- X11 fullscreen toggle fixes with various windowmanagers, multifile playback
- DXR2 support (linux only)
- Output into animated GIF
- "cropdetect" video filter to autodetect needed size to crop
- libavcodec marked as default encoder for MEncoder
- manpage rewrite
- lot of other fixes
- gmplayer uses two new skins: CornerMP-aqua and hwswskin
Also patch-ab is no longer needed and patch-aa has only one hunk remaining.
* Merged the endianness conversion fixes from the GDK version into the
Xlib version; oops (Federico).
* Merged fixes from GTK+ 2.0
* Minor documentation improvements (Federico).
* Fixed endianness conversion in the 16-bit gdk-pixbuf-drawable
functions (Federico).
* Minor fixes for the IBM/AIX compiler (Christian Schaller).
* The image loaders are now linked against the pixbuf and GTK+
libraries so that the Python bindings work (Johan Dahlin).
* Backported the BMP loader from GTK+ 1.3 (Federico).
* Added support for BI_BITFIELDS coding to the BMP loader [Ximian bug
#12125] (Federico).
* Fixed stupid bug in the ICO loader. ICO pixbufs should always have
an alpha channel [Ximian bug #11224]. (Federico)
* Slight tweaks to the documentation Makefile. (Federico)
* Added support for 16-bpp BMPs and ICOs (Federico).
* Added support for 32-bpp ICOs (Federico).
* Use the correct visual and colormap for the pixbuf-demo widgets
* Install the headers in a versioned directory so that they don't
collide with the GNOME 2 platform (Havoc).
- 3DNow optimization of AC3 4ch downmix fixed, optims re-enabled
- DGA error/fail after 1 second of playing fixed
- big A-V desync for some AVI files fixed
- image distortions when seeking in some MPEG files fixed
- libmpdvdkit + gcc 3.x issue solved (our CFLAGS triggered a gcc3 bug...)
- GUI aspect ratio and fullscreen switch fixes, cleanups
- low FPS .wmv playback fixed
- buggy v4l audio grabbing disabled, we cannot fix, but patches are
welcomed!
- various mjpeg improvements in libavcodec
- Grayscale native support (PGM format), plus four binarization methods.
- Internal preprocessor including deskewing, balancing, thresholding and
interpolation (by Giulio Lunati).
- Various new features: border path computing, barcode search,
detection of extremities, PAGE only mode, the flea, the spyhole,
instant threshold, per-depth optimized X code, etc.
- Documentation updated (but not finished). Glossary added.
- Many bugfixes and interface enhancements.
Gri is a language for scientific graphics applications. By 'language'
I mean that it is a command-driven application, as opposed to a
click/point application. It is analogous to latex or tex, and shares
the property that extensive power is the reward for tolerating a
modest learning curve. Gri output is in industry-standard PostScript,
suitable for incorporation in documents prepared by various text
processors.
Gri can make x-y graphs, contour-graphs, and image graphs. In
addition to high-level capabilities, it has enough low-level
capabilities to allow users to achieve a high degree of customization.
Precise control is extended to all aspects of drawing, including
line-widths, colors, and fonts. Text includes a subset of the tex
language, so that it is easy to incorporate Greek letters and
mathematical symbols in labels.
NetBSD-1.5.*, so pull in the build-time dependency on binutils and gcc
conditionally on the version reported by ${AS}. For that case, set
${GCC_EXEC_PREFIX} in the build environment, and pass in an "--as="
argument to "configure", to make absolutely sure the new "gas" gets used.
available). Not tested with the recent gcc and binutils changes.
Only one local patchfile is now needed, which fix a -Wl,-R problem
and reenables the auto detect support for libmpdvdkit on NetBSD.
Where libmpdvdkit is now disabled, due to the css code in that source.
The MPlayer code use libdvdread so it still can read DVDs.
The css code is also the reason why we have to use NO_SRC_ON_FTP.
List of changes, without being complete.
Changes in 0.90pre2:
- hwac3 fixes (try AFMT_AC3 first), bigendian fixes for PCM audio
- 2-pass encoding with libavcodec (-lavcopts vpass=x), B frames support
- faster .ogg seeking (use -forceidx for slow but accurate one)
- audio-only with gui fixed
- -rootwin fixed for x11,xv,xmga,xvidix
- aspect ratio fix for ffmpeg12
- mpeg-es detection fixed
- runtime cpudetect disabled by default
- support for many new win32 codecs, including mss1, tm20, ucod, truespeech
- divx audio dll sig11 with glibc 2.1.(2|3) fixed
- includes libmpdvdkit - patched kit of dvd libraries (enabled for linux+freebsd)
- manpages updated
- 'f' (fullscreen switch) fixed for some WMs, still has problems
- various freebsd and openbsd portability patches
- options -xy,-zoom,-flip usable with mencoder too, -xy keeps aspect
- -vop lavc, using new fast best libavcodec for realtime mpeg1 transcoding
- fixed gtk gui crash with --disable-mencoder
- no more black vobsub subtitles :)
Changes in 0.90pre1:
- 100% GPL - yeah, so what?
- runtime CPU detection code - any x86 MPlayer binary can run on "any"
other x86 CPU
- video filter layer - with numerous plugins (crop, expand, etc, see
the documentation)
- ability to view or encode DVD/vobsub and/or text subtitles into the
output AVI
- prelimenary dvdnav support (experimental)
- Direct Rendering - can't be used with libavcodec yet, sorry
- ability to display subtitles under the image, on black bands
- FFmpeg (included) now supports decoding MPEG4v2 (MP42) and DivX 5.0
files too
- greatly enhanced MPEG4 encoder - in libavcodec. It's not always as good
as DivX4, but is twice as fast - so it's very useful on realtime encoding.
And generally also.
- audio-only (wav/mp3/wma/ogg) file playback (console only)
- playing/encoding audio from separate wav/mp3 file
- support for more fileformats, including the new and very popular .ogg
- multiple files playback from console, including various playlist formats
- and of course, faster than ever... :)
Changes in version 1.8.5:
- various portability fixes
- various SVG fixes and optimizations
Changes in version 1.8.4:
- SVG renderer:
- make graph|node|edge ids unique, particularly for multiedges
- put graph|node|edge names in <title>...</title>
- use some property inheritance to reduce size of output
- fix compile errors when no zlib
- updated DTD reference
- GD renderer:
- Minimal Type1 font support
- look in {${X11BASE},${X11PREFIX}}/lib/X11/fonts/Type1
- look for .pfa or .pfb font files based on fontname
- run gdgen.c through dos2unix - problems with gcc on SuSE
- fix Mac-OSX build problems:
- improve strto[u]ll configure tests
- add -fno-common for extern problem
- function renamed to avoid conflicts (vis -> visibility)
- add configure tests for search.h, malloc.h, getopt.h, errno.h
- improve configure tests for FILE struct features
- add configure tests for lrand48
- add new demo graphs:
- graphs/undirected/Heawood.dot
- graphs/undirected/Petersen.dot
- neato:
- fix for -x implementation in neato (Bug 77)
- fix spline problem (Bug 87)
- fix some divide-by-zero problems
- twopi:
- fix Bug 117
- update man pages for disconnected graphs capability
- added arrowhead or arrowtail = tee
- add dotneato/pack code to twopi
- add contrib/prune to gnu build and install
Changes in version 1.8.3:
- fixed parse error for lines starting with '#' in .dot files
- fixed a recently introduced bug that caused failure of:
digraph G { {rank = same; A -> B; B -> A } }
- updated DOCTYPE header in SVG outputs
- added dotneato/common/xbuf.[ch] for dynamic string handling
to avoid sprintf buffer overruns.
- twopigen - handle special case of graphs with < 3 nodes.
- neato - handle point shapes
- added fontcolor support to svg
- Fixed bug 109
- Removed duplicate definitions for str[n]casecmp
- Added missing declarations needed for Windows
- Cleaned up warning messages from set but unused variables
- Removed use of DOS preprocessor variable; uniformly replaced by MSWIN32
USE_BUILDLINK_ONLY so that the buildlink.mk files included by
Makefile.common will trigger the correct Makefile logic. Also strongly
buildlinkify while I'm here.
* load.c: Fixed reported bug. If Imlib thought it needed a fallback,
bad things will happen.
* utils.c: Removed the open/close_helper functions. A little less
bloat now removed.
Here's a list of added features/bug fixes.
1) Tgif's home has moved from University of Maryland to University of
Southern California.
2) If there's no domain resources specified in the X resource file,
tgif will use the library path specified when tgif is compiled as the
default. This is the directory where the equation symbols are installed.
Thanks to A. Sengupta <osegu@iitk.ac.in> for pointing out the problem.
3) For Red Hat 7.2, /usr/bin/xpmtoppm cannot handle input file whose
format is XPM version 1. Suport for a new compiler option,
-D_XPM3TOPPM, is added so that tgif can generate XPM format version 3 by
default. This is has the same effect as setting Tgif.XPmOutputVersion to 3
and Tgif.UseXPmVersion1ForImageMap to false.
4) Add more information in the tooltip when
Tgif.MeasureTooltipVerbose is set to true.
5) Add a new LaTeX equation symbol file "eq4xpm.sym" in the release.
Instead of generating and embedding an EPS file, this new object
will generate and embed an XPM file with additional controls. Please
open this file with tgif for more details and check out
http://bourbon.usc.edu:8001/tgif/faq/eq4xpm.html.
Thanks to Kai Sattler <kaisat@post.tele.dk> for the contribution.
6) Fix a bug in tgwb when a text object is being edited.
7) Obsolete the Tgif.LandscapePdfSetPageDevice X default (because the
name was misleading) and add support for a new X default,
Tgif.PdfSetPageDevice. As mentioned in the release notes for
tgif-4.1.39, the use of this should
not be necessary in the future (and is considered a bug in
should not be necessary in the future (and is considered a bug in
ps2pdf).
The old commandline option, "-nolandpdfspd" is also obsolete and
is replaced by "-pdfspd=yes" and "-pdfspd=no".
8) Add a new X default, Tgif.NoModeWindow, so that tgif can start
without a mode window.
9) Add a new X default, Tgif.MakeUnsavableInSlideShow, so that the
current file will be made unsavable when slideshow mode is entered. (If
the current file contains auto page numbering objects, the file will
be made
unsavable regardless of the setting of this X default.)
10) Update the Reliable IP-multicast Library (distributed in the
rmcast subdirectory of this distribution). The new version allows
point-to-point connection between 2 whiteboards. Please see the updated
README.tgwb file. Thanks to the team members of the research group lead by
Professor Edmundo de Souze e Silva <edmundo@land.ufrj.br> at the
Federal University of Rio de Janeiro.
in pkg/16016.
Relevant changes since 1.2.0:
* Fixed transparency in the GIF loading code
* Added support for the IFF (LBM) image format
* Fixed transparency in 8-bit PNG files
* Added support for loading XPM image data directly
by me. From the DESCR:
Clara OCR is a free (GPL) Optical Character Recognition (OCR) program
for systems that support the C library and the X windows system (e.g.
most flavours of Unix). The development platform of Clara OCR is
32-bit Intel running GNU/Linux.
Clara OCR is intended for large scale digitalization projects. It
features a powerful GUI and a web interface for cooperative
digitalization of books. Clara OCR development started in 1999 and
is approaching production quality.
Features:
Converts pbm/pgm image files to text (ISO-8859)
Can process scans in batch for large documents
Can run from the command-line
Is relatively easy to train
Non-features:
Is not "omnifont"; you must train it for each document
Does not scan the images
Does not support unicode
Cannot read handwriting
- Build and install shared libraries "libav" and "libavcodec".
- Fix "ffserver" to search its configuration file in "${PKG_SYSCONFDIR}".
- Install "ffserver.conf" configuration to example and not into
documentation directory.
Bump package revision to 1 after these changes.
No proper changes information available since version 1.8.1.
However browsing Graphviz's CVS logs and diff'ing with previous version
one can found:
- bug fixes
- code cleanup
- buffer overflow in dot/neato output generation fixed
- memory leaks fixed (??)
XmBDFEditor is a Motif-based BDF font editor with the following features:
o Multiple fonts can be loaded from the command line.
o Multiple fonts can be open at the same time.
o Cutting and pasting glyphs between fonts.
o Multiple glyph bitmap editors can be open at the same time.
o Cutting and pasting between glyph bitmap editors.
o Export of XBM files from glyph bitmap editors.
o Automatic correction of certain metrics when a font is loaded.
o Generation of XLFD font names for fonts without XLFD names.
o Update an XLFD font name from the font properties.
o Update the font properties from an XLFD font name.
o Font property editor.
o Font comment editor.
o Supports unencoded glyphs (ENCODING of -1).
o Display of glyph encodings in octal, decimal, or hex.
o Builtin on-line help.
o Imports PK/GF fonts.
o Imports HBF (Han Bitmap Font) fonts.
o Imports Linux console fonts (PSF, CP, and FNT).
o Imports Sun console fonts (vfont format).
o Imports fonts from the X server.
o Imports Windows FON/FNT fonts.
o Imports TrueType fonts and collections.
o Exports PSF fonts.
o Exports HEX fonts.
o Edits two and four bits per pixel gray scale fonts.
they're not installed yet. Check for their existance via a separate !=
rule and then use that to wrap the calls to {gtk,glib}-config inside a
.if/.endif clause.
* From tips provided by Philip Newton after a close reading of the manual page
and other docs, several typos and errors were cleared out, and a small
buglet in the error message within bmpsize() was corrected.
* Moved the code that changes a relative filename to an absolute one. This
was being called in all cases, now it is *not* called if cacheing is
disabled (via $NO_CACHE). This is for the sake of applications running in
environments where they may have trouble with full paths (due to "lockbox"
configurations).
* Made a few adjustments to the docs per suggestion, to clarify some of the
usage cases.
pkgsrc changes:
Don't use libtool any longer to reduce patches.
Add USE_PERL, since 4 scripts depend on it.
program changes:
Lots of new converters/modification programs and scripts, e.g. pbmtonokia,
ppmtolj, and pnmtopalm. Better alpha support.
Also, lots of bug fixes.
shared library version, and in the dependendency information for hundreds of
other packages. [Update to png itself was supplied by Thomas Wizner.]
Relevant changes since 1.0.12, extracted from the "CHANGES" file:
Re-enabled PNG_MNG_FEATURES_SUPPORTED and enabled PNG_ASSEMBLER_CODE_SUPPORTED
by default.
Added runtime selection of MMX features.
Added png_set_strip_error_numbers function and related macros.
Added a check for attempts to read or write PLTE in grayscale PNG datastreams.
Enabled user memory function by default.
Modified png_create_struct so it passes user mem_ptr to user memory allocator.
Increased png_mng_features flag from png_byte to png_uint_32.
Check for missing profile length field in iCCP chunk and free chunk_data
in case of truncated iCCP chunk.
Revised contrib/gregbook/rpng*-x.c to avoid a memory leak and to exit cleanly
if user attempts to run it on an 8-bit display.
Updated contrib/gregbook
Use png_malloc instead of png_zalloc to allocate palette in pngset.c
Added some typecasts to eliminate gcc 3.0 warnings. Changed prototypes
of png_write_oFFS width and height from png_uint_32 to png_int_32.
Updated example.c
Revised prototypes for png_debug_malloc and png_debug_free in pngtest.c
Revised contrib/gregbook
Revised pnggccrd.c to conditionally compile some thread-unsafe code only
when PNG_THREAD_UNSAFE_OK is defined.
Added tests to prevent pngwutil.c from writing a bKGD or tRNS chunk with
value exceeding 2^bit_depth-1
Replaced calls to fprintf(stderr,...) with png_warning() in pnggccrd.c
Removed restriction that do_invert_mono only operate on 1-bit opaque files
Changed a png_warning() to png_debug() in pnggccrd.c
Fixed contrib/gregbook/rpng-x.c, rpng2-x.c to avoid crash with XFreeGC().
Include background_1 in png_struct regardless of gamma support.
Revised example.c to provide more details about using row_callback().
Added type cast to each NULL appearing in a function call, except for
WINCE functions.
Removed type casts from all NULLs.
Simplified png_create_struct_2().
Revised png_create_info_struct() and png_creat_struct_2().
Added error message if png_write_info() was omitted.
Type cast NULLs appearing in function calls when _NO_PROTO or
PNG_TYPECAST_NULL is defined.
Type cast NULLs appearing in function calls except when PNG_NO_TYPECAST_NULL
is defined.
Changed typecast of "size" argument to png_size_t in pngmem.c calls to
the user malloc_fn, to agree with the prototype in png.h
Added a pop/push operation to pnggccrd.c, to preserve Eflag (Maxim Sobolev)
Added a pop/push operation to pngvcrd.c, to preserve Eflag.
Always allocate 256-entry internal palette, hist, and trans arrays, to
avoid out-of-bounds memory reference caused by invalid PNG datastreams.
Added a check for prefix_length > data_length in iCCP chunk handler.
all dependencies on packages depending on "png" which contain shared
libraries, all for the (imminent) update to the "png" package.
[List courtesy of John Darrow, courtesy of "bulk-build".]
Collection.
Graphviz is a set of graph drawing tools.
It includes:
- dot: makes hierarchical layouts of directed graphs
- neato: makes "spring" model layouts of undirected graphs
- lefty: a programmable graphics editor
- dotty: a customizable interface written in LEFTY
- tcldot: a customizable graphical interface written in TCL
- libgraph: the base library for graph tools
- various associated utilities
* many bug fixes
* added support for 16bit minisblack/miniswhite images in RGBA interface
* integrated experimental OJPEG support
* updated tiffcp utility
* substantial changes to tiff2ps
List of changes is unknown since the previous version was NetBSD's
own "port" stored in ftp.netbsd.org. This version is from
sourceforge.net and it should work with the new gphoto-2.0.
* libexif/configure.in: Introduce proper versionning.
* libexif: There's only one ByteOrder per ExifData.
* libexif/libexif-entry.c: More tags implemented in
(exif_entry_get_value).
Summary of changes:
- removal of USE_GTEXINFO
- addition of mk/texinfo.mk
- inclusion of this file in package Makefiles requiring it
- `install-info' substituted by `${INSTALL_INFO}' in PLISTs
- tuning of mk/bsd.pkg.mk:
removal of USE_GTEXINFO
INSTALL_INFO added to PLIST_SUBST
`${INSTALL_INFO}' replace `install-info' in target rules
print-PLIST target now generate `${INSTALL_INFO}' instead of `install-info'
- a couple of new patch files added for a handful of packages
- setting of the TEXINFO_OVERRIDE "switch" in packages Makefiles requiring it
- devel/cssc marked requiring texinfo 4.0
- a couple of packages Makefiles were tuned with respect of INFO_FILES and
makeinfo command usage
See -newly added by this commit- section 10.24 of Packages.txt for
further information.
char **argvec = malloc(); vprintf(fmt, argvec); va_end(argvec);
which of cource won't work on anything that does not pass its
arguments on the stack (like PPC). This patch adds a handcrafted
vprintf() lookalike damn_i386_centric_printf() that does TRT.
Changes:
+ Adapted to Python 2.1. Among other things, all uses of the
"regex" module has been repleased with "re".
+ Fixed attribute error when reading large PNG files (this bug
was introduced in maintenance code released after the 1.1.1
release)
+ Ignore non-string objects in sys.path
+ Fixed Image.transform(EXTENT) for negative xoffsets
+ Fixed loading of image plugins if PIL is installed as a package.
(The plugin loader now always looks in the directory where the
Image.py module itself is found, even if that directory isn't on
the standard search path)
+ The Png plugin has been added to the list of preloaded standard
formats
+ Fixed bitmap/text drawing in fill mode.
+ Fixed "getextrema" to work also for multiband images.
+ Added transparency support for L and P images to the PNG codec.
+ Improved support for read-only images. The "load" method now
sets the "readonly" attribute for memory-mapped images. Operations
that modifies an image in place (such as "paste" and drawing operations)
creates an in-memory copy of the image, if necessary. (before this
change, any attempt to modify a memory-mapped image resulted in a
core dump...)
+ Added special cases for lists everywhere PIL expects a sequence.
This should speed up things like "putdata" and drawing operations.
+ The Image.offset method is deprecated. Use the ImageChops.offset
function instead.
+ Changed ImageChops operators to copy palette and info dictionary
from the first image argument.
The Cult3D plugin enables users view interactive 3D objects,
including full transparency, reflectivity, particle systems, bump
mapping, phong shading, movement and sound.
For a random example: http://www.formula1.com/news/3d/s8396.html
