6 git-cvsserver - A CVS server emulator for git
14 export CVS_SERVER=git-cvsserver
15 'cvs' -d :ext:user@server/path/repo.git co <HEAD_name>
17 pserver (/etc/inetd.conf):
20 cvspserver stream tcp nowait nobody /usr/bin/git-cvsserver git-cvsserver pserver
25 'git-cvsserver' [options] [pserver|server] [<directory> ...]
30 All these options obviously only make sense if enforced by the server side.
31 They have been implemented to resemble the linkgit:git-daemon[1] options as
35 Prepend 'path' to requested CVSROOT
38 Don't allow recursing into subdirectories
41 Don't check for `gitcvs.enabled` in config. You also have to specify a list
42 of allowed directories (see below) if you want to use this option.
45 Print version information and exit
48 Print usage information and exit
51 You can specify a list of allowed directories. If no directories
52 are given, all are allowed. This is an additional restriction, gitcvs
53 access still needs to be enabled by the `gitcvs.enabled` config option
54 unless '--export-all' was given, too.
60 This application is a CVS emulation layer for git.
62 It is highly functional. However, not all methods are implemented,
63 and for those methods that are implemented,
64 not all switches are implemented.
66 Testing has been done using both the CLI CVS client, and the Eclipse CVS
67 plugin. Most functionality works fine with both of these clients.
72 Currently cvsserver works over SSH connections for read/write clients, and
73 over pserver for anonymous CVS access.
75 CVS clients cannot tag, branch or perform GIT merges.
77 git-cvsserver maps GIT branches to CVS modules. This is very different
78 from what most CVS users would expect since in CVS modules usually represent
79 one or more directories.
84 1. If you are going to offer anonymous CVS access via pserver, add a line in
89 cvspserver stream tcp nowait nobody git-cvsserver pserver
92 Note: Some inetd servers let you specify the name of the executable
93 independently of the value of argv[0] (i.e. the name the program assumes
94 it was executed with). In this case the correct line in /etc/inetd.conf
98 cvspserver stream tcp nowait nobody /usr/bin/git-cvsserver git-cvsserver pserver
101 No special setup is needed for SSH access, other than having GIT tools
102 in the PATH. If you have clients that do not accept the CVS_SERVER
103 environment variable, you can rename git-cvsserver to cvs.
105 Note: Newer CVS versions (>= 1.12.11) also support specifying
106 CVS_SERVER directly in CVSROOT like
109 cvs -d ":ext;CVS_SERVER=git-cvsserver:user@server/path/repo.git" co <HEAD_name>
111 This has the advantage that it will be saved in your 'CVS/Root' files and
112 you don't need to worry about always setting the correct environment
113 variable. SSH users restricted to git-shell don't need to override the default
114 with CVS_SERVER (and shouldn't) as git-shell understands `cvs` to mean
115 git-cvsserver and pretends that the other end runs the real cvs better.
117 2. For each repo that you want accessible from CVS you need to edit config in
118 the repo and add the following section.
124 # optional for debugging
125 logfile=/path/to/logfile
128 Note: you need to ensure each user that is going to invoke git-cvsserver has
129 write access to the log file and to the database (see
130 <<dbbackend,Database Backend>>. If you want to offer write access over
131 SSH, the users of course also need write access to the git repository itself.
133 [[configaccessmethod]]
134 All configuration variables can also be overridden for a specific method of
135 access. Valid method names are "ext" (for SSH access) and "pserver". The
136 following example configuration would disable pserver access while still
137 allowing access over SSH.
146 3. If you didn't specify the CVSROOT/CVS_SERVER directly in the checkout command,
147 automatically saving it in your 'CVS/Root' files, then you need to set them
148 explicitly in your environment. CVSROOT should be set as per normal, but the
149 directory should point at the appropriate git repo. As above, for SSH clients
150 _not_ restricted to git-shell, CVS_SERVER should be set to git-cvsserver.
154 export CVSROOT=:ext:user@server:/var/git/project.git
155 export CVS_SERVER=git-cvsserver
158 4. For SSH clients that will make commits, make sure their server-side
159 .ssh/environment files (or .bashrc, etc., according to their specific shell)
160 export appropriate values for GIT_AUTHOR_NAME, GIT_AUTHOR_EMAIL,
161 GIT_COMMITTER_NAME, and GIT_COMMITTER_EMAIL. For SSH clients whose login
162 shell is bash, .bashrc may be a reasonable alternative.
164 5. Clients should now be able to check out the project. Use the CVS 'module'
165 name to indicate what GIT 'head' you want to check out. This also sets the
166 name of your newly checked-out directory, unless you tell it otherwise with
167 `-d <dir_name>`. For example, this checks out 'master' branch to the
168 `project-master` directory:
171 cvs co -d project-master master
178 git-cvsserver uses one database per git head (i.e. CVS module) to
179 store information about the repository for faster access. The
180 database doesn't contain any persistent data and can be completely
181 regenerated from the git repository at any time. The database
182 needs to be updated (i.e. written to) after every commit.
184 If the commit is done directly by using git (as opposed to
185 using git-cvsserver) the update will need to happen on the
186 next repository access by git-cvsserver, independent of
187 access method and requested operation.
189 That means that even if you offer only read access (e.g. by using
190 the pserver method), git-cvsserver should have write access to
191 the database to work reliably (otherwise you need to make sure
192 that the database is up-to-date any time git-cvsserver is executed).
194 By default it uses SQLite databases in the git directory, named
195 `gitcvs.<module_name>.sqlite`. Note that the SQLite backend creates
196 temporary files in the same directory as the database file on
197 write so it might not be enough to grant the users using
198 git-cvsserver write access to the database file without granting
199 them write access to the directory, too.
201 You can configure the database backend with the following
202 configuration variables:
204 Configuring database backend
205 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
207 git-cvsserver uses the Perl DBI module. Please also read
208 its documentation if changing these variables, especially
209 about `DBI->connect()`.
212 Database name. The exact meaning depends on the
213 selected database driver, for SQLite this is a filename.
214 Supports variable substitution (see below). May
215 not contain semicolons (`;`).
216 Default: '%Ggitcvs.%m.sqlite'
219 Used DBI driver. You can specify any available driver
220 for this here, but it might not work. cvsserver is tested
221 with 'DBD::SQLite', reported to work with
222 'DBD::Pg', and reported *not* to work with 'DBD::mysql'.
223 Please regard this as an experimental feature. May not
224 contain colons (`:`).
228 Database user. Only useful if setting `dbdriver`, since
229 SQLite has no concept of database users. Supports variable
230 substitution (see below).
233 Database password. Only useful if setting `dbdriver`, since
234 SQLite has no concept of database passwords.
236 gitcvs.dbTableNamePrefix::
237 Database table name prefix. Supports variable substitution
238 (see below). Any non-alphabetic characters will be replaced
241 All variables can also be set per access method, see <<configaccessmethod,above>>.
243 Variable substitution
244 ^^^^^^^^^^^^^^^^^^^^^
245 In `dbdriver` and `dbuser` you can use the following variables:
250 git directory name, where all characters except for
251 alpha-numeric ones, `.`, and `-` are replaced with
252 `_` (this should make it easier to use the directory
253 name in a filename if wanted)
255 CVS module/git head name
257 access method (one of "ext" or "pserver")
259 Name of the user running git-cvsserver.
260 If no name can be determined, the
263 Eclipse CVS Client Notes
264 ------------------------
266 To get a checkout with the Eclipse CVS client:
268 1. Select "Create a new project -> From CVS checkout"
269 2. Create a new location. See the notes below for details on how to choose the
271 3. Browse the 'modules' available. It will give you a list of the heads in
272 the repository. You will not be able to browse the tree from there. Only
274 4. Pick 'HEAD' when it asks what branch/tag to check out. Untick the
275 "launch commit wizard" to avoid committing the .project file.
277 Protocol notes: If you are using anonymous access via pserver, just select that.
278 Those using SSH access should choose the 'ext' protocol, and configure 'ext'
279 access on the Preferences->Team->CVS->ExtConnection pane. Set CVS_SERVER to
280 'git-cvsserver'. Note that password support is not good when using 'ext',
281 you will definitely want to have SSH keys setup.
283 Alternatively, you can just use the non-standard extssh protocol that Eclipse
284 offer. In that case CVS_SERVER is ignored, and you will have to replace
285 the cvs utility on the server with git-cvsserver or manipulate your `.bashrc`
286 so that calling 'cvs' effectively calls git-cvsserver.
288 Clients known to work
289 ---------------------
291 - CVS 1.12.9 on Debian
292 - CVS 1.11.17 on MacOSX (from Fink package)
293 - Eclipse 3.0, 3.1.2 on MacOSX (see Eclipse CVS Client Notes)
299 All the operations required for normal use are supported, including
300 checkout, diff, status, update, log, add, remove, commit.
301 Legacy monitoring operations are not supported (edit, watch and related).
302 Exports and tagging (tags and branches) are not supported at this stage.
304 CRLF Line Ending Conversions
305 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
307 By default the server leaves the '-k' mode blank for all files,
308 which causes the cvs client to treat them as a text files, subject
309 to crlf conversion on some platforms.
311 You can make the server use `crlf` attributes to set the '-k' modes
312 for files by setting the `gitcvs.usecrlfattr` config variable.
313 In this case, if `crlf` is explicitly unset ('-crlf'), then the
314 server will set '-kb' mode for binary files. If `crlf` is set,
315 then the '-k' mode will explicitly be left blank. See
316 also linkgit:gitattributes[5] for more information about the `crlf`
319 Alternatively, if `gitcvs.usecrlfattr` config is not enabled
320 or if the `crlf` attribute is unspecified for a filename, then
321 the server uses the `gitcvs.allbinary` config for the default setting.
322 If `gitcvs.allbinary` is set, then file not otherwise
323 specified will default to '-kb' mode. Otherwise the '-k' mode
324 is left blank. But if `gitcvs.allbinary` is set to "guess", then
325 the correct '-k' mode will be guessed based on the contents of
328 For best consistency with cvs, it is probably best to override the
329 defaults by setting `gitcvs.usecrlfattr` to true,
330 and `gitcvs.allbinary` to "guess".
335 git-cvsserver depends on DBD::SQLite.
337 Copyright and Authors
338 ---------------------
340 This program is copyright The Open University UK - 2006.
344 - Martyn Smith <martyn@catalyst.net.nz>
345 - Martin Langhoff <martin@catalyst.net.nz>
347 with ideas and patches from participants of the git-list <git@vger.kernel.org>.
351 Documentation by Martyn Smith <martyn@catalyst.net.nz>, Martin Langhoff <martin@catalyst.net.nz>, and Matthias Urlichs <smurf@smurf.noris.de>.
355 Part of the linkgit:git[7] suite