primax: Thread: SANE Backend

Hello!

Is anyone working on a Sane Backend for primax parallelport scanners?

If not, I'd like to take your program and turn it into a SANE Backend.
However, I don't have any experience with writing Sane Backends, so it
may take some time...

Or has anybody a good reason against a Sane Backend?

PS: Please excuse my bad english.

-- 
WAR IS PEACE - FREEDOM IS SLAVERY - IGNORANCE IS STRENGTH
2 x 2 = 5
Who controls the past controls the future:
Who controls the present controls the past
                              -- 1984, George Orwell

Christian Biesinger wrote:
> 
> Hello!
> 
> Is anyone working on a Sane Backend for primax parallelport scanners?
If there is anybody, they do not seem to be in contact with this list.

> If not, I'd like to take your program and turn it into a SANE Backend.
> However, I don't have any experience with writing Sane Backends, so it
> may take some time...
I think that it should be possible to "massage" the current code so that
it would be reusable simply as a library.
Also I came accross another parallel port driver for sane which seems
quite
interesting : http://as6edriver.sourceforge.net/

> Or has anybody a good reason against a Sane Backend?
> 
> PS: Please excuse my bad english.
No problem with your english.

If you decide to go ahead with this I would be interested in helping
you.

CP

Hi
 
> Is anyone working on a Sane Backend for primax parallelport scanners?

I do it actually.

It's not a real backend for the scanner. I decided to write a meta backend,
that contacts to a server. The server code is most of the primaxscan code. So
it would be easy to update the backend code when primaxscan changes.

If you want to help me, you're welcome.


-- 
Andre Herms                   ####@####.####
                              http://www.cs.uni-magdeburg.de/~aherms

Hello again

Andre Herms schrieb:
> > Is anyone working on a Sane Backend for primax parallelport scanners?
> 
> I do it actually.
> 
> It's not a real backend for the scanner. I decided to write a meta backend,
> that contacts to a server. The server code is most of the primaxscan code. So
> it would be easy to update the backend code when primaxscan changes.

Ah, interesting way to do it.

There would have been another way to do it:
The Primax Program is a command-line program.
So, the sane plugin could call this program.

OTOH, this may not be a "clean" solution.

> If you want to help me, you're welcome.

I'd like to.

-- 
WAR IS PEACE - FREEDOM IS SLAVERY - IGNORANCE IS STRENGTH
2 x 2 = 5
Who controls the past controls the future:
Who controls the present controls the past
                              -- 1984, George Orwell

On Wed, Aug 30, 2000 at 01:22:28AM +0100, C Hennessy wrote:
> If there is anybody, they do not seem to be in contact with this list.
I remember someone on this list wnating to make the driver give raw output,
this could be the first step towards a SANE driver. (simply a Meta-Backend)

Marco told me some time ago he was working on a SANE backend, but there seems
to be no contact with him anymore...

> > If not, I'd like to take your program and turn it into a SANE Backend.
> > However, I don't have any experience with writing Sane Backends, so it
> > may take some time...
It really seems to be very confusing (at least after I read the SANE docs..
.)

> I think that it should be possible to "massage" the current code so that
> it would be reusable simply as a library.
Well I thought of a Meta-backend which simply reads raw data which is produced
by the "real" backend (our program with added raw funcionality)

> Also I came accross another parallel port driver for sane which seems
> quite
> interesting : http://as6edriver.sourceforge.net/
hmmm... I'll have to check this out.

> > Or has anybody a good reason against a Sane Backend?
not really... but i haven't really understood the SANE stuff, yet ...

> If you decide to go ahead with this I would be interested in helping
> you.
ok. I think i can remember I've posted a quick and dirty guide about what h=
as
to be changed in the program to make it output raw data to this list.

--
Christian Ordig             | Homepage: http://thor.prohosting.com/~chrordig/
Germany                     |    eMail: Christian Ordig ####@####.####

Andre Herms wrote:
> 
> Hi
> 
> > Is anyone working on a Sane Backend for primax parallelport scanners?
> 
> I do it actually.
> 
> It's not a real backend for the scanner. I decided to write a meta backend,
> that contacts to a server. The server code is most of the primaxscan code. So
> it would be easy to update the backend code when primaxscan changes.

So if I understand this better then you mean that we would have both the
sane
demon plus a primaxscan demon running at the same time ?

If this is so, is it not more complicated than integrating primaxscan
into
sane ?
ANd then updates of primaxscan, so that there would be a library which
the
primaxsane backend would simply link against. This would also allow us
to
continue to build the primaxscan binary.

CP

----------  Weitergeleitete Nachricht  ----------
Subject: Re: SANE Backend
Date: Wed, 30 Aug 2000 15:55:18 +0200
From: Andre Herms ####@####.####


> Ah, interesting way to do it.
> 
> There would have been another way to do it:
> The Primax Program is a command-line program.
> So, the sane plugin could call this program.
> 
> OTOH, this may not be a "clean" solution.

I choosed this way because of a special feature (-:.
The lamp needs a certain warmup time. So the server is always running and it
will manage this:
If you start the backend (connect to the server) the lamp will switch on. The
server will get the time. If you start the scan, the server will wait until the
warmup time is over and then start.
After the scan, the lamp won't go off. this will happen after a certain timeout.
So if you restart the scan, there will be no warmup cicle.
Sounds cool ehr ?!

> I'd like to.

I need really help. I need most help with the makefiles, as I don't know much
about it. The most problem seems to be the sane backend (not the server)
because its compiled as shared libary and i don't know how this is done. So I
did just copy n' paste mit the sane makefile, but this isn't clean.

** If anyone could help me with this. **

I will try to put the backend to a stable state, that other can get and test it.

At the moment there is a dummy server. It acts as the primax server should do
later. 

I would attach them, but I don't think it's good to post attachments to a
newsgroup. 

Q: Is there a cvs or ftp server?

so far.

	Andre

-- 
Andre Herms                   ####@####.####
                              http://www.cs.uni-magdeburg.de/~aherms
-------------------------------------------------------

-- 
Andre Herms                   ####@####.####
                              http://www.cs.uni-magdeburg.de/~aherms

C Hennessy schrieb:
> Andre Herms wrote:
> > > Is anyone working on a Sane Backend for primax parallelport scanners?
> >
> > I do it actually.
> >
> > It's not a real backend for the scanner. I decided to write a meta backend,
> > that contacts to a server. The server code is most of the primaxscan code. So
> > it would be easy to update the backend code when primaxscan changes.
> 
> So if I understand this better then you mean that we would have both the
> sane
> demon plus a primaxscan demon running at the same time ?

He wrote to me in a personal mail (though I don't know if he did it
intentionally) that this had the following reason: When a client
connects to the server, the server switches the scanner light on. When
the client disconnects, it remains switched on for some time (I hope I
got it right).

> If this is so, is it not more complicated than integrating primaxscan into sane ?
> ANd then updates of primaxscan, so that there would be a library which the
> primaxsane backend would simply link against. This would also allow us
> to continue to build the primaxscan binary.

I'm not sure if I got you right here.
I think what you want is the following:
A library which both the sane backend and the current binary link
against, so that both versions benefit from improvements and bug fixes.

If this is what you want, I think it's a great idea!

-- 
WAR IS PEACE - FREEDOM IS SLAVERY - IGNORANCE IS STRENGTH
2 x 2 = 5
Who controls the past controls the future:
Who controls the present controls the past
                              -- 1984, George Orwell

On Wed, Aug 30, 2000 at 09:44:05PM +0200, Andre Herms wrote:
> 
> 
> ----------  Weitergeleitete Nachricht  ----------
> Subject: Re: SANE Backend
> Date: Wed, 30 Aug 2000 15:55:18 +0200
> From: Andre Herms ####@####.####
> 
> 
> I choosed this way because of a special feature (-:.
> The lamp needs a certain warmup time. So the server is always running and it
> will manage this:
> If you start the backend (connect to the server) the lamp will switch on. The
> server will get the time. If you start the scan, the server will wait until the
> warmup time is over and then start.
> After the scan, the lamp won't go off. this will happen after a certain timeout.
> So if you restart the scan, there will be no warmup cicle.
> Sounds cool ehr ?!
yep.
 
> I need really help. I need most help with the makefiles, as I don't know much
> about it.
> The most problem seems to be the sane backend (not the server)
> because its compiled as shared libary and i don't know how this is done. 
My problem is more understanding the SANE stuff... building a shared lib
is quite easy. But filling all those damn structs with life isn't really
something you understand after reading 10 minutes ;-))

> ** If anyone could help me with this. **
- code must be relocatable (-fPIC)
- and we want shared libs (-shared)

that's it.

> I will try to put the backend to a stable state, that other can get and test it.
 
> At the moment there is a dummy server. It acts as the primax server should do
> later. 
How do they communicate? What is to be changed in the primaxscan source?
 
> I would attach them, but I don't think it's good to post attachments to a
> newsgroup. 
nope...
 
> Q: Is there a cvs or ftp server?
you could host it at sourceforge (www.sourceforge.net)
 
-- 
Christian Ordig             | Homepage: http://thor.prohosting.com/~chrordig/ 
Germany                     |    eMail: Christian Ordig ####@####.####

Christian Biesinger wrote:
> 
> C Hennessy schrieb:
> > Andre Herms wrote:
> > > > Is anyone working on a Sane Backend for primax parallelport scanners?
> > >
> > > I do it actually.
> > >
> > > It's not a real backend for the scanner. I decided to write a meta backend,
> > > that contacts to a server. The server code is most of the primaxscan code. So
> > > it would be easy to update the backend code when primaxscan changes.
> >
> > So if I understand this better then you mean that we would have both the
> > sane
> > demon plus a primaxscan demon running at the same time ?
> 
> He wrote to me in a personal mail (though I don't know if he did it
> intentionally) that this had the following reason: When a client
> connects to the server, the server switches the scanner light on. When
> the client disconnects, it remains switched on for some time (I hope I
> got it right).
> 
> > If this is so, is it not more complicated than integrating primaxscan into sane ?
> > ANd then updates of primaxscan, so that there would be a library which the
> > primaxsane backend would simply link against. This would also allow us
> > to continue to build the primaxscan binary.
> 
> I'm not sure if I got you right here.
> I think what you want is the following:
> A library which both the sane backend and the current binary link
> against, so that both versions benefit from improvements and bug fixes.
> 
> If this is what you want, I think it's a great idea!

Yes, it is what I meant.

How feasible is this ?

CP