checksum in place). I don't have access to one of these printers
for testing, but the new driver does at least compile.
I nominate this for re-tagging.
Noticed by: luigi
phrase "PATENT INFRINGEMENT" to one hundred (100) disclaimers and
allows full-bleed printing on the Photosmart 100. Therefore do the
PORTVERSION Bump.
Invoke perl more specifically.
Support Microsoft .htm standard for home page.
PR: 34139
Submitted by: KATO Tsuguru <tkato@prontomail.com>
HP fixed "off-by-one error" in the HPIJS sources, that caused hpijs
to core dump at runtime, when being compiled with -O or -O2.
I confirmed, that everything works perfectly again with -O and -O2.
Bumped port revision, to flag this as an important change.
Thanks to David Suffield from HP who took our (and others) reports
seriously, fixed the bug and released the new HPIJS version.
- Important fix, therefore PORTREVISION bumped.
background info: hpijs currently may not compiled with -O or -O2.
Simply leaving out any optimization (like it was default in prior
hpijs versions) cures the problem.
Many thanks to David Suffield from HP-Vancouver.
- deploy DATADIR/DOCSDIR
- don't use internal .mk work variables, rely on supported
ones
_DISTDIR -> DISTDIR/DIST_SUBDIR
o use <machine/endian.h> for checking byte order instead of an
ARCH check
PR: 31077
Submitted by: KATO Tsuguru <tkato@prontomail.com>
- therefore bump port revision
- add a makefile template, so that it will be compiled in
- patch configure.batch and the interactive one
interactive configuration is still not enabled, because every
driver entry needs to be checked thoroughly if its still present,
if more printer models are supported, if there are new printers/drivers,
etc ... This needs time that I currently don't have.
Any taker ???
to CFLAGS
- this saves paper if you device to kill a job quickly because you
forgot something ...
- prevents in UPD driver (reported by Joerg) to make a "clean shutdown",
which is generally a nice idea, but in "real life" leads to a wastage
of paper ...
- portrevision bumped because of this important feature
Submitted by: J Wunsch <j@ida.interface-business.de>
or depend on the X11 libraries, if it is set.
This should allow "make release" to work happily with the documentation
build again.
Approved by: andreas (MAINTAINER)
- created a new distinfo file
- bumped PORTREVISION since this makes sure, that packages are build from
proper source
Submitted by: Christopher Masto <chris@netmonger.net>
this is another showstopper after the variable name change and still
doesn't show up here, since I have the sources in DISTFILES dir...
PORTREVISION still doesn't need a bump, since no package could have
been created and if, there is no new port functionality involved with
this fix. Sorry for any trouble, but its too expensive for me, to re-fetch
all files as final test.
Submitted by: "Jamin A. Brown" <jamin@gwi.net>
I forgot to change HPDJ_DRV to HPDJ_SRC in DISTFILES
PORTREVISION not bumped since building of packages wasn't possible.
Even if port/package could be build, because hpdj source was already
present, no functionality change.
Submitted by: "Justin W. Pauler" <jwpauler@jwpages.com>, "Jamin A. Brown" <jamin@gwi.net>
high quality drivers for Epsons, HPs, ...
- bumped port revision
- modified all target and added a makefile component to build
tool from stp driver (escputil) to do maintenance on epsons
install manpage as well
- variable name changes, new scheme sounds better
- use variables as often as possible in all make targets
to reduce port maintenance as much as possible
- fix two copy and paste mistakes in echo messages
- updated pkg_plist, tested port and package building
as well as removal
- modified configure.batch to build stp driver
- removed port revision, because of new version
- added PCL3 driver (hpdj successor) from Martin Lottermoser
*both* driver, HPDJ and PCL3, are now compiled in
reason: PCL3 is still flagged as BETA and author didn't get much response
Although Martin says, its definitively not ALPHA, its safer to have the
old drivers around, if things fail for some reason.
Martin confirmed, that its possible to have both in one gs binary !!!
- disable configure script temporarily, since dialog needs a driver update
but I don't want to hold back that port any longer.
- enabled as many gs driver as possible
- reorganized MASTER_SITES
- added mirrors
- add a note, that russion mirror isn't up to date, so don't add it
- optimized downloading from 6 different sources by special ordering
- reorganized Makefile completely
- put only ${GS_SOURCES} into EXTRACT_ONLY variable
unpacking other 3rd party src's is handled in post-extract target
- directory creating, patching, unpacking and install actions are
now handeled within the proper make target. This makes port maintenance
much easier, i.e.: if you enter make extract, then only extraction
of things happen, but makefiles are still not patched. Before this
cleanup, things were not so structured and not easy to maintain.
Maybe we get a configure script later, to choose between different
3rd party driver options, then this "sorting" and documenting
will support the migration process...
- silence port by hiding actions with "@"
- insert many informational echoes. This makes port maintenance and
troubleshooting easier
- put more unique comments in front of the different 3rd party modules
- document things better
- use as many environment variables as necessary, so that "numerical"
changes only have to happen on some places and not withing make targets
- use gs version number also in docu directory
- properly install runtime, info and doc files from 3rd party drivers
- put not needed stuff into "NOPORTDOCS" section
- keep old font names to avoid fetching unnecessary things
- updated patches as necessary
- added new patch from Nartin Lottermoser
- removed 2 patches that have been incorporated into gs 6.50
- added needed manual pages
- updated pkg-plist
- finally tested port and package building as well as port removal
new port passes all disciplines ;-)
- http://www.harsch.net/Ghostscript/ghostscript.html
- driver is based on Uli Wortmann's HO DJ850 driver
- supports duplex printing
Note from the author: in duplex mode black is not so intensive as
in normal mode. Maybe HP wants to prevent, that you don't see the
black of the other side of the page too much.
If somebody knows HP's algorithm, how to reduce the black, then
author would try to compensate it with some code...
my dual PIII-700 trying to figure out the sizes of the
primary and secondary caches. The sizes turned out to be
rather wrong (definetly true for the L2 cache, may be L1
is indeed 32Kb). Their algorithm seems bogus to me, but
most importantly, the obtained values are NOT USED
anywhere in the GhostScript code... One of the submitted
patches rips the "cache size" section from the genarch
entirely.
The attached modification and new patches also:
. replace the use of mktemp/fopen with mkstemp/fdopen
. modify the unix-gcc.mak so that there is no need for
GNU-make
PR: 20038
Submitted by: Mikhail Teterin <mi@aldan.algebra.com>
reduce alignment problems. It is only slightly slower, but produces much
nicer print.
- Added $Id$
Obtained from: Harold Bower <HalBower@worldnet.att.net>
