1.0.5:
This has a good chance of being the last one in a series of stable
releases (with stuff backported from our latest and greatest).
Thanks to Pascal for maintaining it! As such, it comes with a short
list of maintenance things as change log:
Update to RawSpeed r438
Update to LibRaw 0.14.7
White balance presets for Nikon Coolpix P7100 and Panasonic GF3
White balance preset updates for Canon EOS 7D, Canon EOS 350D
Standard Color Matrices for Canon EOS 650D, Canon EOS 5D Mark
III, Canon EOS 1D X, Canon PowerShot G1 X, Canon PowerShot
SX220, Nikon D3200, Nikon D4, Nikon D800, Olympus E-M5, Panasonic
GF5, Sony SLT-A37/A57, Leica X1/X2, Sony DSC-RX100
A few memory leaks were resolved
A few generic bugs were resolved
1.0.4:
The changes over darktable 1.0.3 are:
More robust OpenMP compiler detection code
New warming/cooling filter presets for color correction plugin
Lighttable thumbnails should be slightly faster and sharper
Correctly restore panels when using Tab.
Checking if an export target directly is read-only
Writing of hierarchical tags in our .xmp has been improved
Don't list system display profile for anything but the display
profile selection
We disabled scrollwheel scrolling in darkroom mode as it
sometimes conflicts with widget behavior
Lighttable thumbnails are now color managed with some caveats (1)
Improved color rendition:
Nikon D800
Canon EOS 5D Mark III
White balance presets:
Canon EOS 5D Mark III
Canon EOS 7D (updated)
Olympus E-M5
Samsung NX100 (updated)
Olympus E-PL1 (updated)
Olympus E-PL2 (updated)
Olympus E-PL3 (updated)
(1) Lighttable color management caveats:
Our previous releases didn't have any color management in lighttable
mode (and the filmstrip), in this release we have a quick-fix
implementation of color management for lighttable mode. For 1.0.4,
newly imported files use the Preview JPEG by default for the
thumbnail, in that case it's still not color managed. After entering
darkroom mode (after which the thumbnail is regenerated in case
you changed the history stack) it will be color managed. Without
changed history stack the thumbnail will always be the embedded
JPEG. If you need lighttable mode to be color managed all the time
you can disable reading of the Preview JPEGs in our Preferences
dialog "don't use embedded preview jpeg but half-size raw", do note
that this will significantly slow down thumbnail generation and
consequently will slow down new imports. To have darktable regenerate
all old previews (in a color managed fashion) you can delete the
~/.cache/darktable/mipmaps* files.
When tiff was upgraded to 4.0, camlimages stopped building. Both
caml and tiff redefine several common typedefs such as uint32. Unlike
the 3-series of tiff, tiff-4.0 also redefined int64 and uint64. The
existing hack didn't foresee int64 and uint64 getting used, and so
camlimages broke.
One patch was created and another revised to override the caml typedef
definitions with macros before tiff.h is included. The original
tiffread.c patch was reworked to override uint16 and uint32 *again*
after the tiff.h include and not before as it was originally. Very ugly
all around, but I just extended what camlimages was already doing.
* Compute and report sum of critical chunk lengths IHDR, PLTE, IDAT, and IEND,
plus the 8-byte PNG signature instead of just the total IDAT data length.
Simplify finding the lengths from the trial compressions, by replacing
the write function with one that simply counts the bytes that would have
been written to a trial PNG, instead of actually writing a PNG, reading it
back, and counting the IDAT bytes. The savings, while measurable, are
not very significant -- The "-brute" and default pngcrush runs that I
tried are between one and five percent faster. Most of the time is
expended in zlib compression of the IDATs, which is not affected by
the change.
* Removed comments about the system library having to be libpng14 or earlier.
This restriction was fixed in version 1.7.20.
* Ignore all ancillary chunks except during the final trial. This can be
significantly faster when large ancillary chunks such as iCCP are present.
Changes 1.7.32:
* Fixed bug introduced in 1.7.30: Do not call png_set_check_for_invalid_index()
when nosave != 0 (otherwise pngcrush crashes with the "-n" option).
* New translations: cs, eo, hr, sr, uk
* Updated translations: da, de, fi, id, is, it, nl, pl, sk, sv, vi, zh_CN
* Improved the man page
* Prevent NULL pointer dereference on out of memory situation
* Fixed bug that caused read past the end of a buffer (CVE-2012-2845)
* New translations: en_AU, uk
* Updated translations: cs, da, de, en_CA, nl, pl, sk, sv, vi
* Added more supported lens in Canon MakerNote
* Added some defensive NULL pointer checks
* Fixed a number of security and stability issues due to buffer overflows,
bad pointer dereferences and division-by-zero including bug 3434540
and bug 3434545 (CVE-2012-2812, CVE-2012-2813, CVE-2012-2814,
CVE-2012-2836, CVE-2012-2837, CVE-2012-2840, CVE-2012-2841,
CVE-2012-2845)
A heap-based buffer overflow was found in the way OpenJPEG, an
open-source JPEG 2000 codec written in C language, performed parsing of
JPEG2000 having certain number of tiles and tilesizes. A remote
attacker could provide a specially crafted JPEG 2000 file, which when
opened in an application linked against openjpeg would lead to that
application crash, or, potentially arbitrary code execution with the
privileges of the user running the application.
http://code.google.com/p/openjpeg/source/detail?r=1727
Bump PKGREVISION.
pkglint cleanups while here.
Removed scripts/makefile.cegcc from the *.zip and *.7z distributions; it
depends on configure, which is not included in those archives.
Changed "a+w" to "u+w" in Makefile.in to fix CVE-2012-3386.
I don't see CVS-2012-3386 as a vulnerability that applies to pkgsrc,
since to trigger it, you have to run 'make distcheck', and pkgsrc
never does that.
* Dropped *.tar.bz2 from distribution.
* Added a comma that was missing from one of the "usage" strings (error
introduced in version 1.7.29).
Changes 1.7.30:
* Only run the new (in libpng-1.5.10) test of palette indexes during the
first trial.
- When path clipping changes a LINETO to a MOVETO, it also changes
any CLOSEPOLY command to a LINETO to the initial point. This fixes
a problem with pdf and svg where the CLOSEPOLY would then draw a
line to the latest MOVETO position instead of the intended initial
position. - JKS
- The radius labels in polar plots no longer use a fixed padding,
but use a different alignment depending on the quadrant they are
in. This fixes numerical problems when (rmax - rmin) gets too
small. - MGD
changes:
-disable aalib by default
-"--list-all-config" to dump the full config tree + values
-bugfixes
-translation updates
pkgsrc note: dropped aalib support
1.3.16 (June 24, 2012)
==========================
Security Fixes:
* Don't translate 'comment' and 'label' attributes if the request is
made while a file is being read. Only translate such attributes
if they come from the command line or API user.
Bug fixes:
* SWT: SWT reader suffered from a number of implementation errors
which caused it not to work any more. Works again.
* XBM: Fix memory leak observed when reading file in 'ping' mode.
* Support -trim on images which use a consistent (single color)
transparent background. In this case, trim is done based on
opacity rather than foreground color.
* Include <sys/types.h> in order to assure that 'size_t' and
'ssize_t' are declared. This is necessary since
MagickExtentImage() uses these types as part of its definition.
* +repage was not working because parser was insisting that it
should include an argument.
* -units was scaling existing resolution the wrong way around
(i.e. multiplying rather than dividing).
* PerlMagick: Fix compilation with Perl 5.16.
* PingBlob(): PingBlob was not working for all cases. Is now based
on BlobToImage() for assured reliability.
New Features:
None
Feature improvements:
* MAT: Animated movies inside 4D matrices are loaded now.
* PDF: File base name is used as the document title.
* PNG: Fix issues observed specifically with libpng 1.5.10.
Performance Improvements:
* Pixel iterators should be more efficient now if the image uses a
file-backed cache.
* Motion blur algorithm does scale well as cores are added so
include OpenMP support for it by default.
Windows Delegate Updates:
* JPEG: Updated to IJG 8d release.
* PNG: Updated to 1.5.11 release
* TIFF: Updated to 4.0.2 release.
* Zlib: Updated to 1.2.7 release.
* libxml2: Updated to 2.8.0 release.
Behavior Changes:
None
* tif_getimage.c: added support for _SEPARATED CMYK images.
* tif_getimage.c: Added support for greyscale + alpha.
* Added TIFFCreateCustomDirectory() and TIFFCreateEXIFDirectory()
functions.
* tif_print.c: Lots of fixes around printing corrupt or hostile input.
* Improve handling of corrupt ycbcrsubsampling values.
* tif_unix.c: use strerror to get meaningful error messages.
* tif_jpeg.c: fix serious bugs in JPEGDecodeRaw().
* tif_jpeg.c: Fix size overflow (zdi-can-1221,CVE-2012-1173).
doc/Makefile.am. Otherwise, the changes are lost when automake is
rerun, which now (?) happens sometimes (?) during the build. This then
leads to massive PLIST divergence. Bump PKGREVISION in case anyone,
such as perhaps the person who updated it, installed it without
PKG_DEVELOPER set.
