LCDproc development and user support list

Text archives Help


[Lcdproc] ppdev support


Chronological Thread 
  • From: peter AT adpm.de (Peter Marschall)
  • Subject: [Lcdproc] ppdev support
  • Date: Mon Mar 27 15:11:01 2006

Hi Andr=C3=A9s,

On Monday, 27. March 2006 01:50, Andr=C3=A9s Trapanotto wrote:
> A year ago I've proposed some way to give lcdproc ppdev support, but I
> just sent the source files to the list and there are... in the archive :'(
>
> At now I want to try again to commit my work... Could you help me?
>
> I've worked in two files:
> lcdproc/server/drivers/port.h
> lcdproc/server/drivers/port.c

>
> for the first I'm attaching the diff to the current cvs, and I'm
> sending the port.c "as is" because it is new.

Thanks for these files.
Unfortunately they come in at a very unfortunate point of time.
I am currently preparing the next version of LCDproc and do not want to hav=
e=20
larger code changes.

Would you mind to have your patches in the patches/ directory in CVS and=20
applied after the release of the next version ?

> There is something missing to get it work: to fix the automated
> configuration. I've done manually modifying the
> lcdproc/server/drivers/Makefile as you can see (in the attachment) but I
> don't know how to do it in the right way (here is the point were I'm
> claiming for help :) )

Hmmm, as Makefiles are generated at configuration time and there is more th=
ant=20
one. What exactly did you change in which Makefile ?

Then lets see how we can get this automated.

Another question: is ppdev specific to Linux or do other OSses support it a=
s=20
well ?
In the first case we need to shield the other OSses.

> There are some ToDo points to, like key support, several parallel
> ports
> and parallel port configuration (is hardcoded to /dev/parport0 right
> now, but could be changed easily to get the right parameter from the
> configuration file)...

Please don't hesitate to continue working on these issues.

> And then, the explanation of how it works:
>
> Up to now, LCDd access right to the parallel port thru io functions
> (port_in, port_out, etc...). This way needs root permission and the
> known of the parallel port io configuration... And this functions are
> included in-line from port.h to the driver.
>
> I've written another set some functions (pp_init, pp_bye,
> pp_write_data, etc.) in port.c that are not in-line compiled, has a new
> struct pp_driver as argument allowing to pass them more parameters (now
> we need to tell which port, register and data we want to use, read or
> write).
>
> And there is something else that is a wrapper to get old functions
> (port_in, port_out, etc) supported for such drivers that are not
> modified to use new parallel port functions.
>
> So, if you have an LCD attached to your parallel port (like my
> hd44780)
> without keys attached to it, you can apply this patch and just forget
> the io access to the parallel port :)
>
> Well, I hope these could be useful to the project. The patch itself
> has
> some in-line comments but ask me for any doubt.

Regards
Peter

=2D-=20
Peter Marschall
eMail:
peter AT adpm.de




Archive powered by MHonArc 2.6.18.

Top of page