6 git-cvsserver - A CVS server emulator for git
11 export CVS_SERVER=git-cvsserver
12 'cvs' -d :ext:user@server/path/repo.git co <HEAD_name>
17 This application is a CVS emulation layer for git.
19 It is highly functional. However, not all methods are implemented,
20 and for those methods that are implemented,
21 not all switches are implemented.
23 Testing has been done using both the CLI CVS client, and the Eclipse CVS
24 plugin. Most functionality works fine with both of these clients.
29 Currently cvsserver works over SSH connections for read/write clients, and
30 over pserver for anonymous CVS access.
32 CVS clients cannot tag, branch or perform GIT merges.
37 1. If you are going to offer anonymous CVS access via pserver, add a line in
42 cvspserver stream tcp nowait nobody git-cvsserver pserver
45 Note: In some cases, you need to pass the 'pserver' argument twice for
46 git-cvsserver to see it. So the line would look like
49 cvspserver stream tcp nowait nobody git-cvsserver pserver pserver
52 No special setup is needed for SSH access, other than having GIT tools
53 in the PATH. If you have clients that do not accept the CVS_SERVER
54 env variable, you can rename git-cvsserver to cvs.
56 2. For each repo that you want accessible from CVS you need to edit config in
57 the repo and add the following section.
63 # optional for debugging
64 logfile=/path/to/logfile
67 Note: you need to ensure each user that is going to invoke git-cvsserver has
68 write access to the log file and to the database (see
69 <<dbbackend,Database Backend>>. If you want to offer write access over
70 SSH, the users of course also need write access to the git repository itself.
72 [[configaccessmethod]]
73 All configuration variables can also be overriden for a specific method of
74 access. Valid method names are "ext" (for SSH access) and "pserver". The
75 following example configuration would disable pserver access while still
76 allowing access over SSH.
85 3. On the client machine you need to set the following variables.
86 CVSROOT should be set as per normal, but the directory should point at the
87 appropriate git repo. For example:
90 For SSH access, CVS_SERVER should be set to git-cvsserver
95 export CVSROOT=:ext:user@server:/var/git/project.git
96 export CVS_SERVER=git-cvsserver
99 4. For SSH clients that will make commits, make sure their .bashrc file
100 sets the GIT_AUTHOR and GIT_COMMITTER variables.
102 5. Clients should now be able to check out the project. Use the CVS 'module'
103 name to indicate what GIT 'head' you want to check out. Example:
106 cvs co -d project-master master
113 git-cvsserver uses one database per git head (i.e. CVS module) to
114 store information about the repository for faster access. The
115 database doesn't contain any persitent data and can be completly
116 regenerated from the git repository at any time. The database
117 needs to be updated (i.e. written to) after every commit.
119 If the commit is done directly by using git (as opposed to
120 using git-cvsserver) the update will need to happen on the
121 next repository access by git-cvsserver, independent of
122 access method and requested operation.
124 That means that even if you offer only read access (e.g. by using
125 the pserver method), git-cvsserver should have write access to
126 the database to work reliably (otherwise you need to make sure
127 that the database if up-to-date all the time git-cvsserver is run).
129 By default it uses SQLite databases in the git directory, named
130 `gitcvs.<module_name>.sqlite`. Note that the SQLite backend creates
131 temporary files in the same directory as the database file on
132 write so it might not be enough to grant the users using
133 git-cvsserver write access to the database file without granting
134 them write access to the directory, too.
136 You can configure the database backend with the following
137 configuration variables:
139 Configuring database backend
140 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
142 git-cvsserver uses the Perl DBI module. Please also read
143 its documentation if changing these variables, especially
144 about `DBI->connect()`.
147 Database name. The exact meaning depends on the
148 used database driver, for SQLite this is a filename.
149 Supports variable substitution (see below). May
150 not contain semicolons (`;`).
151 Default: '%Ggitcvs.%m.sqlite'
154 Used DBI driver. You can specify any available driver
155 for this here, but it might not work. cvsserver is tested
156 with 'DBD::SQLite', reported to work with
157 'DBD::Pg', and reported *not* to work with 'DBD::mysql'.
158 Please regard this as an experimental feature. May not
159 contain double colons (`:`).
163 Database user. Only useful if setting `dbdriver`, since
164 SQLite has no concept of database users. Supports variable
165 substitution (see below).
168 Database password. Only useful if setting `dbdriver`, since
169 SQLite has no concept of database passwords.
171 All variables can also be set per access method, see <<configaccessmethod,above>>.
173 Variable substitution
174 ^^^^^^^^^^^^^^^^^^^^^
175 In `dbdriver` and `dbuser` you can use the following variables:
180 git directory name, where all characters except for
181 alpha-numeric ones, `.`, and `-` are replaced with
182 `_` (this should make it easier to use the directory
183 name in a filename if wanted)
185 CVS module/git head name
187 access method (one of "ext" or "pserver")
189 Name of the user running git-cvsserver.
190 If no name can be determined, the
193 Eclipse CVS Client Notes
194 ------------------------
196 To get a checkout with the Eclipse CVS client:
198 1. Select "Create a new project -> From CVS checkout"
199 2. Create a new location. See the notes below for details on how to choose the
201 3. Browse the 'modules' available. It will give you a list of the heads in
202 the repository. You will not be able to browse the tree from there. Only
204 4. Pick 'HEAD' when it asks what branch/tag to check out. Untick the
205 "launch commit wizard" to avoid committing the .project file.
207 Protocol notes: If you are using anonymous access via pserver, just select that.
208 Those using SSH access should choose the 'ext' protocol, and configure 'ext'
209 access on the Preferences->Team->CVS->ExtConnection pane. Set CVS_SERVER to
210 'git-cvsserver'. Not that password support is not good when using 'ext',
211 you will definitely want to have SSH keys setup.
213 Alternatively, you can just use the non-standard extssh protocol that Eclipse
214 offer. In that case CVS_SERVER is ignored, and you will have to replace
215 the cvs utility on the server with git-cvsserver or manipulate your .bashrc
216 so that calling 'cvs' effectively calls git-cvsserver.
218 Clients known to work
219 ---------------------
221 - CVS 1.12.9 on Debian
222 - CVS 1.11.17 on MacOSX (from Fink package)
223 - Eclipse 3.0, 3.1.2 on MacOSX (see Eclipse CVS Client Notes)
229 All the operations required for normal use are supported, including
230 checkout, diff, status, update, log, add, remove, commit.
231 Legacy monitoring operations are not supported (edit, watch and related).
232 Exports and tagging (tags and branches) are not supported at this stage.
234 The server will set the -k mode to binary when relevant. In proper GIT
235 tradition, the contents of the files are always respected.
236 No keyword expansion or newline munging is supported.
241 git-cvsserver depends on DBD::SQLite.
243 Copyright and Authors
244 ---------------------
246 This program is copyright The Open University UK - 2006.
250 - Martyn Smith <martyn@catalyst.net.nz>
251 - Martin Langhoff <martin@catalyst.net.nz>
253 with ideas and patches from participants of the git-list <git@vger.kernel.org>.
257 Documentation by Martyn Smith <martyn@catalyst.net.nz>, Martin Langhoff <martin@catalyst.net.nz>, and Matthias Urlichs <smurf@smurf.noris.de>.
261 Part of the gitlink:git[7] suite