Genesys experimental status update

classic Classic list List threaded Threaded
12 messages Options
Reply | Threaded
Open this post in threaded view
|

Genesys experimental status update

Stéphane VOLTZ
Hello,

        here's the status of the genesys experimental backend:
        - full support for HP2300C : 1200, 600, 300, 150 and 75 dpi at 8 or
          16 bits, in color, grey or lineart mode
        - full support for MD5345/MD6228/MD6471 : 2400, 1200, 600, 500, 400,
          300, 250, 200, 150, 100 and 50 dpi at 8 or 16 bits, in color, grey
          or lineart mode
        - memleaks fixed
        - includes have been checked
        - files have been indented
        - exported functions are named sanei_genesys_*
        - passes 'make libcheck'
        - passes 'scanimage -T'
        - passes 'tstbackend -l 3' (this awfully long at 1200 dpi ...)
        - tested with saned
        - endianness has been taken into account (although untested due to
          lack of hardware)
        - man page and desc files have been created (scanners not supported
          yet  will be put in unsupported.desc)


        So I believe it is in a good shape for regular CVS inclusion. However
there are 2 problems left:
        - the backend sometimes hang at init, depending on how much data has
          been read in the last scan done before leaving the frontend.
        - the gl841 part has 'fprintf' to be removed, needs the 'function
          pointer' code and some warnings to be fixed.

        If noone objects, or noone hasn't have allready done it in its dev
environnement, I'll fixed the genesys_gl841.c issues. The hang is more
annoying since I don't know how to fix it yet. I guess we'll have to count
low-level packets and make sure there is an even number of them ...

Regards,
        Stef

--
sane-devel mailing list: [hidden email]
http://lists.alioth.debian.org/mailman/listinfo/sane-devel
Unsubscribe: Send mail with subject "unsubscribe your_password"
             to [hidden email]
Reply | Threaded
Open this post in threaded view
|

Re: Genesys experimental status update

Nikolas Arend
St?phane VOLTZ wrote:

>Hello,
>
>        here's the status of the genesys experimental backend:
>        - full support for HP2300C : 1200, 600, 300, 150 and 75 dpi at 8 or
>          16 bits, in color, grey or lineart mode
>        - full support for MD5345/MD6228/MD6471 : 2400, 1200, 600, 500, 400,
>          300, 250, 200, 150, 100 and 50 dpi at 8 or 16 bits, in color, grey
>          or lineart mode
>        - memleaks fixed
>        - includes have been checked
>        - files have been indented
>        - exported functions are named sanei_genesys_*
>        - passes 'make libcheck'
>        - passes 'scanimage -T'
>        - passes 'tstbackend -l 3' (this awfully long at 1200 dpi ...)
>        - tested with saned
>        - endianness has been taken into account (although untested due to
>          lack of hardware)
>        - man page and desc files have been created (scanners not supported
>          yet  will be put in unsupported.desc)
>
>
>        So I believe it is in a good shape for regular CVS inclusion. However
>there are 2 problems left:
>        - the backend sometimes hang at init, depending on how much data has
>          been read in the last scan done before leaving the frontend.
>        - the gl841 part has 'fprintf' to be removed, needs the 'function
>          pointer' code and some warnings to be fixed.
>
>        If noone objects, or noone hasn't have allready done it in its dev
>environnement, I'll fixed the genesys_gl841.c issues. The hang is more
>annoying since I don't know how to fix it yet. I guess we'll have to count
>low-level packets and make sure there is an even number of them ...
>
>Regards,
>        Stef
>
>  
>

Hi,

Following the status-update posts I'm not quite clear about the status
of the gl841 part of the backend. Does it do s.th. yet and are scanners
with
that chipset already supposed to be recognized? I checked out the
experimental cvs code and tried both genesys and genesys-new, but despite
sane-find-scanner detected my CanoScan Lide 80 and the chipset, it does
not get listed by scanimage -L.
Could one of the devs please post a short gl841-specific status report
so I know what the backend is supposed to do?
Thanks a ton for your work on the backend!!

Best,   Nick.

--
sane-devel mailing list: [hidden email]
http://lists.alioth.debian.org/mailman/listinfo/sane-devel
Unsubscribe: Send mail with subject "unsubscribe your_password"
             to [hidden email]
Reply | Threaded
Open this post in threaded view
|

Saned

Andi McLean
Hi,

I'm the maintainer of the JSane-net project.

I have an issue with the Sane netowork protocol.

To find the size / resolution and pixel depth of an image, I need to call
sane_get_parameters()
If I do this before calling sane_start() the values are estimates, which can
cause problems. If i try to call it diretly after sane_start() but before
reading the data from another port. the call is not serviced.

DO I have to call, sane_start() read the data into a buffer (File or memory)
then call sane_getparameters()?
Or is there another way?

Any help appricated
Andi

attachment0 (196 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Re: Genesys experimental status update

Stéphane VOLTZ
In reply to this post by Nikolas Arend
Le Dimanche 12 Juin 2005 12:53, Nikolas Arend a écrit :

> Hi,
>
> Following the status-update posts I'm not quite clear about the status
> of the gl841 part of the backend. Does it do s.th. yet and are scanners
> with
> that chipset already supposed to be recognized? I checked out the
> experimental cvs code and tried both genesys and genesys-new, but despite
> sane-find-scanner detected my CanoScan Lide 80 and the chipset, it does
> not get listed by scanimage -L.
> Could one of the devs please post a short gl841-specific status report
> so I know what the backend is supposed to do?
> Thanks a ton for your work on the backend!!
>
> Best,   Nick.

        Hello,

        with the current code in experimental CVS, the GL841 cannot work. It
is detected but get stuck early. I don't know for other source repositories.
At least Gerhard Jaeger and Philipp Schmid have expressed interest on this
list to get it work. Maybe they could give some hints on how they are
progressing ?

Regards,
        Stef

--
sane-devel mailing list: [hidden email]
http://lists.alioth.debian.org/mailman/listinfo/sane-devel
Unsubscribe: Send mail with subject "unsubscribe your_password"
             to [hidden email]
Reply | Threaded
Open this post in threaded view
|

Re: Re: Genesys experimental status update

Jens Luedicke
In reply to this post by Nikolas Arend
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Nikolas Arend wrote:

> Following the status-update posts I'm not quite clear about the status
> of the gl841 part of the backend. Does it do s.th. yet and are scanners
> with
> that chipset already supposed to be recognized? I checked out the
> experimental cvs code and tried both genesys and genesys-new, but despite
> sane-find-scanner detected my CanoScan Lide 80 and the chipset, it does
> not get listed by scanimage -L.
> Could one of the devs please post a short gl841-specific status report
> so I know what the backend is supposed to do?

I'm not a dev, but my CanoScan LiDE 50 is listed by sane-find-scanners
and detected by scanimage -L, but scanning doesn't work at all.

> Thanks a ton for your work on the backend!!
Indeed! Thanks a lot! I can't wait for the day I can ditch my windows ...

- --
Jens Luedicke
web: http://perldude.de

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.1 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iD8DBQFCrELCiCvkc4H0gTERAqnmAJ0U2c0b/IKPlAj2AsgWABzBY2F8swCfej4J
RmjxHXQy5WcUNZSYcCgPz2E=
=kEOp
-----END PGP SIGNATURE-----

--
sane-devel mailing list: [hidden email]
http://lists.alioth.debian.org/mailman/listinfo/sane-devel
Unsubscribe: Send mail with subject "unsubscribe your_password"
             to [hidden email]
Reply | Threaded
Open this post in threaded view
|

Re: Re: Genesys experimental status update

Philipp Schmid
In reply to this post by Stéphane VOLTZ

St?phane VOLTZ wrote:

>Le Dimanche 12 Juin 2005 12:53, Nikolas Arend a ?crit :
>
>  
>
>>Hi,
>>
>>Following the status-update posts I'm not quite clear about the status
>>of the gl841 part of the backend. Does it do s.th. yet and are scanners
>>with
>>that chipset already supposed to be recognized? I checked out the
>>experimental cvs code and tried both genesys and genesys-new, but despite
>>sane-find-scanner detected my CanoScan Lide 80 and the chipset, it does
>>not get listed by scanimage -L.
>>Could one of the devs please post a short gl841-specific status report
>>so I know what the backend is supposed to do?
>>Thanks a ton for your work on the backend!!
>>
>>Best,   Nick.
>>    
>>
>
> Hello,
>
> with the current code in experimental CVS, the GL841 cannot work. It
>is detected but get stuck early. I don't know for other source repositories.
>At least Gerhard Jaeger and Philipp Schmid have expressed interest on this
>list to get it work. Maybe they could give some hints on how they are
>progressing ?
>
>Regards,
> Stef
>
>  
>
Hello,

currently I don't devel at genesys cause I've to spend a lot of time for the class
tests in school.

I'm aiming to do some developpment on the GL841 wenn I've got more time.

But everybody who's also interested in devel for GL841 is welcomed.

Bye
        Philipp


--
sane-devel mailing list: [hidden email]
http://lists.alioth.debian.org/mailman/listinfo/sane-devel
Unsubscribe: Send mail with subject "unsubscribe your_password"
             to [hidden email]
Reply | Threaded
Open this post in threaded view
|

Re: Saned

Giuseppe Sacco
In reply to this post by Andi McLean
Il giorno dom, 12-06-2005 alle 12:26 +0100, Andi McLean ha scritto:

> Hi,
>
> I'm the maintainer of the JSane-net project.
>
> I have an issue with the Sane netowork protocol.
>
> To find the size / resolution and pixel depth of an image, I need to call
> sane_get_parameters()
> If I do this before calling sane_start() the values are estimates, which can
> cause problems. If i try to call it diretly after sane_start() but before
> reading the data from another port. the call is not serviced.
[...]

Hi Andi,
I started asking about this problem in this list a week ago. You may
have a look at the complete thread at
http://lists.alioth.debian.org/pipermail/sane-devel/2005-June/013745.html
but, sadly, I have still to find the solution.

Bye,
Giuseppe


--
sane-devel mailing list: [hidden email]
http://lists.alioth.debian.org/mailman/listinfo/sane-devel
Unsubscribe: Send mail with subject "unsubscribe your_password"
             to [hidden email]
Reply | Threaded
Open this post in threaded view
|

Re: Saned

Henning Meier-Geinitz
In reply to this post by Andi McLean
Hi,

On Sun, Jun 12, 2005 at 12:26:22PM +0100, Andi McLean wrote:
> I have an issue with the Sane netowork protocol.
>
> To find the size / resolution and pixel depth of an image, I need to call
> sane_get_parameters()
> If I do this before calling sane_start() the values are estimates, which can
> cause problems. If i try to call it diretly after sane_start() but before
> reading the data from another port. the call is not serviced.

What do you mean by "the call is not serviced"?

> DO I have to call, sane_start() read the data into a buffer (File or memory)
> then call sane_getparameters()?
> Or is there another way?

See SANE standard for how to do a scan:

sane_start ()
sane_get_parameters()
sane_read()

Bye,
  Henning


--
sane-devel mailing list: [hidden email]
http://lists.alioth.debian.org/mailman/listinfo/sane-devel
Unsubscribe: Send mail with subject "unsubscribe your_password"
             to [hidden email]
Reply | Threaded
Open this post in threaded view
|

Re: Saned

Henning Meier-Geinitz
In reply to this post by Giuseppe Sacco
Hi,

On Tue, Jun 14, 2005 at 09:56:15AM +0200, Giuseppe Sacco wrote:
> I started asking about this problem in this list a week ago. You may
> have a look at the complete thread at
> http://lists.alioth.debian.org/pipermail/sane-devel/2005-June/013745.html
> but, sadly, I have still to find the solution.

???

I think I have given an answer. Maybe I miss the point but the way I
described is just the normal way to scan an image. That works also
with saned (and has been working for years).

Example with saned and the test backend on localhost:

SANE_DEBUG_NET=255 scanimage -v -v -d net:localhost:test -T
[sanei_debug] Setting debug level of net to 255.
[net] sane_init: authorize = 0x80492c0, version_code = 0xbfffe7bc
[net] sane_init: SANE net backend version 1.0.13 (AF-indep+IPv6) from sane-backends 1.0.15-cvs
[...]
[net] sane_start
[net] sane_start: remote start
[net] sane_start: server has little endian byte order
[net] sane_start: remote start finished, data at port 32776
[net] sane_start: done (Success)
[net] sane_get_parameters
[net] sane_get_parameters: remote get parameters
[net] sane_get_parameters: returned status Success
scanimage: scanning image of size 157x196 pixels at 8 bits/pixel
scanimage: acquiring gray frame, 8 bits/sample
scanimage: reading one scanline, 157 bytes...
[net] sane_read: handle=0x80553e8, data=0x805af88, max_length=157, length=0xbfffe7cc
[net] sane_read: reading paket length
[net] sane_read: read 4 bytes, 0 from 4 total
[net] sane_read: next record length=8188 bytes
[net] sane_read: 157 bytes read, 8031 remaining
[...]

Bye,
  Henning

--
sane-devel mailing list: [hidden email]
http://lists.alioth.debian.org/mailman/listinfo/sane-devel
Unsubscribe: Send mail with subject "unsubscribe your_password"
             to [hidden email]
Reply | Threaded
Open this post in threaded view
|

Re: Saned

Andi McLean
In reply to this post by Henning Meier-Geinitz
On Wednesday 15 Jun 2005 17:49, Henning Meier-Geinitz wrote:

> Hi,
>
> On Sun, Jun 12, 2005 at 12:26:22PM +0100, Andi McLean wrote:
> > I have an issue with the Sane netowork protocol.
> >
> > To find the size / resolution and pixel depth of an image, I need to call
> > sane_get_parameters()
> > If I do this before calling sane_start() the values are estimates, which
> > can cause problems. If i try to call it diretly after sane_start() but
> > before reading the data from another port. the call is not serviced.
>
> What do you mean by "the call is not serviced"?

>
> > DO I have to call, sane_start() read the data into a buffer (File or
> > memory) then call sane_getparameters()?
> > Or is there another way?
>
> See SANE standard for how to do a scan:
>
> sane_start ()
> sane_get_parameters()
> sane_read()
There is no sane_read() in the network protocol. When you send sane_start()
a port number is returned. You read data from this port, (The data is in the
same format as sane_read) . The original port you sent the request is blocked
until all image data has been read.

>
> Bye,
>   Henning

attachment0 (196 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Saned

Henning Meier-Geinitz
Hi,

On Wed, Jun 15, 2005 at 06:36:57PM +0100, Andi McLean wrote:
> There is no sane_read() in the network protocol. When you send sane_start()
> a port number is returned. You read data from this port, (The data is in the
> same format as sane_read) . The original port you sent the request is blocked
> until all image data has been read.

Maybe I still miss something but doesn't the protocol I sent in the
other mail show that there is no blocking? The whole idea of using two
ports is that you can still send RPC during the read. See "do_scan"
function in saned.c. I haven't checked the details, maybe there is a
timing issue that involves blocking in your case?

In this case the test with "scanimage -d net:localhost:test -t"
shouldn't work as scanimage does sane_get_parameters before sane_read.

Bye,
  Henning

--
sane-devel mailing list: [hidden email]
http://lists.alioth.debian.org/mailman/listinfo/sane-devel
Unsubscribe: Send mail with subject "unsubscribe your_password"
             to [hidden email]
Reply | Threaded
Open this post in threaded view
|

Re: Saned

Andi McLean
Thanks for your time and effort on this.
Ok I've worked it out, You need to open the connection to the scanner host.
After this has been done you can then perform other requests.

Sorts of makes sense. Maybe you'll want to include this tip in a faq somewhere

Andi

attachment0 (196 bytes) Download Attachment