[sane-devel] hp7400c won't work with latest sane ?

[René Rebe]

Tymoteusz wrote:
> René Rebe pisze:
>> Hi,
>>
>> Tymoteusz wrote:
>>> René Rebe pisze:
>>>> Tymoteusz wrote:
>>>>> Rene Rebe pisze:
>>>>>> Hi,
>>>>>>
>>>>>> On 04.04.2008, at 01:05, Tymoteusz wrote:
>>>>>>
>>>>>>> m. allan noah pisze:
>>>>>>>> On 4/3/08, Tymoteusz <tymoteusz.drozd at gmail.com> wrote:
>>>>>>>>
>>>>>>>>>> Yeah, just that X segfaulted, does not tell much.
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> However, you could test with scanimage in a xterm (or variant) 
>>>>>>>>>> or even
>>>>>>>>>> the console text terminal to check your basic image processing
>>>>>>>>>> functionality ...
>>>>>>>>>>
>>>>>>>>>> Yours,
>>>>>>>>>>
>>>>>>>>>> -- 
>>>>>>>>>>  René Rebe - ExactCODE GmbH - Europe, Germany, Berlin
>>>>>>>>>>  http://exactcode.de | http://t2-project.org | 
>>>>>>>>>> http://rene.rebe.name
>>>>>>>>>>
>>>>>>>>> Well scanimage doesn't  crashing xorg. In this moment i've 
>>>>>>>>> compiled old
>>>>>>>>> sane-backends 1.0.18
>>>>>>>>> and my scaner had return to previous state, it's working for 
>>>>>>>>> now, but i
>>>>>>>>> think that this situation is  not so fine. Is  there  anyway to 
>>>>>>>>> solve
>>>>>>>>> this problem ?
>>>>>>>>>
>>>>>>>> yes- get a debug log like i asked for previously....
>>>>>>>>
>>>>>>>> allan
>>>>>>>>
>>>>>>>> when i catch some more time i will try to investigate a
>>>>>>>>
>>>>>>>>> bit more this issue.
>>>>>>>>>
>>>>>>>>> Regards
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> -- 
>>>>>>>>> sane-devel mailing list: sane-devel at lists.alioth.debian.org
>>>>>>>>> http://lists.alioth.debian.org/mailman/listinfo/sane-devel
>>>>>>>>> Unsubscribe: Send mail with subject "unsubscribe your_password"
>>>>>>>>>             to sane-devel-request at lists.alioth.debian.org
>>>>>>>>>
>>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>> oups sorry i had forgotten to add ur email adres to my previous post
>>>>>>> log is here http://atelier1.website.pl/N-A
>>>>>> It is just inf. waiting for light warmup. I just yesterday found a 
>>>>>> similar
>>>>>> problem with a newer hardware and will provide a workaround when 
>>>>>> done.
>>>>>>
>>>>>> Can you test if you can scan successfully if you let the scanner warm
>>>>>> it's light before you scan with the Avision SANE backend?
>>>>>>
>>>>>> Yours,
>>>>>>
>>>>> Well i have left scanner for warmup but this didn't give any 
>>>>> difference.
>>>>> Regards.
>>>>
>>>> Where is the log to take a look?
>>>>
>>>> Btw. I just noticed in the source I have on the screen the 7400
>>>> is marked as AV_LIGHT_CHECK_BOGUS, which essentially ignores the
>>>> light warmup state in the code. Can you review if your backend
>>>> source you installed from contains this?
>>>>
>>>> Yours,
>>>>
>>> Hmm
>>>
>>> AV_LIGHT_CHECK_BOGUS appears in:
>>>
>>> /backend/
>>>    767:avision.c
>>>    2477:avision.c
>>>    107:avision.h
>>>
>>>
>>> log is here http://atelier1.website.pl/N-A
>>> 7400_old.log is previous log
>>> 7400.log is new one.
>>>
>>> also i'm sure that something had triggered lamp warmup because after 
>>> 40min scanner was very hot (much warmer then after 30 scan's 
>>> committed with old version of avison).
>>
>> So, still waiting for the light. Can you add a return SANE_STATUS_GOODM;
>> in the beginning of your wait_4_light() function in avision.c?
>>
>> If I find some time I'll send you some workaround for testing, as
>> maybe the silicon would report a correct status if polled less often.
>>
>> If you have free time you could also test upfront if changing the
>> sleep(1) that should be in your function to sleep(10) or sleep(20)
>> fixes light detection for your scanner.
>>
>> Yours,
>>
> hmm i need some more info becuse i had
> 
> avision.c: In function 'wait_4_light':
> avision.c:2439: error: 'SANE_STATUS_GOODM' undeclared (first use in this 
> function)
> 
> 
> i had puted this here:
> 
> wait_4_light (Avision_Scanner* s)
> {
>  return SANE_STATUS_GOODM; <<
>  Avision_Device* dev = s->hw;
> 
>  /* read stuff */
>  struct command_read rcmd;
> 
> 
> Regards

Sorry, the trailing M was a typo from my side. SANE_STATUS_GOOD
(taking a 2nd look) it should be.

Depending on your compiler you might have to move the return after
the variable declarations ...

Yours,

-- 
   René Rebe - ExactCODE GmbH - Europe, Germany, Berlin
   http://exactcode.de | http://t2-project.org | http://rene.rebe.name