suggested in the previous section may be new to you. You do not\r
have to worry. git supports "shared public repository" style of\r
cooperation you are probably more familiar with as well.</p>\r
-<p>For this, set up a public repository on a machine that is\r
-reachable via SSH by people with "commit privileges". Put the\r
-committers in the same user group and make the repository\r
-writable by that group. Make sure their umasks are set up to\r
-allow group members to write into directories other members\r
-have created.</p>\r
-<p>You, as an individual committer, then:</p>\r
-<ul>\r
-<li>\r
-<p>\r
-First clone the shared repository to a local repository:\r
-</p>\r
-</li>\r
-</ul>\r
-<div class="listingblock">\r
-<div class="content">\r
-<pre><tt>$ git clone repo.shared.xz:/pub/scm/project.git/ my-project\r
-$ cd my-project\r
-$ hack away</tt></pre>\r
-</div></div>\r
-<ul>\r
-<li>\r
-<p>\r
-Merge the work others might have done while you were hacking\r
- away:\r
-</p>\r
-</li>\r
-</ul>\r
-<div class="listingblock">\r
-<div class="content">\r
-<pre><tt>$ git pull origin\r
-$ test the merge result</tt></pre>\r
-</div></div>\r
-<div class="admonitionblock">\r
-<table><tr>\r
-<td class="icon">\r
-<div class="title">Note</div>\r
-</td>\r
-<td class="content">\r
-<p>The first <tt>git clone</tt> would have placed the following in\r
-<tt>my-project/.git/remotes/origin</tt> file, and that's why this and\r
-the next step work.</p>\r
-<div class="listingblock">\r
-<div class="content">\r
-<pre><tt>URL: repo.shared.xz:/pub/scm/project.git/ my-project\r
-Pull: master:origin</tt></pre>\r
-</div></div>\r
-</td>\r
-</tr></table>\r
-</div>\r
-<ul>\r
-<li>\r
-<p>\r
-push your work as the new head of the shared\r
- repository.\r
-</p>\r
-</li>\r
-</ul>\r
-<div class="listingblock">\r
-<div class="content">\r
-<pre><tt>$ git push origin master</tt></pre>\r
-</div></div>\r
-<p>If somebody else pushed into the same shared repository while\r
-you were working locally, <tt>git push</tt> in the last step would\r
-complain, telling you that the remote <tt>master</tt> head does not\r
-fast forward. You need to pull and merge those other changes\r
-back before you push your work when it happens.</p>\r
-<p>The <tt>git push</tt> command without any explicit refspec parameter\r
-pushes the refs that exist both in the local repository and the\r
-remote repository. So the last <tt>push</tt> can be done with either\r
-one of these:</p>\r
-<div class="listingblock">\r
-<div class="content">\r
-<pre><tt>$ git push origin\r
-$ git push repo.shared.xz:/pub/scm/project.git/</tt></pre>\r
-</div></div>\r
-<p>as long as the shared repository does not have any branches\r
-other than <tt>master</tt>.\r
-[NOTE]</p>\r
-<div class="exampleblock">\r
-<div class="exampleblock-content">\r
-<p>If you created your shared repository by cloning from somewhere\r
-else, you may have the <tt>origin</tt> branch. Your developers\r
-typically do not use that branch; remove it. Otherwise, that\r
-would be pushed back by the <tt>git push origin</tt> because your\r
-developers' repository would surely have <tt>origin</tt> branch to keep\r
-track of the shared repository, and would be counted as "exist\r
-on both ends".</p>\r
-</div></div>\r
-</div>\r
-<h2>Advanced Shared Repository Management</h2>\r
-<div class="sectionbody">\r
-<p>Being able to push into a shared repository means being able to\r
-write into it. If your developers are coming over the network,\r
-this means you, as the repository administrator, need to give\r
-each of them an SSH access to the shared repository machine.</p>\r
-<p>In some cases, though, you may not want to give a normal shell\r
-account to them, but want to restrict them to be able to only\r
-do <tt>git push</tt> into the repository and nothing else.</p>\r
-<p>You can achieve this by setting the login shell of your\r
-developers on the shared repository host to <tt>git-shell</tt> program.</p>\r
-<div class="admonitionblock">\r
-<table><tr>\r
-<td class="icon">\r
-<div class="title">Note</div>\r
-</td>\r
-<td class="content">Most likely you would also need to list <tt>git-shell</tt> program in\r
-<tt>/etc/shells</tt> file.</td>\r
-</tr></table>\r
-</div>\r
-<p>This restricts the set of commands that can be run from incoming\r
-SSH connection for these users to only <tt>receive-pack</tt> and\r
-<tt>upload-pack</tt>, so the only thing they can do are <tt>git fetch</tt> and\r
-<tt>git push</tt>.</p>\r
-<p>You still need to create UNIX user accounts for each developer,\r
-and put them in the same group. Make sure that the repository\r
-shared among these developers is writable by that group.</p>\r
-<ol>\r
-<li>\r
-<p>\r
-Initializing the shared repository with <tt>git-init-db —shared</tt>\r
-helps somewhat.\r
-</p>\r
-</li>\r
-<li>\r
-<p>\r
-Run the following in the shared repository:\r
-</p>\r
-<div class="listingblock">\r
-<div class="content">\r
-<pre><tt>$ chgrp -R $group repo.git\r
-$ find repo.git -type d -print | xargs chmod ug+rwx,g+s\r
-$ GIT_DIR=repo.git git repo-config core.sharedrepository true</tt></pre>\r
-</div></div>\r
-</li>\r
-</ol>\r
-<p>The above measures make sure that directories lazily created in\r
-<tt>$GIT_DIR</tt> are writable by group members. You, as the\r
-repository administrator, are still responsible to make sure\r
-your developers belong to that shared repository group and set\r
-their umask to a value no stricter than 027 (i.e. at least allow\r
-reading and searching by group members).</p>\r
-<p>You can implement finer grained branch policies using update\r
-hooks. There is a document ("control access to branches") in\r
-Documentation/howto by Carl Baldwin and JC outlining how to (1)\r
-limit access to branch per user, (2) forbid overwriting existing\r
-tags.</p>\r
+<p>See <a href="cvs-migration.txt">git for CVS users</a> for the details.</p>\r
</div>\r
<h2>Bundling your work together</h2>\r
<div class="sectionbody">\r
</div>\r
<div id="footer">\r
<div id="footer-text">\r
-Last updated 22-Jan-2006 23:54:24 PDT\r
+Last updated 30-Jan-2006 23:10:20 PDT\r
</div>\r
</div>\r
</body>\r
have to worry. git supports "shared public repository" style of
cooperation you are probably more familiar with as well.
-For this, set up a public repository on a machine that is
-reachable via SSH by people with "commit privileges". Put the
-committers in the same user group and make the repository
-writable by that group. Make sure their umasks are set up to
-allow group members to write into directories other members
-have created.
-
-You, as an individual committer, then:
-
-- First clone the shared repository to a local repository:
-------------------------------------------------
-$ git clone repo.shared.xz:/pub/scm/project.git/ my-project
-$ cd my-project
-$ hack away
-------------------------------------------------
-
-- Merge the work others might have done while you were hacking
- away:
-------------------------------------------------
-$ git pull origin
-$ test the merge result
-------------------------------------------------
-[NOTE]
-================================
-The first `git clone` would have placed the following in
-`my-project/.git/remotes/origin` file, and that's why this and
-the next step work.
-------------
-URL: repo.shared.xz:/pub/scm/project.git/ my-project
-Pull: master:origin
-------------
-================================
-
-- push your work as the new head of the shared
- repository.
-------------------------------------------------
-$ git push origin master
-------------------------------------------------
-If somebody else pushed into the same shared repository while
-you were working locally, `git push` in the last step would
-complain, telling you that the remote `master` head does not
-fast forward. You need to pull and merge those other changes
-back before you push your work when it happens.
-
-The `git push` command without any explicit refspec parameter
-pushes the refs that exist both in the local repository and the
-remote repository. So the last `push` can be done with either
-one of these:
-------------
-$ git push origin
-$ git push repo.shared.xz:/pub/scm/project.git/
-------------
-as long as the shared repository does not have any branches
-other than `master`.
-[NOTE]
-============
-If you created your shared repository by cloning from somewhere
-else, you may have the `origin` branch. Your developers
-typically do not use that branch; remove it. Otherwise, that
-would be pushed back by the `git push origin` because your
-developers' repository would surely have `origin` branch to keep
-track of the shared repository, and would be counted as "exist
-on both ends".
-============
-
-Advanced Shared Repository Management
--------------------------------------
-
-Being able to push into a shared repository means being able to
-write into it. If your developers are coming over the network,
-this means you, as the repository administrator, need to give
-each of them an SSH access to the shared repository machine.
-
-In some cases, though, you may not want to give a normal shell
-account to them, but want to restrict them to be able to only
-do `git push` into the repository and nothing else.
-
-You can achieve this by setting the login shell of your
-developers on the shared repository host to `git-shell` program.
-
-[NOTE]
-Most likely you would also need to list `git-shell` program in
-`/etc/shells` file.
-
-This restricts the set of commands that can be run from incoming
-SSH connection for these users to only `receive-pack` and
-`upload-pack`, so the only thing they can do are `git fetch` and
-`git push`.
-
-You still need to create UNIX user accounts for each developer,
-and put them in the same group. Make sure that the repository
-shared among these developers is writable by that group.
-
-. Initializing the shared repository with `git-init-db --shared`
-helps somewhat.
-
-. Run the following in the shared repository:
-+
-------------
-$ chgrp -R $group repo.git
-$ find repo.git -type d -print | xargs chmod ug+rwx,g+s
-$ GIT_DIR=repo.git git repo-config core.sharedrepository true
-------------
-
-The above measures make sure that directories lazily created in
-`$GIT_DIR` are writable by group members. You, as the
-repository administrator, are still responsible to make sure
-your developers belong to that shared repository group and set
-their umask to a value no stricter than 027 (i.e. at least allow
-reading and searching by group members).
-
-You can implement finer grained branch policies using update
-hooks. There is a document ("control access to branches") in
-Documentation/howto by Carl Baldwin and JC outlining how to (1)
-limit access to branch per user, (2) forbid overwriting existing
-tags.
-
+See link:cvs-migration.txt[git for CVS users] for the details.
Bundling your work together
---------------------------
</div>\r
<div id="preamble">\r
<div class="sectionbody">\r
-<p>Ok, so you're a CVS user. That's ok, it's a treatable condition, and the\r
-first step to recovery is admitting you have a problem. The fact that\r
-you are reading this file means that you may be well on that path\r
-already.</p>\r
-<p>The thing about CVS is that it absolutely sucks as a source control\r
-manager, and you'll thus be happy with almost anything else. git,\r
-however, may be a bit <em>too</em> different (read: "good") for your taste, and\r
-does a lot of things differently.</p>\r
-<p>One particular suckage of CVS is very hard to work around: CVS is\r
-basically a tool for tracking <em>file</em> history, while git is a tool for\r
-tracking <em>project</em> history. This sometimes causes problems if you are\r
-used to doing very strange things in CVS, in particular if you're doing\r
-things like making branches of just a subset of the project. git can't\r
-track that, since git never tracks things on the level of an individual\r
-file, only on the whole project level.</p>\r
-<p>The good news is that most people don't do that, and in fact most sane\r
-people think it's a bug in CVS that makes it tag (and check in changes)\r
-one file at a time. So most projects you'll ever see will use CVS\r
-<em>as if</em> it was sane. In which case you'll find it very easy indeed to\r
-move over to git.</p>\r
-<p>First off: this is not a git tutorial. See\r
-<a href="tutorial.html">Documentation/tutorial.txt</a> for how git\r
-actually works. This is more of a random collection of gotcha's\r
-and notes on converting from CVS to git.</p>\r
-<p>Second: CVS has the notion of a "repository" as opposed to the thing\r
-that you're actually working in (your working directory, or your\r
-"checked out tree"). git does not have that notion at all, and all git\r
-working directories <em>are</em> the repositories. However, you can easily\r
-emulate the CVS model by having one special "global repository", which\r
-people can synchronize with. See details later, but in the meantime\r
-just keep in mind that with git, every checked out working tree will\r
-have a full revision control history of its own.</p>\r
-</div>\r
-</div>\r
-<h2>Importing a CVS archive</h2>\r
-<div class="sectionbody">\r
-<p>Ok, you have an old project, and you want to at least give git a chance\r
-to see how it performs. The first thing you want to do (after you've\r
-gone through the git tutorial, and generally familiarized yourself with\r
-how to commit stuff etc in git) is to create a git'ified version of your\r
-CVS archive.</p>\r
-<p>Happily, that's very easy indeed. git will do it for you, although git\r
-will need the help of a program called "cvsps":</p>\r
-<div class="literalblock">\r
-<div class="content">\r
-<pre><tt>http://www.cobite.com/cvsps/</tt></pre>\r
-</div></div>\r
-<p>which is not actually related to git at all, but which makes CVS usage\r
-look almost sane (ie you almost certainly want to have it even if you\r
-decide to stay with CVS). However, git will want <em>at least</em> version 2.1\r
-of cvsps (available at the address above), and in fact will currently\r
-refuse to work with anything else.</p>\r
-<p>Once you've gotten (and installed) cvsps, you may or may not want to get\r
-any more familiar with it, but make sure it is in your path. After that,\r
-the magic command line is</p>\r
-<div class="literalblock">\r
-<div class="content">\r
-<pre><tt>git cvsimport -v -d <cvsroot> -C <destination> <module></tt></pre>\r
-</div></div>\r
-<p>which will do exactly what you'd think it does: it will create a git\r
-archive of the named CVS module. The new archive will be created in the\r
-subdirectory named <destination>; it'll be created if it doesn't exist.\r
-Default is the local directory.</p>\r
-<p>It can take some time to actually do the conversion for a large archive\r
-since it involves checking out from CVS every revision of every file,\r
-and the conversion script is reasonably chatty unless you omit the <em>-v</em>\r
-option, but on some not very scientific tests it averaged about twenty\r
-revisions per second, so a medium-sized project should not take more\r
-than a couple of minutes. For larger projects or remote repositories,\r
-the process may take longer.</p>\r
-<p>After the (initial) import is done, the CVS archive's current head\r
-revision will be checked out — thus, you can start adding your own\r
-changes right away.</p>\r
-<p>The import is incremental, i.e. if you call it again next month it'll\r
-fetch any CVS updates that have been happening in the meantime. The\r
-cut-off is date-based, so don't change the branches that were imported\r
-from CVS.</p>\r
-<p>You can merge those updates (or, in fact, a different CVS branch) into\r
-your main branch:</p>\r
-<div class="literalblock">\r
-<div class="content">\r
-<pre><tt>git resolve HEAD origin "merge with current CVS HEAD"</tt></pre>\r
-</div></div>\r
-<p>The HEAD revision from CVS is named "origin", not "HEAD", because git\r
-already uses "HEAD". (If you don't like <em>origin</em>, use cvsimport's\r
-<em>-o</em> option to change it.)</p>\r
-</div>\r
-<h2>Emulating CVS behaviour</h2>\r
-<div class="sectionbody">\r
-<p>So, by now you are convinced you absolutely want to work with git, but\r
-at the same time you absolutely have to have a central repository.\r
-Step back and think again. Okay, you still need a single central\r
-repository? There are several ways to go about that:</p>\r
-<ol>\r
+<p>So you're a CVS user. That's ok, it's a treatable condition. The job of\r
+this document is to put you on the road to recovery, by helping you\r
+convert an existing cvs repository to git, and by showing you how to use a\r
+git repository in a cvs-like fashion.</p>\r
+<p>Some basic familiarity with git is required. This\r
+<a href="tutorial.html">tutorial introduction to git</a> should be sufficient.</p>\r
+<p>First, note some ways that git differs from CVS:</p>\r
+<ul>\r
<li>\r
<p>\r
-Designate a person responsible to pull all branches. Make the\r
-repository of this person public, and make every team member\r
-pull regularly from it.\r
+Commits are atomic and project-wide, not per-file as in CVS.\r
</p>\r
</li>\r
<li>\r
<p>\r
-Set up a public repository with read/write access for every team\r
-member. Use "git pull/push" as you used "cvs update/commit". Be\r
-sure that your repository is up to date before pushing, just\r
-like you used to do with "cvs commit"; your push will fail if\r
-what you are pushing is not up to date.\r
+Offline work is supported: you can make multiple commits locally,\r
+ then submit them when you're ready.\r
</p>\r
</li>\r
<li>\r
<p>\r
-Make the repository of every team member public. It is the\r
-responsibility of each single member to pull from every other\r
-team member.\r
+Branching is fast and easy.\r
</p>\r
</li>\r
-</ol>\r
+<li>\r
+<p>\r
+Every working tree contains a repository with a full copy of the\r
+ project history, and no repository is inherently more important than\r
+ any other. However, you can emulate the CVS model by designating a\r
+ single shared repository which people can synchronize with; see below\r
+ for details.\r
+</p>\r
+</li>\r
+</ul>\r
+</div>\r
+</div>\r
+<h2>Importing a CVS archive</h2>\r
+<div class="sectionbody">\r
+<p>First, install version 2.1 or higher of cvsps from\r
+<a href="http://www.cobite.com/cvsps/">http://www.cobite.com/cvsps/</a> and make\r
+sure it is in your path. The magic command line is then</p>\r
+<div class="listingblock">\r
+<div class="content">\r
+<pre><tt>$ git cvsimport -v -d <cvsroot> -C <destination> <module></tt></pre>\r
+</div></div>\r
+<p>This puts a git archive of the named CVS module in the directory\r
+<destination>, which will be created if necessary. The -v option makes\r
+the conversion script very chatty.</p>\r
+<p>The import checks out from CVS every revision of every file. Reportedly\r
+cvsimport can average some twenty revisions per second, so for a\r
+medium-sized project this should not take more than a couple of minutes.\r
+Larger projects or remote repositories may take longer.</p>\r
+<p>The main trunk is stored in the git branch named <tt>origin</tt>, and additional\r
+CVS branches are stored in git branches with the same names. The most\r
+recent version of the main trunk is also left checked out on the <tt>master</tt>\r
+branch, so you can start adding your own changes right away.</p>\r
+<p>The import is incremental, so if you call it again next month it will\r
+fetch any CVS updates that have been made in the meantime. For this to\r
+work, you must not modify the imported branches; instead, create new\r
+branches for your own changes, and merge in the imported branches as\r
+necessary.</p>\r
+</div>\r
+<h2>Development Models</h2>\r
+<div class="sectionbody">\r
+<p>CVS users are accustomed to giving a group of developers commit access to\r
+a common repository. In the next section we'll explain how to do this\r
+with git. However, the distributed nature of git allows other development\r
+models, and you may want to first consider whether one of them might be a\r
+better fit for your project.</p>\r
+<p>For example, you can choose a single person to maintain the project's\r
+primary public repository. Other developers then clone this repository\r
+and each work in their own clone. When they have a series of changes that\r
+they're happy with, they ask the maintainer to pull from the branch\r
+containing the changes. The maintainer reviews their changes and pulls\r
+them into the primary repository, which other developers pull from as\r
+necessary to stay coordinated. The Linux kernel and other projects use\r
+variants of this model.</p>\r
+<p>With a small group, developers may just pull changes from each other's\r
+repositories without the need for a central maintainer.</p>\r
+</div>\r
+<h2>Emulating the CVS Development Model</h2>\r
+<div class="sectionbody">\r
+<p>Start with an ordinary git working directory containing the project, and\r
+remove the checked-out files, keeping just the bare .git directory:</p>\r
+<div class="listingblock">\r
+<div class="content">\r
+<pre><tt>$ mv project/.git /pub/repo.git\r
+$ rm -r project/</tt></pre>\r
+</div></div>\r
+<p>Next, give every team member read/write access to this repository. One\r
+easy way to do this is to give all the team members ssh access to the\r
+machine where the repository is hosted. If you don't want to give them a\r
+full shell on the machine, there is a restricted shell which only allows\r
+users to do git pushes and pulls; see <a href="git-shell.html">git-shell(1)</a>.</p>\r
+<p>Put all the committers should in the same group, and make the repository\r
+writable by that group:</p>\r
+<div class="listingblock">\r
+<div class="content">\r
+<pre><tt>$ chgrp -R $group repo.git\r
+$ find repo.git -mindepth 1 -type d |xargs chmod ug+rwx,g+s\r
+$ GIT_DIR=repo.git git repo-config core.sharedrepository true</tt></pre>\r
+</div></div>\r
+<p>Make sure committers have a umask of at most 027, so that the directories\r
+they create are writable and searchable by other group members.</p>\r
+<p>Suppose this repository is now set up in /pub/repo.git on the host\r
+foo.com. Then as an individual commiter you can clone the shared\r
+repository:</p>\r
+<div class="listingblock">\r
+<div class="content">\r
+<pre><tt>$ git clone foo.com:/pub/repo.git/ my-project\r
+$ cd my-project</tt></pre>\r
+</div></div>\r
+<p>and hack away. The equivalent of <tt>cvs update</tt> is</p>\r
+<div class="listingblock">\r
+<div class="content">\r
+<pre><tt>$ git pull origin</tt></pre>\r
+</div></div>\r
+<p>which merges in any work that others might have done since the clone\r
+operation.</p>\r
+<div class="admonitionblock">\r
+<table><tr>\r
+<td class="icon">\r
+<div class="title">Note</div>\r
+</td>\r
+<td class="content">\r
+<p>The first <tt>git clone</tt> places the following in the\r
+<tt>my-project/.git/remotes/origin</tt> file, and that's why the previous step\r
+and the next step both work.</p>\r
+<div class="listingblock">\r
+<div class="content">\r
+<pre><tt>URL: foo.com:/pub/project.git/ my-project\r
+Pull: master:origin</tt></pre>\r
+</div></div>\r
+</td>\r
+</tr></table>\r
+</div>\r
+<p>You can update the shared repository with your changes using:</p>\r
+<div class="listingblock">\r
+<div class="content">\r
+<pre><tt>$ git push origin master</tt></pre>\r
+</div></div>\r
+<p>If someone else has updated the repository more recently, <tt>git push</tt>, like\r
+<tt>cvs commit</tt>, will complain, in which case you must pull any changes\r
+before attempting the push again.</p>\r
+<p>In the <tt>git push</tt> command above we specify the name of the remote branch\r
+to update (<tt>master</tt>). If we leave that out, <tt>git push</tt> tries to update\r
+any branches in the remote repository that have the same name as a branch\r
+in the local repository. So the last <tt>push</tt> can be done with either of:</p>\r
+<div class="listingblock">\r
+<div class="content">\r
+<pre><tt>$ git push origin\r
+$ git push repo.shared.xz:/pub/scm/project.git/</tt></pre>\r
+</div></div>\r
+<p>as long as the shared repository does not have any branches\r
+other than <tt>master</tt>.</p>\r
+<div class="admonitionblock">\r
+<table><tr>\r
+<td class="icon">\r
+<div class="title">Note</div>\r
+</td>\r
+<td class="content">\r
+<p>Because of this behaviour, if the shared repository and the developer's\r
+repository both have branches named <tt>origin</tt>, then a push like the above\r
+attempts to update the <tt>origin</tt> branch in the shared repository from the\r
+developer's <tt>origin</tt> branch. The results may be unexpected, so it's\r
+usually best to remove any branch named <tt>origin</tt> from the shared\r
+repository.</p>\r
+</td>\r
+</tr></table>\r
+</div>\r
+</div>\r
+<h2>Advanced Shared Repository Management</h2>\r
+<div class="sectionbody">\r
+<p>Git allows you to specify scripts called "hooks" to be run at certain\r
+points. You can use these, for example, to send all commits to the shared\r
+repository to a mailing list. See <a href="hooks.txt">Hooks used by git</a>.</p>\r
+<p>You can enforce finer grained permissions using update hooks. See\r
+<a href="howto/update-hook-example.txt">Controlling access to branches using\r
+update hooks</a>.</p>\r
</div>\r
<h2>CVS annotate</h2>\r
<div class="sectionbody">\r
</div>\r
<div id="footer">\r
<div id="footer-text">\r
-Last updated 06-Jan-2006 17:12:56 PDT\r
+Last updated 30-Jan-2006 23:10:24 PDT\r
</div>\r
</div>\r
</body>\r
git for CVS users
=================
-Ok, so you're a CVS user. That's ok, it's a treatable condition, and the
-first step to recovery is admitting you have a problem. The fact that
-you are reading this file means that you may be well on that path
-already.
-
-The thing about CVS is that it absolutely sucks as a source control
-manager, and you'll thus be happy with almost anything else. git,
-however, may be a bit 'too' different (read: "good") for your taste, and
-does a lot of things differently.
-
-One particular suckage of CVS is very hard to work around: CVS is
-basically a tool for tracking 'file' history, while git is a tool for
-tracking 'project' history. This sometimes causes problems if you are
-used to doing very strange things in CVS, in particular if you're doing
-things like making branches of just a subset of the project. git can't
-track that, since git never tracks things on the level of an individual
-file, only on the whole project level.
-
-The good news is that most people don't do that, and in fact most sane
-people think it's a bug in CVS that makes it tag (and check in changes)
-one file at a time. So most projects you'll ever see will use CVS
-'as if' it was sane. In which case you'll find it very easy indeed to
-move over to git.
-
-First off: this is not a git tutorial. See
-link:tutorial.html[Documentation/tutorial.txt] for how git
-actually works. This is more of a random collection of gotcha's
-and notes on converting from CVS to git.
-
-Second: CVS has the notion of a "repository" as opposed to the thing
-that you're actually working in (your working directory, or your
-"checked out tree"). git does not have that notion at all, and all git
-working directories 'are' the repositories. However, you can easily
-emulate the CVS model by having one special "global repository", which
-people can synchronize with. See details later, but in the meantime
-just keep in mind that with git, every checked out working tree will
-have a full revision control history of its own.
+So you're a CVS user. That's ok, it's a treatable condition. The job of
+this document is to put you on the road to recovery, by helping you
+convert an existing cvs repository to git, and by showing you how to use a
+git repository in a cvs-like fashion.
+Some basic familiarity with git is required. This
+link:tutorial.html[tutorial introduction to git] should be sufficient.
-Importing a CVS archive
------------------------
-
-Ok, you have an old project, and you want to at least give git a chance
-to see how it performs. The first thing you want to do (after you've
-gone through the git tutorial, and generally familiarized yourself with
-how to commit stuff etc in git) is to create a git'ified version of your
-CVS archive.
-
-Happily, that's very easy indeed. git will do it for you, although git
-will need the help of a program called "cvsps":
-
- http://www.cobite.com/cvsps/
-
-which is not actually related to git at all, but which makes CVS usage
-look almost sane (ie you almost certainly want to have it even if you
-decide to stay with CVS). However, git will want 'at least' version 2.1
-of cvsps (available at the address above), and in fact will currently
-refuse to work with anything else.
-
-Once you've gotten (and installed) cvsps, you may or may not want to get
-any more familiar with it, but make sure it is in your path. After that,
-the magic command line is
-
- git cvsimport -v -d <cvsroot> -C <destination> <module>
-
-which will do exactly what you'd think it does: it will create a git
-archive of the named CVS module. The new archive will be created in the
-subdirectory named <destination>; it'll be created if it doesn't exist.
-Default is the local directory.
+First, note some ways that git differs from CVS:
-It can take some time to actually do the conversion for a large archive
-since it involves checking out from CVS every revision of every file,
-and the conversion script is reasonably chatty unless you omit the '-v'
-option, but on some not very scientific tests it averaged about twenty
-revisions per second, so a medium-sized project should not take more
-than a couple of minutes. For larger projects or remote repositories,
-the process may take longer.
+ * Commits are atomic and project-wide, not per-file as in CVS.
-After the (initial) import is done, the CVS archive's current head
-revision will be checked out -- thus, you can start adding your own
-changes right away.
+ * Offline work is supported: you can make multiple commits locally,
+ then submit them when you're ready.
-The import is incremental, i.e. if you call it again next month it'll
-fetch any CVS updates that have been happening in the meantime. The
-cut-off is date-based, so don't change the branches that were imported
-from CVS.
+ * Branching is fast and easy.
-You can merge those updates (or, in fact, a different CVS branch) into
-your main branch:
+ * Every working tree contains a repository with a full copy of the
+ project history, and no repository is inherently more important than
+ any other. However, you can emulate the CVS model by designating a
+ single shared repository which people can synchronize with; see below
+ for details.
- git resolve HEAD origin "merge with current CVS HEAD"
-
-The HEAD revision from CVS is named "origin", not "HEAD", because git
-already uses "HEAD". (If you don't like 'origin', use cvsimport's
-'-o' option to change it.)
-
-
-Emulating CVS behaviour
+Importing a CVS archive
-----------------------
+First, install version 2.1 or higher of cvsps from
+link:http://www.cobite.com/cvsps/[http://www.cobite.com/cvsps/] and make
+sure it is in your path. The magic command line is then
+
+-------------------------------------------
+$ git cvsimport -v -d <cvsroot> -C <destination> <module>
+-------------------------------------------
+
+This puts a git archive of the named CVS module in the directory
+<destination>, which will be created if necessary. The -v option makes
+the conversion script very chatty.
+
+The import checks out from CVS every revision of every file. Reportedly
+cvsimport can average some twenty revisions per second, so for a
+medium-sized project this should not take more than a couple of minutes.
+Larger projects or remote repositories may take longer.
+
+The main trunk is stored in the git branch named `origin`, and additional
+CVS branches are stored in git branches with the same names. The most
+recent version of the main trunk is also left checked out on the `master`
+branch, so you can start adding your own changes right away.
+
+The import is incremental, so if you call it again next month it will
+fetch any CVS updates that have been made in the meantime. For this to
+work, you must not modify the imported branches; instead, create new
+branches for your own changes, and merge in the imported branches as
+necessary.
+
+Development Models
+------------------
+
+CVS users are accustomed to giving a group of developers commit access to
+a common repository. In the next section we'll explain how to do this
+with git. However, the distributed nature of git allows other development
+models, and you may want to first consider whether one of them might be a
+better fit for your project.
+
+For example, you can choose a single person to maintain the project's
+primary public repository. Other developers then clone this repository
+and each work in their own clone. When they have a series of changes that
+they're happy with, they ask the maintainer to pull from the branch
+containing the changes. The maintainer reviews their changes and pulls
+them into the primary repository, which other developers pull from as
+necessary to stay coordinated. The Linux kernel and other projects use
+variants of this model.
+
+With a small group, developers may just pull changes from each other's
+repositories without the need for a central maintainer.
+
+Emulating the CVS Development Model
+-----------------------------------
+
+Start with an ordinary git working directory containing the project, and
+remove the checked-out files, keeping just the bare .git directory:
+
+------------------------------------------------
+$ mv project/.git /pub/repo.git
+$ rm -r project/
+------------------------------------------------
+
+Next, give every team member read/write access to this repository. One
+easy way to do this is to give all the team members ssh access to the
+machine where the repository is hosted. If you don't want to give them a
+full shell on the machine, there is a restricted shell which only allows
+users to do git pushes and pulls; see gitlink:git-shell[1].
+
+Put all the committers should in the same group, and make the repository
+writable by that group:
+
+------------------------------------------------
+$ chgrp -R $group repo.git
+$ find repo.git -mindepth 1 -type d |xargs chmod ug+rwx,g+s
+$ GIT_DIR=repo.git git repo-config core.sharedrepository true
+------------------------------------------------
+
+Make sure committers have a umask of at most 027, so that the directories
+they create are writable and searchable by other group members.
+
+Suppose this repository is now set up in /pub/repo.git on the host
+foo.com. Then as an individual commiter you can clone the shared
+repository:
+
+------------------------------------------------
+$ git clone foo.com:/pub/repo.git/ my-project
+$ cd my-project
+------------------------------------------------
+
+and hack away. The equivalent of `cvs update` is
+
+------------------------------------------------
+$ git pull origin
+------------------------------------------------
+
+which merges in any work that others might have done since the clone
+operation.
+
+[NOTE]
+================================
+The first `git clone` places the following in the
+`my-project/.git/remotes/origin` file, and that's why the previous step
+and the next step both work.
+------------
+URL: foo.com:/pub/project.git/ my-project
+Pull: master:origin
+------------
+================================
+
+You can update the shared repository with your changes using:
-So, by now you are convinced you absolutely want to work with git, but
-at the same time you absolutely have to have a central repository.
-Step back and think again. Okay, you still need a single central
-repository? There are several ways to go about that:
+------------------------------------------------
+$ git push origin master
+------------------------------------------------
-1. Designate a person responsible to pull all branches. Make the
-repository of this person public, and make every team member
-pull regularly from it.
+If someone else has updated the repository more recently, `git push`, like
+`cvs commit`, will complain, in which case you must pull any changes
+before attempting the push again.
-2. Set up a public repository with read/write access for every team
-member. Use "git pull/push" as you used "cvs update/commit". Be
-sure that your repository is up to date before pushing, just
-like you used to do with "cvs commit"; your push will fail if
-what you are pushing is not up to date.
+In the `git push` command above we specify the name of the remote branch
+to update (`master`). If we leave that out, `git push` tries to update
+any branches in the remote repository that have the same name as a branch
+in the local repository. So the last `push` can be done with either of:
-3. Make the repository of every team member public. It is the
-responsibility of each single member to pull from every other
-team member.
+------------
+$ git push origin
+$ git push repo.shared.xz:/pub/scm/project.git/
+------------
+as long as the shared repository does not have any branches
+other than `master`.
+
+[NOTE]
+============
+Because of this behaviour, if the shared repository and the developer's
+repository both have branches named `origin`, then a push like the above
+attempts to update the `origin` branch in the shared repository from the
+developer's `origin` branch. The results may be unexpected, so it's
+usually best to remove any branch named `origin` from the shared
+repository.
+============
+
+Advanced Shared Repository Management
+-------------------------------------
+
+Git allows you to specify scripts called "hooks" to be run at certain
+points. You can use these, for example, to send all commits to the shared
+repository to a mailing list. See link:hooks.txt[Hooks used by git].
+
+You can enforce finer grained permissions using update hooks. See
+link:howto/update-hook-example.txt[Controlling access to branches using
+update hooks].
CVS annotate
------------
projects / git.git / commitdiff