript drivers, it is possible to change the units that `hscale' and `vscale' are given in. This can be done by redefining `@scaleunit' in `SDict' by a TeX command such as
\special{! /@scaleunit 1 def}
The `@scaleunit' variable, which is by default 100, is what `hscale' and `vscale' are divided by to yield an absolute scale factor.
All of the methods for including graphics we have described so far enclose the graphic in a PostScript save/restore pair, guaranteeing that the figure will have no effect on the rest of the document. Another type of special command allows literal PostScript instructions to be inserted without enclosing them in this protective shield; users of this feature are supposed to understand what they are doing (and they shouldn't change the PostScript graphics state unless they are willing to take the consequences). This command can take many forms, because it has had a torturous history; any of the following will work:
\special{ps:text}
\special{ps::text}
\special{ps::[begin]text}
\special{ps::[end]text}
(with longer forms taking precedence over shorter forms, when they are used). Note that `ps::' and `ps::[end]' do not do any positioning, so they can be used to continue PostScript literals started with `ps:' or `ps::[begin]'. There is also the command
\special{ps: plotfile filename}
which will copy the commands from filename verbatim into the output (but omitting lines that begin with %). An example of the proper use of literal specials can be found in the file `rotate.tex' which makes it easy to typeset text turned 90 degrees.
To finish off this section, the following examples are presented without explanation:
\def\rotninety{\special{ps:currentpoint currentpoint translate 90
rotate neg exch neg exch translate}}\font\huge=cmbx10 at 14.4truept
\setbox0=\hbox to0pt{\huge A\hss}\vskip16truept\centerline{\copy0
\special{ps:gsave}\rotninety\copy0\rotninety\copy0\rotninety
\box0\special{ps:grestore}}\vskip16truept
\vbox to 2truein{\special{ps:gsave 0.3 setgray}\hrule height 2in
width\hsize\vskip-2in\special{ps:grestore}\font\big=cminch\big
\vss\special{ps:gsave 1 setgray}\vbox to 0pt{\vskip2pt
\line{\hss\hskip4pt NEAT\hss}\vss}\special{ps:0 setgray}%
\hbox{\raise2pt\line{\hss NEAT\hss}\special{ps:grestore}}\vss}
Some caveats are in order when using the above forms. Make sure that each `gsave' on a page is matched with a `grestore' on the same page. Do not use `save' or `restore'. Use of these macros can interact with the PostScript generated by dvips if care is not taken; try to understand what the above macros are doing before writing your own. The `\rotninety' macro especially has a useful trick that appears again and again.
PostScript is an excellent page description language--but it does tend to be rather verbose. Compressing PostScript graphics files can often reduce them by more than a factor of five. For this reason, if the filename parameter to one of the graphics inclusion techniques starts with a backtick (``'), the filename is instead interpreted as a command to execute that will send the actual file to standard output. Thus,
\special{psfile="`zcat foo.ps.Z"}
will include the uncompressed version of `foo.ps'. Since such a command is not accessible to TeX, if you use this facility with the `EPSF' macros, you need to supply the bounding box parameter yourself, as in
\epsffile[72 72 540 720]{"`zcat screendump.ps.Z"}
to include `screendump.ps'. Of course, the commands to be executed can be anything, including using a file conversion utility such as `tek2ps' or whatever is appropriate.
This extension is not portable to other DVI-to-PostScript translators.
Thanks to Donald E. Knuth, the dvips driver now supports PostScript fonts through the virtual font capability. PostScript fonts are (or should be) accompanied by a font metric file such as `Times-Roman.afm', which describes characteristics of a font called Times-Roman. To use such fonts with TeX, we need `tfm' files that contain similar information. These can be generated from `afm' files by running the program `afm2tfm', supplied with dvips. This program also creates virtual fonts with which you can use normal plain TeX conventions. See Donald E. Knuth, TUGboat v. 11, no. 1, Apr. 1990, pp. 13--23, "Virtual Fonts: More Fun for Grand Wizards", for a general introduction to virtual fonts.
Non-resident downloaded PostScript fonts tend to use a great deal of printer virtual memory. PostScript printers typically have between 130,000 and 400,000 bytes of available virtual memory; each downloaded font will require approximately 30,000 bytes of that. For many applications, bitmapped fonts work much better, even at typesetter resolutions (with the `-Z' option).
Even resident PostScript fonts can take a fair amount of memory, but less with this release of dvips than previously. Also, bitmapped fonts tend to image faster than PostScript fonts.
In its simplest form, the `afm2tfm' program converts an Adobe `afm' file (e.g., `Times-Roman.afm') into a `raw' TeX TFM file (say, `rptmr.tfm' with the command
afm2tfm Times-Roman rptmr
This file is raw because it does no character remapping; it simply converts the character information on a one-to-one basis to TeX characters with the same code. (It would be better to have eschewed the use of the `r' prefix in favor of using a variant letter to specify the encoding; but we didn't think of that at the time, and it's too late to change existing fonts. See section `Top' in Filenames for TeX fonts.)
In the following examples, we will use the font Times-Roman to illustrate the conversion process. For the standard 35 LaserWriter fonts, however, it is highly recommended that you use the supplied `tfm' and `vf' files that come with dvips, as these files contain some additional changes that make them work better with TeX than they otherwise would.
Standard PostScript fonts have a different encoding scheme from that of plain TeX. Although both schemes are based on ASCII, special characters such as ligatures and accents are handled quite differently. Therefore we obtain best results by using a `virtual font' interface, which makes TeX see a standard TeX encoding for the PostScript font. Such a virtual font can be obtained, for example, by the command
afm2tfm Times-Roman -v ptmr rptmr
This produces two files as output, namely the `virtual property list' file `ptmr.vpl', and the raw font metric file `rptmr.tfm'. (The upper case `-V' also produces a `vpl' file, but maps a set of small caps into the lower case alphabet.) To use the font in TeX, you should first run
vptovf ptmr.vpl ptmr.vf ptmr.tfm
and then install the virtual font file `ptmr.vf' where Dvips will see it, and install `ptmr.tfm' and `rptmr.tfm' where TeX and Dvips will see them.
You can also make more complex virtual fonts by editing `ptmr.vpl' before running `vptovf'; such editing might add the uppercase Greek characters in the standard TeX positions, for instance. (This has already been done for the font files that come with Dvips.) Once this has been done, you're all set. You can use code like this in TeX henceforth:
\font\myfont = ptmr at 12pt \myfont Hello, I am being typeset in 12-point Times-Roman.
Thus we have two fonts, one actual (`rptmr', which is analogous to the font in the printer) and one virtual (`ptmr', which has been remapped to the standard TeX encoding (almost), and has typesetting know-how added. You could also say
\font\PTR = rptmr at 10pt
and typeset directly with that, but then you would have no ligatures or kerning, and you would have to use Adobe character positions for special letters like The virtual font called `ptmr' not only has ligatures and kerning, and most of the standard accent conventions of TeX, it also has a few additional features not present in the Computer Modern fonts. For example, it includes all the Adobe characters (such as the Polish ogonek and the French guillemots). The only things you lose from ordinary TeX text fonts are the dotless `j' (which can be hacked into the VPL file with literal PostScript specials if you have the patience) and uppercase Greek letters (which just don't exist unless you buy them separately).
The remapped p* fonts mostly follow plain TeX conventions for
accents. The exceptions: the Hungarian umlaut (which is at position
`0x7D' in `cmr10', but position `0xCD' in `ptmr');
the dot accent (at positions `0x5F' and `0xC7', respectively);
and `\AA', which neds different tweaking. In order to use these
accents with PostScript fonts or in math mode when `\textfont0' is
a PostScript font, you will need to use the following definitions.
These definitions will not work with the Computer Modern fonts for the
relevant accents. They are already part of the distributed
`psfonts.sty' for use with LaTeX.
\def\H#1{{\accent"CD #1}}
\def\.#1{{\accent"C7 #1}}
\def\dot{\mathaccent"70C7 }
\newdimen\aadimen
\def\AA{\leavevmode\setbox0\hbox{h}\aadimen\ht0
\advance\aadimen-1ex\setbox0\hbox{A}\rlap{\raise.67\aadimen
\hbox to \wd0{\hss\char'27\hss}}A}
These PostScript fonts can be scaled to any size. Go wild! Using PostScript fonts, however, does use up a great deal of the printer's memory and it does take time. You may find downloading bitmapped fonts to be faster than using the built-in PostScript fonts!
The `afm2tfm' program also allows you to specify a different encoding for a PostScript font. This can be done at two levels.
You can specify a different output encoding with `-t'. This only applies when you are building a virtual font, and it tells `afm2tfm' to attempt to remap the font to match the output encoding as closely as possible. In such an output encoding, you can also specify ligature pairs and kerning information that will be used in addition to the information in the `afm' file.
You can also specify a different PostScript encoding with `-p'. This changes both the raw `tfm' file created by Afm2tfm and the `vf' and `tfm' files created by VPtoVF. It also requires that the encoding file be downloaded as part of any document that uses this font. This is the only way to access characters in a PostScript font that are neither encoded in the `afm' file nor built from other characters (constructed characters). For instance, `Times-Roman' contains the extra characters `trademark' and `registered' (among others) that can only be accessed through such a PostScript reencoding. Any ligature or kern information specified in the PostScript encoding is ignored by `afm2tfm'.
To combine the effects of `-p' and `-t', use `-T'. If you make regular use of a private non-standard reencoding `-T' is usually the better idea, since you can get unexpected inconsistencies in mapping otherwise.
Afm2tfm's encoding files have precisely the same format as an encoding vector in any PostScript font. Specifically:
% Comments are ignored, unless the first word after the percent sign
% is LIGKERN -- see below.
/MyEncoding [ /Alpha /Beta /Gamma /Delta ...
/A /B ... /Z % exactly 256 entries, each with a / at the front
/wfooaccent /xfooaccent /yfooaccent /zfooaccent ] def
Comments, which start with a percent sign and continue until the end
of the line, are ignored unless they start with `LIGKERN'. The
first `word' of the file must start with a forward slash (a PostScript
literal name) and defines the name of the encoding. The next word must
be an left bracket `['. Following that must be precisely 256
character names; use `/.notdef' for any that you do not want
defined. Finally, there must be a matching right bracket ]. A
final `def' token is optional. All names are case sensitive.
Any ligature or kern information is given in the comments. If the first word after the percent sign is `LIGKERN', then the entire rest of the line is parsed for ligature and kern information. This ligature and kern information is given in groups of words, each group of which must be terminated by a semicolon (with a space before and after it, unless it occurs at the end of a line.)
In these `LIGKERN' statements, three types of information may be specified. These three types are ligature pairs, kerns to remove or ignore, and the character value of this font's boundary character. Which of the types the particular set of words corresponds to is automatically determined by the allowable syntax.
Throughout a `LIGKERN' statement, the boundary character is specified as `||'.
To set the boundary character value, a command such as `|| = 39 ;' must be used.
To indicate a kern to remove, give the names of the two characters (without the leading slash) separated by `{}', as in `one {} one ;'. This is similar to the way you might use `{}' in a TeX file to turn off ligatures or kerns at a particular location. Either or both of the character names can be given as `*', which is a wild card matching any character; thus, all kerns can be removed with `* {} * ;'.
To specify a ligature, specify the names of the pair of characters, followed by the ligature `operation' (as in METAFONT), followed by the replacing character name. Either (but not both) of the first two characters can be `||' to indicate a word boundary.
Normally the `operation' is `=:' meaning that both characters are removed and replaced by the third character, but by adding `|' characters on either side of the `=:', you can specify which of the two leading characters to retain. In addition, by suffixing the ligature operation with one or two `>' signs, you can indicate that the ligature scanning operation should skip that many characters before proceeding. This works just like in METAFONT. A typical ligature might be specified with `ff i =: ffi ;'. A more convoluted ligature is `one one |=:|>> exclam ;' which indicates that every pair of adjacent `1''s should be separated by an exclamation point, and then two of the resulting characters should be skipped over before continuing searching for ligatures and kerns. You cannot give more `>''s in an ligature operation as you did `|', so there are a total of eight possible ligature operations:
=: |=: |=:> =:| =:|> |=:| |=:|> |=:|>>
The default set of ligatures and kerns built in to `afm2tfm' can be specified with:
% LIGKERN question quoteleft =: questiondown ;
% LIGKERN exclam quoteleft =: exclamdown ;
% LIGKERN hyphen hyphen =: endash ; endash hyphen =: emdash ;
% LIGKERN quoteleft quoteleft =: quotedblleft ;
% LIGKERN quoteright quoteright =: quotedblright ;
% LIGKERN space {} * ; * {} space ; 0 {} * ; * {} 0 ;
% LIGKERN 1 {} * ; * {} 1 ; 2 {} * ; * {} 2 ; 3 {} * ; * {} 3 ;
% LIGKERN 4 {} * ; * {} 4 ; 5 {} * ; * {} 5 ; 6 {} * ; * {} 6 ;
% LIGKERN 7 {} * ; * {} 7 ; 8 {} * ; * {} 8 ; 9 {} * ; * {} 9 ;
Afm2tfm can do other manipulations as well. For example, to make an obliqued variant of Times Roman:
afm2tfm Times-Roman -s .167 -v ptmro rptmro
which creates `ptmro.vpl' and `rptmro.tfm'. To use this, put the line
rptmro Times-Roman ".167 SlantFont"
@flindex psfonts.map
into `psfonts.map'. Then `rptmro' (our name for the obliqued
Times) will act as if it were a resident font, although it is actually
constructed from Times-Roman via the PostScript routine SlantFont
(which slants everything 1/6 to the right, in this case).
Similarly, you can get an expanded font with
afm2tfm Times-Roman -e 1.2 -v ptmrre rptmrre
and by recording the pseudo-resident font
rptmrre Times-Roman "1.2 ExtendFont"
in `psfonts.map'.
You can also create a small caps font with a command such as
afm2tfm Times-Roman -V ptmrc rptmrc
This will generate a set of small caps mapped into the usual lower case positions and scaled down to 0.8 of the normal cap dimensions. You can also specify the scaling as something other than the default 0.8:
afm2tfm Times-Roman -c 0.7 -V ptmrc rptmrc
It is unfortunately not possible to increase the width of the small caps independently of the rest of the font. If you want a really professional looking set of small caps you will have to make up a custom pair of ``tfm'' and ``vpl'' files using the `-e' option, change the name and checksum of the D 1 font in `ptmrc.vpl' to match the customized raw `tfm', and replace the small caps at the lower case mappings with the small caps in the customized file. You can then throw away the customized `vpl' file, but not the customized `tfm'. That must be identified with an appropriate line in `psfonts.map'.
If you change the PostScript encoding of a font, you must specify the input file as a header file, as well as give a reencoding command. For instance, let us say we are using Times-Roman reencoded according to the encoding `MyEncoding' (stored in the file `myenc.enc') as `rptmrx'. In this case, our `psfonts.map' entry would look like
rptmrx Times-Roman "MyEncoding ReEncodeFont" <myenc.enc
The `afm2tfm' program prints out the precise line you need to add to `psfonts.map' to use that font, assuming the font is resident in the printer; if the font is not resident, you must add the header command to download the font yourself. Each identical line only needs to be specified once in the `psfonts.map' file, even though many different fonts (small caps variants, or ones with different output encodings) may be based on it.
If you want to use a non-printer-resident PostScript font for which you have a `pfb' or `pfa' file (an Adobe Type 1 font program), you can make it act like a resident font by putting a `<' sign and the name of the `pfb' or `pfa' file just after the font name in the `psfonts.map' @flindex psfonts.map file entry. For example,
rpstrn StoneInformal <StoneInformal.pfb
will cause dvips to include `StoneInformal.pfb' in your document as if it were a header file, whenever the pseudo-resident font StoneInformal is used in a document. Similarly, you can generate transformed fonts and include lines like
rpstrc StoneInformal <StoneInformal.pfb ".8 ExtendFont"
in `psfonts.map', in which case `StoneInformal.pfb' will be loaded whenever StoneInformal-Condensed is used. Each header file is loaded at most once per document.
If you are using a `pfb' file that is also has a different PostScript encoding, you need to download multiple header files in `psfonts.map'. If, for instance, `Optima' was both non-resident and you wanted to reencode it in PostScript with `MyEncoding' stored in `myenc.enc', you need:
rpstrnx Optima "MyEncoding ReEncodeFont" <myenc.enc <Optima.pfb
When using PFB files, dvips is smart enough to unpack the binary PFB format into printable ASCII so there is no need to perform this conversion yourself. In addition, it will scan the font to determine its memory usage, as it would for any header file.
Here is a brief description of the contents of `psfonts.map'. If a line is empty or begins with a space, asterisk, semicolon, or hash mark, it is ignored. Otherwise, the line is separated into words, where words are separated by spaces or tabs, except that if a word begins with a double quote, it extends until the next double quote or the end of the line. If a word starts with a less than character, it is treated as a font header file (or a downloaded PostScript font). There can be more than one such header for a given font. If a word starts with a double quote, it is a special instruction for generating that font. Otherwise it is a name. The first such name is the TFM file that a virtual font file can refer to. If there is another name, it is used as the PostScript name; if there is only one name, it is used for both the TeX name and the PostScript name.
The command line switches to `afm2tfm' are:
Here are the `CODINGSCHEMES' that result from the various possible choices for reencoding.
(CODINGSCHEME TeX text + AdobeStandardEncoding)
(CODINGSCHEME TeX text + DCEncoding)
(CODINGSCHEME DCEncoding + AdobeStandardEncoding)
(CODINGSCHEME DCEncoding + DCEncoding)
The dvips driver has a plethora of command line options. Reading through this section will give a good idea of the capabilities of the driver.
Many of the parameterless options listed here can be turned off by immediately suffixing the option with a zero (0); for instance, to turn off page reversal if it is turned on by default, use `-r0'. The options that can be turned off in this way are `a', `f', `k', `i', `m', `q', `r', `s', `E', `F', `K', `M', `N', `U', and `Z'.
This is a handy summary of the options; it is printed out when you run dvips with no arguments.
This is dvipsk version Copyright 1986, 1993 Radical Eye Software
Usage: dvips [options] filename[.dvi]
a* Conserve memory, not time y # Multiply by dvi magnification
b # Page copies, for posters e.g. A Print only odd (TeX) pages
c # Uncollated copies B Print only even (TeX) pages
d # Debugging C # Collated copies
e # Maxdrift value D # Resolution
f* Run as filter E* Try to create EPSF
h f Add header file F* Send control-D at end
i* Separate file per section K* Pull comments from inclusions
k* Print crop marks M* Don't make fonts
l # Last page N* No structured comments
m* Manual feed O c Set/change paper offset
n # Maximum number of pages P s Load config.$s
o f Output file R Run securely
p # First page S # Max section size in pages
q* Run quietly T c Specify desired page size
r* Reverse order of pages U* Disable string param trick
s* Enclose output in save/restore V* Send downloadable PS fonts as PK
t s Paper format X # Horizontal resolution
x # Override dvi magnification Y # Vertical resolution
Z* Compress bitmap fonts
pp #-# First-last page
# = number f = file s = string * = suffix, `0' to turn off
c = comma-separated dimension pair (e.g., 3.2in,-32.1cm)
mtpk or pstopk or some combination of the two in order
to generate the required bitmap fonts; neither of these programs are
supplied with Dvips.
The dvips program has a system of loading configuration files such that parameters can be set globally across the system, on a per-printer basis, or by the user.
When dvips starts up, first the global `config.ps' file is searched for and loaded. This file is looked for along the path for configuration files, which is set in the `Makefile' at compilation. After this master configuration file is loaded, a file by the name of @flindex .dvipsrc `.dvipsrc' is loaded from the current user's home directory, if such a file exists. This file is loaded in exactly the same way as the global configuration file, and it can override any options set in the global file.
Then the command line is read and parsed. If the `-P' option is encountered, at that point in the command line a configuration file for that printer is read in. Thus, the printer configuration file can override anything in the global or user configuration file, and it can also override anything seen in the command line up to the point that the `-P' option was encountered.
After the command line has been completely scanned, if there was no `-P' option selected, and also the `-o' and `-f' command line options were not used, a `PRINTER' environment variable is searched for. If this variable exists, and a configuration file for the printer mentioned in it exists, this configuration file is loaded last of all.
Note that because the printer-specific configuration files are read after the user's configuration file, the user cannot override things in the printer configuration files. On the other hand, the configuration path usually includes the current directory, and can be set to include the user's home directory (or any other directory of the user), so the user can always provide his or her own printer-specific configuration files that will be found before the system global ones.
It is best to give a METAFONT mode as well as a resolution in the printer configuration file. Also make sure that METAFONT knows about the mode, by entering it into your local `mode_def' file. (For example, the file `modes.mf' @flindex modes.mf which is available from `ftp.cs.umb.edu' in `pub/tex/modes.mf'.)
The most common problem in generating fonts with METAFONT is that this file with the mode definitions is not included when creating the `plain.base' file.
Most of the configuration file options are similar to the command line options, but there are a few new ones.
Again, many may be turned off by suffixing the letter with a zero (0). These options are `a', `f', `q', `r', `K', `N', `U', and `Z'.
Within a configuration file, any empty line or line starting with a space, asterisk, equal sign, or a pound sign is ignored.
@ letterSize 8.5in 11in @ letter 8.5in 11in @+ %%BeginPaperSize: Letter @+ letter @+ %%EndPaperSize @ legal 8.5in 14in @+ ! %%DocumentPaperSizes: Legal @+ %%BeginPaperSize: Legal @+ legal @+ %%EndPaperSize
Note that you can even include structured comments in the configuration file! When dvips has a paper format name given on the command line, it looks for a match by the name; when it has a `papersize' special, it looks for a match by dimensions. The first match found (in the order the paper size information is found in the configuration file) is used. If nothing matches, a warning is printed and the first paper size given is used, so the first paper size should always be the default. The dimensions must match within a quarter of an inch. Landscape mode for all of the paper sizes are automatically supported. If your printer has a command to set a special paper size, then give dimensions of `0in 0in'; the PostScript code that sets the paper size can refer to the dimensions the user requested as `hsize' and `vsize'; these will be macros defined in the PostScript that return the requested size in default PostScript units. Note that virtually all of the PostScript commands you use here are device dependent and degrade the portability of the file; that is why the default first paper size entry should not send any PostScript commands down (although a structured comment or two would be okay). Also, some printers want `BeginPaperSize' comments and paper size setting commands; others (such as the NeXT) want `PaperSize' comments and they will handle setting the paper size. There is no solution I could find that works for both (except maybe specifying both.) See the supplied `config.ps' file for a more realistic example.
%! Hey, we're PostScript /Times-Roman findfont 30 scalefont setfont 144 432 moveto vmstatus exch sub 40 string cvs show pop showpage
Note that the number returned by this file is the total memory free; it is often a good idea to tell dvips that the printer has somewhat less memory. This is because many programs download permanent macros that can reduce the memory in the printer. In general, a memory size of about `300000' is plenty, since dvips can automatically split a document if required. It is unfortunate that PostScript printers with much less virtual memory still exist. Some systems or printers can dynamically increase the memory available to a PostScript interpreter, in which case this file might return a ridiculously low number; the NeXT computer is such a machine. For these systems, a value of one million works well.
@flindex testpage.tex
This is useful for a printer that consistently offsets output pages by a
certain amount. You can use the file `testpage.tex' to determine
the correct value for your printer. Be sure to do several runs with the
same O value--some printers vary widely from run to run.
One major problem with TeX and the Computer Modern fonts is the huge amount of disk space a full set of high resolution fonts can take. Dvips solves this problem by creating fonts on demand, so only those fonts that are actually used are stored on disk. At a typical site, less than one-fifth of the full set of Computer Modern fonts are used over a long period, so this saves a great deal of disk space.
Furthermore, the addition of dynamic font generation allows fonts to be used at any size, including typesetter resolutions and extremely huge banner sizes. Nothing special needs to be done; the fonts will be automatically created and installed as needed.
The downside is that it does take a certain amount of time to create a new font if it has never been used before. But once a font is created, it will exist on disk, and the next time that document is printed it will print very quickly.
It is the `MakeTeXPK' shell script that is responsible for making these fonts. It must echo the filename of the new font (and nothing else) to standard output. Use standard error for commentary. `MakeTeXPK' is passed various arguments, the first of which is the font it's supposed to make. You can override the other argument conventions with environment variables or at compilation time; see the Kpathsea documentation.
The `MakeTeXPK' script supplied invokes Metafont (using the Sauter scripts first, if necessary and if they are installed) to create the font and then copies the resultant `pk' file to a world-writable font cache area.
`MakeTeXPK' can be customized to do other things to get the font. For instance, if you are installing dvips to replace (or run alongside) an existing PostScript driver, and that driver demands `gf' fonts, you can easily modify `MakeTeXPK' to invoke `gftopk' to convert the `gf' files to `pk' files for dvips. This provides the same space savings listed above.
Because dvips (and thus `MakeTeXPK') is run by a wide variety of users, there must be a system-wide place to put the cached font files. In order for everyone to be able to supply fonts, the directory must be world writable. If your system administrator considers this a security hole, `MakeTeXPK' can write to `/tmp/pk' or some such directory, and periodically the cached fonts can be moved to a more general system area. The cache directory must exist on the `pk' file search path in order for `MakeTeXPK' to work.
Dvips reads a certain set of environment variables to configure its operation. The path variables are read as needed, after all configuration files are read, so they override values in the configuration files. (Except for the `TEXCONFIG' variable.)
See section `Path specifications' in Kpathsea library, for details of interpretation of environment variable values, and the common environment variables. Only the environment specific to `dvips' are mentioned here.
For special effects, if any of the macros `bop-hook', `eop-hook', `start-hook', or `end-hook' are defined in the PostScript `userdict', they will be executed at the beginning of a page, end of a page, start of the document, and end of a document, respectively.
When these macros are executed, the default PostScript coordinate system and origin is in effect. Such macros can be defined in headers added by the `-h' option or the `header=' special, and might be useful for writing, for instance, DRAFT across the entire page, or, with the aid of a shell script, dating the document. These macros are executed outside of the save/restore context of the individual pages, so it is possible for them to accumulate information, but if a document must be divided into sections because of memory constraints, such added information will be lost across section breaks.
The single argument to `bop-hook' is the sequence number of the page in the file; the first page gets zero, the second one, etc. The procedure must leave this number on the stack. None of the other hooks are (currently) given parameters, although this may change in the future.
As an example of what can be done, the following special will write a light DRAFT across each page in the document:
\special{!userdict begin /bop-hook{gsave 200 30 translate
65 rotate /Times-Roman findfont 216 scalefont setfont
0 0 moveto 0.7 setgray (DRAFT) show grestore}def end}
Note that using `bop-hook' or `eop-hook' in any way that preserves information across pages will break compliance with the Adobe document structuring conventions, so if you use any such tricks, it is recommended that you also use the `-N' option to turn off structured comments.
Several of the above tricks can be used nicely together, and it is not necessary that a `printer configuration file' be used only to set printer defaults. For instance, a `-P' file can be set up to print the date on each page; the particular configuration file will execute a command to put the date into a header file, which is then included with a `h' line in the configuration file. Multiple `-P' options can be used.
If you have any experience with `dvipsk' under DOS, karl@cs.umb.edu would like to hear about it.
See section `Reporting bugs' in Kpathsea, for the bug reporting address and information.
To compile and install Dvipsk:
configure; for example,
changing the installation directories. Alternatively, override the Make
variables on the command line when you run Make.
sh configure in the top-level directory. This tries to
figure out system dependencies and the installation prefix.
See section `The configure script' in Kpathsearch library, for options and other information about configure.
configure created.)
make; e.g., just type `make' in the top-level directory.
Barring configuration and compiler bugs, this will compile all the programs.
@flindex MACHINES
See the `./MACHINES' for possibly helpful system-dependent information.
make install. If you want to install only the executables, do
make install-exec; for only the data files, make
install-data. And if you don't want to install the fonts (perhaps
because your directory structure is different from the default), set
the Make variable install_fonts=false.
configdir.
afm2tfm (see section The Afm2tfm Program) to make TFM and VF files so the fonts
will be available in the same encoding as the fonts distributed with
Dvips. Also check `psfonts.map' to be sure the fonts are listed
there (see section Non-resident PostScript Fonts).
Here are the typical locations for vendor-supplied fonts:
You've gone through all the trouble of installing dvips, carefully read all the instructions in this manual, and still can't get something to work. This is all too common, and is usually caused by some broken PostScript application out there. The following sections provide some helpful hints if you find yourself in such a situation.
In all cases, you should attempt to find the smallest file that causes the problem. This will not only make debugging easier, it will also reduce the number of possible interactions among different parts of the system.
The `-d' flag to dvips is very useful for helping to track down certain errors. The parameter to this flag is an integer that tells what errors are currently being tracked. To track a certain class of debug messages, simply provide the appropriate number given below; if you wish to track multiple classes, sum the numbers of the classes you wish to track. The classes are:
If you are not getting any output at all, even from the simplest one-character file (for instance, `\ \bye'), then something is very wrong. Practically any file sent to a PostScript laser printer should generate some output, at the very least a page detailing what error occurred, if any. Talk to your system administrator about downloading a PostScript error handler. (Adobe distributes a good one called `ehandler.ps'.) @flindex ehandler.ps
It is possible, especially if you are using non-Adobe PostScript, that your PostScript interpreter is broken. Even then it should generate an error message. I've tried to work around as many bugs as possible in common non-Adobe PostScript interpreters, but I'm sure I've missed a few.
If dvips gives any strange error messages, or compilation on your machine generated a lot of warnings, perhaps the dvips program itself is broken. Carefully check the types in `dvips.h' and the declarations in the `Makefile', and try using the debug options to determine where the error occurred.
It is possible your spooler is broken and is misinterpreting the structured comments. Try the `-N' flag to turn off structured comments and see what happens.
If some documents come out inverted or too small, your spooler is not supplying an end of job indicator at the end of each file. (This happens a lot on small machines that don't have spoolers.) You can force dvips to do this with the `-F' flag, but note that this generates files with a binary character (control-D) in them. You can also try using the `-s' flag to enclose the entire job in a save/restore pair.
If your printer returns error messages, the error message gives very good information on what might be going wrong. One of the most common error messages is `bop undefined'. This is caused by old versions of Transcript and other spoolers that do not properly parse the setup section of the PostScript. To fix this, turn off structured comments with the `-N' option, but make sure you get your spooling software updated.
Another error message is `VM exhausted'. (Some printers indicate this error by locking up; others quietly reset.) This is caused by telling dvips that the printer has more memory than it actually does, and then printing a complicated document. To fix this, try lowering the parameter to `m' in the configuration file; use the debug option to make sure you adjust the correct file.
Other errors may indicate that the graphics you are trying to include don't nest properly in other PostScript documents, or any of a number of other possibilities. Try the output on a QMS PS-810 or other Adobe PostScript printer; it might be a problem with the printer itself.
This common error is caused by not editing the `config.ps' file to reflect the correct resolution for your site. You can use the debug flags (`-d64') to see what files are actually being read.
This is usually caused by incorrectly specifying the amount of memory the printer has in `config.ps'; see the description above.
The reasons why graphics inclusions fail are too numerous to mention. The most common problem is an incorrect bounding box; read the section on bounding boxes and check your PostScript file. Complain very loudly to whoever wrote the software that generated the file if the bounding box is indeed incorrect.
Another possible problem is that the figure you are trying to include does not nest properly; there are certain rules PostScript applications should follow when generating files to be included. The dvips program includes work-arounds for such errors in Adobe Illustrator and other programs, but there are certainly applications that haven't been tested.
One possible thing to try is the `-K' flag, to strip the comments from an included figure. This might be necessary if the PostScript spooling software does not read the structuring comments correctly. Use of this flag will break graphics from some applications, though, since some applications read the PostScript file from the input stream looking for a particular comment.
Any application which generates graphics output containing raw binary (not ASCII hex) will probably fail with dvips.
If dvips complains that it cannot find certain font files, it is possible that the paths haven't been set up correctly for your system. Use the debug flags to determine precisely what fonts are being looked for and make sure these match where the fonts are located on your system.
This happens a lot if either `MakeTeXPK' hasn't been properly edited and installed, or if the local installation of METAFONT isn't correct.
`MakeTeXPK' must echo the generated filename (and nothing else) to standard output. See section Automatic Font Generation.
If METAFONT isn't found when `MakeTeXPK' is running, then you need to install it. Retrieve it from, e.g., `ftp.cs.umb.edu' in `pub/tex/web2c.tar.gz' and `pub/tex/web.tar.gz'.
If METAFONT runs but generates fonts that are too large (and prints out the name of each character as well as just a character number), then your METAFONT base file probably hasn't been made properly. To make a proper `plain.base', assuming the local mode definitions are contained in `modes.mf' type the following command (assuming UNIX):
inimf "plain; input modes; dump"Then copy the `plain.base' file from the current directory to where the base files are stored on your system.
By the way, the macros defined in `cmbase' will break fonts that do not use `cmbase'; such fonts include the LaTeX fonts. Loading the `cmbase' macros when they are needed is done automatically and takes less than a second--an insignificant fraction of the total run time of METAFONT for a font, especially when the possibility of generating incorrect fonts is taken into account. If you create the LaTeX font `circle10', for instance, with the `cmbase' macros loaded, the characters will have incorrect widths.
This new feature of `dvips' is somewhat experimental so your experiences and comments are welcome. Initially added by Jim Hafner, IBM Research, `hafner@almaden.ibm.com', the color support has gone through many changes by Tomas Rokicki. Besides the changes to the source code itself, there are additional TeX macro files: `colordvi.tex' and `blackdvi.tex'. There are also `.sty' versions of these files that can be used with LaTeX and other similar macro packages. This feature adds one-pass multi-color printing of TeX documents on any color PostScript device.
In this section we describe the use of color from the document preparer's point of view and then add some simple instructions on installation for the system administrator.
All the color macro commands are defined in `colordvi.tex' (or `colordvi.sty'). To access these macros simply add to the top of your TeX file the command
\input colordvi
or, if your document uses style files like LaTeX, add the `colordvi' style option as in
\documentstyle[12pt,colordvi]{article}
There are basically two kinds of color macros, ones for local color changes to, say, a few words or even one symbol and one for global color changes. Note that all the color names use a mixed case scheme. There are 68 predefined colors, with names taken primarily from the Crayola crayon box of 64 colors, and one pair of macros for the user to set his own color pattern. More on this extra feature later. You can browse the file `colordvi.tex' for a list of the predefined colors. The comments in this file also show a rough correspondence between the crayon names and PANTONEs.
A local color command is in the form
\ColorName{this will print in color}
Here `ColorName' is the name of a predefined color. As this example shows, this type of command takes one argument which is the text that is to print in the selected color. This can be used for nested color changes since it restores the original color state when it completes. For example, suppose you were writing in green and want to switch temporarily to red, then blue, back to red and restore green. Here is one way that you can do this:
This text is green but here we are \Red{switching to red,
\Blue{nesting blue}, recovering the red} and back to
original green.
In principle there is no limit to the nesting level, but it is not advisable to nest too deep lest you loose track of the color history. The global color command has the form
\textColorName
This macro takes no arguments and immediately changes the default color from that point on to the specified color. This of course can be overriden globally by another such command or locally by local color commands. For example, expanding on the example above, we might have
\textGreen
This text is green but here we are \Red{switching to red,
\Blue{nesting blue}, recovering the red} and back to
original green.
\textCyan
The text from here on will be cyan unless
\Yellow{locally changed to yellow}. Now we are back to cyan.
The color commands will even work in math mode and across math mode
boundaries. This means that if you have a color before going into math
mode, the mathematics will be set in that color as well. More
importantly however, in alignment environments like `\halign',
`tabular' or `eqnarray', local color commands cannot extend
beyond the alignment characters.
Because local color commands respect only some environment and
deliminator changes besides their own, care must be taken in setting
their scope. It is best not to have then stretch too far.
At the present time there are no macros for color environments in LaTeX
which might have a larger range. This is primarily to keep the TeX
and LaTeX use compatible.
There are two ways for the user to specify colors not already defined. For local changes, there is the command `\Color' which takes two arguments. The first argument is a quadruple of numbers between zero and one and specifies the intensity of cyan, magenta, yellow and black (CMYK) in that order. The second argument is the text that should appear in the given color. For example, suppose you want the words "this color is pretty" to appear in a color which is 50% cyan, 85% magenta, 40% yellow and 20% black. You would use the command
\Color{.5 .85 .4 .2}{this color is pretty}
For global color changes, there is a command `\textColor' which takes one argument, the CMYK quadruple of relative color intensities. For example, if you want the default color to be as above, then the command
\textColor{.5 .85 .4 .2}
The text from now on will be this pretty color
will do the trick. Making a global color change in the midst of a nested local colors is highly discouraged. Consequently, dvips will give you warning message and do its best to recover by discarding the current color history.
These color macros are defined by use of specialized `\special' keywords. As such, they are put in the `.dvi' file only as explicit message strings to the driver. The (unpleasant) result is that certain unprotected regions of the text can have unwanted color side effects. For example, if a color region is split by TeX across a page boundary, then the footers of the current page (e.g., the page number) and the headers of the next page can inherit that color. To avoid this effect globally, users should make sure that these special regions of the text are defined with their own local color commands. For example in TeX, to protect the header and footer, use
\headline{\Black{My Header}}
\footline{\Black{\hss\tenrm\folio\hss}}
This warning also applies to figures and other insertions, so be careful!
Of course, in LaTeX, this is much more difficult to do because of the complexity of the macros that control these regions. This is unfortunate, but is somehow inevitable because TeX and LaTeX were not written with color in mind.
Even when writing your own macros, much care must be taken. The color macros that `colorize' a portion of the text work by prefixing the text with a special command to turn the color on and postfixing it with a different special command to restore the original color. It is often useful to ensure that TeX is in horizontal mode before the first special command is issued; this can be done by prefixing the color command with `\leavevmode'.
If you have a TeX or LaTeX document written with color macros and you want to print it in black and white there are two options. On all (good) PostScript devices, printing a color file will print in corresponding grey-levels. This is useful since in this way you can get a rough idea of where the colors are changing without using expensive color printing devices. The second option is to replace the call to input `colordvi.tex' with `blackdvi.tex' (and similarly for the `.sty' files). So in the above example, replacing the word `colordvi' with `blackdvi' suffices. This file defines the color macros as no-ops, and so will produce normal black/white printing. By this simple mechanism, the user can switch to all black/white printing without having to ferret out the color commands. Also, some device drivers, particularly non-PostScript ones like screen previewers, will simply ignore the color commands and so print in normal black/white. Hopefully, in the future screen previewers for color displays will be compatible with some form of color support.
To configure dvips for a particular color device you need to fine tune the color parameters to match your devices color rendition. To do this, you will need a PANTONE chart for your device. The header file `color.lpro' shows a (rough) correspondence between the Crayola crayon names and the PANTONE numbers and also defines default CMYK values for each of the colors. Note that these colors must be defined in CMYK terms and not RGB as dvips outputs PostScript color commands in CMYK. This header file also defines (if they are not known to the interpreter) the PostScript commands `setcmykcolor' and `currentcmykcolor' in terms of a RGB equivalent so if your device only understands RGB, there should be no problem.
The parameters set in this file were determined by comparing the PANTONE chart of a Tektronics PHASER printer with the actual Crayola Crayons. Because these were defined for a particular device, the actual color rendition on your device may be very different. There are two ways to adjust this. One is to use the PANTONE chart for your device to rewrite `color.lpro' prior to compilation and installation. A better alternative, which supports multiple devices, is to add a header file option in the configuration file for each device that defines, in `userdict', the color parameters for those colors that need redefining.
For example, if you need to change the parameters defining `Goldenrod' (approximately PANTONE 109) for your device `mycolordev', do the following. In the PANTONE chart for your device, find the CMYK values for PANTONE 109. Let's say they are `{\ 0 0.10 0.75 0.03 }'. Then create a header file named `mycolordev.pro' with the commands
userdict begin
/Goldenrod { 0 0.10 0.75 0.03 setcmykcolor} bind def
Finally, in `config.mycolordev' add the line
h mycolordev.pro
This will then define `Goldenrod' in your device's CMYK values in `userdict' which is checked before defining it in `TeXdict' by `color.pro'.
This mechanism, together with additions to `colordvi.tex' and `blackdvi.tex' (and the `.sty' files), can also be used to predefine other colors for your users.
To support color, dvips recognizes a certain set of specials. These specials all start with the keyword `color' or the keyword `background'.
We will describe `background' first, since it is the simplest. The `background' keyword must be followed by a color specification. That color specification is used as a fill color for the background. The last `background' special on a page is the one that gets issued, and it gets issued at the very beginning of the page, before any text or specials are sent. (This is possible because the prescan phase of dvips notices all of the color specials so that the appropriate information can be written out during the second phase.)
Ahh, but what is a color specification? It is one of three things. First, it might be a PostScript procedure as defined in a PostScript header file. The `color.pro' file defines 64 of these, including `Maroon'. This PostScript procedure must set the current color to be some value; in this case, `Maroon' is defined as `0 0.87 0.68 0.32 setcmykcolor'.
The second possibility is the name of a color model (initially, one of `rgb', `hsb', `cmyk', or `gray') followed by the appropriate number of parameters. When dvips encounters such a macro, it sends out the parameters first, followed by the string created by prefixing `TeXcolor' to the color model. Thus, the color specification `rgb 0.3 0.4 0.5' would generate the PostScript code `0.3 0.4 0.5 TeXrgbcolor'. Note that the case of zero arguments is disallowed, as that is handled by the single keyword case above (where no changes to the name are made before it is sent to the PostScript file.)
The third and final type of color specification is a double quote followed by any sequence of PostScript. The double quote is stripped from the output. For instance, the color specification `"AggiePattern setpattern' will set the `color' to the Aggie logo pattern (assuming such exists.)
So those are the three types of color specifications. The same type of specifications are used by both the `background' special and the `color' special. The `color' special itself has three forms. The first is just `color' followed by a color specification. In this case, the current global color is set to that color; the color stack must be empty when such a command is executed.
The second form is `color push' followed by a color specification. This saves the current color on the color stack and sets the color to be that given by the color specification. This is the most common way to set a color.
The final version of the `color' special is just `color pop', with no color specification; this says to pop the color last pushed on the color stack from the color stack and set the current color to be that color.
The `dvips' program correctly handles these color specials across pages, even when the pages are repeated or reversed.
These color specials can be used for things such as patterns or screens as well as simple colors. However, note that in the PostScript, only one `color specification' can be active at a time. For instance, at the beginning of a page, only the bottommost entry on the color stack is sent; also, when a color is `popped', all that is done is that the color specification from the previous stack entry is sent. No `gsave' or `grestore' is used. This means that you cannot easily mix usage of the `color' specials for screens and colors, just one or the other. This may be addressed in the future by adding support for different `categories' of color-like state.