Age | Commit message (Collapse) | Author | Files | Lines |
|
|
|
* Restored support for reading and writing GIF images
* Fix for a possible problem in gdImageTrueColorToPalette
|
|
ok'd a while back at pkgsrcCon by agc and wiz
|
|
|
|
|
|
* several build fixes (#144632, #144780)
* documentation fixes (#144267)
* avoid use of uninitialized values in GimpColorArea widget (#144786)
* fixed Levels tool on grayscale images with alpha channel (#141930)
* fixed Save/Load functionality in channel-mixer plug-in (#145380)
* fixed brokeness in GimpContext that caused tool options and device
settings to be saved incorrectly
* fixed bug in Colorize tool
* fixed undo stack corruption that occured with some undo settings (#145379)
* fixed long-standing bug that caused the tablet pointer to change
inadvertently (#115774)
* allow to undo script 'Round Corners' (#146344)
* implemented gimp-path-get-point-at-dist PDB function (#138754)
* don't crash while loading JPEG files with corrupt EXIF data (#145212)
* fixed loading of CMYK TIFF images on big-endian hosts (#147328)
* fixed palette handling in MNG plug-in (#139947)
* fixed bug in Script-Fu console (#139200)
* fixed corruption when saving RLE-encoded BMPs on big-endian hosts (#147759)
|
|
|
|
Fixes
* Potential memory overruns in gdImageFilled Polygon
* The sign of Y-axis values returned in the bounding box by gdImageStringFT
|
|
* Make the EGifPutExtension{First,Next,Last} family of functions use WRITE
so user defined WRITE methods will output them correctly.
* Modify EGifSpew and EGifPutComment to use EGifPutExtension{First,Next,Last}
so we won't output broken GIFs when dealing with GIFs with multiple
subblocks.
* More -Wall fixes revealed while testing on Solaris and FreeBSD.
* Updated the gif_lib.html documentation to not use EGifPutExtension when
dealing with multiple subblocks. Use EGifPutExtension{First,Next,Last}
instead.
* Some Windows code from the old CVS repository now available in the windows
subdirectory. I don't have a Windows environment to test and maintain this
but maybe someone out there will find it useful. Caveat hacker.
Changes 4.1.2:
* Numerous bug fixes from people on the old libungif mailing list.
* GIF_ERROR and GIF_MESSAGE are on the deprecation list as they are also
utility helper functions rather than essential to the functioning of the
library.
* Complete deprecation list is now in the README file
* Audited the sources with gcc -Wall. Everything detectable has now been fixed.
* Ran the library code through indent.
Changes 4.1.1:
* libungif is now hosted on sourceforge:
http://sourceforge.net/projects/libungif
* Merge in many bug fixes that were sent in while I was hiking the
Appalachian Trail.
* The qprintf methods of the library are now deprecated. Do not use
GifQuietPrint or GifQprintf. These should have been pushed out into the
utility helper library instead of sitting around in the library proper at
the same time as the getarg functions were moved out. Getting rid of these
will let us get rid of our dependence on stdarg.h/varargs.h (Which a Gif
reading library has no business requiring.)
|
|
o only include it in the BUILD_DEFS if the current platform is concerned.
For now that only includes i386, but we might find amd64 users in the
future.
o test against [Yy][Ee][Ss], for consistency.
Fixes part of PR 26243.
|
|
As usual, this is a maintenance release, aimed at bug-fixing;
no new features are being added to the 4.0 branch.
The main purpose of this release is compatibility with recent GTK+
releases along with other fixes.
|
|
|
|
Fix PLIST while here.
|
|
Should fix bulk build.
|
|
against libm for fabs().
this is the same fix as is already in netpbm-latest.
|
|
Bump PKGREVISION, since now the pkgconfig file contains the right rpath.
|
|
|
|
* bug fixes
|
|
on a server without X11.
Bump PKGREVISION.
|
|
|
|
|
|
|
|
family of C++ bindings for gnome libraries
fortunately, these families can coexist
|
|
|
|
removed from the distribution in the meantime. Pointed out by
Klaus Klein in PR pkg/21351.
|
|
included program "hpcdtoppm" has a restrictive license. This fixes
PR pkg/21351 by Klaus Klein.
|
|
|
|
David Griffith in PR 25237.
|
|
|
|
|
|
|
|
(can coexist with older versions)
|
|
revision to 0.2.2. from the web page:
This release is only to fix a few bugs with sound in swf_play and a
few problems with the GdkPixbuf plugin.
|
|
|
|
changes:
- bugfixes
|
|
pixman_image_set_clip_region expects to take ownership of the
passed-in region, so we create a copy to give it.
bump PKGREVISION to 1
|
|
avoid derefering NULL pointers
bump PKGREVISION to 1
|
|
|
|
* Windows build uses __stdcall calling convention
* Even more complete freetype version checking
* Binary transparency is now handled correctly in gdImageToPalette
|
|
|
|
Patch supplied by Kouichirou Hiratsuka <hira (at) po6.nsk.ne.jp> in PR#26111.
Thanks.
|
|
jakarta-tomcat4 has other issues independant of this - will patch next
|
|
|
|
* Fix API breakage for stock_save_as vs. stock_save-as
* Fix build so that make distcheck works
* Bugs fixed: #142001, #142988
|
|
|
|
|
|
before importing the package to avoid having to commit the patches is
a good idea. Committing them anyway in a temp dir is not...
|
|
cambevao grabs images from NetBSD's or FreeBSD's /dev/bktr and ov511+ based
USB webcams and saves them as jpegs. It includes a built-in mini webserver
for static and streaming images.
|
|
Packages changes:
* Update HOMEPAGE.
* Note that patch files were ordered so that the info file does
not appear newer than its Texinfo source.
Changes since last packaged version:
2002-03-01 Alessandro Rubini <rubini@linux.it>
* library.c (Barcode_Encode): handle bitsets separately, like
in Barcode_Print (otherwise, you might loose NO_CHECKSUM).
* debian/changelog: moved to 0.98 (but should fix all debian/ stuff)
* contrib/barcode-for-delphi.tar.gz: new, contributed by Michael Geddes
* contrib/barcode-for-delphi.tar.gz.README: short info for file above
* main.c (main): allow "-g" to set size of code in a table as well
-- suggested and initiallt implemented by Joachim Reichelt.
(main): fixed missing "Page: 1 1" in postscript for tables
(main): for tables don't use Encode_and_Print, use a local
unrolled version, so the default margin can be removed and
we avoid problems with negative xoff in the output phase
(first column may be offset, as noted by Karl Magnus Kolstoe)
* doc/doc.barcode: documented change above.
* README: removed refernces to systemy.it, I'm no more there
Added link to FreeBSD ports tree and GLabels
Organized as sections with titles
* contrib/barcode-SVG.tar.gz: new contribution, by David J. Humphreys
* contrib/barcode-SVG.tar.gz.README: short description of file above
* contrib/grab-0.0.4.tar.gz: new program, by Tuukka Toivonen.
* contrib/grab-0.0.4.tar.gz.README: short description of file above
2002-03-01 Boszormenyi Zoltan <zboszor@externet.hu>
* barcode.spec: new file
2002-02-27 Alessandro Rubini <rubini@linux.it>
* bookland/bookland.py: updated to version 0.92 by Judah Milgram
2001-11-13 Alessandro Rubini <rubini@linux.it>
* README: Added a pointer to the Windows port.
2001-11-09 Cloyce D. Spradling <cloyce@headgear.org>
* code128.c (Barcode_128b_verify): prototype fix (unsigned char).
* ps.c (Barcode_ps_print): the comment for ascii codes must only
be printed if ascii is being printed (benign bug).
2001-11-06 Christoph Pross <c.pross@beck-ipc.com>
* pcl.c (Barcode_pcl_print): bugfix: bars were misplaced
2001-10-16 Nathan D. Holmes <nh412124@ursa.cosd.fedex.com>
====> Version 0.97 released to ftp.systemy.it and ftp.gnu.org
* code93.c: new file (I modified something as well - Alessandro)
* doc/doc.barcode (Supported Encodings): added docs for code93
* main.c: added code93 names
* Makefile.in (LIBOBJECTS): added code93.o
* library.c: added code93 entry points
2001-10-16 Alessandro Rubini <rubini@linux.it>
* doc/manpager: added (trivial) support for @itemize, added copyright
* sample.c (main): accept -P to create PCL output
* doc/doc.barcode (PCL Output): added the section, using text
provided by Andrea Scopece and turning it into texinfo.
Removed also a few extra newlines that made bad man pages
2001-10-16 Andrea Scopece <a.scopece@vizzavi.it>
* barcode.h: added BARCODE_OUT_PCL and BARCODE_OUT_PCL_III
* library.c (Barcode_Print): use BARCODE_OUT_PCL to select pcl_print()
* Makefile.in (LIBOBJECTS): added pcl.o
* pcl.c: new file, with PCL output support
2001-10-16 Alessandro Rubini <rubini@linux.it>
* doc/doc.barcode (Supported Encodings): added documentation for
EAN and UPC with checksum, as coded by mr. Böszörményi (below)
2001-10-16 Boszormenyi Zoltan <zboszor@externet.hu>
* ean.c:
I found that barcode-0.96 does not correctly encodes UPC-E.
This was mainly because in upc_e_to_a() the switch() branches
all ended without "break;".
Printing UPC-E did not print the leading encoding number and
the trailing UPC-A checksum. Now it does.
Accept EAN-13 and EAN-8 with and without the checksum.
So always treat 7 and 12 digit sequences as valid EAN codes
and accept 8 and 13 digit sequences as valid EAN codes when
the last digit is the same as the calculated one.
Accept UPC-A as above (11 or 12 digit sequences.)
Accepts UPC-E as:
6 digit sequence: the middle part of the code (leading '0' assumed)
7 digit sequence: the middle part and either a leading '0' or '1',
or a trailing UPC-A checksum. In that case, leading '0' assumed.
8 digit sequence: a leading '0' or '1', middle part, checksum
UPC-A, UPC-E, EAN-13, and EAN-8 may all include
an additional barcode to the right of the main barcode.
As a 7 or a 12 digit sequence may be ambiguous, please remember
that EAN is checked before UPC if you specified BARCODE_ANY.
2001-08-20 Alessandro Rubini <rubini@linux.it>
* sample.c (main): added upc-e and ean-8 samples
2001-08-11 Boszormenyi Zoltan <zboszor@externet.hu>
* ps.c (Barcode_ps_print): don't output "showpage" in EPS mode
2001-08-11 Alessandro Rubini <rubini@linux.it>
* README: Changed host names, added barcode-commit list.
2000-11-29 Alessandro Rubini <rubini@morgana.systemy.it>
* doc/doc.barcode (The Field List): 1pt == 0.352mm, not 0.0352
(thanks to Stephen Irven for reporting)
|
|
the BUILDLINK_DEPENDS >= 0.1.1 in cairo's Makefile which is the only
consumer anyway. Or, in other words: simplify
|