Boca filter and driver working on Suze Linux, some problems andquestions on Mac OS X
Johan Henselmans
johan at netsense.nl
Fri Dec 9 14:32:39 PST 2005
Michael Sweet wrote:
> Johan Henselmans wrote:
>> I have built a boca driver and filter for the boca labelprinters.
>>
>> After some hairpulling I finally found out how the cups raster was
>> built, so I could do some magic bit swapping to get the bits on all
>> the right places.
>>
>> In the testing, most of it worked from a Linux Suze distribution.
>>
>> One problem there: The printer is a USB printer and some artefacts
>> seem to remain on the top of the page. Might that be an unclean
>> buffer? How should I proceed to get these artefacts removed?
>
> How are you reading the raster data?
From pstoraster. I get, because of the driver definitions a raw bw
bitmap. I read the 8 lines at a time. I had some discussions with the
boca guys. They thought that perhaps the artefacts were caused by
wraparounds to the next 'page', and were remnants of a buffer that came
out dirty after filling the output line. The height of the page is not a
multitude of 8, see.
>
>> I also got reckless, what the heck, let's do it on Mac OS X.
>>
>> So I compiled cups-1.1.23, fixed one error in the filter compilation
>> with 'lcc_dynamic' by replacing this with 'gcc'.
>
> 10.4's GCC 4 is missing the supc++ library as well as cc_dynamic.
> The fix is to use g++ and drop those libraries, as Apple will not
> be providing them anymore ("working as advertised" is the resolution
> for the Radar bug I opened on this...)
>
> This is reflected in the CUPS 1.2 sources as well as Apple's CUPS
> 1.1.x Darwin sources.
>
>> ...
>> /usr/lib64/cups/filter/texttops 1 johan test 1 options testfile.txt |
>> /usr/lib64/cups/filter/pstoraster | ./rastertoboca 1 johan test 1
>> options > /dev/usb/lp0
>>
>> How am I supposed to do that on Mac Os X?
>
> You have to use the backend - there are no character devices for USB
> printers on OSX.
>
Aaah, so the only way I can test the stuff is by sending it to the
backend, and hope that everything turns out well?
So where can I find some clue where things go wrong? The odd thing is,
the filter works fine using the webinterface it seems: the test page
gets partially printed. But using the 'normal' route seems to fail with
the described error.
Johan
More information about the cups
mailing list