[sane-devel] Canon LiDE 25 and Oct 2nd snapshot

[Henning Meier-Geinitz]

Hi,

On Sun, Oct 02, 2005 at 10:58:52PM +1000, Luke Kendall wrote:
> I took the snapshots from http://www.sane-project.org/snapshots/
> which were:
>     sane-backends-2005-10-02.tar.gz
>     sane-frontends-2005-10-02.tar.gz
> 
> I also grabbed libusb-0.1.10a.tar.gz from sourceforge and isnatlled
> that.  Though it installs as 0.1, which seems a strange numbering
> scheme:

That's normal for libusb.

> I noticed an oddity when compiling the frontend: 
> 
> .... /usr/local/include -I/usr/include/gtk-1.2 -I/usr/include/glib-1.2 -I/usr/lib/glib/include -I/usr/X11R6/include gtkglue.c
> cc1: warning: changing search order for system directory "/usr/local/include"
> cc1: warning:   as it has already been specified as a non-system directory
> make[1]: *** No rule to make target `../sanei/libsanei.a', needed by `xscanimage'.  Stop.
> make[1]: Leaving directory `/home/luke/linux/scanner/sane-frontends/src'
> make: *** [install-recursive] Error 1
> 
> Perhaps that was because I just did a "make install", and should have
> made all first?  I went into sanei and just did a make in there and
> continued on.

Always do "make" and then "make install". "make install" does not
implicitely call "make".

> After I'd made install in sane-backend then sane-frontend (and moved
> aside the older versions and re-ren ldconfig), I did a "modprobe
> scanner"

Why? You just installed libusb so you NOT want the scanner module!

> then "mount -t usbdevfs none /proc/bus/usb".  Then sane-find-scanner
> reported the scanner:

> ....
> found USB scanner (vendor=0x04a9 [Canon], product=0x2220 [CanoScan], chip=LM9832/3) at libusb:001:003
> ....
> 
> But scanimage -L didn't find it (as root).  I'll confess this is on a
> very old system (RH 7.2, running kernel 2.4.28).

Maybe because it's blocked by the scanner module. Try without (rmmod
scanner).

If that doesn't help, send a log file (export SANE_DEBUG_PLUSTEK=255;
scanimage -L).

> I then had a look for Stephan February's changes (in his post of Sept
> 29), but it looks like they didn't make it into the Oct2 2nd snapshots?

Yes, they are nor yet included. But these patches are independent from
your problem of not detecting the scanner.

> So I applied the changes manually to backend/plustek-usbdevs.c.

> I was surprised that a "make" didn't compile anything.  I then looked
> through the output from make and saw that plustek-usbdevs.c had never
> been compiled in the first place.

As Gerhard already mentioned:
"NOTE: before compiling, always do a touch backend/plustek.c to make sure
      the backend gets compiled..."

> So I think I'm a bit out of my depth now, since I don't understand how 
> plustek-usbdevs.c gets compiled (it only appears in the DISTFILES list).

It's #included in the backend file.

> I did notice a few mildly interesting warnings though:

The development snapshots have quite a lot of warnings enabled. They
are unrelated.

> Any ideas what I've done wrong?

Check the logfile as mentioned above.

Bye,
  Henning