Big changes:
- Allows the symbolic links which point to directories, and treats
the names of struct, union, enum and enum member as definitions.
(5.7)
- Uses GPLv3 and LGPLv3 license. (5.6)
- Becomes GNU BOKINware. (5.5)
Details:
http://cvs.savannah.gnu.org/viewvc/global/global/NEWS?view=markup
Bunch of incompatible format changes. DON'T FORGET TO RE-GENERATE TAGS!
VAAPI (Video Acceleration API) enables hardware accelerated video
decode/encode at various entry-points (VLD, IDCT, Motion Compensation etc.)
for the prevailing coding standards today (MPEG-2, MPEG-4 ASP/H.263, MPEG-4
AVC/H.264, and VC-1/VMW3).
CHANGES IN CUPS V1.4.6
- Fixed a "make check" issue on Solaris (STR #3729)
- Regression: The pstops filter did not support landscape printing of
PostScript files (STR #3722)
- The scheduler killed retried (fax) jobs after restarting them
(STR #3697)
- The cupsAdminSetServerSettings() function disabled sharing when
debug logging was enabled (STR #3712)
CHANGES IN CUPS V1.4.5
- Documentation fixes (STR #3542, STR #3650)
- Localization fixes (STR #3635, STR #3636, STR #3647, STR #3666)
- Security: Fixed a memory corruption bug reported in CVE-2010-2941
(STR #3648)
- The CUPS API incorrectly mapped the HTTP_UNAUTHORIZED status to the
IPP_NOT_AUTHORIZED status code, when IPP_NOT_AUTHENTICATED would be
the correct mapping (STR #3684)
- The scheduler would restart jobs while shutting down (STR #3679)
- Fixed a PPD loader bug that could cause a crash in cupsd (STR #3680)
- Improved the mapping of non-standard PPD and PWG names (STR #3671)
- The scheduler did not initialize Kerberos in all cases (STR #3662)
- cupsAdminSetServerSettings duplicated Listen and Order lines
(STR #3645)
- Added DeviceN colorspace support to the CUPS Raster format (STR #3419)
- ppdMarkDefaults() did not clear the marked field of the previous
choices (STR #3642)
- The serial backend would not allow a raw job to be canceled
(STR #3649)
- The socket backend could go into an infinite loop with certain
printers (STR #3622)
- Setting the PRINTER or LPDEST environment variables to "name/instance"
did not work (STR #3485)
- The scheduler did not handle the JobRetryLimit setting properly
(STR #3466)
- The lpstat command always showed a remote job ID of 0 for shared
printers (STR #3627)
- Increased the write timeout for the libusb-based USB backend to 5
minutes (STR #3595)
- The libusb-base USB backend did not check whether the printer has a
serial number (STR #3590)
- The lpadmin command did not support setting of custom option values
(STR #3631)
- The lpadmin command did not support setting of the location or
description of a class (STR #3613)
- The cupsaddsmb command did not give up after too many failed attempts
(STR #3615)
- The CUPS library no longer uses certain problematic ctype macros that
change based on the locale's character set.
- PJL value substitution of more than 9 values was broken (STR #3621)
- Custom options with missing string values caused ppdEmit* to segfault
(STR #3620)
- Fixed an issue with the Italian version of the web interface
(STR #3624)
- Fixed the Solaris SMF configuration file for cups-lpd (STR #3611)
- The scheduler did not set the notify-subscribed-event attribute when
delivering printer-added or printer-modified events (STR #3608)
- The mailto notifier could get into an infinite loop (STR #3609)
- Date/time information was not shown in banner pages.
- Relational operators were broken in #if/#elif/#else/#endif expressions
for the PPD compiler.
- Moving a job via the web interface failed without asking for
authentication (STR #3559)
- The scheduler now clears the printer-state-reasons when the driver is
changed (STR #3570)
- The web interface did not allow a user to change the driver
(STR #3537, STR #3601)
- The scheduler was not setting the PATH_INFO environment variable when
needed (STR #3600)
- The scheduler incorrectly set the CUPSD_AUTH_TYPE environment
variable instead of AUTH_TYPE (STR #3599)
- Fixed a buffer overrun in the PPD compiler (STR #3594)
- Fixed some additional IPP job template attribute mapping issues in the
scheduler.
Someone who groks linker issues should try to make the asm code
position independent. Until then I suspect we will end up using
--disable-asm everywhere.
Upstream changes:
Compared to the previous version, 2.3.2, nothing exciting can be found, but,
among other things, a couple of serious bugs has been fixed ? everybody is
welcome to upgrade.
Detailed list of changes can be found in the shortlog:
Evan Niessen-Derry (1):
Support newer versions of mp4v2
Gregory Petrosyan (4):
Fix format_print()'s buffer width calculation
Fix replaygain muting the sound when softvol is not active
mp4: read genre tags
cmus 2.3.3
Xavier Chantry (4):
uchar : fix build with clang
Revert "Change sample_rate to uint32_t in mp4 and aac plugins"
ffmpeg : avcodec_decode_audio2 is deprecated
ffmpeg : fix seeking
pkgsrc changes:
* patch-ac no longer required, integrated upstream.
upstream changes:
* Video support for most major video formats
* Publish videos to major Web services, including YouTube, Flickr,
Facebook, and PicasaWeb.
* Runtime monitoring of library directory and auto-import of newly
created files.
* Background writing of metadata (tags, titles, ratings, orientation,
and exposure date) to master files (user-configurable).
* Flagging photos for batch operations.
* Set multiple photos to desktop background slideshow.
* Numerous translation updates.
Changes:
- Compatibility fix: \bibAnnote and \bibAnnoteFile are provided only
with merging, so they do not conflict with alternative definitions
that are sometimes found in certain BibTeX styles.
destdir enabled (now the default)! To do this, remove the logic from
our custom Makefile and make the logic shipped by the package do its
thing (which is smarter than us in this cse).
This lets libgtop work again, which in turn fixes stuff like the GNOME
system monitor or bug-buddy.
Bump PKGREVISION.
- make the packet contain explicit 32bit data for compatibility with
clients on machines with 32bit time_t.
- replace ctime() calls on such changed former time_t values by a
shim that does the translation.
This does not start to work after 2038 yet - that would need a change
of the collection protocol.
