Translators and developers should count this as the string freeze unless
  something REALLY essential comes up. Send in patches to these translations
  when you get a chance (and patches are appreciated, as they are much easier
  to deal with).

  All tests are passing except 1 known to fail (upgrade051).

* Added some {}

  specified
* Use db->path when appropriate
* Commented out the FAKEROOT checks in libalpm.  This should never ever be done.
  TODO test this quite a bit, as this will never cause the transactions to fail
  if RW operations are requested... right now it is totally up to the front end
  to decide when to fail
* Use realpath() to canonicalize the root path when specified, so
  _alpm_makepath() doesn't freak out
* Fixed some output/indent of MDFile and SHAFile algorithms
* More efficient sprintf() usage in MDFile/SHAFile
* Added real error output to _alpm_makepath

  occasions where some alpm stuff could be used without initializing the
  library (vercmp is one).  TODO make these functions (handle accessors)
  better by returning "library not initialized" instead of failing.
* Removed NoUpgrade lines from pacman.conf - we need to test this!
* Re-corrected the lib targets for src/util/*
* make dist seems to have updated the po files

  in but did it right by duplicating the package first.

  anyway.
* Spelling/commenting etc.

  Giovanni Scafora <linuxmania@gmail.com>
* Updated German translation
  Pierre Schmitz <pierre@archlinux.de>

  catch a bug in 4.19.

  it. Eventually we'll make progress.
* Rewrote the _alpm_splitdep function to behave more like all our other
  function calls. Use heap instead of stack allocation for the depend struct,
  so now it needs to be freed by the caller.

* Moved entirely to alpm_pkg_get_* accessors, to read data on demand
* Mostly removed the INFRQ_ parameters from outside the be_files backend (making
  the backend more extensible in the long run)
* packages created from _alpm_db_scan now have the db and origin set (making
  accessors actually work for these packages)
* removed _alpm_db_ensure_pkgcache

* totally revamped the _alpm_checkconflicts function, making it cleaner and
  easier to read (and thus fix in the long run) - maintainable code ftw
  NOTE: feel free to rename the functions... I couldn't think of anything better

* removed an extra loop in sync.c:find_replacements - no sense in looping over
  an entire DB while strcmp'ing the name, when we have get_pkgfromcache

Other:
* package struct "license" -> "licenses"
* Created _alpm_sync_find (duplicate code in some places, find_pkginsync
* Minor const correctness changes along the way
* fixed a couple extra '/' pathing issues (non-issues really)
* removed a duplicate pkg_cmp function

  Giovanni Scafora <linuxmania@gmail.com>

  allows for the same bit operators to be used across the board on pkg_get
  operations.
* Changed name of INFRQ_NONE -> INFRQ_BASE to more clearly reflect what it
  does (loads pkg name and version).
* Added a few missing things on package functions, such as SYMEXPORT and
  ALPM_LOG_FUNC.
* Slight updates to pmenv to print 'pass' and 'fail' instead of 'passed' and
  'failed'. Keeps output a bit more concise.
* Fixed a doxygen comment spelling error. :P

* Added some provision switching hackery.  This could probably use some
  refactoring,.. it solves the following case:

    pkg1 and pkg2 provide 'foo' and are both installed
    pkg3 depends on 'foo' and so lists 'pkg1' in the REQUIREDBY db section
    pkg1 is upgraded and no longer provides 'foo'
    ** This code ensures that the REQUIREDBY of pkg3 is updated to require pkg2
       now instead of pkg1

  Giovanni Scafora <linuxmania@gmail.com>
* Fixed FS #6442 - don't list directories in -Ql output.

  requiredby for 'already installed' packages, so the test didn't work
  properly. Once these errors were fixed, both tests pass fine.

  about bigger DB changes later, but lets not screw anything up for release.
* Removed some weird uses of "not ... ==" usage in pactest- correct me if I'm
  wrong, but isn't "!=" a lot more clean and concise?
* Print description of failed tests in the pactest summary. This could get
  dirty with a lot of failed tests though, so watch out.

  together while using makepkg -g.

  for and is a dependency of another package. (upgrade057)
* Removed man2html stuff from doc/Makefile.am.
* Patch: improved depends checking in _alpm_checkdeps
  Nagy Gabor <ngaba@petra.hos.u-szeged.hu>
* Slight revisions to pmrule.py to allow for a PKG_PROVIDES test.
* Very quick updates to the pactest README.
* Added rules to some of the newer upgrade tests. Note: upgrade055 now fails
  due to requiredby entries not being written packages that are dependencies
  in a provides role.

  deal with. Currently, libalpm with AND without Nagy's patch fails it.
* Slight spacing update to pmenv.py.

  tar: *: Cannot stat: No such file or directory
  tar: Error exit delayed from previous errors

* Added the actual "Nagy's bug" as upgrade055.py.
* Updated the Makefile to distribute the pactest readme.
* Removed COPYING from the pactest directory- see COPYING at the root for the
  same license.

  to set the umask that some people may find helpful with an 'abs' group (but
  it isn't enabled by default).

  .FILELIST

  add the "requiredby" information that libalpm would have written.
  Now it succeeds.

  provides, and conflicts.
  - upgrade051 is known to fail- but we want to eventually make it work.
  - upgrade052 is Nagy's bug as reported on the ML- I think. If it isn't,
    let me know.

  per package.

* Makefile.am updates for Aaron's move of re-pacman to contrib.