HELP WANTED: kdeprint with GNUlpr

[Kurt Pfeifle]

James Richard Tyrer wrote:

> I have volunteered to work on kdeprint's LPR support.
> 

You must be a hero....    ;-)

(You won't have many users to service....)

> After a delay in getting my new system up and running -- the only thing 
> left is to get KOffice-1.3 working -- I will get started on this.
> 
> The most important thing is to get the kjobviewer to work.  The code 
> looks basically OK, so this is a debugging issue.
> 
> Then before doing any more, the question is.  Do we want LPR support?
> 

This is the most important question!

> The GNUlpr project is currently inactive. :-( 

"Currently"? It is so for a very long time already, counting *years*....

> Therefore, if we are 
> going to have LPR support, KDE is going to have to adopt the project and 
> fork it. 

Not that my opinion does count much here... (I am not a coder, so my
comments do come from the sideline.)

* GNUlpr is dead.

* CUPS is alive and kicking.

* Most Linux distros ship with CUPS as their default printing system,
   and LPRng as the second choice. GNUlpr doesn't stand any chance to
   reach more than a few hundred Linux users around the globe, and
   almost none in the Unix world.

* If any other printing system should get stronger support in KDEPrint,
   it is LPRng. First off, it has come back to live. Patrick Powell
   seems to be back coding. (Last I saw was that he is very active
   helping to cleanup Foomatic for the upcoming 3.0.1 release and put
   better LPRng support into that.). And LPRng has a much stronger
   traditional user base than GNUlpr. It is also used in commercial
   proprietary Unix environments, which GNUlpr is not (and where KDE
   is also appearing on desktop systems).

* And having to *fork* the project to make it work with KDEPrint
   just isn't worth the effort. There are better things to do for
   KDE and KDEPrint. (I am going to post my personal wishlist sometime
   in the near future -- but this centers very much around CUPS.)

(I may be wrong with some of my assumptions, and I am not at all
familiar with the GNUlpr features [apart from the fact that they
have been lifted off the early CUPS source code], so I may be missing
some important benefit of that package. If so, then teach me better,
please.)

> To do this, we need an ace C++ coder (I am an ace programmer 
> but a C++ newbie) to get the current code release:
> 
> https://sourceforge.net/project/showfiles.php?group_id=3800&release_id=63447 
> 
> 
> http://prdownloads.sourceforge.net/lpr/gnulpr-1.0.1.tar.gz?download
> 
> to build with the current GCC release.  I worked at this for a while, 
> but didn't succeed in getting it to work.
> 
> This would then become the Klpr project and KDE would have a native 
> print system.
> 
> Also I would appreciate anyone willing to use GNUlpr as their print 
> system to assist in getting the KDE support working.  I managed to 
> install: lpr-0.72 from source but had to install the RPMs for some of 
> the support stuff that wouldn't build on the new compiler.
> 
> You will also need to install GhostScript-8.11 built from source as an 
> SO -- make so && make soinstall.  And install the gsk patch to your 
> source code for KDE:
> 
> http://home.earthlink.net/~tyrerj/files/gs-fix/gsk-1.0.patch.bz2
> 
> The gsk executable is currently just a script.  This will need to be 
> expanded into a C/C++ executable that will call the GhostScript library 
> directly.
> 
> And, patch Qt so that you can use PostScript data files without the 
> fonts embedded.
> 
> http://home.earthlink.net/~tyrerj/files/qt-x11-free-3.2.1-PSfontname.patch.bz2 
> 
> 
> This will patch qt-3.2.2 but has NOT been tested on qt-3.2.2 yet (I have 
> to have KOffice for that).  Updates will be posted the same place:
> 
> http://home.earthlink.net/~tyrerj/files/
> 

This sounds like an awefull lot of work to do to just get up and
walking.

> Any comments?
> 

I don't want to demoralize you at all -- but these are my thoughts.
And you asked for comments. Wouldn't you have more fun and rewards
of all kinds if you put your energy behind making LPRNg support more
complete and make the CUPS support perfect?

CUPS 1.2 will have a *lot* of new features, and the first betas to
work with and create a world class GUI frontend for will be appearing
around the year's end. This is where the music plays and this is not
as frustrating as trying to kiss a dead horse back to life.

Cheers,
Kurt    [ just my 2 cents... ]