[OpenJDK 2D-Dev] jdk6 and jdk7: wrong class cast in sun.print.IPPPrintService (in Solaris classes)
Jennifer Godinez
jennifer.godinez at oracle.com
Wed Nov 17 19:28:13 UTC 2010
Nico,
The Bug ID is 7000940.
Jennifer
Nico R. wrote:
> Hi,
>
> Jennifer Godinez wrote:
>
>> Thank you for the very thorough explanation. I will file the bug but I
>> need the following information:
>> OS
>>
>
> Looks like this does not matter at all, as long as it runs CUPS. Anyway:
> Linux kernel (2.6.36-tuxonice) on a Gentoo system (Gentoo Base System
> release 2.0.1-r1)
>
>
>> JDKversion
>>
>
> Reproduced with dev-java/icedtea-6.1.9.1 (USE="doc examples hs19
> javascript nio2 nsplugin nss pulseaudio xrender -cacao -debug
> -elibc_FreeBSD -systemtap -zero").
>
> Looks like it applies to current JDK6 or JDK7 revisions in hg, see my
> earlier mail.
>
>
>> hardware
>>
>
> Irrelevant, this is a software problem.
>
> If it cannot be reproduced by adding printers or changing their
> availability, the method using the remote.cache file for CUPS can be
> used to simulate it. (You can also use the printers.conf configuration
> file for CUPS, it is usually placed in /etc/cups/.) See my earlier mail.
>
>
>> CUPS version
>>
>
> net-print/cups-1.4.4-r2
>
> You may want to have a look at
> <URL:http://cups.org/documentation.php/doc-1.4/ref-cupsd-conf.html#FileDevice>,
> which describes that "file" URIs are legal in CUPS.
>
>
> Some further notes on CUPS device URIs:
>
> DeviceURIs can be "ipp" URIs, but also, for example, start with
> "usb://". As far as I can tell, OpenJDK does not support "usb" URIs, so
> the code I mentioned earlier throws an IOException, and the
> ClassCastException does not happen. "file" URIs /are/ supported by
> OpenJDK, so the ClassCastException is thrown.
>
>
> The build descriptions (ebuild files) for the mentioned Gentoo packages
> are available from
> <URL:http://sources.gentoo.org/cgi-bin/viewvc.cgi/gentoo-x86/>.
>
>
> Please let us know the bug number after you’ve filed it.
>
> By the way, no need to Cc me, I’m a subscriber of this list.
>
> Thanks
>
More information about the 2d-dev
mailing list