[sane-devel] HP-scanjet 8250 (avision-backend) ADF-not supported

[Johannes Goecke]

Hi Mike,

on Mon, Nov 29, 2010 at 09:31:44AM +0900, Mike Kelly wrote:
> Thanks to the support of a few testers, especially S. Lipinski, I have  
> been able to isolate the command sequence which gets the ADF working.   

OK I'll do my best ;-)

after unsuccessful attempts to build (debian-sid)-Packages
with the latest git-sources the avision.5.patch, I went the 
non distribution way and compiled, and installed manually.

I can't guarantee i did everything 100% correct, but lets see:

in short I did the following:


# git clone git://git.debian.org/sane/sane-backends.git

(this was at the 11.12.2010)

# cd sane-backends
# patch -p1 < avision.5.patch

there were 2 rejects 
the first the BACKEND_BUILD Define now 292
and the second were text changes in "DEB" debug outputs. 
so nothing critical.

# ./configure
# make
# make install

(and remove the original debian libsane-package)

# export SANE_DEBUG_AVISION=7
# export SANE_CONFIG_DIR=/usr/local/etc/sane.d

# scanimage -L
[sanei_debug] Setting debug level of avision to 7.
[avision] sane_init:(Version: 1.0 Build: 292)
	...
[avision] attach:
[avision] attach: opening libusb:001:005
	...
[avision] attach: Inquiry gives mfg=HP, model=C9930A, product revision=1.03.
	...
[avision] attach: Checking model: 68
[avision] attach: Scanner matched entry: 68: "HP", "C9930A", 0x3f0, 0xb01
	...
device `avision:libusb:001:005' is a Hewlett-Packard ScanJet 8200 flatbed scanner

	So I what i guess the scanner is detected correctly.

# scanimage -vvv -p -d avision:libusb:001:005  --format=tiff  > test.tiff 2> err2.log
	... 
[avision] Timeouts: write: 30000, read: 30000, status: 10000
[avision] avision_usb_status: timeout 10000, 1 retries
[avision] ==> (bulk read) going down ...
[avision] <== (bulk read) got: 1, status: 0
[avision] get_accessories_info: raw data:
[avision]   [0] 00000000b   0o   0d  0x
[avision]   [1] 00000000b   0o   0d  0x
[avision]   [2] 00000001b   1o   1d  1x
[avision]   [3] 00000000b   0o   0d  0x
[avision]   [4] 00000000b   0o   0d  0x
[avision]   [5] 00000000b   0o   0d  0x
[avision]   [6] 00000000b   0o   0d  0x
[avision]   [7] 00000000b   0o   0d  0x
[avision] get_accessories_info: [0]  ADF: 0
[avision] get_accessories_info: [1]  Light Box: 0
[avision] get_accessories_info: [2]  ADF model: 1 (Oodles)
[avision] get_accessories_info: ADF model number but no ADF? Invalid!
[avision] fix_half_missing_adf
[avision] Timeouts: write: 30000, read: 30000, status: 10000
[avision] avision_usb_status: timeout 10000, 1 retries
[avision] ==> (bulk read) going down ...
[avision] <== (bulk read) got: 1, status: 0
[avision] fix_half_missing_adf: write 1 complete.
[avision] Timeouts: write: 30000, read: 30000, status: 10000
[avision] avision_usb_status: timeout 10000, 1 retries
[avision] ==> (bulk read) going down ...
[avision] <== (bulk read) got: 1, status: 0
[avision] fix_half_missing_adf: raw data:
[avision]   [0] 00000011b   3o   3d  3x
[avision]   [1] 00000111b   7o   7d  7x
[avision]   [2] 00000101b   5o   5d  5x
[avision] fix_half_missing_adf: didn't get expected value 0x040204.
[avision] sane_open: additional probe failed: Invalid argument
[avision] sane_open: status: 4 == 3 && try: 3
scanimage: open of device avision:libusb:001:005 failed: Invalid
argument
Calling sane_exit
[avision] sane_exit:
scanimage: finished


	so at this time the scanner doesn't even want to scan.
Thats all i can do in the Moment. 


> Basically, the driver needs to send a reset command to the scanner,  
> potentially solving both the initial ADF not working issue and the ADF  
> failure after scanning a few pages.  Although I now know the right  
> sequence to send, I still have to work through a rather annoying trial  
> and error process of testing.  This process is being tracked in the  
> Alioth bug system.
>
> If you are able to download, patch, compile, and install sane-backends, 
> please try out patch #5 from the bug, and post the results of the debug 
> log to the bug system.

Ok i just have seen this, I try to upload the complete logfile tomorrow
morning CET.

>
> This link should take you to the correct bug entry:
> https://alioth.debian.org/tracker/?func=detail&atid=410366&aid=312718&group_id=30186
>
> Thanks,
>
> Mike
> (:
>

Greetins
Johannes