Roger Leigh wrote:
None at all for ijsgimpprint. There was a problem a while ago (well
over a year and a half IIRC, probably over two), but that's been long
fixed. Note that ijsgimpprint links with libijs, using the Debian
libijs-dev package, and so will not be susceptible to IJS bugs at the
server side. gs does not yet use the shared libijs, and so will be
outdated. Getting a properly versioned libijs.so is on my todo list
(since Bordeaux last year!), as is getting gs to link with it.
Unfortunately, I'm still fighting a losing battle with the baroque gs
build system...
It is not a problem in libijs.so/libijs.a - but that HP made a KRGB patch
which modifies how ghostscript's ijs device works. The idea is that when
the ijs client (in gs) detects the server (the device-specific part) supports
KRGB, the ijs client is supposed "peels off" the black color and send that off
separately to get better blacks. The problem is that the detection
was broken, so after the patch, ghostscript gets very confused
about the color model and crashes when the ijs server says it only
supports b/w (for the current task, not necessarily the whole driver).
The patch is in ghostscript/src/gdevijs.c, not in libijs . It is fairly
generic, and if any of the printers supported by ijsgimpprint prints in
b/w (if only as one of its modes, I believe), ghostscript would segfault.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
