Re: [ANNOUNCE] GIT 1.2.0

Re: [ANNOUNCE] GIT 1.2.0

[H. Peter Anvin]

Junio C Hamano wrote:
> The latest feature release GIT 1.2.0 is available at the
> usual places:
> 
> 	http://www.kernel.org/pub/software/scm/git/
> 
> 	git-1.2.0.tar.{gz,bz2}			(tarball)
> 	RPMS/$arch/git-*-1.2.0-1.$arch.rpm	(RPM)
> 
> Right now binary RPM is available only for x86_64, because I do
> not have an access to RPM capable i386 box.
> 

You can build the i386 binary rpms on hera as such:

rpmbuild --rebuild --target i386 git-1.2.0-1.src.rpm

I had to install openssl-devel from the i386 distribution, which for 
some reason wasn't part of the x86-64 distribution, but that's now taken 
care of.

	-hpa

Re: [ANNOUNCE] GIT 1.2.0

[Junio C Hamano]

"H. Peter Anvin" <hpa@zytor.com> writes:

> Junio C Hamano wrote:
>> The latest feature release GIT 1.2.0 is available at the
>> usual places:
>> 	http://www.kernel.org/pub/software/scm/git/
>> 	git-1.2.0.tar.{gz,bz2}			(tarball)
>> 	RPMS/$arch/git-*-1.2.0-1.$arch.rpm	(RPM)
>> Right now binary RPM is available only for x86_64, because I do
>> not have an access to RPM capable i386 box.
>>
>
> You can build the i386 binary rpms on hera as such:
>
> rpmbuild --rebuild --target i386 git-1.2.0-1.src.rpm
>
> I had to install openssl-devel from the i386 distribution, which for
> some reason wasn't part of the x86-64 distribution, but that's now
> taken care of.

OK.  Maybe I will when I find time, but not right now.

By the way, did I screw up the original announce?  I do not see
the message come back to me from vger...

Re: [ANNOUNCE] GIT 1.2.0

[Petr Baudis]

Dear diary, on Sun, Feb 12, 2006 at 10:53:13PM CET, I got a letter
where Junio C Hamano <junkio@cox.net> said that...
> By the way, did I screw up the original announce?  I do not see
> the message come back to me from vger...

I didn't receive it either, and neither my git-bisect patch.

-- 
				Petr "Pasky" Baudis
Stuff: http://pasky.or.cz/
Of the 3 great composers Mozart tells us what it's like to be human,
Beethoven tells us what it's like to be Beethoven and Bach tells us
what it's like to be the universe.  -- Douglas Adams

[ANNOUNCE] GIT 1.2.0

[Junio C Hamano]

A resend.  I am suspecting vger does not want two lists hosted
there both named on To: line...

-- >8 --
The latest feature release GIT 1.2.0 is available at the
usual places:

	http://www.kernel.org/pub/software/scm/git/

	git-1.2.0.tar.{gz,bz2}			(tarball)
	RPMS/$arch/git-*-1.2.0-1.$arch.rpm	(RPM)

Right now binary RPM is available only for x86_64, because I do
not have an access to RPM capable i386 box.


What's new in 1.2.0

The new release 1.2.0 took about one month since 1.1.0 was released.
It has literally _tons_ of updates since the last maintenance release
1.1.6.  While people who have been following the master branch are
probably familiar with most of them, I expect others would be quite
surprised by the amount of changes.  Hopefully the surprise is of a
pleasant kind.

Here is the list of all the good stuff (not counting what are
already in the 1.1.X maintenance series).

* Major features

 - Checkout -m option.  This allows you to switch branches when
   you have local changes to paths that are different in the
   current branch and new branch.  The local changes are merged to
   the version from the branch you are switching to and the merge
   result is left in your working tree (you may see conflicts).

 - Add -c and --cc to diff-tree and diff-files.  They give the
   "combined diff" output that shows merges more human readably,
   and the "git diff" wrapper uses --cc by default.  This release
   also comes with updated gitk (Paul Mackerras) that takes advantage
   of this feature.

 - git-commit updates.

   - Allow git-commit from a subdirectory.

   - Aborted "git-commit -a" leaves the index as it was.

   - "git commit --only paths..." checks in changes to only
     named paths.

     This will become the default, instead of the traditional
     --include semantics, during development track towards
     1.3.0, so if you are used to the --include semantics,
     please start training your fingers to explicitly say
     "git commit --include paths...".

   - Add --author='A U Thor <author@example.com>' command
     line option.

   - "git commit -v" seeds the end of the commit log message editor
     with the patch to be committed.

 - git-status updates.  Now it takes the exactly same set of parameters
   as git-commit takes, and serves as a way to preview what you would
   be committing.  The -v option to get diff output is also available.

* Portability

  FreeBSD support in the main Makefile (Alecs King, Alex Riesen)
  Run GIT-VERSION-GEN with $(SHELL), not sh (Jason Riedy)
  compat/unsetenv.c (Jason Riedy)
  stat() to work around nonportable EEXIST (Jason Riedy)
  Asciidoc futureproofing (Pavel)
  Disable USE_SYMLINK_HEAD by default (Pavel)
  Various fixes for Cygwin around dirent and sockaddr_storage.

* Foreign SCM interface

  cvs-import updates (Andreas Ericsson and Martin)
  cvs-exportcommit fixes and docs (Martin)
  send-email various cleanups (Ryan)
  svnimport internal arglist fix (Sasha Khapyorsky)
  svnimport relative path fix (Christian Biesinger)

* Samples and Docs

  New tutorial (J. Bruce Fields)
  Update hook example (Andreas Ericsson)
  Topic-branch howto updates (Tony Luck)
  Other misc. cleanups (Florian Weimer, J. Bruce Fields, Petr Baudis, Jon Loeliger)
  Document git-diff-tree --always (Petr Baudis)
  Basic documentation for git-show (Petr Baudis)

* Cleanups

  Use tree_desc tree parser in tar-tree (Daniel, Linus)
  Use sha1_file.c's mkdir-like routine in apply.c (Jason Riedy)
  Use adler32() from zlib instead of defining our own (Peter Eriksen)
  Assorted cleanups (Uwe Zeisberger)
  Make --abbrev more consistent with diff-tree
  Omit duplicated parents from rev-list --parents output
  Clean up rev-parse error checking (Linus)

* Usability improvements

  git-apply -pN (Daniel)
  git-rev-list -n1 and -1 (Eric Wong)
  diff-tree --always flag (Linus)
  git-fetch -k (Tom Prince)
  git-describe: default to HEAD (Andreas Ericsson and me)
  git-clone --bare
  git-show and git-whatchanged (Linus and me)
  git-daemon user-relative paths updates (Mark Wooding and me)
  git-rerere reuse recorded resolve.
  git-fmt-merge-msg updates.
  git-format-patch: always --mbox and show sane Date (Eric W. Biederman et al)
  git-show-branch assorted updates
  git-commit seeds the message with list of conflicted files (Linus)
  git-clone and git-fetch shows progress information (Linus and me)
  git-repo-config allows type specifiers, does not dump core if bool (Pasky)

* Fixes

  git-format-patch -s (Eric W. Biederman)
  git-whatchanged: exit out early on errors (Linus)
  Various http-fetch fixes (Mark Wooding, Nick Hengeveld)
  Local push/pull env cleanup (Matt Draisey)
  Exec git programs without using PATH (Michal Ostrowski)
  Pass --upload-pack to fetch-pack and friends (Michal Ostrowski)
  Allow diff and index commands to be interrupted (Petr Baudis)
  Fix "git-push --tags".
  Parse sha1^12 for 12th parent correctly.
  git-diff-tree --stdin uses given parents by git-rev-list --parents.
  Show grafted parents in rev-list --parents.
  git-ls-files -o reads from --exclude-per-dir from upper directories.
  do not allow empty author/committer name/email.
  call setup_git_directory() before git_config().
  futureproof and optimize delta unpacker (Nicolas Pitre)

* Performance

  merge-recursive (Fredrik)
  git-read-tree --aggressive
  optimize objects bookkeeping for 5x speedup (Johannes)

For the full list of changes above, see:

	$ git log ^v1.1.6 v1.1.0..v1.2.0

Re: [ANNOUNCE] GIT 1.2.0

[Junio C Hamano]

Petr Baudis <pasky@suse.cz> writes:

> Dear diary, on Sun, Feb 12, 2006 at 10:53:13PM CET, I got a letter
> where Junio C Hamano <junkio@cox.net> said that...
>> By the way, did I screw up the original announce?  I do not see
>> the message come back to me from vger...
>
> I didn't receive it either, and neither my git-bisect patch.

Hmph, maybe there is a hiccup at the majordomo?  I did not hear
much about that from HPA either but hopefully mine I now know
how to work it around.  Earlier I had git on To: and kernel on CC:
but this time I felt the announcement was important enough that
I had both on To: which I did not see come back.  A later resend
in my usual way made it through, so I'll stick to it as a
workaround for now.

Your patch e-mail, I _did_ get two copies, one direct and one
via vger, so it might be something else.

Anyhow, I am done for the day (it turned out to be a loooong
one), and I'll reopen "master" and "maint" after I get some
rest.

    X-From-Line: git-owner@vger.kernel.org Sun Feb 12 16:27:05 2006
    Received: from pop.west.cox.net (68.6.19.2) by osiris with POP3 for
      <postmaster@osiris>; 12 Feb 2006 16:27:05 -0000
    Return-Path: <git-owner@vger.kernel.org>
    Received: from fed1rmimpi03.cox.net ([70.169.32.70])
              by fed1rmmtai06.cox.net
              (InterMail vM.6.01.05.02 201-2131-123-102-20050715) with ESMTP
              id <20060212160519.RZSC2846.fed1rmmtai06.cox.net@fed1rmimpi03.cox.net>;
              Sun, 12 Feb 2006 11:05:19 -0500
    Received: from vger.kernel.org ([209.132.176.167])
            by fed1rmimpi03.cox.net with IMP
            id wg5C1T01W3d4pq90000000
            for nrsolis@cox.net; Sun, 12 Feb 2006 11:05:17 -0500
    Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
            id S1751138AbWBLQFj (ORCPT <rfc822;junkio@cox.net> + 2 others);
            Sun, 12 Feb 2006 11:05:39 -0500
    Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751139AbWBLQFj
            (ORCPT <rfc822;git-outgoing>); Sun, 12 Feb 2006 11:05:39 -0500
    Received: from w241.dkm.cz ([62.24.88.241]:40140 "EHLO machine.or.cz")
            by vger.kernel.org with ESMTP id S1751138AbWBLQFi (ORCPT
            <rfc822;git@vger.kernel.org>); Sun, 12 Feb 2006 11:05:38 -0500
    Received: (qmail 512 invoked by uid 2001); 12 Feb 2006 17:06:14 +0100
    Date:	Sun, 12 Feb 2006 17:06:14 +0100
    From:	Petr Baudis <pasky@suse.cz>
    To:	junkio@cox.net
    Cc:	git@vger.kernel.org
    Subject: [PATCH] Properly git-bisect reset after bisecting from non-master head
    Message-ID: <20060212160614.GV31278@pasky.or.cz>

Re: [ANNOUNCE] GIT 1.2.0

[Petr Baudis]

Dear diary, on Sun, Feb 12, 2006 at 11:14:43PM CET, I got a letter
where Junio C Hamano <junkio@cox.net> said that...
> For the full list of changes above, see:
> 
> 	$ git log ^v1.1.6 v1.1.0..v1.2.0

Actually, you can do just

 	$ git log v1.1.6..v1.2.0

since v1.1.6 is a superset of v1.1.0. It is like "list all new patches I
will get when I update from v1.1.6 to v1.2.0".

-- 
				Petr "Pasky" Baudis
Stuff: http://pasky.or.cz/
Of the 3 great composers Mozart tells us what it's like to be human,
Beethoven tells us what it's like to be Beethoven and Bach tells us
what it's like to be the universe.  -- Douglas Adams

Re: [ANNOUNCE] GIT 1.2.0

[Johannes Schindelin]

Hi,

On Sun, 12 Feb 2006, Junio C Hamano wrote:

> What's new in 1.2.0
> 
> The new release 1.2.0 took about one month since 1.1.0 was released.
> It has literally _tons_ of updates since the last maintenance release
> 1.1.6.

Wow. It did not seem like soooo much reading the mailing list (... and I 
thought git development would slow down a bit after 1.0 ;-)). Thank you 
very much, Junio. You are doing a very good job!

>   optimize objects bookkeeping for 5x speedup (Johannes)

Note: it is only 2x here. Also, thank you very much for this wonderful 
word, which contains three consecutive pairs of letters!

Ciao,
Dscho