World: r3wp

I suspect that by the time I finish it, Gabriele will have come out 
with a pdf-maker which will render mine obsolete!

The idea here is to produce postscript as the last thing, just before 
sending the data to the printer. We shouldn't produce postscript, 
that we then need to preview before printing, because we then need 
to solve the problem of converting postscript to something viewable 
on the screen. It's better to work with DRAW or whatever developers 
use to visualize their content and have preview at that level. Then 
in the end, the user wants to print and now, finally the content 
is converted to postscript to be sent to a printer.


Having a postscript printer is a huge benefit, because you can say 
goodbye to printer drivers and all the trouble, those cause. The 
REBOL postscript dialect is your driver. :-)

A postscript printer is a printer, that understands the postscript 
format. It reads the text commands, a postscript file is, and print. 
Data like images is inside the postscript file as hex or something.

Then the million dollar question: Why do printer manufacturers make 
printer drivers filling up 10s or 100s of MB on users harddiscs? 
Why do those drivers need to be so huge, if it's such a simple task 
to print (using postscript)?

(Of course if the printer isn't a postscript printer, and you want 
to print a postscript file, you need a driver to convert from postscript 
to the format, the printer understands. So people without a postscript 
printer can still print from postscript-producing REBOL programs. 
They then just need a printer driver.)

Theoretically HTML+CSS should have everything you need to print with 
a fairly high degree of control, but all browser implementations 
suck.  Badly.

there's no support for text rotation or hyphenation in CSS either

Most people producing print expect WYSIWYG. HTML+CSS is not a WYSIWYG 
standard (contrary to what many seems to think). So trying to produce 
WYSIWYG output from a browser (or expecting WYSIWYG output from a 
browser) is doomed from the beginning.

hence the stupidiy of expecting traditional web apps to be able to 
do anything. :-(

and why javascript should have been promoted more for the last 10 
years.

a bit off topic: has any one noticed how often when printing HTML 
pages that the last page that comes out is just empty or useless?

How well does OO with such things?

Yes, it happens alot to me too and is a waste of paper. Good that 
many websites has that little printer-friendly-version icon.

Volker, OpenOffice doesn't include a browser (the versions I've used), 
while StarOffice did (when I used that). So opening a HTML doc in 
OO gives you the wordprocessor with the concept of paper-pages, and 
then you print, what you see.

But how good is that html->page-maker?

oops, I remembered wrong. Opening a HTML-file in OpenOffice gives 
you an HTML editor (without the concept of paper-pages). I'll make 
a test...

IIRC OO has multiple views, like pages, continuos. MAybe you can 
simply change that

Opening a tv-program site from a danish tv-station made NeoOffice 
(a Mac version of OO) loop. I'll try with www.rebol.com

geomol, is "Astronomically Slow" still the middle name of NeoOffice?

yes :D

www.rebol.com opened OK in NeoOffice, and I could change to "Print 
Layout". Right-margin is not perfect (as often with HTML, because 
it isn't a wysiwyg format), and it can probably be configured. Saving 
as PDF gave me exactly what was seen on the screen.

Volker, so my first impression is, that OO is OK to handle HTML, 
but not perfect (and how could any program be with all that misunderstanding 
of HTML out there).

H: I think there is hyphenation in CSS 3, iirc.

Some small progress .. I can now get floating text tables, and text 
that flows from one textbox on one page to the next page.  But not 
completely error free yet :(

This is in Draw .. haven't tackled this in PostScript yet.

do http://www.compkarori.com/reb/psd2draw5.r

this is an alpha dialect that just does draw at present.

I have two textboxes on the first page, and one on the second.  Text 
flows from the first page to the second, sort of justified.

A textbox is used also to centre the large titles on each page.

I have no idea why the small text looks as though it is in bold.

the point of having this dialect is to create page templates for 
viewing data from databases or whatever.  The text is replaced either 
by evaluation or by compose, and you have your draw layout.

sure you can create VID layouts for viewing data from databases, 
but you can't print it easily.

Need to generate bar codes in postscript ?  Here's some software 
on a MIT/X-Consortium license http://www.terryburton.co.uk/barcodewriter/

so, as the author says, you can include a barcode as simply as

50 450 moveto (1-86074-271) (includetext) isbn barcode

I know about ez-plot being capable of doing very useful plots ( barring 
sequential date data ).

but of course, there's no great way of printing out these plots.

This is a simple graph ( multivariable ) graph using the postscript 
dialect.  http://www.compkarori.com/emr/graph.png

That's how it looks in draw and also when printed to a postscript 
printer.

It would be nice to have vertically oriented text for the y-axis 
label .. but I'm going to have to think again about text rotation 
in draw so that it can be mapped easily also to postscript.

is that graph drawn with AGG?

Draw is AGG is it not?

that graph is output from my EMR program.

the fonts look a little smeared.

is that because they are vectorized?

I guess so.

But they print fine to the printer.

that's good

I haven't followed it, but did you solve font positioning in DRAW 
vs. postscript?

dunno what kind of graphs you are using, but sequential data, maybe 
mrtg or rrdtool would be usefull? Of course it would not probably 
solve your problem with printing ... what you have apparently works 
for you ...