mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Linus Torvalds <torvalds@osdl.org>
To: Paul Jackson <pj@engr.sgi.com>
Cc: junkio@cox.net, rddunlap@osdl.org, ross@jose.lug.udel.edu,
	linux-kernel@vger.kernel.org
Subject: Re: more git updates..
Date: Sun, 10 Apr 2005 13:57:33 -0700 (PDT)	[thread overview]
Message-ID: <Pine.LNX.4.58.0504101338360.1267@ppc970.osdl.org> (raw)
In-Reply-To: <20050410115055.2a6c26e8.pj@engr.sgi.com>



On Sun, 10 Apr 2005, Paul Jackson wrote:
> 
> Ah ha - that explains the read-tree and write-tree names.
> 
> The read-tree pulls stuff out of this file system into
> your working files, clobbering local edits.  This is like
> the read(2) system call, which clobbers stuff in your
> read buffer.

Yes. Except it's a two-stage thing, where the staging area is always the 
"current directory cache".

So a "read-tree" always reads the tree information into the directory 
cache, but does not actually _update_ any of the files it "caches". To do 
that, you need to do a "checkout-cache" phase.

Similarly, "write-tree" writes the current directory cache contents into a
set of tree files. But in order to have that match what is actually in
your directory right now, you need to have done a "update-cache"  phase
before you did the "write-tree".

So there is always a staging area between the "real contents" and the 
"written tree". 

> That way of thinking really doesn't work well here.
> 
> I will have to look more closely at pasky's GIT toolkit
> if I want to see an SCM style interface.

Yes. You really should think of GIT as a filesystem, and of me as a 
_systems_ person, not an SCM person. In fact, I tend to detest SCM's. I 
think the reason I worked so well with BitKeeper is that Larry used to do 
operating systems. He's also a systems person, not really an SCM person. 
Or at least he's in between the two.

My operations are like the "system calls". Useless on their own: they're
not real applications, they're just how you read and write files in this
really strange filesystem. You need to wrap them up to make them do
anything sane.

For example, take "commit-tree" - it really just says that "this is the 
new tree, and these other trees were its parents". It doesn't do any of 
the actual work to _get_ those trees written.

So to actually do the high-level operation of a real commit, you need to
first update the current directory cache to match what you want to commit
(the "update-cache" phase).

Then, when your directory cache matches what you want to commit (which is
NOT necessarily the same thing as your actual current working area - if
you don't want to commit some of the changes you have in your tree, you
should avoid updating the cache with those changes), you do stage 2, ie
"write-tree". That writes a tree node that describes what you want to
commit.

Only THEN, as phase three, do you do the "commit-tree". Now you give it 
the tree you want to commit (remember - that may not even match your 
current directory contents), and the history of how you got here (ie you 
tell commit what the previous commit(s) were), and the changelog. 

So a "commit" in SCM-speak is actually three totally separate phases in my
filesystem thing, and each of the phases (except for the last
"commit-tree" which is the thing that brings it all together) is actually
in turn many smaller parts (ie "update-cache"  may have been called
hundreds of times, and "write-tree" will write several tree objects that
point to each other).

Similarly, a "checkout" really is about first finding the tree ID you want
to check out, and then bringing it into the "directory cache" by doing a
"read-tree" on it. You can then actually update the directory cache 
further: you might "read-tree" _another_ project, or you could decide that 
you want to keep one of the files you already had.

So in that scneario, after doing the read-tree you'd do an "update-cache"
on the file you want to keep in your current directory structure, which
updates your directory cache to be a _mix_ of the original tree you now
want to check out _and_ of the file you want to use from your current
directory. Then doing a "checkout-cache -a" will actually do the actual
checkout, and only at that point does your working directory really get
changed.

Btw, you don't even have to have any working directory files at all. Let's
say that you have two independent trees, and you want to create a new
commit that is the join of those two trees (where one of the trees take
precedence). You'd do a "read-tree <a> <b>", which will create a directory
cache (but not check out) that is the union of the <a> and <b> trees (<b>
will overrride). And then you can do a "write-tree" and commit the
resulting tree - without ever having _any_ of those files checked out. 

		Linus

  reply	other threads:[~2005-04-10 20:55 UTC|newest]

Thread overview: 179+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2005-04-09 19:45 Linus Torvalds
2005-04-09 19:56 ` Linus Torvalds
2005-04-09 20:07 ` Petr Baudis
2005-04-09 21:00   ` Linus Torvalds
2005-04-09 21:00     ` tony.luck
2005-04-10 16:01       ` Linus Torvalds
2005-04-12 17:34         ` Helge Hafting
2005-04-10 18:19       ` Paul Jackson
2005-04-10 23:04         ` Bernd Eckenfels
2005-04-11  9:27           ` Anton Altaparmakov
2005-04-09 21:08     ` Linus Torvalds
2005-04-09 23:31       ` Linus Torvalds
2005-04-10  2:41         ` Petr Baudis
2005-04-10 16:27           ` [ANNOUNCE] git-pasky-0.1 Petr Baudis
2005-04-10 16:55             ` Linus Torvalds
2005-04-10 19:49               ` Sean
2005-04-10 17:33             ` Ingo Molnar
2005-04-10 17:42               ` Willy Tarreau
2005-04-10 17:45                 ` Ingo Molnar
2005-04-10 18:45                   ` Petr Baudis
2005-04-10 19:13                     ` Willy Tarreau
2005-04-10 21:27                       ` Petr Baudis
2005-04-10 20:38                     ` Linus Torvalds
2005-04-10 21:39                       ` Linus Torvalds
2005-04-10 23:49                         ` Petr Baudis
2005-04-10 22:27                       ` Petr Baudis
2005-04-10 23:10                         ` Linus Torvalds
2005-04-10 23:26                           ` Petr Baudis
2005-04-10 23:46                             ` Linus Torvalds
2005-04-10 23:56                               ` Petr Baudis
2005-04-11  0:20                                 ` GIT license (Re: Re: Re: Re: Re: [ANNOUNCE] git-pasky-0.1) Linus Torvalds
2005-04-11  0:27                                   ` Petr Baudis
2005-04-11  7:45                                   ` Ingo Molnar
2005-04-11  8:40                                     ` Florian Weimer
2005-04-11 10:52                                       ` Petr Baudis
2005-04-11 16:05                                         ` Florian Weimer
2005-04-10 23:23                         ` [ANNOUNCE] git-pasky-0.1 Paul Jackson
2005-04-11  0:15                           ` Randy.Dunlap
2005-04-11  0:30                       ` Re: " Petr Baudis
2005-04-11  1:11                         ` Linus Torvalds
2005-04-10 20:41                     ` Paul Jackson
2005-04-11  1:58             ` [ANNOUNCE] git-pasky-0.2 Petr Baudis
2005-04-11  2:46               ` Daniel Barkalow
2005-04-11 10:17                 ` Petr Baudis
2005-04-11  8:50               ` Ingo Molnar
2005-04-11 10:16                 ` Petr Baudis
2005-04-11 13:57               ` [ANNOUNCE] git-pasky-0.3 Petr Baudis
2005-04-12 12:47                 ` Martin Schlemmer
2005-04-12 13:02                   ` Petr Baudis
2005-04-12 13:13                     ` Martin Schlemmer
2005-04-12 13:23                       ` Petr Baudis
2005-04-12 13:07                 ` David Woodhouse
2005-04-13  8:47                   ` Russell King
2005-04-13  8:59                     ` Petr Baudis
2005-04-13  9:06                       ` H. Peter Anvin
2005-04-13  9:09                         ` David Woodhouse
2005-04-13  9:25                       ` David Woodhouse
2005-04-13  9:42                         ` Petr Baudis
2005-04-13 10:24                           ` David Woodhouse
2005-04-13 17:01                           ` Daniel Barkalow
2005-04-13 18:07                             ` Petr Baudis
2005-04-13 18:22                               ` git mailing list (Re: Re: Re: Re: [ANNOUNCE] git-pasky-0.3) Linus Torvalds
2005-04-13 18:38                               ` Re: Re: Re: [ANNOUNCE] git-pasky-0.3 Daniel Barkalow
2005-04-13 12:43                         ` Xavier Bestel
2005-04-13 16:48                           ` H. Peter Anvin
2005-04-13 18:15                             ` Xavier Bestel
2005-04-13 23:05                           ` bd
2005-04-13 14:38                         ` Linus Torvalds
2005-04-13 14:47                           ` David Woodhouse
2005-04-13 14:59                             ` Linus Torvalds
2005-04-13  9:35                 ` Russell King
2005-04-13  9:38                   ` Russell King
2005-04-13  9:49                     ` Petr Baudis
2005-04-13 11:02                       ` Ingo Molnar
2005-04-13 14:50                         ` Linus Torvalds
2005-04-13  9:46                   ` Petr Baudis
2005-04-13 10:28                     ` Russell King
2005-04-13 19:03                   ` Russell King
2005-04-13 19:13                     ` Petr Baudis
2005-04-13 19:21                       ` Russell King
2005-04-13 19:23                         ` H. Peter Anvin
2005-04-10  6:53         ` more git updates Christopher Li
2005-04-10 11:48           ` Ralph Corderoy
2005-04-10 19:23           ` Paul Jackson
2005-04-10 18:42             ` Christopher Li
2005-04-10 22:30               ` Petr Baudis
2005-04-11 13:58           ` H. Peter Anvin
2005-04-20 20:29             ` Kai Henningsen
2005-04-24  0:42               ` Paul Jackson
2005-04-24  1:29                 ` Bernd Eckenfels
2005-04-24  4:13                   ` Paul Jackson
2005-04-24  4:38                     ` Bernd Eckenfels
2005-04-24  4:53                       ` Paul Jackson
2005-04-25 11:57                       ` Theodore Ts'o
2005-04-25 16:40                         ` David Wagner
2005-04-25 20:35                         ` Bernd Eckenfels
2005-04-24 16:52                   ` Horst von Brand
2005-04-24  8:00                 ` Kai Henningsen
     [not found]               ` <6f6293f10504210220744af114@mail.gmail.com>
2005-04-24  8:01                 ` Kai Henningsen
2005-04-11 11:35         ` [rfc] git: combo-blobs Ingo Molnar
2005-04-11 14:45           ` Paul Jackson
2005-04-11 15:12             ` Ingo Molnar
2005-04-11 15:32               ` Linus Torvalds
2005-04-11 15:39                 ` Ingo Molnar
2005-04-11 15:57                   ` Ingo Molnar
2005-04-11 16:01                   ` Linus Torvalds
2005-04-11 16:33                     ` Ingo Molnar
2005-04-12  5:42                       ` Barry K. Nathan
2005-04-11 18:13                     ` Chris Wedgwood
2005-04-11 18:30                       ` Linus Torvalds
2005-04-11 20:18                         ` Linus Torvalds
2005-04-11 18:40                       ` Petr Baudis
2005-04-11 17:50               ` Paul Jackson
2005-04-11 15:28             ` Ingo Molnar
2005-04-11 15:31               ` Ingo Molnar
2005-04-12  4:05         ` more git updates David Eger
2005-04-12  8:16           ` Petr Baudis
2005-04-12 20:44             ` David Eger
2005-04-12 21:21               ` Linus Torvalds
2005-04-12 22:29                 ` Krzysztof Halasa
2005-04-12 22:49                   ` Linus Torvalds
2005-04-13  4:32                     ` Matthias Urlichs
2005-04-12 22:36                 ` David Eger
2005-04-12 23:48                   ` Panagiotis Issaris
2005-04-12 23:40                 ` Andrea Arcangeli
2005-04-12 23:45                   ` Linus Torvalds
2005-04-13  0:14                     ` Andrea Arcangeli
2005-04-13  1:10                       ` Linus Torvalds
2005-04-13 10:59                         ` Andrea Arcangeli
2005-04-13 20:44                         ` Matt Mackall
2005-04-13 23:42                           ` Krzysztof Halasa
2005-04-14  0:13                             ` Matt Mackall
2005-04-13  9:30                     ` Russell King
2005-04-13 10:20                       ` Andrea Arcangeli
2005-04-13 14:43                       ` Linus Torvalds
2005-04-10  2:07     ` Paul Jackson
2005-04-10  2:20       ` Paul Jackson
2005-04-10  2:09     ` Paul Jackson
2005-04-10  7:51     ` Junio C Hamano
2005-04-10  5:53       ` Christopher Li
2005-04-10  9:28         ` Junio C Hamano
2005-04-10  7:06           ` Christopher Li
2005-04-10 11:38             ` tony.luck
2005-04-10  9:48           ` Petr Baudis
2005-04-10  9:40         ` Wichert Akkerman
2005-04-10  9:41         ` Petr Baudis
2005-04-10  7:09           ` Christopher Li
2005-04-10 11:21       ` Proposal for shell-patch-format [was: Re: more git updates..] Rutger Nijlunsing
2005-04-10 15:44       ` more git updates Linus Torvalds
2005-04-10 17:00         ` Rutger Nijlunsing
2005-04-10 18:50         ` Paul Jackson
2005-04-10 20:57           ` Linus Torvalds [this message]
2005-04-10 19:03             ` Christopher Li
2005-04-10 22:38               ` Linus Torvalds
2005-04-10 19:53                 ` Christopher Li
2005-04-10 23:21                   ` Linus Torvalds
2005-04-10 21:28                     ` Christopher Li
2005-04-12  5:14                       ` David Lang
2005-04-12  6:00                         ` Paul Jackson
2005-04-12  7:05                         ` Barry K. Nathan
2005-04-11  6:57                 ` bert hubert
2005-04-11  7:20                   ` Christer Weinigel
2005-04-10 23:14             ` Paul Jackson
2005-04-10 23:38               ` Linus Torvalds
2005-04-11  0:19                 ` Paul Jackson
2005-04-11 15:49                 ` Randy.Dunlap
2005-04-11 18:30                   ` Petr Baudis
2005-04-11  0:10               ` Petr Baudis
2005-04-09 22:00 ` Paul Jackson
2005-04-09 23:21 ` Ralph Corderoy
2005-04-10  0:39   ` Paul Jackson
2005-04-10  1:14     ` Bernd Eckenfels
2005-04-10  1:33       ` Paul Jackson
2005-04-10 10:22     ` Ralph Corderoy
2005-04-10 17:30       ` Paul Jackson
2005-04-10 17:31 ` Rik van Riel
2005-04-10 17:35   ` Ingo Molnar
2005-04-11 16:46 ` ross
2005-04-10 22:07 Luck, Tony

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=Pine.LNX.4.58.0504101338360.1267@ppc970.osdl.org \
    --to=torvalds@osdl.org \
    --cc=junkio@cox.net \
    --cc=linux-kernel@vger.kernel.org \
    --cc=pj@engr.sgi.com \
    --cc=rddunlap@osdl.org \
    --cc=ross@jose.lug.udel.edu \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®