update for new base rev.

(drochner)

update to 1.1.15
This release contains some security fixes, notably a DoS via corrupted
Ogg files (CVE-2008-3231), some related fixes, and fixes for a few
possible buffer overflows.
The other changes include recognition of AMR audio and Snow video.

(drochner)

Fix PLIST. Bump revision.

(joerg)

Fix DESTDIR support.

(joerg)

Distfile changed (not the contents, it was just repackaged), thus do teh
DIST_SUBDIR dance.  Noted by joerg.

(ghen)

Do not use OPSYS for the platform rules, they are an enumeration of all
supported platforms. From PR 39407.

(joerg)

Add and enable: bzr-gtk bzr-svn bzrtools

(epg)

Added devel/bzrtools version 1.6.0 [epg 2008-08-26]
Added devel/bzr-gtk version 0.95.0 [epg 2008-08-26]
Added devel/bzr-svn version 0.4.11~rc1 [epg 2008-08-26]

(epg)

Import bzr-svn-0.4.11~rc1.

bzr-svn is a plugin that allows Bazaar direct access to Subversion
repositories. It allows most bzr commands to work directly against
Subversion repositories, as if you were using bzr with a native
bzr repository.

Status:

Vendor Tag: TNF
Release Tags: pkgsrc-base

(epg)

Import bzr-gtk-0.95.0.

This is a plugin for bzr that contains various GTK+ frontends to Bazaar
commands. It currently contains a tool to see the history and
relationships between the revisions visually and one to view
annotated files.

Status:

Vendor Tag: TNF
Release Tags: pkgsrc-base

(epg)

Import bzrtools-1.6.0.

This is a set of plugins for Bazaar.

Status:

Vendor Tag: TNF
Release Tags: pkgsrc-base

(epg)

Add common settings for bzr plugins (soon to be used by bzrtools, bzr-gtk, and
bzr-svn packages).

(epg)

Updated devel/bzr to 1.6 [epg 2008-08-26]

(epg)

- Update to 1.6.
- pkgsrc changes:
  - Take maintainership.
  - 'make check' was running tests twice; use custom do-test instead.
  - Don't install basically empty index.txt document.

bzr 1.6 2008-08-25
------------------

Finally, the long awaited bzr 1.6 has been released. This release includes
new features like Stacked Branches, improved weave merge, and an updated
server protocol (now on v3) which will allow for better cross version
compatibility. With this release we have deprecated Knit format
repositories, and recommend that users upgrade them, we will continue to
support reading and writing them for the forseeable future, but we will
not be tuning them for performance as pack repositories have proven to be
better at scaling. This will also be the first release to bundle
TortoiseBzr in the standalone Windows installer.

bzr 1.6rc5 2008-08-19
---------------------

  BUG FIXES:

    * Disable automatic detection of stacking based on a containing
      directory of the target. It interacted badly with push, and needs a
      bit more work to get the edges polished before it should happen
      automatically. (John Arbash Meinel, #259275)

bzr 1.6rc4 2008-08-18
---------------------

  BUG FIXES:

    * Fix a regression in knit => pack fetching.  We had a logic
      inversion, causing the fetch to insert fulltexts in random order,
      rather than preserving deltas.  (John Arbash Meinel, #256757)

bzr 1.6rc3 2008-08-14
---------------------

  CHANGES:

    * Disable reading ``.bzrrules`` as a per-branch rule preferences
      file. The feature was not quite ready for a full release.
      (Robert Collins)

  IMPROVEMENTS:

    * Update the windows installer to bundle TortoiseBzr and ``qbzr``
      into the standalone installer. This will be the first official
      windows release that installs Tortoise by default.
      (Mark Hammond)

  BUG FIXES:

    * Fix a regression in ``bzr+http`` support. There was a missing
      function (``_read_line``) that needed to be carried over from
      ``bzr+ssh`` support. (Andrew Bennetts)

    * ``GraphIndex`` objects will internally read an entire index if more
      than 1/20th of their keyspace is requested in a single operation.
      This largely mitigates a performance regression in ``bzr log FILE``
      and completely corrects the performance regression in ``bzr log``.
      The regression was caused by removing an accomodation which had been
      supporting the index format in use. A newer index format is in
      development which is substantially faster. (Robert Collins)

bzr 1.6rc2 2008-08-13
---------------------

This release candidate has a few minor bug fixes, and some regression
fixes for Windows.

  BUG FIXES:

    * ``bzr upgrade`` on remote branches accessed via bzr:// and
      bzr+ssh:// now works.  (Andrew Bennetts)

    * Change the ``get_format_description()`` strings for
      ``RepositoryFormatKnitPack5`` et al to be single line messages.
      (Aaron Bentley)

    * Fix for a regression on Win32 where we would try to call
      ``os.listdir()`` on a file and not catch the exception properly.
      (Windows raises a different exception.) This would manifest in
      places like ``bzr rm file`` or ``bzr switch``.
      (Mark Hammond, John Arbash Meinel)

    * ``Inventory.copy()`` was failing to set the revision property for
      the root entry. (Jelmer Vernooij)

    * sftp transport: added missing ``FileExists`` case to
      ``_translate_io_exception`` (Christophe Troestler, #123475)

    * The help for ``bzr ignored`` now suggests ``bzr ls --ignored`` for
      scripting use. (Robert Collins, #3834)

    * The default ``annotate`` logic will now always assign the
      last-modified value of a line to one of the revisions that modified
      it, rather than a merge revision. This would happen when both sides
      claimed to have modified the line resulting in the same text. The
      choice is arbitrary but stable, so merges in different directions
      will get the same results.  (John Arbash Meinel, #232188)

bzr 1.6rc1 2008-08-06
---------------------

This release candidate for bzr 1.6 solidifies the new branch stacking
feature.  Bazaar now recommends that users upgrade all knit repositories,
because later formats are much faster.  However, we plan to continue read/write and
upgrade support for knit repostories for the forseeable future.  Several
other bugs and performance issues were fixed.

  CHANGES:

    * Knit format repositories are deprecated and bzr will now emit
      warnings whenever it encounters one.  Use ``bzr upgrade`` to upgrade
      knit repositories to pack format.  (Andrew Bennetts)

  IMPROVEMENTS:

    * ``bzr check`` can now be told which elements at a location it should
      check.  (Daniel Watkins)

    * Commit now supports ``--exclude`` (or ``-x``) to exclude some files
      from the commit. (Robert Collins, #3117)

    * Fetching data between repositories that have the same model but no
      optimised fetcher will not reserialise all the revisions, increasing
      performance. (Robert Collins, John Arbash Meinel)

    * Give a more specific error when target branch is not reachable.
      (James Westby)

    * Implemented a custom ``walkdirs_utf8`` implementation for win32.
      This uses a pyrex extension to get direct access to the
      ``FindFirstFileW`` style apis, rather than using ``listdir`` +
      ``lstat``. Shows a very strong improvement in commands like
      ``status`` and ``diff`` which have to iterate the working tree.
      Anywhere from 2x-6x faster depending on the size of the tree (bigger
      trees, bigger benefit.) (John Arbash Meinel)

    * New registry for log properties handles  and the method in
      LongLogFormatter to display the custom properties returned by the
      registered handlers. (Guillermo Gonzalez, #162469)

  BUG FIXES:

    * Add more tests that stacking does not create deltas spanning
      physical repository boundaries.
      (Martin Pool, #252428)

    * Better message about incompatible repositories.
      (Martin Pool, #206258)

    * ``bzr branch --stacked`` ensures the destination branch format can
      support stacking, even if the origin does not.
      (Martin Pool)

    * ``bzr export`` no longer exports ``.bzrrules``.
      (Ian Clatworthy)

    * ``bzr serve --directory=/`` now correctly allows the whole
      filesystem to be accessed on Windows, not just the root of the drive
      that Python is running from.
      (Adrian Wilkins, #240910)

    * Deleting directories by hand before running ``bzr rm`` will not
      cause subsequent errors in ``bzr st`` and ``bzr commit``.
      (Robert Collins, #150438)

    * Fix a test case that was failing if encoding wasn't UTF-8.
      (John Arbash Meinel, #247585)

    * Fix "no buffer space available" error when branching with the new
      smart server protocol to or from Windows.
      (Andrew Bennetts, #246180)

    * Fixed problem in branching from smart server.
      (#249256, Michael Hudson, Martin Pool)

    * Handle a file turning in to a directory in TreeTransform.
      (James Westby, #248448)

  API CHANGES:

    * ``MutableTree.commit`` has an extra optional keywork parameter
      ``exclude`` that will be unconditionally supplied by the command
      line UI - plugins that add tree formats may need an update.
      (Robert Collins)

    * The API minimum version for plugin compatibility has been raised to
      1.6 - there are significant changes throughout the code base.
      (Robert Collins)

    * The generic fetch code now uses three attributes on Repository objects
      to control fetch. The streams requested are controlled via :
      ``_fetch_order`` and ``_fetch_uses_deltas``. Setting these
      appropriately allows different repository implementations to recieve
      data in their optimial form. If the ``_fetch_reconcile`` is set then
      a reconcile operation is triggered at the end of the fetch.
      (Robert Collins)

    * The ``put_on_disk`` and ``get_tar_item`` methods in
      ``InventoryEntry`` were deprecated. (Ian Clatworthy)

    * ``Repository.is_shared`` doesn't take a read lock. It didn't
      need one in the first place (nobody cached the value, and
      ``RemoteRepository`` wasn't taking one either). This saves a round
      trip when probing Pack repositories, as they read the ``pack-names``
      file when locked. And during probe, locking the repo isn't very
      useful. (John Arbash Meinel)

  INTERNALS:

    * ``bzrlib.branchbuilder.BranchBuilder`` is now much more capable of
      putting together a real history without having to create a full
      WorkingTree. It is recommended that tests that are not directly
      testing the WorkingTree use BranchBuilder instead.  See
      ``BranchBuilder.build_snapshot`` or
      ``TestCaseWithMemoryTree.make_branch_builder``.  (John Arbash Meinel)

    * ``bzrlib.builtins.internal_tree_files`` broken into two giving a new
      helper ``safe_relpath_files`` - used by the new ``exclude``
      parameter to commit. (Robert Collins)

    * Make it easier to introduce new WorkingTree formats.
      (Ian Clatworthy)

    * The code for exporting trees was refactored not to use the
      deprecated ``InventoryEntry`` methods. (Ian Clatworthy)

    * RuleSearchers return () instead of [] now when there are no matches.
      (Ian Clatworthy)

bzr 1.6beta3 2008-07-17
-----------------------

This release adds a new 'stacked branches' feature allowing branches to
share storage without being in the same repository or on the same machine.
(See the user guide for more details.)  It also adds a new hook, improved
weaves, aliases for related locations, faster bzr+ssh push, and several
bug fixes.

  FEATURES:

    * New ``pre_change_branch_tip`` hook that is called before the
      branch tip is moved, while the branch is write-locked.  See the User
      Reference for signature details.  (Andrew Bennetts)

    * Rule-based preferences can now be defined for selected files in
      selected branches, allowing commands and plugins to provide
      custom behaviour for files matching defined patterns.
      See ``Rule-based preferences`` (part of ``Configuring Bazaar``)
      in the User Guide and ``bzr help rules`` for more information.
      (Ian Clatworthy)

    * Sites may suggest a branch to stack new branches on.  (Aaron Bentley)

    * Stacked branches are now supported. See ``bzr help branch`` and
      ``bzr help push``.  Branches must be in the ``development1`` format
      to stack, though the stacked-on branch can be of any format.
      (Robert Collins)

  IMPROVEMENTS:

    * ``bzr export --format=tgz --root=NAME -`` to export a gzipped tarball
      to stdout; also ``tar`` and ``tbz2``.
      (Martin Pool)

    * ``bzr (re)merge --weave`` will now use a standard Weave algorithm,
      rather than the annotation-based merge it was using. It does so by
      building up a Weave of the important texts, without needing to build
      the full ancestry. (John Arbash Meinel, #238895)

    * ``bzr send`` documents and better supports ``emacsclient`` (proper
      escaping of mail headers and handling of the MUA Mew).
      (Christophe Troestler)

    * Remembered locations can be specified by aliases, e.g. :parent, :public,
      :submit.  (Aaron Bentley)

    * The smart protocol now has improved support for setting branches'
      revision info directly.  This makes operations like push
      faster.  The new request method name is
      ``Branch.set_last_revision_ex``.  (Andrew Bennetts)

  BUG FIXES:

    * Bazaar is now able to be a client to the web server of IIS 6 and 7.
      The broken implementations of RFC822 in Python and RFC2046 in IIS
      combined with boundary-line checking in Bazaar previously made this
      impossible. (NB, IIS 5 does not suffer from this problem).
      (Adrian Wilkins, #247585)

    * ``bzr log --long`` with a ghost in your mainline now handles that
      ghost properly. (John Arbash Meinel, #243536)

    * ``check`` handles the split-up .bzr layout correctly, so no longer
      requires a branch to be present.
      (Daniel Watkins, #64783)

    * Clearer message about how to set the PYTHONPATH if bzrlib can't be
      loaded.
      (Martin Pool, #205230)

    * Errors about missing libraries are now shown without a traceback,
      and with a suggestion to install the library.  The full traceback is
      still in ``.bzr.log`` and can be shown with ``-Derror``.
      (Martin Pool, #240161)

    * Fetch from a stacked branch copies all required data.
      (Aaron Bentley, #248506)

    * Handle urls such as ftp://user@host.com@www.host.com where the user
      name contains an @.
      (Neil Martinsen-Burrell, #228058)

    * ``needs_read_lock`` and ``needs_write_lock`` now suppress an error during
      ``unlock`` if there was an error in the original function. This helps
      most when there is a failure with a smart server action, since often the
      connection closes and we cannot unlock.
      (Andrew Bennetts, John Arbash Meinel, #125784)

    * Obsolete hidden command ``bzr fetch`` removed.
      (Martin Pool, #172870)

    * Raise the correct exception when doing ``-rbefore:0`` or ``-c0``.
      (John Arbash Meinel, #239933)

    * You can now compare file revisions in Windows diff programs from
      Cygwin Bazaar.
      (Matt McClure, #209281)

    * revision_history now tolerates mainline ghosts for Branch format 6.
      (Aaron Bentley, #235055)

    * Set locale from environment for third party libs.
      (Martin von Gagern, #128496)

  DOCUMENTATION:

    * Added *Using stacked branches* to the User Guide.
      (Ian Clatworthy)

    * Updated developer documentation.
      (Martin Pool)

  TESTING:

  * ``-Dmemory`` will cause /proc/PID/status to be catted before bzr
    exits, allowing low-key analysis of peak memory use. (Robert Collins)

  * ``TestCaseWithTransport.make_branch_and_tree`` tries harder to return
    a tree with a ``branch`` attribute of the right format.  This was
    preventing some ``RemoteBranch`` tests from actually running with
    ``RemoteBranch`` instances.  (Andrew Bennetts)

  API CHANGES:

    * Removed ``Repository.text_store``, ``control_store``, etc.  Instead,
      there are new attributes ``texts, inventories, revisions,
      signatures``, each of which is a ``VersionedFiles``.  See the
      Repository docstring for more details.
      (Robert Collins)

    * ``Branch.pull`` now accepts an ``_override_hook_target`` optional
      parameter.  If you have a subclass of ``Branch`` that overrides
      ``pull`` then you should add this parameter.  (Andrew Bennetts)

    * ``bzrlib.check.check()`` has been deprecated in favour of the more
      aptly-named ``bzrlib.check.check_branch()``.
      (Daniel Watkins)

    * ``Tree.print_file`` and ``Repository.print_file`` are deprecated.
      These methods are bad APIs because they write directly to sys.stdout.
      bzrlib does not use them internally, and there are no direct tests
      for them. (Alexander Belchenko)

  INTERNALS:

    * ``cat`` command no longer uses ``Tree.print_file()`` internally.
      (Alexander Belchenko)

    * New class method ``BzrDir.open_containing_tree_branch_or_repository``
      which eases the discovery of the tree, the branch and the repository
      containing a given location.
      (Daniel Watkins)

    * New ``versionedfile.KeyMapper`` interface to abstract out the access to
      underlying .knit/.kndx etc files in repositories with partitioned
      storage. (Robert Collins)

    * Obsolete developer-use command ``weave-join`` has been removed.
      (Robert Collins)

    * ``RemoteToOtherFetcher`` and ``get_data_stream_for_search`` removed,
      to support new ``VersionedFiles`` layering.
      (Robert Collins)

bzr 1.6beta2 2008-06-10
-----------------------

This release contains further progress towards our 1.6 goals of shallow
repositories, and contains a fix for some user-affecting bugs in the
repository layer.  Building working trees during checkout and branch is
now faster.

  BUG FIXES:

    * Avoid KnitCorrupt error extracting inventories from some repositories.
      (The data is not corrupt; an internal check is detecting a problem
      reading from the repository.)
      (Martin Pool, Andrew Bennetts, Robert Collins, #234748)

    * ``bzr status`` was breaking if you merged the same revision twice.
      (John Arbash Meinel, #235407)

    * Fix infinite loop consuming 100% CPU when a connection is lost while
      reading a response body via the smart protocol v1 or v2.
      (Andrew Bennetts)

    * Inserting a bundle which changes the contents of a file with no trailing
      end of line, causing a knit snapshot in a 'knits' repository will no longer
      cause KnitCorrupt. (Robert Collins)

    * ``RemoteBranch.pull`` needs to return the ``self._real_branch``'s
      pull result. It was instead just returning None, which breaks ``bzr
      pull``. (John Arbash Meinel, #238149)

    * Sanitize branch nick before using it as an attachment filename in
      ``bzr send``. (Luk�  Lalinsk�, #210218)

    * Squash ``inv_entry.symlink_target`` to a plain string when
      generating DirState details. This prevents from getting a
      ``UnicodeError`` when you have symlinks and non-ascii filenames.
      (John Arbash Meinel, #135320)

  IMPROVEMENTS:

    * Added the 'alias' command to set/unset and display aliases. (Tim Penhey)

    * ``added``, ``modified``, and ``unknowns`` behaviour made consistent (all three
      now quote paths where required). Added ``--null`` option to ``added`` and
      ``modified`` (for null-separated unknowns, use ``ls --unknown --null``)
      (Adrian Wilkins)

    * Faster branching (1.09x) and lightweight checkouts (1.06x) on large trees.
      (Ian Clatworthy, Aaron Bentley)

  DOCUMENTATION:

    * Added *Bazaar Zen* section to the User Guide. (Ian Clatworthy)

  TESTING:

    * Fix the test HTTPServer to be isolated from chdir calls made while it is
      running, allowing it to be used in blackbox tests. (Robert Collins)

  API CHANGES:

    * ``WorkingTree.set_parent_(ids/trees)`` will now filter out revisions
      which are in the ancestry of other revisions. So if you merge the same
      tree twice, or merge an ancestor of an existing merge, it will only
      record the newest. (If you merge a descendent, it will replace its
      ancestor). (John Arbash Meinel, #235407)

    * ``RepositoryPolicy.__init__`` now requires stack_on and stack_on_pwd,
      through the derived classes do not.  (Aaron Bentley)

  INTERNALS:

    * ``bzrlib.bzrdir.BzrDir.sprout`` now accepts ``stacked`` to control
      creating stacked branches. (Robert Collins)

    * Knit record serialisation is now stricter on what it will accept, to
      guard against potential internal bugs, or broken input. (Robert Collins)

bzr 1.6beta1 2008-06-02
-----------------------

Commands that work on the revision history such as push, pull, missing,
uncommit and log are now substantially faster.  This release adds a
translation of some of the user documentation into Spanish.  (Contributions of
other translations would be very welcome.)  Bazaar 1.6beta1 adds a new network
protocol which is used by default and which allows for more efficient transfers
and future extensions.

  NOTES WHEN UPGRADING:

    * There is a new version of the network protocol used for bzr://, bzr+ssh://
      and bzr+http:// connections.  This will allow more efficient requests and
      responses, and more graceful fallback when a server is too old to
      recognise a request from a more recent client.  Bazaar 1.6 will
      interoperate with 0.16 and later versions, but servers should be upgraded
      when possible.  Bazaar 1.6 no longer interoperates with 0.15 and earlier via
      these protocols.  Use alternatives like SFTP or upgrade those servers.
      (Andrew Bennetts, #83935)

  CHANGES:

    * Deprecation warnings will not be suppressed when running ``bzr selftest``
      so that developers can see if their code is using deprecated functions.
      (John Arbash Meinel)

  FEATURES:

    * Adding ``-Derror`` will now display a traceback when a plugin fails to
      load. (James Westby)

  IMPROVEMENTS:

    * ``bzr branch/push/pull -r XXX`` now have a helper function for finding
      the revno of the new revision (``Graph.find_distance_to_null``). This
      should make something like ``bzr branch -r -100`` in a shared, no-trees
      repository much snappier. (John Arbash Meinel)

    * ``bzr log --short -r X..Y`` no longer needs to access the full revision
      history. This makes it noticeably faster when logging the last few
      revisions. (John Arbash Meinel)

    * ``bzr ls`` now accepts ``-V`` as an alias for ``--versioned``.
      (Jerad Cramp, #165086)

    * ``bzr missing`` uses the new ``Graph.find_unique_ancestors`` and
      ``Graph.find_differences`` to determine missing revisions without having
      to search the whole ancestry. (John Arbash Meinel, #174625)

    * ``bzr uncommit`` now uses partial history access, rather than always
      extracting the full revision history for a branch. This makes it
      resolve the appropriate revisions much faster (in testing it drops
      uncommit from 1.5s => 0.4s). It also means ``bzr log --short`` is one
      step closer to not using full revision history.
      (John Arbash Meinel, #172649)

  BUGFIXES:

    * ``bzr merge --lca`` should handle when two revisions have no common
      ancestor other than NULL_REVISION. (John Arbash Meinel, #235715)

    * ``bzr status`` was breaking if you merged the same revision twice.
      (John Arbash Meinel, #235407)

    * ``bzr push`` with both ``--overwrite`` and ``-r NNN`` options no longer
      fails.  (Andrew Bennetts, #234229)

    * Correctly track the base URL of a smart medium when using bzr+http://
      URLs, which was causing spurious "No repository present" errors with
      branches in shared repositories accessed over bzr+http.
      (Andrew Bennetts, #230550)

    * Define ``_remote_is_at_least_1_2`` on ``SmartClientMedium`` so that all
      implementations have the attribute.  Fixes 'PyCurlTransport' object has no
      attribute '_remote_is_at_least_1_2' attribute errors.
      (Andrew Bennetts, #220806)

    * Failure to delete an obsolete pack file should just give a warning
      message, not a fatal error.  It may for example fail if the file is still
      in use by another process.
      (Martin Pool)

    * Fix MemoryError during large fetches over HTTP by limiting the amount of
      data we try to read per ``recv`` call.  The problem was observed with
      Windows and a proxy, but might affect other environments as well.
      (Eric Holmberg, #215426)

    * Handle old merge directives correctly in Merger.from_mergeable.  Stricter
      get_parent_map requirements exposed a latent bug here.  (Aaron Bentley)

    * Issue a warning and ignore passwords declared in authentication.conf when
      used for an ssh scheme (sftp or bzr+ssh).
      (Vincent Ladeuil, #203186)

    * Make both http implementations raise appropriate exceptions on 403
      Forbidden when POSTing smart requests.
      (Vincent Ladeuil, #230223)

    * Properly *title* header names in http requests instead of capitalizing
      them.
      (Vincent Ladeuil, #229076)

    * The "Unable to obtain lock" error message now also suggests using
      ``bzr break-lock`` to fix it.  (Martin Albisetti, #139202)

    * Treat an encoding of '' as ascii; this can happen when bzr is run
      under vim on Mac OS X.
      (Neil Martinsen-Burrell)

    * ``VersionedFile.make_mpdiffs()`` was raising an exception that wasn't in
      scope. (Daniel Fischer #235687)

  DOCUMENTATION:

    * Added directory structure and started translation of docs in spanish.
      (Martin Albisetti, Lucio Albenga)

    * Incorporate feedback from Jelmer Vernooij and Neil Martinsen-Burrell
      on the plugin and integration chapters of the User Guide.
      (Ian Clatworthy)

    * More Bazaar developer documentation about packaging and release process,
      and about use of Python reprs.
      (Martin Pool, Martin Albisetti)

    * Updated Tortise strategy document. (Mark Hammond)

  TESTING:

    * ``bzrlib.tests.adapt_tests`` was broken and unused - it has been fixed.
      (Robert Collins)

    * Fix the test HTTPServer to be isolated from chdir calls made while it is
      running, allowing it to be used in blackbox tests. (Robert Collins)

    * New helper function for splitting test suites
      ``split_suite_by_condition``. (Robert Collins)

  INTERNALS:

    * ``Branch.missing_revisions`` has been deprecated. Similar functionality
      can be obtained using ``bzrlib.missing.find_unmerged``. The api was
      fairly broken, and the function was unused, so we are getting rid of it.
      (John Arbash Meinel)

  API CHANGES:

    * ``Branch.abspath`` is deprecated; use the Tree or Transport
      instead.  (Martin Pool)

    * ``Branch.update_revisions`` now takes an optional ``Graph``
      object. This can be used by ``update_revisions`` when it is
      checking ancestry, and allows callers to prefer request to go to a
      local branch.  (John Arbash Meinel)

    * Branch, Repository, Tree and BzrDir should expose a Transport as an
      attribute if they have one, rather than having it indirectly accessible
      as ``.control_files._transport``.  This doesn't add a requirement
      to support a Transport in cases where it was not needed before;
      it just simplifies the way it is reached.  (Martin Pool)

    * ``bzr missing --mine-only`` will return status code 0 if you have no
      new revisions, but the remote does. Similarly for ``--theirs-only``.
      The new code only checks one side, so it doesn't know if the other
      side has changes. This seems more accurate with the request anyway.
      It also changes the output to print '[This|Other] branch is up to
      date.' rather than displaying nothing.  (John Arbash Meinel)

    * ``LockableFiles.put_utf8``, ``put_bytes`` and ``controlfilename``
      are now deprecated in favor of using Transport operations.
      (Martin Pool)

    * Many methods on ``VersionedFile``, ``Repository`` and in
      ``bzrlib.revision``  deprecated before bzrlib 1.5 have been removed.
      (Robert Collins)

    * ``RevisionSpec.wants_revision_history`` can be set to False for a given
      ``RevisionSpec``. This will disable the existing behavior of passing in
      the full revision history to ``self._match_on``. Useful for specs that
      don't actually need access to the full history. (John Arbash Meinel)

    * The constructors of ``SmartClientMedium`` and its subclasses now require a
      ``base`` parameter.  ``SmartClientMedium`` implementations now also need
      to provide a ``remote_path_from_transport`` method.  (Andrew Bennetts)

    * The default permissions for creating new files and directories
      should now be obtained from ``BzrDir._get_file_mode()`` and
      ``_get_dir_mode()``, rather than from LockableFiles.  The ``_set_file_mode``
      and ``_set_dir_mode`` variables on LockableFiles which were advertised
      as a way for plugins to control this are no longer consulted.
      (Martin Pool)

    * ``VersionedFile.join`` is deprecated. This method required local
      instances of both versioned file objects and was thus hostile to being
      used for streaming from a smart server. The new get_record_stream and
      insert_record_stream are meant to efficiently replace this method.
      (Robert Collins)

    * ``WorkingTree.set_parent_(ids/trees)`` will now filter out revisions
      which are in the ancestry of other revisions. So if you merge the same
      tree twice, or merge an ancestor of an existing merge, it will only
      record the newest. (If you merge a descendent, it will replace its
      ancestor). (John Arbash Meinel, #235407)

    * ``WorkingTreeFormat2.stub_initialize_remote`` is now private.
      (Martin Pool)

(epg)

Note initial addition of p5-ExtUtils-Command version 1.14 as
devel/p5-ExtUtils-Command into The NetBSD Packages Collection.

The Perl 5 module ExtUtils::Command is used to replace common UNIX
commands. In all cases the provided functions work from @ARGV rather
than taking arguments. This makes them easier to deal with in
Makefiles. Call them like this:
  perl -MExtUtils::Command -e some_command some files to work on
The module Shell::Command is also included in this package: it
provides the same functions but take arguments normally.

(seb)

Add & enable p5-ExtUtils-Command

(seb)

Initial import of p5-ExtUtils-Command version 1.14 in the NetBSD
Packages Collection.

The Perl 5 module ExtUtils::Command is used to replace common UNIX
commands. In all cases the provided functions work from @ARGV rather
than taking arguments. This makes them easier to deal with in
Makefiles. Call them like this:
  perl -MExtUtils::Command -e some_command some files to work on
The module Shell::Command is also included in this package: it
provides the same functions but take arguments normally.

Status:

Vendor Tag: TNF
Release Tags: pkgsrc-base

(seb)

Updated math/maxima to 5.16.3

(ahoka)

Update to 5.15

Submitted by Aleksej Saushev in PR pkg/39404.

Changes since Maxima 5.15

Major changes:

* ECL can build Maxima via autoconf machinery

Other revisions:

* Revision and extension of code for exponential integral functions
* Revision and extension of code for hypergeometric functions
* share/contrib/amatrix: revisions
* share/contrib/gf: bug fixes
* share/contrib/sarag: revisions
* share/contrib/zeilberger: revisions
* share/contrib/graphs: revisions
* share/contrib/distrib: revisions and new test script
* share/contrib/maximaMathML: revisions
* share/tensor: additional demonstrations
* Patches for Lispworks and MCL
* Revisions and extensions of reference manual and its translations

Plus various bug fixes.

(ahoka)

Note initial addition of p5-Jifty-DBI version 0.49 as
databases/p5-Jifty-DBI into The NetBSD Packages Collection.

The Perl 5 module Jifty::DBI deals with databases, so that you
don't have to.  This module provides an object-oriented mechanism
for retrieving and updating data in a DBI-accessible database.
This module is the direct descendent of DBIx::SearchBuilder. If
you're familiar with SearchBuilder, Jifty::DBI should be quite
familiar to you.

(seb)

Add & enable p5-Jifty-DBI

(seb)

Initial import of p5-Jifty-DBI version 0.49 in the NetBSD Packages
Collection.

The Perl 5 module Jifty::DBI deals with databases, so that you
don't have to.  This module provides an object-oriented mechanism
for retrieving and updating data in a DBI-accessible database.
This module is the direct descendent of DBIx::SearchBuilder. If
you're familiar with SearchBuilder, Jifty::DBI should be quite
familiar to you.

Status:

Vendor Tag: TNF
Release Tags: pkgsrc-base

(seb)

Note initial addition of p5-Time-Duration-Parse version 0.06 as
time/p5-Time-Duration-Parse into The NetBSD Packages Collection.

Time::Duration::Parse is a Perl 5 module to parse human readable
duration strings like 2 minutes and 3 seconds to seconds.  It does
the opposite of duration_exact function in Time::Duration.

(seb)

Add & enable p5-Time-Duration-Parse

(seb)

Initial import of p5-Time-Duration-Parse version 0.06 in the NetBSD
Packages Collection.

Time::Duration::Parse is a Perl 5 module to parse human readable
duration strings like 2 minutes and 3 seconds to seconds.  It does
the opposite of duration_exact function in Time::Duration.

Status:

Vendor Tag: TNF
Release Tags: pkgsrc-base

(seb)

Add description for fftw and threads support options

(bjs)

iozone update to 3.308 is done.

(bjs)

+ Updated benchmarks/iozone to 3.308 [bjs 2008-08-25]

(bjs)

Update to iozone 3.308.

pkgsrc changes:
- add 'tee' to USE_TOOLS.
- support an optional build with threads support and also support this in the
  'benchmark' target.

Changes since version 2.263

-Bug fix for -k used in conjunction with -t and content
validation.
-Add DragonFly target.
-Put PER_VECTOR_OFFSET in for HP-UX
-Fix compiler warning messages
-Enforce minimum file size of page_size
-Minor fixes.
-Check fsync and close for errors.
-Adding support for testing block devices. Will be done is phases.
This is phase 1. (Single threaded mode only)
-Adding 4 token support to client_list. Each entry may now contain
4 tokens and the new one is the absolute path to the temp
file for testing.
-Add support for external monitor start & stop for throughput
tests.
-IMON_START and IMON_STOP environmental variables used.
-Bug fix for systems without O_DIRECT. Fall through in switch
statement.
-Fix for -c -t over NFS and initial writer close() when told by
another to stop
-Add Benny Halevy to contributors list.
-Fix for Cygwin environment.
-Code cleanup, and add arg to external trigger.
-Code fixes for macosx
-Add support for building with Sun's Studio 11 compiler
-Bug fix for fread/fwrite with > 2Gig files.
-Add O_DIRECT for Windows
-Add -+w dedup testing mode.
-Make remaining non-dedup data unique.
-Make non-dedupable more unique.
-Bug fix for non-dedup.
-Make random offsets unique, using Knuth shuffle.
-free memory used for random offset uniqueness.
-Make unique/random offsets 64bits.
-Add fallback for random/unique.
-Make non-dedup region more unique
-Add -+y ## to set percentage of interior dedup.
-Add -+y ## to set percentage of interior dedup.
-Bug fixes for -+w and -+y
-Minor fix for dedup
-Adding -+C to set percent of dedupable within a file.
-Add solaris to read sync O_RSYNC
-Add space to avoid field output touching each other.
-Add check for config file exceeding MAXSTREAMS.
-Add new contributor's name.

(bjs)

pkg_install-20080825:
Fix an inverted test for fake mode. The check stopped +REQUIRED_BY and
friends from being moved to the new package, thereby breaking both the
delete and the dependency tree.

Fix a fexec to fexec_skipempty to silence
pkg_delete: package '' is not installed, /var/db/pkg//+CONTENTS missing
when updating packages with prefix or destdir being set.

(joerg)

If the file removal fails for e.g. due to checksum mismatch, still
update the pkgdb. This restores the expectation of installing a package
over it working fine.

(joerg)

Nuke pkgdb debugging code, it hasn't been used for ages.

(joerg)

+ Added graphics/vigra version 1.6.0 [bjs 2008-08-25]

(bjs)

Add vigra to graphics/Makefile

(bjs)

Import vigra-1.6.0.

VIGRA is a computer vision library that puts its main emphasize on
flexible algorithms, because algorithms represent the principle know-how
of this field. The library was consequently built using generic
programming as introduced by Stepanov and Musser and exemplified in the
C++ Standard Template Library. By writing a few adapters (image iterators
and accessors) you can use VIGRA's algorithms on top of your data
structures, within your environment. Alternatively, you can also use the
data structures provided within VIGRA, which can be easily adapted to a
wide range of applications. VIGRA's flexibility comes almost for free:
Since the design uses compile-time polymorphism (templates), performance
of the compiled program approaches that of a traditional, hand tuned,
inflexible, solution.

Status:

Vendor Tag: TNF
Release Tags: pkgsrc-base

(bjs)

Fix png backend with libpng 1.2.30. Older libpng versions flushed the
output at the end of the write and this was restored with 1.2.30, but
cairo didn't override the flush function, resulting in fflush on a
non-stream. This in turn resulted in crashes of graphviz when compiling
doxygen. Bump revision.

(joerg)

Fix build on amd64, add destdir support.

(joerg)

Fix build on amd64, make destdir safe.

(joerg)

Note update of www/drupal6-translations package to 20080826.

(taca)

Drop translation files which are in development status.
It was rashed to chase these development versions.

Call it 20080826 version now.

(taca)

+ Updated devel/cmake to 2.6.1 [bjs 2008-08-25]

(bjs)

Update patchsum and remove trailing whitespace in Makefile.

(bjs)

Update to cmake-2.6.1.  I could not find any release notes concise enough
to include here; the changelog for this release is here:

<http://www.cmake.org/pipermail/cmake/2008-August/023188.html>

For changes prior to this release, please see ${WRKSRC}/ChangeLog.

While here, add a list of *.cmake files in which to replace
/usr/${X11R6,local} with X11BASE and LOCALBASE, respectively.
Also, the [pkgsrc-relative] API, e.g. 2.6 is now defined by a variable
so that it can be used in pathnames for the build and the PLIST.

(bjs)

Remove ftp.scarlet.be from MASTER_SITE_KDE, service suspended over a month.
Noticed by Zafer Aydogan via private mail.

(obache)

Uses xmkmf in configure script; USE_TOOLS+=xmkmf.
Noted by Hasso Tepper in PR 39286.

(dholland)

Another set of updates of p5-* packages:

Updated net/p5-Net-Pcap to 0.16
Updated net/p5-Net-Write to 1.03
Updated net/p5-Nmap-Parser to 1.16
Updated net/p5-POE-Component-Client-DNS to 1.00
Added devel/p5-Term-Cap version 1.12
Updated devel/p5-POE to 1.003

(he)

Update from version 0.3202nb1 to 1.003,
  discussed in general terms with maintainer.

Pkgsrc changes:
o Add commented-out HOMEPAGE to search.cpan.org as an alternative
o Adjust dependencies

Upstream changes (there may be more before 0.9999, but that was the end
  of the change log):

==================================
2008-07-06T00:20:23.186069Z v1_003
==================================

  2008-07-06 00:19:32 (r2392) by rcaputo; poe/lib/POE.pm M

    Let's try this again.

  2008-07-05 22:50:47 (r2391) by rcaputo
  poe/mylib/Makefile-5004.pm M; poe/mylib/Makefile-5005.pm M

    Touch mylib/gen-tests.perl during Makefile.PL execution. This
    triggers gen-tests.perl during "make test".

  2008-07-05 21:53:37 (r2390) by rcaputo
  poe/mylib/Makefile-5004.pm M; poe/mylib/Makefile-5005.pm M

    Move test generation to PL_FILES. Requires a separate "make" for the
    tests to be generated.

  2008-07-05 20:15:25 (r2389) by rcaputo; poe/mylib/gen-meta.perl M

    Add build_requires and rename config_requires to configure_requires.

==================================
2008-07-05T18:21:13.947381Z v1_002
==================================

  2008-07-05 18:01:55 (r2387) by rcaputo
  poe/mylib/Makefile-5004.pm M; poe/mylib/gen-meta.perl M;
  poe/lib/POE/Wheel/ReadLine.pm M; poe/mylib/Makefile-5005.pm M;
  poe/lib/POE/Filter/HTTPD.pm M; poe/mylib/preprocessor.perl M;
  poe/lib/POE.pm M; poe/docs/Makefile M; poe/mylib/PoeBuildInfo.pm M;
  poe/mylib/gen-tests.perl M; poe/lib/POE/Filter/Reference.pm M;
  poe/lib/POE/NFA.pm M

    Add POE::Test::Loops to config_requires, so modern systems can
    install it before Makefile.PL runs. Allow gen-tests.pl to fail
    gracefully if POE::Test::Loops isn't installed. Remove a bunch of
    tabs from unrelated files.

==================================
2008-07-05T02:39:32.500880Z v1_001
==================================

  2008-07-05 02:38:47 (r2385) by rcaputo; poe/lib/POE.pm M

    Time for a new release.

  2008-07-05 02:35:08 (r2384) by rcaputo; poe/mylib/PoeBuildInfo.pm M

    Use the bleeding-edge POE::Test::Loops.

  2008-07-04 21:47:28 (r2379) by rcaputo; poe/t/30_loops D

    Remove empty directories.

  2008-07-04 21:31:06 (r2378) by rcaputo; poe/MANIFEST M

    Remove the removed files from the MANIFEST.

  2008-07-04 21:21:12 (r2377) by lotr; poe/mylib/PoeBuildInfo.pm M

    update version

  2008-07-04 21:20:36 (r2376) by lotr
  poe/t/30_loops/00_base/k_signals_rerun.pm D;
  poe/t/30_loops/00_base/wheel_run.pm D;
  poe/t/30_loops/00_base/wheel_curses.pm D;
  poe/t/30_loops/00_base/k_alarms.pm D;
  poe/t/30_loops/00_base/k_aliases.pm D;
  poe/t/30_loops/00_base/ses_session.pm D;
  poe/t/30_loops/00_base/wheel_sf_unix.pm D;
  poe/t/30_loops/00_base/k_sig_child.pm D;
  poe/t/30_loops/00_base/sbk_signal_init.pm D;
  poe/t/30_loops/00_base/wheel_sf_ipv6.pm D; poe/mylib/gen-tests.perl M;
  poe/t/30_loops/00_base/wheel_sf_tcp.pm D;
  poe/t/30_loops/00_base/k_detach.pm D;
  poe/t/30_loops/00_base/wheel_sf_udp.pm D;
  poe/t/30_loops/00_base/all_errors.pm D;
  poe/t/30_loops/00_base/comp_tcp.pm D;
  poe/t/30_loops/00_base/comp_tcp_concurrent.pm D;
  poe/t/30_loops/00_base/wheel_readline.pm D;
  poe/t/30_loops/00_base/k_signals.pm D;
  poe/t/30_loops/00_base/wheel_readwrite.pm D;
  poe/t/30_loops/00_base/k_selects.pm D;
  poe/t/30_loops/00_base/wheel_tail.pm D;
  poe/t/30_loops/00_base/ses_nfa.pm D; poe/mylib/PoeBuildInfo.pm M;
  poe/t/30_loops/00_base/wheel_accept.pm D

    use POE::Test::Loops

===================================
2008-06-26T05:25:54.564733Z v1_0003
===================================

  2008-06-26 05:25:26 (r2360) by rcaputo; poe/lib/POE.pm M

    Bump the version for release.

  2008-06-26 04:55:35 (r2359) by rcaputo
  poe/t/10_units/01_pod/02_pod_coverage.t M

    Note some deprecated method names in the POD coverage tests.

  2008-06-26 04:51:13 (r2358) by rcaputo; poe/lib/POE/Wheel/Run.pm M

    Minor NAME update.

  2008-06-20 17:41:54 (r2357) by rcaputo
  poe/lib/POE/Component/Client/TCP.pm M; poe/lib/POE/Wheel/Run.pm M;
  poe/lib/POE/Loop.pm M; poe/t/20_resources/00_base/filehandles.pm M;
  poe/examples/create.perl M; poe/lib/POE/Kernel.pm M;
  poe/examples/sessions.perl M; poe/lib/POE/Filter/Reference.pm M;
  poe/lib/POE/Filter.pm M; poe/mylib/Makefile-5005.pm M;
  poe/lib/POE/Wheel/ReadLine.pm M; poe/lib/POE/Resource/Signals.pm M;
  poe/t/30_loops/00_base/wheel_readwrite.pm M; poe/mylib/Devel/Null.pm M;
  poe/lib/POE/Component/Server/TCP.pm M;
  poe/t/10_units/03_base/15_kernel_internal.t M; poe/lib/POE.pm M;
  poe/t/20_resources/00_base/events.pm M; poe/lib/POE/Wheel.pm M;
  poe/examples/thrash.perl M

    Apply Chris Fedde's massive doc patch, resolving rt.cpan.org ticket
    36394.

  2008-06-20 07:43:46 (r2356) by nothingmuch; poe/lib/POE/Session.pm M

    Fixed instances of the package_states incorrectly referred to as
    class_states in docs

  2008-06-20 02:31:51 (r2355) by rcaputo
  poe/lib/POE/Loop/IO_Poll.pm M; poe/lib/POE/Loop/Select.pm M

    Make the same fix to POE::Loop::IO_Poll as was done in
    POE::Loop::Select. Add an ASSERT_DATA test to both, to see if some
    code is ever executed.

  2008-06-20 02:02:56 (r2354) by rcaputo; poe/lib/POE/Loop/Select.pm M

    Apply Tony Cook's patch to fix the updated time watcher methods.
    Avoids a nasty loop spin when there are no pending events.

  2008-06-18 13:13:12 (r2353) by nothingmuch
  poe/lib/POE/Wheel/Run.pm M; poe/lib/POE/Session.pm M

    replace C<> with C<< >> when the formatted text contains a > symbol

  2008-06-16 02:48:57 (r2352) by rcaputo
  poe/lib/POE/Wheel/ReadLine.pm M; poe/lib/POE/Wheel/SocketFactory.pm M

    Redocument POE::Wheel::ReadLine. Remove tabs from
    POE::Wheel::SocketFactory.

  2008-06-16 00:22:43 (r2351) by rcaputo
  poe/lib/POE/Wheel/SocketFactory.pm M

    Redocument.

  2008-06-13 20:34:15 (r2350) by nothingmuch
  poe/lib/POE/Resource/Signals.pm M

    clarify trace message when polling for signals under USE_SIGCHLD

  2008-06-05 06:00:55 (r2349) by rcaputo; poe/lib/POE/Kernel.pm M

    Rename POE_LOOP to POE_EVENT_LOOP.

  2008-06-05 05:58:33 (r2348) by rcaputo
  poe/lib/POE/Kernel.pm M; poe/lib/POE.pm M;
  poe/lib/POE/Wheel/SocketFactory.pm M

    Partially redocument POE::Wheel::SocketFactory. Add support for
    setting the event loop with an environment variable.

  2008-06-01 18:40:12 (r2347) by rcaputo
  poe/lib/POE/Component/Client/TCP.pm M;
  poe/lib/POE/Component/Server/TCP.pm M; poe/lib/POE/Wheel/FollowTail.pm
  M; poe/lib/POE/Wheel/ListenAccept.pm M;
  poe/lib/POE/Wheel/SocketFactory.pm M

    Redocument POE::Wheel::ListenAccept. Add examples for
    POE::Component::Server::TCP and POE::Wheel::SocketFactory. Tweak
    POE::Wheel::FollowTail to match some patterns that emerged now that a
    second wheel has been redocumented.

  2008-05-31 21:57:34 (r2346) by rcaputo
  poe/lib/POE/Component/Server/TCP.pm M; poe/lib/POE/Wheel/FollowTail.pm
  M

    Redocument.

  2008-05-29 06:37:47 (r2343) by rcaputo; poe/lib/POE/Kernel.pm M

    Support XS loops in POE::Kernel's import() syntax.

  2008-05-29 06:03:46 (r2342) by rcaputo
  poe/lib/POE/Wheel/Curses.pm M; poe/t/10_units/05_filters/50_stackable.t
  M

    Fix a test that relied on unqualified stacked filter class names.

    Redocument POE::Wheel::Cursesl

  2008-05-29 06:02:19 (r2341) by rcaputo; poe/lib/POE/Loop/Select.pm M

    Eliminate a call to get_next_event_time().

  2008-05-29 05:57:52 (r2340) by rcaputo; poe/lib/POE/Filter/Line.pm M

    Fix a bug introduced during redocumenting.

  2008-05-29 04:02:50 (r2339) by rcaputo; poe/lib/POE/Filter/HTTPD.pm M

    Redocument.

  2008-05-28 07:57:32 (r2338) by rcaputo
  poe/lib/POE/Filter/Line.pm M; poe/lib/POE/Filter/Reference.pm M

    Redocument.

  2008-05-27 07:09:15 (r2337) by rcaputo
  poe/lib/POE/Filter/Line.pm M; poe/lib/POE/Filter/RecordBlock.pm M;
  poe/lib/POE/Filter/Stackable.pm M; poe/lib/POE/Filter/Stream.pm M;
  poe/lib/POE/Filter/Grep.pm M; poe/lib/POE/Filter/Map.pm M

    Redocument.

  2008-05-26 22:01:01 (r2336) by rcaputo
  poe/lib/POE/Filter/Stackable.pm M; poe/lib/POE/Filter/Block.pm M;
  poe/lib/POE/Filter/Grep.pm M; poe/lib/POE/Filter/Map.pm M;
  poe/lib/POE/Filter.pm M

    Redocument.

  2008-05-26 18:39:15 (r2335) by rcaputo
  poe/lib/POE/Resource/Extrefs.pm M; poe/lib/POE/Resource/Signals.pm M;
  poe/lib/POE/Resource.pm M; poe/lib/POE/Resource/Aliases.pm M;
  poe/lib/POE/Resource/SIDs.pm M; poe/lib/POE/Resources.pm M;
  poe/lib/POE/Resource/Events.pm M; poe/lib/POE/Resource/Sessions.pm M;
  poe/lib/POE/Kernel.pm M; poe/lib/POE/Resource/Statistics.pm M;
  poe/t/10_units/03_base/09_resources.t M; poe/lib/POE/Driver/SysRW.pm M;
  poe/lib/POE/Resource/FileHandles.pm M

    Redocument.

  2008-05-26 06:48:40 (r2334) by rcaputo
  poe/lib/POE/Loader.pm M; poe/lib/POE/Session.pm M; poe/lib/POE/NFA.pm M

    Redocument.

  2008-05-26 02:59:48 (r2333) by rcaputo; poe/lib/POE.pm M

    Fix Jos Boumans' credit.

  2008-05-26 02:53:21 (r2332) by rcaputo
  poe/lib/POE/Pipe/TwoWay.pm M; poe/lib/POE/Pipe.pm M;
  poe/lib/POE/Pipe/OneWay.pm M

    Redocument.

  2008-05-26 01:19:11 (r2331) by rcaputo
  poe/lib/POE/Driver.pm M; poe/lib/POE/Driver/SysRW.pm M

    Redocument.

  2008-05-25 23:06:14 (r2330) by rcaputo; poe/lib/POE/Loop.pm M

    Point out POE::Test::Loops.

  2008-05-25 23:01:58 (r2329) by rcaputo
  poe/lib/POE/Loop/IO_Poll.pm M; poe/lib/POE/Loop/Tk.pm M;
  poe/lib/POE/Loop/Select.pm M; poe/lib/POE/Loop/Event.pm M;
  poe/lib/POE/Loop/PerlSignals.pm M; poe/lib/POE/Loop/Gtk.pm M;
  poe/lib/POE/Loop/TkCommon.pm M; poe/lib/POE/Loop/TkActiveState.pm M

    Redocument the specific POE::Loop classes.

  2008-05-21 06:03:58 (r2328) by rcaputo; poe/lib/POE/Queue.pm M

    Fix a typo.

  2008-05-21 05:42:02 (r2327) by rcaputo
  poe/lib/POE/Queue/Array.pm M; poe/lib/POE/Queue.pm M

    Redocument.

===================================
2008-05-11T22:30:51.597024Z v1_0002
===================================

  2008-05-11 22:29:43 (r2325) by rcaputo; poe/lib/POE.pm M

    Bump version for release.

  2008-05-11 21:28:17 (r2324) by rcaputo
  poe/lib/POE/Pipe.pm M; poe/lib/POE/Resource/FileHandles.pm M

    Clean up the F_GETFL/F_SETFL mess a little.

  2008-05-11 20:51:44 (r2323) by rcaputo
  poe/t/30_loops/00_base/wheel_run.pm M

    Increase the timeout a little. Seeing some test failures due to
    timeouts.

  2008-05-11 20:45:25 (r2322) by rcaputo
  poe/t/30_loops/00_base/wheel_accept.pm M

    Increase timeout for my slow VirtualPC test system.

  2008-05-11 20:36:05 (r2321) by rcaputo
  poe/t/90_regression/merijn-sigchld-system.t M

    Actually, we must skip 4 tests if /bin/true does not exist.

  2008-04-28 18:10:04 (r2320) by rcaputo; poe/lib/POE/Kernel.pm M

    Link the sig_child() documentation to the USE_SIGCHLD documentation
    per Marc Lehmann's feedback.

  2008-04-28 06:16:20 (r2319) by rcaputo; poe/lib/POE/Kernel.pm M

    Per Marc Lehmann's feedback, I have documented how to avoid a
    potential race condition when catching child processes with
    sig_child(). Basically, if you want sig_child() to catch a process,
    call the method in the same event handler that forked the process.
    Otherwise POE::Kernel may reap the process before sig_child() is
    called.

  2008-04-26 09:13:50 (r2318) by rcaputo; poe/lib/POE/Kernel.pm M

    Document how to avoid the run-wasn't-called warning. Thanks to Marc
    Lehmann for reminding me to document it.

  2008-04-20 21:10:54 (r2317) by rcaputo
  poe/t/30_loops/00_base/wheel_run.pm M

    Increase the timeout in a test to avoid false errors.

===================================
2008-04-19T22:26:06.023137Z v1_0001
===================================

  2008-04-19 22:25:25 (r2315) by rcaputo; poe/mylib/svn-log.perl M

    Use a much faster svn-log.perl from the latest repo-tools.

  2008-04-19 22:18:35 (r2314) by rcaputo; poe/lib/POE.pm M

    Bump for release.

  2008-04-19 20:01:20 (r2313) by rcaputo; poe/lib/POE/Kernel.pm M

    Resolve rt.cpan.org ticket 34803. Apocalypse pointed out that
    $_[KERNEL]->signal(DIE => $_[KERNEL]) would lock up and chew up
    memory and CPU. Very bad bug. Resolved now, thanks!

  2008-04-19 05:49:16 (r2312) by rcaputo
  poe/lib/POE/Resource/Sessions.pm M; poe/lib/POE/Resource/FileHandles.pm
  M

    Removal of POE::API::ResLoader broke POE::API::Peek. This change
    re-initializes $poe_kernel members that stopped being set by the
    previous change. Thanks to Niko Tyni of the Debian Perl Group for
    pointing out the problem.

  2008-04-14 04:46:59 (r2311) by rcaputo; poe/lib/POE/Component.pm M

    First draft of the new docs.

  2008-04-14 03:47:44 (r2310) by rcaputo; poe/lib/POE/Filter.pm M

    First stab at a doc rewrite.

  2008-04-02 16:37:14 (r2309) by rcaputo; poe/lib/POE/Wheel/Run.pm M

    Change the name of Setpgrp to SetPgrp to follow the form of SetSid.

  2008-04-02 14:08:34 (r2308) by bingosnet; poe/lib/POE/Wheel/Run.pm M

    Added setpgrp() to POE::Wheel::Run to change process group of child
    process, if supported by the OS.

  2008-03-31 11:46:07 (r2307) by nothingmuch
  poe/lib/POE/Resource/Signals.pm M;
  poe/t/90_regression/merijn-sigchld-system.t M;
  poe/lib/POE/Loop/PerlSignals.pm M; poe/lib/POE/Kernel.pm M;
  poe/lib/POE/Resource/FileHandles.pm M

    Introduce optional (see USE_SIGCHLD) $SIG{CHLD} based child polling
    in addition to interval based polling

  2008-03-27 18:57:08 (r2306) by nothingmuch; poe/lib/POE/Kernel.pm M

    Document all the "other" constants in their own section of
    POE::Kernel

  2008-03-27 18:39:59 (r2305) by nothingmuch
  poe/lib/POE/Resource/Signals.pm M; poe/lib/POE/Kernel.pm M

    Introduce the CHILD_POLLING_INTERVAL constant

  2008-03-27 18:38:20 (r2304) by nothingmuch; poe/lib/POE/Kernel.pm M

    Accept values from POE_([A-Z]+) environment variables for any defined
    constant, not just ASSERT/TRACE

  2008-03-27 18:17:26 (r2303) by rcaputo; poe/lib/POE/Kernel.pm M

    LARGE_QUEUE_SIZE isn't needed anymore. Linear scanning wasn't that
    much of an optimization.

===================================
2008-03-25T20:09:15.042886Z v1_0000
===================================

  2008-03-25 20:08:47 (r2301) by rcaputo; poe/lib/POE.pm M

    At long last, it is time.

  2008-03-23 01:43:22 (r2298) by rcaputo; poe/lib/POE.pm M

    Addressed several of the CPAN testers failures. Releasing a new test
    version to flush out any new issues.

  2008-03-23 01:42:40 (r2297) by rcaputo
  poe/t/10_units/04_drivers/01_sysrw.t M

    Simulate a driver get() error by reading from a write-only
    filehandle. I suspect that Perl 5.6.2 is reporting the act of reading
    from a closed fielhandle as an EOF rather than an error, so reading
    from a write-only filehandle may trigger different behavior.

  2008-03-23 01:31:21 (r2296) by rcaputo
  poe/t/10_units/05_filters/03_http.t M

    Looks like Perl 5.8.7 mistreats my @x = $o->method() =~ m//g; as an
    lvalue subroutine motification, or something. Split it into a scalar
    assign, and bind the match to the scalar rather than a method return
    value.

  2008-03-23 01:28:18 (r2295) by rcaputo
  poe/t/30_loops/00_base/wheel_sf_ipv6.pm M

    Can't use AF_INET6 directly. Needed to fully qualify its package.

  2008-03-23 01:23:36 (r2294) by rcaputo; poe/lib/POE/Wheel/Run.pm M

    Contrary to the docs, IO::Pty's clone_winsize_from() seems to die
    noisily if the source filehandle isn't a tty.

  2008-03-23 01:19:48 (r2293) by rcaputo; poe/lib/POE/Wheel/ReadLine.pm M

    Fall back to an 80x25 terminal size if Term::ReadKey::GetTerminalSize
    fails.

  2008-03-22 22:22:40 (r2292) by rcaputo; poe/lib/POE.pm M

    Bump up the version for the 2nd pre-release test.

  2008-03-22 22:21:42 (r2291) by rcaputo
  poe/t/30_loops/00_base/wheel_run.pm M

    Tweak the debugging output in a vain attempt to address rt 28804.

  2008-03-19 21:09:44 (r2290) by nothingmuch
  poe/lib/POE/Wheel/Run.pm M; poe/lib/POE/Kernel.pm M

    Demonstrate usage of POE::Kernel->stop in L<POE::Wheel::Run/TIPS AND
    TRICKS>

  2008-03-19 15:38:58 (r2289) by rcaputo
  poe/lib/POE/Kernel.pm M; poe/t/10_units/01_pod/02_pod_coverage.t M

    Solve POD test coverage issues.

  2008-03-14 18:55:13 (r2288) by rcaputo; poe/lib/POE.pm M

    Packaging up a release for the testers.

  2008-03-10 19:59:47 (r2287) by cfedde; poe/Makefile.PL M

    backing out -r2277:2278 sense of network tests restored

  2008-03-10 08:32:44 (r2286) by rcaputo; poe/mylib/Makefile-5005.pm M

    Remove the AF_INET6 warnings when running Makefile.PL on MSWin32.

  2008-03-10 08:11:30 (r2285) by rcaputo
  poe/t/30_loops/00_base/comp_tcp_concurrent.pm M

    Apply SREZIC's patch to work around an issue on 64-bit FreeBSD 6.2.
    Resolves rt.cpan.org #32034.

  2008-03-10 08:04:27 (r2284) by rcaputo
  poe/lib/POE/Resource/Events.pm M; poe/lib/POE/Kernel.pm M

    Apply RDB's patches to update tracing. Closes rt.cpan.org 30342.

  2008-03-10 07:55:29 (r2283) by rcaputo
  poe/lib/POE/Wheel/Run.pm M; poe/t/30_loops/00_base/wheel_readline.pm M;
  poe/t/30_loops/00_base/wheel_curses.pm M

    Use IO::Pty helper methods to condition the slave pty rather than the
    APITUE based low-level calls. Sorry, Mr. Stevens.

  2008-03-10 07:02:30 (r2282) by rcaputo; poe/lib/POE.pm M

    Add a reference to POE's subversion repository, per Zoffix's
    suggestion. Resolves rt.cpan.org #33835.

  2008-03-10 06:33:00 (r2281) by rcaputo; poe/README M

    Minor tweaks.

  2008-03-10 06:08:01 (r2280) by rcaputo
  poe/lib/POE/Loop.pm M; poe/lib/POE/Loop/PerlSignals.pm M;
  poe/lib/POE/Wheel.pm M

    Redocument POE::Loop.

  2008-03-09 21:53:28 (r2279) by rcaputo
  poe/t/30_loops/00_base/k_signals_rerun.pm M;
  poe/t/30_loops/00_base/wheel_run.pm M;
  poe/t/30_loops/00_base/wheel_sf_ipv6.pm M

    Socket6 and Socket both export AF_INET6, but their prototypes don't
    match. Ask Socket6 to keep its symbols to itself.

    Skip even more tests that crash perl on Win32.

  2008-03-08 03:54:34 (r2278) by cfedde; poe/Makefile.PL M

    Make the sense of the network test use positive logic interactive use
    defaults to "do not grant permission" non interactive defaults to
    "grant permission"

  2008-03-08 03:51:14 (r2277) by cfedde
  poe/t/90_regression/agaran-filter-httpd.t M

    #33901: Patch to re-work exit in
    t/90_regression/agaran-filter-httpd.t thanks pravus

  2008-02-29 20:38:17 (r2276) by cfedde
  poe/t/90_regression/cfedde-filter-httpd.t M

    RT#33718: pravus' patch for mst's error report

  2008-02-29 09:00:54 (r2275) by rcaputo
  poe/t/90_regression/steinert-recursive-signal.t D;
  poe/t/30_loops/00_base/k_signals_rerun.pm M;
  poe/t/90_regression/whelan-dieprop.t A; poe/lib/POE/Kernel.pm M;
  poe/MANIFEST M

    Make SIGDIE asynchronous. It will no longer interrupt long chains of
    events (such as _parent/_child chains during session creation and
    destruction).

    Exceptions in _stop is currently untrappable, as the asynchronous
    SIGDIE will have nowhere to be delivered. They will cause POE::Kernel
    to "gracefully" die at the end of _stop's dispatch.

    SIGDIE propagation follows the normal rules with one new addition:
    The signal also proragates from the target session upwards through
    parent, grandparent, and so on. It will not be dispatched to children
    of ancestor sessions. It will not be dispatched to POE::Kernel
    itself.

    Nearly all regression tests pass. The notable exception is Jonathan
    Steinert's recursive signal tests. They have become moot since SIGDIE
    is not recursive anymore.

  2008-02-25 05:56:36 (r2274) by rcaputo
  poe/lib/POE/Pipe.pm M; poe/lib/POE/Resource/FileHandles.pm M

    Add prototypes to F_GETFL and F_SETFL.

  2008-02-20 10:45:43 (r2273) by bingosnet
  poe/t/10_units/02_pipes/02_oneway.t M;
  poe/t/90_regression/cfedde-filter-httpd.t M;
  poe/t/20_resources/00_base/filehandles.pm M;
  poe/t/10_units/02_pipes/03_twoway.t M

    Patch from Martin Ferrari to fix tests not honouring the "don't use
    the network" flag. RT #33276

  2008-02-20 10:18:36 (r2272) by bingosnet; poe/lib/POE/Filter/HTTPD.pm M

    Patch from agaran to fix a problem with his original patch to
    Filter-HTTPD. Adjusted so the tests pass.

  2008-02-19 20:21:48 (r2271) by rcaputo
  poe/lib/POE/Loader.pm M; poe/t/90_regression/rt23181-sigchld-rc.t M;
  poe/t/90_regression/whjackson-followtail.t M

    Add svn:keywords where they are missing.

  2008-02-18 10:45:45 (r2270) by bingosnet
  poe/t/90_regression/agaran-filter-httpd.t A;
  poe/lib/POE/Filter/HTTPD.pm M; poe/MANIFEST M

    Applied a patch to Filter::HTTPD from Maciej Pijanka (agaran).
    Created applicable regression test.

  2008-01-21 19:09:06 (r2269) by rcaputo
  poe/lib/POE/Kernel.pm M; poe/lib/POE/Filter.pm M

    Comment on sub-second accuracy requiring Time::HiRes::time(). Based
    on a discussion with and test case from Sjors Gielen.

  2008-01-12 12:36:44 (r2268) by bingosnet
  poe/MANIFEST M; poe/mylib/PoeBuildInfo.pm M;
  poe/t/90_regression/bingos-followtail.t A

    Added a regression test for the FollowTail problem on *BSD.

  2008-01-11 15:50:02 (r2267) by bingosnet
  poe/lib/POE/Driver.pm M; poe/lib/POE/Wheel/FollowTail.pm M

    Fixed a pod error that was making pod test fail.

    Applied a patch from pravus to fix double InputEvents being triggered
    when tailing an empty file on *BSD.

    <+pravus> basically what i am seeing is that if you create a file and
    stat() it before you

    write to it, rdev will be 0. then, once the file is written to, the
    rdev changes. <+pravus> that's causing Wheel::FollowTail to read
    until EOF (getting the line you just wrote),

    reset the file and then re-read (getting the line again).

  2008-01-07 06:09:02 (r2266) by rcaputo; poe/lib/POE/Driver.pm M

    Redocument.

  2008-01-07 05:23:42 (r2265) by rcaputo; poe/lib/POE/Wheel.pm M

    See also: POE::Driver and POE::Filter.

  2008-01-07 05:21:25 (r2264) by rcaputo; poe/lib/POE/Wheel.pm M

    Discuss how filters and drivers interact with wheels.

  2008-01-06 05:38:06 (r2263) by rcaputo; poe/lib/POE/Wheel.pm M

    Redocument!

  2007-12-12 03:18:11 (r2262) by gwyn17; poe/lib/POE/Session.pm M

    Mid-sized review of POE::Session documentation

  2007-12-12 01:33:51 (r2261) by gwyn17; poe/lib/POE/Session.pm M

    Small POE::Session doco patch from pravus.

  2007-12-09 08:45:26 (r2260) by rcaputo
  poe/lib/POE/Kernel.pm M; poe/lib/POE/Session.pm M

    Finish the first redocumentation pass on POE::Session. Migrate some
    stuff to POE::Kernel. Align Philip Gwyn's changes to POE::Kernel with
    standard style guidelines.

  2007-12-08 12:10:57 (r2259) by gwyn17; poe/lib/POE.pm M

    All L<module> to L<module|module> to prevent pod2html outputing 'the
    module manapge'

  2007-12-08 11:50:48 (r2258) by gwyn17; poe/lib/POE/Kernel.pm M

    Middle-sized documentation review

  2007-12-08 02:27:54 (r2257) by gwyn17; poe/lib/POE.pm M

    Mid-sized documenation update

  2007-12-03 06:45:05 (r2256) by rcaputo; poe/lib/POE/Session.pm M

    Initial stab at _default docs.

  2007-12-03 01:53:58 (r2255) by rcaputo; poe/lib/POE/Session.pm M

    Incorporate some older event parameter constant documentation into
    the recently rewritten version. Mostly by adopting the older
    examples.

  2007-12-02 19:09:24 (r2254) by rcaputo; poe/lib/POE/Session.pm M

    Rewrite docs for postback(), callback() and get_heap().

  2007-12-01 21:05:54 (r2253) by rcaputo; poe/lib/POE/Session.pm M

    Rewrite option() documentation.

  2007-11-23 03:45:35 (r2252) by rcaputo; poe/lib/POE/Session.pm M

    Make more progress on POE::Session

  2007-11-14 03:52:49 (r2251) by immute; poe/MANIFEST M

    Removed the test from MANIFEST.

  2007-11-02 17:44:18 (r2250) by bingosnet; poe/lib/POE/Session.pm M

    Borland C++ doesn't specify ENOSYS in Errno.h. Check for existence of
    ENOSYS and substitute EIO if not found.

  2007-10-26 20:47:22 (r2249) by immute
  poe/t/90_regression/immute-server-tcp-filter.t D

    Removed my old, godawful slow test.

  2007-10-26 07:38:17 (r2248) by rcaputo
  poe/lib/POE/Kernel.pm M; poe/lib/POE.pm M; poe/lib/POE/Session.pm M

    Minor edits to POE and POE::Kernel. Major revisions to POE::Session.
    Still nowhere near done, however.

  2007-10-13 05:36:32 (r2247) by rcaputo; poe/Makefile.PL M

    Oops... reversed the sense of the want-to-run-tests question.

  2007-10-13 05:21:52 (r2246) by rcaputo; poe/MANIFEST M

    Remove some obsolete tests from the manifest.

  2007-10-13 05:18:39 (r2245) by rcaputo
  poe/lib/POE/Resource/Signals.pm M; poe/lib/POE/Resource/Aliases.pm M;
  poe/lib/POE/API/ResLoader.pm D;
  poe/t/10_units/03_base/02_api_resloader.t D;
  poe/lib/POE/Resource/SIDs.pm M; poe/lib/POE/Resource/Sessions.pm M;
  poe/MANIFEST M; poe/lib/POE/Resource/FileHandles.pm M

    Remove POE::API::ResLoader.

  2007-10-13 01:39:58 (r2244) by rcaputo
  poe/lib/POE/Resource/Controls.pm D; poe/lib/POE/Resources.pm M;
  poe/lib/POE/API/Ctl.pm D; poe/t/10_units/03_base/02_api_ctl.t D;
  poe/lib/POE/Kernel.pm M; poe/lib/POE.pm M;
  poe/t/20_resources/00_base/controls.pm D; poe/MANIFEST M

    Remove POE::Api::Ctl per the warning on the mailing list.

  2007-10-08 06:45:59 (r2243) by rcaputo; poe/lib/POE/Kernel.pm M

    Add a print() to the SYNOPSIS.

  2007-10-08 05:26:29 (r2242) by rcaputo; poe/lib/POE/Kernel.pm M

    Finish first pass through POE::Kernel. There are a lot of TODOs
    however.

  2007-10-07 22:06:48 (r2241) by rcaputo; poe/lib/POE/Kernel.pm M

    Document why sessions remain active.

  2007-10-07 03:38:21 (r2240) by rcaputo; poe/lib/POE/Kernel.pm M

    Redocument Kernel accessors and lesser-used utility methods.

  2007-10-07 02:59:02 (r2239) by rcaputo; poe/lib/POE/Kernel.pm M

    Redocument Kernel state accessors.

  2007-10-07 01:43:17 (r2238) by rcaputo; poe/lib/POE/Kernel.pm M

    Redocument reference counters.

  2007-10-03 08:15:02 (r2237) by rcaputo; poe/lib/POE/Kernel.pm M

    Fix a typo.

  2007-10-03 08:13:24 (r2236) by rcaputo; poe/lib/POE/Kernel.pm M

    Fix a couple typos.

  2007-10-03 07:54:55 (r2235) by rcaputo; poe/lib/POE/Kernel.pm M

    Rewrite state() docs.

  2007-10-01 04:12:00 (r2234) by rcaputo; poe/lib/POE/Kernel.pm M

    Make editorial decisions on Matt Sickler's doc contributions (thanks,
    Matt!).

    Do the initial rewrite of the signals docs. There are still some
    straggling TODOs.

  2007-10-01 04:09:10 (r2233) by rcaputo; poe/Makefile.PL M

    Clean up the prompts a little.

  2007-09-27 17:36:26 (r2232) by bingosnet
  poe/lib/POE/Wheel/ReadWrite.pm M

    Fixed a mistake in the documentation

  2007-09-18 03:41:45 (r2231) by immute; poe/lib/POE/Kernel.pm M

    Documentation edits.

  2007-09-09 00:15:32 (r2230) by rcaputo; poe/lib/POE/Kernel.pm M

    Rewrite the I/O watcher methods.

  2007-09-08 08:13:31 (r2229) by rcaputo; poe/lib/POE/Kernel.pm M

    Make a little more progress on the docs.

  2007-08-20 17:34:39 (r2228) by rcaputo; poe/lib/POE/Kernel.pm M

    Make a little more progress on POE::Kernel before going to work.

  2007-08-20 05:38:13 (r2227) by rcaputo; poe/lib/POE/Kernel.pm M

    Partial revision. Documentation is currently complete but the
    structure is broken.

  2007-08-20 05:32:39 (r2226) by teknikill; poe/lib/POE.pm M

    add blurb about myself to the contributers list

  2007-08-19 18:15:59 (r2225) by rcaputo; poe/lib/POE.pm M

    Move the POE.pm use lines to POE::Kernel. Mark that contributors need
    to be documented.

  2007-08-19 13:42:54 (r2224) by bingosnet; poe/lib/POE/Wheel.pm M

    Minor documentation fix to amend typo of filter when wheel was meant

  2007-08-19 05:57:08 (r2223) by rcaputo; poe/lib/POE.pm M

    Document "use POE" magic.

  2007-08-19 05:02:19 (r2222) by rcaputo
  poe/lib/POE/Loop/IO_Poll.pm M; poe/lib/POE/Resource/Extrefs.pm M;
  poe/lib/POE/Wheel/Run.pm M; poe/lib/POE/Resource/Aliases.pm M;
  poe/lib/POE/Filter/HTTPD.pm M; poe/lib/POE/Filter/Line.pm M;
  poe/lib/POE/Kernel.pm M; poe/lib/POE/Session.pm M;
  poe/lib/POE/Loop/TkActiveState.pm M; poe/lib/POE/Wheel/SocketFactory.pm
  M; poe/lib/POE/Wheel/ReadLine.pm M; poe/lib/POE/Resource/Signals.pm M;
  poe/lib/POE/Loop/Select.pm M; poe/lib/POE/Resource/Events.pm M;
  poe/lib/POE/Resource/Sessions.pm M; poe/lib/POE/Queue/Array.pm M;
  poe/lib/POE/Resource/FileHandles.pm M

    Replace the symbol of Eris with the usual TODO.

  2007-08-19 02:45:16 (r2221) by rcaputo; poe/lib/POE/Filter/Block.pm M

    Technically it is not a TODO.

  2007-08-19 00:59:33 (r2220) by rcaputo
  poe/lib/POE/Component/Client/TCP.pm M; poe/lib/POE/Loader.pm M;
  poe/lib/POE/Wheel/Curses.pm M; poe/lib/POE/Filter/HTTPD.pm M;
  poe/lib/POE/Loop.pm M; poe/lib/POE/Resources.pm M;
  poe/lib/POE/Filter/Stackable.pm M; poe/lib/POE/Wheel/FollowTail.pm M;
  poe/lib/POE/Kernel.pm M; poe/lib/POE/Filter/Stream.pm M;
  poe/lib/POE/Resource/Statistics.pm M; poe/lib/POE/Wheel/ListenAccept.pm
  M; poe/lib/POE/Session.pm M; poe/lib/POE/Filter.pm M;
  poe/lib/POE/Wheel/ReadLine.pm M; poe/lib/POE/API/ResLoader.pm M;
  poe/lib/POE/Resource/SIDs.pm M; poe/lib/POE/API/Ctl.pm M;
  poe/lib/POE/Resource/Events.pm M; poe/lib/POE/Pipe.pm M;
  poe/lib/POE/Resource/Sessions.pm M; poe/lib/POE/Component/Server/TCP.pm
  M; poe/lib/POE.pm M; poe/lib/POE/Queue/Array.pm M;
  poe/lib/POE/Resource/Extrefs.pm M; poe/lib/POE/Resource/Aliases.pm M;
  poe/lib/POE/Wheel/Run.pm M; poe/lib/POE/Filter/Line.pm M;
  poe/lib/POE/Filter/RecordBlock.pm M; poe/lib/POE/Filter/Block.pm M;
  poe/lib/POE/Driver/SysRW.pm M; poe/lib/POE/Filter/Grep.pm M;
  poe/lib/POE/Filter/Reference.pm M; poe/lib/POE/Wheel/SocketFactory.pm
  M; poe/lib/POE/Queue.pm M; poe/lib/POE/Pipe/TwoWay.pm M;
  poe/lib/POE/Resource/Signals.pm M; poe/lib/POE/Resource.pm M;
  poe/lib/POE/Resource/Controls.pm M; poe/lib/POE/Wheel/ReadWrite.pm M;
  poe/lib/POE/Driver.pm M; poe/lib/POE/Pipe/OneWay.pm M;
  poe/lib/POE/Filter/Map.pm M; poe/lib/POE/Component.pm M;
  poe/lib/POE/NFA.pm M; poe/lib/POE/Wheel.pm M;
  poe/lib/POE/Resource/FileHandles.pm M

    Mark files that need to be redocumented.

  2007-08-19 00:54:55 (r2219) by rcaputo
  poe/lib/POE/Loop/IO_Poll.pm M; poe/lib/POE/Loop/Tk.pm M;
  poe/lib/POE/Loop/Select.pm M; poe/lib/POE/Loop/Event.pm M;
  poe/lib/POE/Loop/PerlSignals.pm M; poe/lib/POE/Loop/Gtk.pm M;
  poe/lib/POE/Loop/TkCommon.pm M; poe/lib/POE/Loop/TkActiveState.pm M

    Need to redocument these. Also added configuration info. for the
    common loop tests when they are distributed separately.

  2007-08-18 06:26:43 (r2218) by rcaputo; poe/lib/POE.pm M

    Revise documentation in a major way.

  2007-08-11 09:10:38 (r2208) by rcaputo
  poe/t/30_loops/00_base/comp_tcp_concurrent.pm M;
  poe/t/30_loops/00_base/all_errors.pm M;
  poe/t/30_loops/00_base/comp_tcp.pm M

    Set proper skip messages.

  2007-07-25 04:51:09 (r2207) by rcaputo; poe/mylib/PoeBuildInfo.pm M

    According to at least one CPAN tester we cannot assume Storable
    exists. Explicitly require it. Partially addresses
    http://www.nntp.perl.org/group/perl.cpan.testers/2007/06/msg515661.ht
    ml

  2007-07-25 04:44:21 (r2206) by rcaputo
  poe/t/30_loops/00_base/comp_tcp_concurrent.pm M;
  poe/t/90_regression/cfedde-filter-httpd.t M;
  poe/t/30_loops/00_base/wheel_accept.pm M;
  poe/t/90_regression/immute-server-tcp-filter.t M;
  poe/t/30_loops/00_base/comp_tcp.pm M

    According to the CPAN testers (bless their souls), POE 0.9989 failed
    at least four times due to bind() errors. Mainly addresses already in
    use, which implies some rapid-fire testing. Here I'm patching the
    tests that failed to bind to a random free port rather than hardcoded
    numbers. They should be a lot more resilient to this sort of error.

  2007-07-24 15:44:37 (r2205) by rcaputo
  poe/lib/POE/Wheel/SocketFactory.pm M

    Applied Markus Jansen's patch to prevent a memory leak when Socket6
    is not present on Solaris 2.6.

  2007-07-23 18:08:46 (r2204) by rcaputo
  poe/t/30_loops/00_base/k_alarms.pm M

    Attempt to make the timing more stable in t/30_loops/*/k_alarms.t
    tests. Slow machines, especially running emulators within
    virtualizations, get the time wrong sometimes.

===================================
2007-07-23T07:25:45.773662Z v0_9999
===================================

  2007-07-23 07:25:09 (r2202) by rcaputo; poe/lib/POE.pm M

    Bump up the version.

  2007-07-23 07:22:57 (r2201) by rcaputo
  poe/lib/POE/Resource/Events.pm M; poe/lib/POE/Loop/TkCommon.pm M

    Resolve a CPU spinning issue when using the Tk event loop without any
    event watchers. Resolves ticket 27359 reported by Craig at Lucent.
    Others have also reported this issue, but not in a ticket. :)

  2007-07-23 01:37:08 (r2200) by rcaputo
  poe/lib/POE/Loop/TkActiveState.pm M

    Avoid a warning about ambiguous use of select().

  2007-07-22 08:37:07 (r2199) by rcaputo
  poe/lib/POE/Wheel/ReadWrite.pm M

    Apply Gareth's patch for an optional synchronous put() and a flush()
    method with which to flush the wheel's buffer on command. Resolves
    rt.cpan.org ticket 20328.

  2007-07-22 06:15:17 (r2198) by rcaputo
  poe/t/30_loops/00_base/wheel_run.pm M

    Add a brief sleep to the child program to allow pipes to be flushed
    before the program exits. Andreas Koenig brought this to my attention
    via CPAN test reports and rt.cpan.org ticket 27478.

  2007-07-22 02:21:32 (r2197) by rcaputo; poe/lib/POE/Queue/Array.pm M

    Mark Jason Dominus' binary insert example (Higher Order Perl, p. 292)
    inspired me to benchmark POE::Queue::Array against his algorithm. The
    linear seeks become slower after queues of depth 4. They're
    neck-and-neck for shorter ones. POE::Queue::Array's algorithm is a
    little faster otherwise, probably due to the special cases it
    implements.

    This patch removes the linear search cases for "small" queues. They
    were not being helpful. It also simplifies the binary search code,
    which is always a plus (Perl runs faster if it runs less code).
    Finally, it simplifies the remaining necessary linear search code...
    but that's after a binary seek anyway.

    Net loss: 94 lines of code. All tests still pass.

  2007-07-21 07:04:55 (r2196) by rcaputo
  poe/lib/POE/Component/Server/TCP.pm M

    Resolve rt.cpan.org ticket 27833. Philip Gwyn found a case where the
    connection count for a server could go negative. Not-shutting-down
    hilarity ensues in this case. I've just applied his patch.

  2007-07-21 06:52:55 (r2195) by rcaputo
  poe/t/30_loops/00_base/wheel_sf_ipv6.pm M

    Eric Hacker found a case where the IPv6 tests failed when AF_INET6
    isn't supplied by Socket6. This commit wraps the use of that constant
    in a block eval and skips all the tests if there's a failure. At
    least I hope that's the case. Should resolve rt.cpan.org ticket
    25879.

  2007-07-21 06:51:26 (r2194) by rcaputo
  poe/lib/POE/Wheel/SocketFactory.pm M

    Oops. Remove tabs.

  2007-07-21 06:40:31 (r2193) by rcaputo
  poe/lib/POE/Wheel/SocketFactory.pm M

    Per STRO's suggestion, test and provide dummies for the IPv6
    constants separately. Resolves rt.cpan.org ticket 27250.

  2007-07-04 21:31:15 (r2192) by rcaputo
  poe/lib/POE/Resource/Signals.pm M

    Yuval Kogman discovered that sig_child() events were not triggering
    after a second POE::Kernel->run() in the same program. We tracked
    this down to a static compile-time initialization. Moving that into
    POE::Resource::Signals::_data_sig_initialize() fixed the problem
    right up.

  2007-06-14 01:26:41 (r2191) by rcaputo; poe/mylib/PoeBuildInfo.pm M

    Rob Bloodgood found a new deprecation. This change works around it.

  2007-05-24 05:52:46 (r2190) by immute; poe/lib/POE/Wheel/ReadWrite.pm M

    Added handle accessors to Wheel::ReadWrite

  2007-05-08 20:32:50 (r2189) by rcaputo
  poe/lib/POE/Wheel/SocketFactory.pm M

    Shevek suggested that Fcntl be used to import F_GETFL, F_SETFL and
    O_NONBLOCK since POSIX doesn't always do it. Tests here show that
    using both doesn't throw warnings, so there doesn't seem to be harm
    in it.

  2007-04-29 06:38:11 (r2188) by rcaputo
  poe/lib/POE/Filter/HTTPD.pm M; poe/t/10_units/05_filters/03_http.t M

    Marlon Bailey discovered that POE::Filter::HTTPD threw incorrect
    objects when methods were of mixed case. Such as "GEt". This commit
    includes a test for the proper behavior and a fix. NB: You'll get
    back an HTTP::Request object with a mixed-case method. The filter
    does NOT force a particular case here.

  2007-04-19 21:35:33 (r2187) by rcaputo
  poe/mylib/Makefile-5004.pm M; poe/lib/POE/Wheel/ReadLine.pm M;
  poe/mylib/Makefile-5005.pm M; poe/lib/POE/Filter/HTTPD.pm M;
  poe/lib/POE/Filter/Block.pm M; poe/MANIFEST M;
  poe/lib/POE/Driver/SysRW.pm M; poe/lib/POE/Filter/Reference.pm M;
  poe/lib/POE/Wheel/SocketFactory.pm M

    I read on perl5-porters this trick to conditionally "use bytes"
    without source filtration. It works! And it's lexical! ZOMG! This
    commit replaces the stinky make-time source filter we were using with
    the compile-time trick. Thanks go out to Phil Harvey for coming up
    with the idea.

  2007-04-06 19:53:33 (r2186) by rcaputo
  poe/lib/POE/Wheel/SocketFactory.pm M

    blocking() appears to be a FileHandle method, not an IO::Handle
    method. Use FileHandle explicitly to ensure it's loaded.

  2007-03-30 05:09:23 (r2185) by rcaputo
  poe/lib/POE/Component/Server/TCP.pm M

    Cosmetic fixes in the high/low water mark code.

(he)

Add p5-Term-Cap.

(he)

Import p5-Term-Cap version 1.12.

This is a perl module providing an interface to the termcap database.


Status:

Vendor Tag: TNV
Release Tags: pkgsrc-base

(he)

Update darkstat to 3.0.711 by mainteiner update request via PR 39401.

Changes since 3.0.708:
- Split --debug into --verbose and --no-daemon
- Include launchd config and instructions for running darkstat
  on Mac OS X.  Contributed by Damien Clauzel.
- Implement PPPoE decoding on ethernet iface. (--pppoe)
- Web: Add automatic reload button.  Thanks Dennis!
- Web: Add a graph legend with min/avg/max.
- Web: Remove hashtable stats pages.

(obache)

Update from version 0.98nb1 to 1.00, discussed in general terms with maintainer.

Pkgsrc changes:
o Canonicalize HOMEPAGE
o Adjust dependencies to match Makefile.PL

Upstream changes (at least I *think* these are the ones for 0.99 and 1.00):

======================================
9999-99-99 99:99:99.999999Z (untagged)
======================================

  2007-01-06 18:12:47 (r64) by rcaputo; DNS.pm M

    By the power of PAUSEskull, this is a fine morning! Release 1.00!

  2007-01-06 18:11:52 (r63) by rcaputo; DNS.pm M

    Expose the underlying Net::DNS::Resolver with a get_resolver()
    accessor. Thanks to Chris Williams for kinda suggesting it. :)

  2006-11-06 19:23:35 (r62) by rcaputo; Makefile.PL M

    Set a LICENSE. Cheap kwalitee points!

  2006-10-17 16:35:24 (r61) by rcaputo; t/06_hosts.t M

    Don't try to call a method on a failed request. Resolves rt.cpan.org
    21190 by Alexandr Ciornii.

  2006-10-17 15:05:34 (r60) by rcaputo; Makefile.PL M

    Some rt.cpan.org tickets refer to issues in Net::DNS. Require the
    latest version of that dependency in the hopes that some things are
    fixed. Too bad I don't have test cases for those issues... there's no
    way for me to tell if anything's fixed by this simple change.

=========================
2006-05-21 20:45:46 v0_99
=========================

  2006-05-21 20:45:11 (r58) by rcaputo; DNS.pm M

    Bump up the version.

  2006-05-21 20:44:39 (r57) by rcaputo; DNS.pm M

    Fix shutdown(). It was not performing nearly enough cleanup.

    Replace some post() calls with call() to avoid race conditions. In
    general, method interfaces should use call() rather than post() so
    that they affect internal structures synchronously.

  2006-05-21 20:43:08 (r56) by rcaputo
  t/02_tag_args.t M; t/03_api_3.t M; t/05_api_4.t M; t/06_hosts.t M

    Turn on ASSERT_DEFAULT for some of the tests. Add _stop handlers so
    they don't fail with all POE asserts on.

  2006-03-24 03:58:51 (r55) by rcaputo; DNS.pm M

    Belatedly bump up the version.

(he)

Updated print/teTeX3-bin to 3.0nb18.

(tonnerre)

If mkdtemp is not defined, define our own version and deploy it. Also, don't
just add mkdtemp to configure.in but also make sure that it is heard.
PKGREVISION++.

(tonnerre)