Wordwrapping Demo (5k) (was PS wordwrap)
David Byram-Wigfield
%%Title: Basic PostScript Linewrapping
%%BoundingBox: 0 0 612 842 % for some viewers
%%Creator: by...@cappella-archive.com
%%For: Cappella Archive
%%Date: 22/4/01
%%EndComments
% comment lines introduced by a % must not linebreak
% Drop over GS, distill, or download to a PS printer
%%BeginSetup
% Default text margins: top, right, bottom, left
% top margin = text height: right margin = text width
/TM { /tm exch def } def 680 TM % 9.5"
/RM { /rm exch def } def 450 RM % 6.5"
/BM { /bm exch def } def 0 BM
/LM { /lm exch def } def 0 LM
/LG { /lg exch def } def 12 LG % linespacing
/textbox { 680 TM 450 RM 0 BM 0 LM 12 LG 10 rom lm tm moveto } def
% knife and fork linewrapping procedures
/find { search { pop 3 -1 roll 1 add 3 1 roll } { pop exit } ifelse } def
% look for spaces
/spacecount { 0 exch ( ) { find } loop } def
% compare with page textwidth (the measure)
/toofar? { ( ) search pop dup stringwidth pop currentpoint pop add rm gt } def
% Mark-up codes
% full line advance
/L { lg tm exch sub TM lm tm moveto } def
% place text and linewrap: 2 'shows' here for a word and its space
/S { dup spacecount { toofar? { L show show } { show show } ifelse }
repeat pop } bind def
% linewrap text: advance to next line
/P { S L } bind def
% font abbreviation
/F { findfont exch scalefont setfont } def
% substitute any Type 1 font: use your o/s PostScript names
/rom { /Times-Roman F } def
/bol { /Times-Bold F } def
/it { /Times-Italic F } def
/ss { /Helvetica F } def
/si { /Helvetica-Oblique F } def
% single page format
/1upA4 { /asyouwere save def 72 72 translate textbox } def
% print the last page: close the file
/close { showpage grestore clear asyouwere restore } def
%%EndSetup
%%BeginScript
1upA4 % open the page format
10 bol (Linewrapping Text ) P
L
10 it (Michael \<mog...@mlug.missouri.edu\> wrote: ) P
(I am trying to figure out how I'd make text on a page wordwrap properly
and allow you to switch fonts mid-sentence.
. . I tried a couple example word wrap functions from my books and they
either seem to start each type of text block on a new line or when you
change fonts they stop wrapping. Anyone know a way to do this that would
allow multiple fonts in the same line and wrap correctly? ) P
L 10 rom
(As no one at the time of writing has actually answered the question, this
demo should explain it more fully. ) P
(There are two methods of end of line wordwrapping text. The first checks
the position of each trailing space against the required width of the line
of text on the page \(the `measure' in typographical parlance\). If there
is sufficient room, each word and its accompanying space is placed with a
'show' instruction until the measure is full and a line advance then
carries over any additional text to the following line. This method allows
the font size and style to be easily altered as the text is typed. Like
this: ) P
L 10 bol (This text is 10 point bold. ) S
9 it (This text is 9 point italic. ) S
8 ss (This is 8 point sans serif. ) S
10 it 0 0 1 setrgbcolor (You may also add some colour which is useful for
PDF pages. ) S
10 rom 0 setgray (Here the text returns to 10 point roman and a line
advance creates the starting point for a new paragraph. ) S
L 10 it
(Notice the empty space in the file before the final bracket delimiter, as
a trailing space must always accompany the previous word, which will
otherwise disappear into digital limbo. The TinyDict has a procedure which
glues the two together after measurement and also page-jumps when the last
line is reached. The traditional UK book measure is 24 ems i.e. 288
points, so I use 300 points to `print' the last space. ) P
L 10 rom
(Obviously this method is not suitable for setting text that is flush
justified to left and right. In this case each complete line of text must
be stored before it can be assessed and the spaces then stretched by
`widthshow' to make it fit the measure. The `naive' method is to justify
each line as it is completed, like traditional letterpress, and this is
the method I have adopted in the TinyDict. ) P
L
(The head-banging way is to read in an entire paragraph, split the text
according to the measure, and place each change of font, style and size,
and its associated text in a series of arrays and then adjust the word
space value for the relevant line for each one. If hyphenation algorithms
and hanging punctuation searches are added, the method becomes processor
intensive and not particularly suitable for on-the-fly parsing. PostScript
manuals are very coy about this more sophisticated procedure and my
information was gleaned from discussions with the Engineering Division of
Adobe Systems Europe. ) P
L
10 rom (For a more detailed discussion of full justification see the free
Cappella Archive ebook: ) P
10 it (Practical PostScript: A Beginner's Guide to Digital Typesetting, ) S
9 ss (90 PDF pages \(416k\) ) P
(http://www.cappella-archive.com/bookpdfs/pracpost.pdf ) P
(or noframes: http://www.cappella.demon.co.uk/bookpdfs/pracpost.pdf ) P
%%EndScript
close
%%EOF
Cappella Archive, Book on Demand Editions
The TinyDict mark-up for self-printing books
www.cappella-archive.com