[Gnome-print] Re: RFC: A draft proposal ...

[Damien Diederen <dash linuxbe org>]

Hello !

On Tue, Apr 18, 2000 at 01:59:32PM +0200, Mathieu Lacage wrote:
> Derek Simkowiak <dereks@kd-dev.com> writes:
> 
[clip]
> > 	I'm working on a Gtk+-only text editor (I'm currently bogged down
> > in writing a new text widget :), and for my printing system I plan to pipe
> > the text file through "mpage" (which converts text into Postscript).  
> > Type "man mpage" and you'll immediately see the huge variety of print
> > options I'll be able to offer in my program.  All I need to do is put
> > together a GUI dialog box around the common/useful mpage command-line
> > options and boom! I've got a full-blown printing system that works with
> > any printer supported by Red Hat Linux (read: Ghostscript filters).
> 
> Ghostscript rasterizing system is not THAT good. Raph has some sample
> prints which prove this.

While I agree that GhostScript rasterizing is not always perfect, I wonder
if this is not "fixable". Is the GhostScript codebase unmaintainable ? 
Wouldn't it be possible to take the parser and the PostScript interpreter
out of it, and to interface it with libart (or whatever you want to use
as a rendering engine ?).

This would also improve the print quality for the vast number of users that
are using GhostScript to print on their non-PostScript printers with 
non-gnome applications (and the large base of PostScript manipulation tools
will probably not be ported to Gnome, while they could be used by Gnome
applications -- have a look at the example Derek gave about his text editor
and mpage).

> > 
> > 	One point not mentioned in the RFC is all the pre-existing Open
> > Source code, algorithms, and documentation built around Postscript.  If we
> > don't go with Postscript, we would not only need to re-invent the wheel,
> > but we'd need to throw out a mature, stable, well-supported wheel at the
> > same time.
> > 
> > 	...Not to mention that many printers support Postscript right out
> > of the box.
>
> which is NOT the case for most low-cost printers which most people buy.
> 
> It happens sadly that our target users (desktop users) don't usually
> buy PS printers. (My mother has no PS printer)

Rasterizing still has to be done, even if you don't use PostScript, as
a vast majority of application printouts is vector data.

> 
> > 
> > 	Also, using Postscript is a proven model.  If Postscript works for
> > GNUStep, I don't see why it shouldn't work for Gnome.
> 
> Postscript imaging model misses some stuff. First one to come to my mind
> is transparency (ie: alpha compositing) which can be simulated in PS 
> though, I know.

I don't quite grasp why a GhostScript/Libart frankeinstein couldn't serve
our need (the extension could be described in the XPD, and emulation
could be provided for true PostScript printers -- this needs to be done,
anyways).

...
0.5 1 0 0 gs_setrgbacolor
eofill
...

> Mathieu
> 

Cu,
Damien.

-- 
* Pétition contre les brevets logiciels : http://lln.udev.org/sign.php3
-- 
Damien Diederen
dash@linuxbe.org
http://users.swing.be/diederen/