ostree v2014.1

[Colin Walters <walters verbum org>]

Hi all,

Now that it's a new year, time for a new OSTree release, and it's also
time to roll over that first big digit in the OSTree version number.  In
case you haven't noticed, I believe version numbers are meaningless.
They are an attempt by our software industry to convince ourselves that
newer is better, when in reality it's often not.  Not without deep,
rigorous investment in automated testing, focus on quality over
flexibility, and high speed feedback from real world deployments.
Deployments need the ability to try new code, and roll back - something
OSTree was designed from the start to make easy and painless, in
contrast to dpkg/rpm which are obsessed with incrementing these version
numbers.

But philosophy aside =) OSTree v2014.1 is now available:

https://git.gnome.org/browse/ostree/tag/?id=v2014.1
http://ftp.acc.umu.se/pub/GNOME/sources/ostree/

Let me highlight a few things.  First of all, I'm pretty sure I finally
squashed a race condition in the HTTP pull code:

https://bugzilla.gnome.org/show_bug.cgi?id=706456

This wouldn't leave your system corrupted exactly, but it would require
running a few commands as root to unwedge it.  I know OSTree comes from
the Linux world where running obscure commands as root to deal with
upgrade fallout is the norm, but I'm happy to show progress on breaking
with that tradition ;)

I blogged about "ostree admin switch":
http://blog.verbum.org/2014/01/21/ostree-in-action-rpm-ostree-and-switching-trees/

Let's see...it's also cool to see the contribution from Daniel for Arch
Linux's "mkinitcpio".  What is he doing with that?  Is Arch Linux
engineering investigating OSTree?  I have no idea...that's the fun thing
about Free Software, you often never know who all is using it and why.

There is some initial support for extended metadata from Jeremy; I've
been holding off on more of this to think about it.  At present, I'm now
thinking we may be able to get by with "static deltas".  More on that in
https://bugzilla.gnome.org/show_bug.cgi?id=721799
and
https://bugzilla.gnome.org/show_bug.cgi?id=709050

Thanks to all contributors!
 
Colin Walters (50):
      Update libgsystem
      Add OstreeAsyncProgress, use it for ostree_repo_pull
      libostree: Squash two g-ir-scanner warnings
      libostree: Squash a compiler warning
      libostree: Remove extra : in comment
      main: Treat default osname more consistently
      Update libgsystem
      fetcher: Display incremental download progress
      ostree-remount.service: Run before systemd-random-seed.service
      Move xattr handling into libgsystem, fix sysroot to handle directory ownership/perms
      Update libgsystem
      TODO: Add a few items
      repo-pull: Remove leftover unused %s
      repo: Provide a friendlier error message if no signatures are found
      prune: Remove redundant 'bytes' in format string
      repo: Introspect _sign_commit() and _verify_commit()
      ostree/remote-add: Add --no-gpg-verify option
      Remove 'write-refs' builtin
      build: Add Makefile-decls.am
      Makefile-decls.am: Some minor fixes from hotssh
      sysroot: Error out on deploy --os=<unknown> which has not been initialized
      pull: Ensure we begin a status line
      build: Don't use += for ACLOCAL_AMFLAGS
      deploy: Write bootloader config even if just kernel arguments changed
      admin: Add --karg-proc-cmdline argument
      Update libgsystem
      Unify uboot and syslinux test cases
      COPYING: Update to latest FSF with current address
      deploy: Rework kernel arguments, add --karg-append to "admin deploy"
      commit: Reject non-regular/non-symlinks earlier with better error message
      packaging: Update infrastructure
      libotutil: Fix a possible uninitialized free() in error path
      Squash some harmless compiler warnings
      admin/switch: New builtin to switch between trees
      test-sysroot: Update for API change
      deploy/main: Unify some bits between admin-switch and admin-upgrade
      Update .gitignore
      tests: Small tweaks to pull corruption test
      pull: Add a few more g_debug() and assertions
      libostree: Ensure we set standard::type when querying files
      tests/pull-corruption: Ensure we corrupt an object to be pulled
      pull: Hopefully squash race where we would exit early
      pull: Close another race
      repo: Document ostree_repo_sign_commit(), and add introspection data
      README: Just link to wiki, move most of it to README-historical.md
      pull: Add remotename:ref syntax
      tests: Only install test-corruption if gjs is found
      TODO: Update
      pull: Be less chatty with G_MESSAGES_DEBUG=all
      Release 2014.1

Daniel Narvaez (3):
      Add support for mkinitcpio
      main/remote: Add a show-url operation to the remote command
      Fix crash when deploying with implicit os name

Jeremy Whiting (1):
      core: Add size information to commit metadata

Roddy Shuler (1):
      core: Set mtime of content objects to 0