A. Guide to CVS commands

This appendix describes the overall structure of CVS commands, and describes some commands in detail (others are described elsewhere; for a quick reference to CVS commands, see section Quick reference to CVS commands).

A.1 Overall structure of CVS commands

The overall format of all CVS commands is:

cvs [ cvs_options ] cvs_command [ command_options ] [ command_args ]

cvs

The name of the CVS program.

cvs_options

Some options that affect all sub-commands of CVS. These are described below.

cvs_command

One of several different sub-commands. Some of the commands have aliases that can be used instead; those aliases are noted in the reference manual for that command. There are only two situations where you may omit `cvs_command': `cvs -H' elicits a list of available commands, and `cvs -v' displays version information on CVS itself.

command_options

Options that are specific for the command.

command_args

Arguments to the commands.

There is unfortunately some confusion between cvs_options and command_options. When given as a cvs_option, some options only affect some of the commands. When given as a command_option it may have a different meaning, and be accepted by more commands. In other words, do not take the above categorization too seriously. Look at the documentation instead.

A.2 CVS's exit status

CVS can indicate to the calling environment whether it succeeded or failed by setting its exit status. The exact way of testing the exit status will vary from one operating system to another. For example in a unix shell script the `$?' variable will be 0 if the last command returned a successful exit status, or greater than 0 if the exit status indicated failure.

If CVS is successful, it returns a successful status; if there is an error, it prints an error message and returns a failure status. The one exception to this is the cvs diff command. It will return a successful status if it found no differences, or a failure status if there were differences or if there was an error. Because this behavior provides no good way to detect errors, in the future it is possible that cvs diff will be changed to behave like the other CVS commands.

A.3 Default options and the ~/.cvsrc file

There are some command_options that are used so often that you might have set up an alias or some other means to make sure you always specify that option. One example (the one that drove the implementation of the `.cvsrc' support, actually) is that many people find the default output of the `diff' command to be very hard to read, and that either context diffs or unidiffs are much easier to understand.

The `~/.cvsrc' file is a way that you can add default options to cvs_commands within cvs, instead of relying on aliases or other shell scripts.

The format of the `~/.cvsrc' file is simple. The file is searched for a line that begins with the same name as the cvs_command being executed. If a match is found, then the remainder of the line is split up (at whitespace characters) into separate options and added to the command arguments before any options from the command line.

If a command has two names (e.g., checkout and co), the official name, not necessarily the one used on the command line, will be used to match against the file. So if this is the contents of the user's `~/.cvsrc' file:

log -N
diff -uN
rdiff -u
update -Pd
checkout -P
release -d

the command `cvs checkout foo' would have the `-P' option added to the arguments, as well as `cvs co foo'.

With the example file above, the output from `cvs diff foobar' will be in unidiff format. `cvs diff -c foobar' will provide context diffs, as usual. Getting "old" format diffs would be slightly more complicated, because diff doesn't have an option to specify use of the "old" format, so you would need `cvs -f diff foobar'.

In place of the command name you can use cvs to specify global options (see section Global options). For example the following line in `.cvsrc'

cvs -z6

causes CVS to use compression level 6.

A.4 Global options

The available `cvs_options' (that are given to the left of `cvs_command') are:

--allow-root=rootdir

May be invoked multiple times to specify one legal CVSROOT directory with each invocation. Also causes CVS to preparse the configuration file for each specified root, which can be useful when configuring write proxies, See Setting up the server for password authentication & Distributing load across several CVS servers.

-a

Authenticate all communication between the client and the server. Only has an effect on the CVS client. As of this writing, this is only implemented when using a GSSAPI connection (see section Direct connection with GSSAPI). Authentication prevents certain sorts of attacks involving hijacking the active TCP connection. Enabling authentication does not enable encryption.

-b bindir

In CVS 1.9.18 and older, this specified that RCS programs are in the bindir directory. Current versions of CVS do not run RCS programs; for compatibility this option is accepted, but it does nothing.

-T tempdir

Use tempdir as the directory where temporary files are located. Overrides the setting of the $TMPDIR environment variable and any precompiled directory. This parameter should be specified as an absolute pathname. (When running client/server, `-T' affects only the local process; specifying `-T' for the client has no effect on the server and vice versa.)

-d cvs_root_directory

Use cvs_root_directory as the root directory pathname of the repository. Overrides the setting of the $CVSROOT environment variable. See section The Repository.

-e editor

Use editor to enter revision log information. Overrides the setting of the $CVSEDITOR and $EDITOR environment variables. For more information, see Committing your changes.

-f

Do not read the `~/.cvsrc' file. This option is most often used because of the non-orthogonality of the CVS option set. For example, the `cvs log' option `-N' (turn off display of tag names) does not have a corresponding option to turn the display on. So if you have `-N' in the `~/.cvsrc' entry for `log', you may need to use `-f' to show the tag names.

-H

--help

Display usage information about the specified `cvs_command' (but do not actually execute the command). If you don't specify a command name, `cvs -H' displays overall help for CVS, including a list of other help options.

-R

Turns on read-only repository mode. This allows one to check out from a read-only repository, such as within an anoncvs server, or from a CD-ROM repository.

Same effect as if the CVSREADONLYFS environment variable is set. Using `-R' can also considerably speed up checkouts over NFS.

-n

Do not change any files. Attempt to execute the `cvs_command', but only to issue reports; do not remove, update, or merge any existing files, or create any new files.

Note that CVS will not necessarily produce exactly the same output as without `-n'. In some cases the output will be the same, but in other cases CVS will skip some of the processing that would have been required to produce the exact same output.

-Q

Cause the command to be really quiet; the command will only generate output for serious problems.

-q

Cause the command to be somewhat quiet; informational messages, such as reports of recursion through subdirectories, are suppressed.

-r

Make new working files read-only. Same effect as if the $CVSREAD environment variable is set (see section All environment variables which affect CVS). The default is to make working files writable, unless watches are on (see section Mechanisms to track who is editing files).

-s variable=value

Set a user variable (see section Expansions in administrative files).

-t

Trace program execution; display messages showing the steps of CVS activity. Particularly useful with `-n' to explore the potential impact of an unfamiliar command.

-v

--version

Display version and copyright information for CVS.

-w

Make new working files read-write. Overrides the setting of the $CVSREAD environment variable. Files are created read-write by default, unless $CVSREAD is set or `-r' is given.

-x

Encrypt all communication between the client and the server. Only has an effect on the CVS client. As of this writing, this is only implemented when using a GSSAPI connection (see section Direct connection with GSSAPI) or a Kerberos connection (see section Direct connection with Kerberos). Enabling encryption implies that message traffic is also authenticated. Encryption support is not available by default; it must be enabled using a special configure option, `--enable-encryption', when you build CVS.

-z gzip-level

Set the compression level. Valid levels are 1 (high speed, low compression) to 9 (low speed, high compression), or 0 to disable compression (the default). Only has an effect on the CVS client.

A.5 Common command options

This section describes the `command_options' that are available across several CVS commands. These options are always given to the right of `cvs_command'. Not all commands support all of these options; each option is only supported for commands where it makes sense. However, when a command has one of these options you can almost always count on the same behavior of the option as in other commands. (Other command options, which are listed with the individual commands, may have different behavior from one CVS command to the other).

Note: the `history' command is an exception; it supports many options that conflict even with these standard options.

-D date_spec

Use the most recent revision no later than date_spec. date_spec is a single argument, a date description specifying a date in the past.

The specification is sticky when you use it to make a private copy of a source file; that is, when you get a working file using `-D', CVS records the date you specified, so that further updates in the same directory will use the same date (for more information on sticky tags/dates, see section Sticky tags).

`-D' is available with the annotate, checkout, diff, export, history, ls, rdiff, rls, rtag, tag, and update commands. (The history command uses this option in a slightly different way; see section history options).

For a complete description of the date formats accepted by CVS, Date input formats.

Remember to quote the argument to the `-D' flag so that your shell doesn't interpret spaces as argument separators. A command using the `-D' flag can look like this:

$ cvs diff -D "1 hour ago" cvs.texinfo

-f

When you specify a particular date or tag to CVS commands, they normally ignore files that do not contain the tag (or did not exist prior to the date) that you specified. Use the `-f' option if you want files retrieved even when there is no match for the tag or date. (The most recent revision of the file will be used).

Note that even with `-f', a tag that you specify must exist (that is, in some file, not necessary in every file). This is so that CVS will continue to give an error if you mistype a tag name.

`-f' is available with these commands: annotate, checkout, export, rdiff, rtag, and update.

WARNING: The commit and remove commands also have a `-f' option, but it has a different behavior for those commands. See commit options, and Removing files.

-k kflag

Override the default processing of RCS keywords other than `-kb'. See section Keyword substitution, for the meaning of kflag. Used with the checkout and update commands, your kflag specification is sticky; that is, when you use this option with a checkout or update command, CVS associates your selected kflag with any files it operates on, and continues to use that kflag with future commands on the same files until you specify otherwise.

The `-k' option is available with the add, checkout, diff, export, import and update commands.

WARNING: Prior to CVS version 1.12.2, the `-k' flag overrode the `-kb' indication for a binary file. This could sometimes corrupt binary files. See section Merging and keywords, for more.

-l

Local; run only in current working directory, rather than recursing through subdirectories.

Available with the following commands: annotate, checkout, commit, diff, edit, editors, export, log, rdiff, remove, rtag, status, tag, unedit, update, watch, and watchers.

-m message

Use message as log information, instead of invoking an editor.

Available with the following commands: add, commit and import.

-n

Do not run any tag program. (A program can be specified to run in the modules database (see section The modules file); this option bypasses it).

Note: this is not the same as the `cvs -n' program option, which you can specify to the left of a cvs command!

Available with the checkout, commit, export, and rtag commands.

-P

Prune empty directories. See Removing directories.

-p

Pipe the files retrieved from the repository to standard output, rather than writing them in the current directory. Available with the checkout and update commands.

-R

Process directories recursively. This is the default for all CVS commands, with the exception of ls & rls.

Available with the following commands: annotate, checkout, commit, diff, edit, editors, export, ls, rdiff, remove, rls, rtag, status, tag, unedit, update, watch, and watchers.

-r tag

Use the revision specified by the tag argument instead of the default head revision. As well as arbitrary tags defined with the tag or rtag command, two special tags are always available: `HEAD' refers to the most recent version available in the repository, and `BASE' refers to the revision you last checked out into the current working directory.

The tag specification is sticky when you use this with checkout or update to make your own copy of a file: CVS remembers the tag and continues to use it on future update commands, until you specify otherwise (for more information on sticky tags/dates, see section Sticky tags).

The tag can be either a symbolic or numeric tag, as described in Tags-Symbolic revisions, or the name of a branch, as described in Branching and merging.

Specifying the `-q' global option along with the `-r' command option is often useful, to suppress the warning messages when the RCS file does not contain the specified tag.

Note: this is not the same as the overall `cvs -r' option, which you can specify to the left of a CVS command!

`-r' is available with the annotate, checkout, commit, diff, history, export, rdiff, rtag, and update commands.

-W

Specify file names that should be filtered. You can use this option repeatedly. The spec can be a file name pattern of the same type that you can specify in the `.cvswrappers' file. Available with the following commands: import, and update.

* getdate.texi file. * * *** DO NOT EDIT THIS FILE DIRECTLY *** * * Edit getdate.texi instead. */

A.6 Date input formats

First, a quote:

Our units of temporal measurement, from seconds on up to months, are so complicated, asymmetrical and disjunctive so as to make coherent mental reckoning in time all but impossible. Indeed, had some tyrannical god contrived to enslave our minds to time, to make it all but impossible for us to escape subjection to sodden routines and unpleasant surprises, he could hardly have done better than handing down our present system. It is like a set of trapezoidal building blocks, with no vertical or horizontal surfaces, like a language in which the simplest thought demands ornate constructions, useless particles and lengthy circumlocutions. Unlike the more successful patterns of language and science, which enable us to face experience boldly or at least level-headedly, our system of temporal calculation silently and persistently encourages our terror of time.

… It is as though architects had to measure length in feet, width in meters and height in ells; as though basic instruction manuals demanded a knowledge of five different languages. It is no wonder then that we often look into our own immediate past or future, last Tuesday or a week from Sunday, with feelings of helpless confusion. …

-- Robert Grudin, Time and the Art of Living.

This section describes the textual date representations that GNU programs accept. These are the strings you, as a user, can supply as arguments to the various programs. The C interface (via the get_date function) is not described here.

A.6.1 General date syntax

A date is a string, possibly empty, containing many items separated by whitespace. The whitespace may be omitted when no ambiguity arises. The empty string means the beginning of today (i.e., midnight). Order of the items is immaterial. A date string may contain many flavors of items:

• calendar date items
• time of day items
• time zone items
• day of the week items
• relative items
• pure numbers.

We describe each of these item types in turn, below.

A few ordinal numbers may be written out in words in some contexts. This is most useful for specifying day of the week items or relative items (see below). Among the most commonly used ordinal numbers, the word `last' stands for -1, `this' stands for 0, and `first' and `next' both stand for 1. Because the word `second' stands for the unit of time there is no way to write the ordinal number 2, but for convenience `third' stands for 3, `fourth' for 4, `fifth' for 5, `sixth' for 6, `seventh' for 7, `eighth' for 8, `ninth' for 9, `tenth' for 10, `eleventh' for 11 and `twelfth' for 12.

When a month is written this way, it is still considered to be written numerically, instead of being "spelled in full"; this changes the allowed strings.

In the current implementation, only English is supported for words and abbreviations like `AM', `DST', `EST', `first', `January', `Sunday', `tomorrow', and `year'.

The output of the date command is not always acceptable as a date string, not only because of the language problem, but also because there is no standard meaning for time zone items like `IST'. When using date to generate a date string intended to be parsed later, specify a date format that is independent of language and that does not use time zone items other than `UTC' and `Z'. Here are some ways to do this:

$ LC_ALL=C TZ=UTC0 date
Mon Mar  1 00:21:42 UTC 2004
$ TZ=UTC0 date +'%Y-%m-%d %H:%M:%SZ'
2004-03-01 00:21:42Z
$ date --iso-8601=ns  # a GNU extension
2004-02-29T16:21:42,692722128-0800
$ date --rfc-2822  # a GNU extension
Sun, 29 Feb 2004 16:21:42 -0800
$ date +'%Y-%m-%d %H:%M:%S %z'  # %z is a GNU extension.
2004-02-29 16:21:42 -0800
$ date +'@%s.%N'  # %s and %N are GNU extensions.
@1078100502.692722128

Alphabetic case is completely ignored in dates. Comments may be introduced between round parentheses, as long as included parentheses are properly nested. Hyphens not followed by a digit are currently ignored. Leading zeros on numbers are ignored.

A.6.2 Calendar date items

A calendar date item specifies a day of the year. It is specified differently, depending on whether the month is specified numerically or literally. All these strings specify the same calendar date:

1972-09-24     # ISO 8601.
72-9-24        # Assume 19xx for 69 through 99,
               # 20xx for 00 through 68.
72-09-24       # Leading zeros are ignored.
9/24/72        # Common U.S. writing.
24 September 1972
24 Sept 72     # September has a special abbreviation.
24 Sep 72      # Three-letter abbreviations always allowed.
Sep 24, 1972
24-sep-72
24sep72

The year can also be omitted. In this case, the last specified year is used, or the current year if none. For example:

9/24
sep 24

Here are the rules.

For numeric months, the ISO 8601 format `year-month-day' is allowed, where year is any positive number, month is a number between 01 and 12, and day is a number between 01 and 31. A leading zero must be present if a number is less than ten. If year is 68 or smaller, then 2000 is added to it; otherwise, if year is less than 100, then 1900 is added to it. The construct `month/day/year', popular in the United States, is accepted. Also `month/day', omitting the year.

Literal months may be spelled out in full: `January', `February', `March', `April', `May', `June', `July', `August', `September', `October', `November' or `December'. Literal months may be abbreviated to their first three letters, possibly followed by an abbreviating dot. It is also permitted to write `Sept' instead of `September'.

When months are written literally, the calendar date may be given as any of the following:

day month year
day month
month day year
day-month-year

Or, omitting the year:

month day

A.6.3 Time of day items

A time of day item in date strings specifies the time on a given day. Here are some examples, all of which represent the same time:

20:02:00.000000
20:02
8:02pm
20:02-0500      # In EST (U.S. Eastern Standard Time).

More generally, the time of day may be given as `hour:minute:second', where hour is a number between 0 and 23, minute is a number between 0 and 59, and second is a number between 0 and 59 possibly followed by `.' or `,' and a fraction containing one or more digits. Alternatively, `:second' can be omitted, in which case it is taken to be zero.

If the time is followed by `am' or `pm' (or `a.m.' or `p.m.'), hour is restricted to run from 1 to 12, and `:minute' may be omitted (taken to be zero). `am' indicates the first half of the day, `pm' indicates the second half of the day. In this notation, 12 is the predecessor of 1: midnight is `12am' while noon is `12pm'. (This is the zero-oriented interpretation of `12am' and `12pm', as opposed to the old tradition derived from Latin which uses `12m' for noon and `12pm' for midnight.)

The time may alternatively be followed by a time zone correction, expressed as `shhmm', where s is `+' or `-', hh is a number of zone hours and mm is a number of zone minutes. When a time zone correction is given this way, it forces interpretation of the time relative to Coordinated Universal Time (UTC), overriding any previous specification for the time zone or the local time zone. The minute part of the time of day may not be elided when a time zone correction is used. This is the best way to specify a time zone correction by fractional parts of an hour.

Either `am'/`pm' or a time zone correction may be specified, but not both.

A.6.4 Time zone items

A time zone item specifies an international time zone, indicated by a small set of letters, e.g., `UTC' or `Z' for Coordinated Universal Time. Any included periods are ignored. By following a non-daylight-saving time zone by the string `DST' in a separate word (that is, separated by some white space), the corresponding daylight saving time zone may be specified.

Time zone items other than `UTC' and `Z' are obsolescent and are not recommended, because they are ambiguous; for example, `EST' has a different meaning in Australia than in the United States. Instead, it's better to use unambiguous numeric time zone corrections like `-0500', as described in the previous section.

If neither a time zone item nor a time zone correction is supplied, time stamps are interpreted using the rules of the default time zone (see section Specifying time zone rules).

A.6.5 Day of week items

The explicit mention of a day of the week will forward the date (only if necessary) to reach that day of the week in the future.

Days of the week may be spelled out in full: `Sunday', `Monday', `Tuesday', `Wednesday', `Thursday', `Friday' or `Saturday'. Days may be abbreviated to their first three letters, optionally followed by a period. The special abbreviations `Tues' for `Tuesday', `Wednes' for `Wednesday' and `Thur' or `Thurs' for `Thursday' are also allowed.

A number may precede a day of the week item to move forward supplementary weeks. It is best used in expression like `third monday'. In this context, `last day' or `next day' is also acceptable; they move one week before or after the day that day by itself would represent.

A comma following a day of the week item is ignored.

A.6.6 Relative items in date strings

Relative items adjust a date (or the current date if none) forward or backward. The effects of relative items accumulate. Here are some examples:

1 year
1 year ago
3 years
2 days

The unit of time displacement may be selected by the string `year' or `month' for moving by whole years or months. These are fuzzy units, as years and months are not all of equal duration. More precise units are `fortnight' which is worth 14 days, `week' worth 7 days, `day' worth 24 hours, `hour' worth 60 minutes, `minute' or `min' worth 60 seconds, and `second' or `sec' worth one second. An `s' suffix on these units is accepted and ignored.

The unit of time may be preceded by a multiplier, given as an optionally signed number. Unsigned numbers are taken as positively signed. No number at all implies 1 for a multiplier. Following a relative item by the string `ago' is equivalent to preceding the unit by a multiplier with value -1.

The string `tomorrow' is worth one day in the future (equivalent to `day'), the string `yesterday' is worth one day in the past (equivalent to `day ago').

The strings `now' or `today' are relative items corresponding to zero-valued time displacement, these strings come from the fact a zero-valued time displacement represents the current time when not otherwise changed by previous items. They may be used to stress other items, like in `12:00 today'. The string `this' also has the meaning of a zero-valued time displacement, but is preferred in date strings like `this thursday'.

When a relative item causes the resulting date to cross a boundary where the clocks were adjusted, typically for daylight saving time, the resulting date and time are adjusted accordingly.

The fuzz in units can cause problems with relative items. For example, `2003-07-31 -1 month' might evaluate to 2003-07-01, because 2003-06-31 is an invalid date. To determine the previous month more reliably, you can ask for the month before the 15th of the current month. For example:

$ date -R
Thu, 31 Jul 2003 13:02:39 -0700
$ date --date='-1 month' +'Last month was %B?'
Last month was July?
$ date --date="$(date +%Y-%m-15) -1 month" +'Last month was %B!'
Last month was June!

Also, take care when manipulating dates around clock changes such as daylight saving leaps. In a few cases these have added or subtracted as much as 24 hours from the clock, so it is often wise to adopt universal time by setting the TZ environment variable to `UTC0' before embarking on calendrical calculations.

A.6.7 Pure numbers in date strings

The precise interpretation of a pure decimal number depends on the context in the date string.

If the decimal number is of the form yyyymmdd and no other calendar date item (see section Calendar date items) appears before it in the date string, then yyyy is read as the year, mm as the month number and dd as the day of the month, for the specified calendar date.

If the decimal number is of the form hhmm and no other time of day item appears before it in the date string, then hh is read as the hour of the day and mm as the minute of the hour, for the specified time of day. mm can also be omitted.

If both a calendar date and a time of day appear to the left of a number in the date string, but no relative item, then the number overrides the year.

A.6.8 Seconds since the Epoch

If you precede a number with `@', it represents an internal time stamp as a count of seconds. The number can contain an internal decimal point (either `.' or `,'); any excess precision not supported by the internal representation is truncated toward minus infinity. Such a number cannot be combined with any other date item, as it specifies a complete time stamp.

Internally, computer times are represented as a count of seconds since an epoch--a well-defined point of time. On GNU and POSIX systems, the epoch is 1970-01-01 00:00:00 UTC, so `@0' represents this time, `@1' represents 1970-01-01 00:00:01 UTC, and so forth. GNU and most other POSIX-compliant systems support such times as an extension to POSIX, using negative counts, so that `@-1' represents 1969-12-31 23:59:59 UTC.

Traditional Unix systems count seconds with 32-bit two's-complement integers and can represent times from 1901-12-13 20:45:52 through 2038-01-19 03:14:07 UTC. More modern systems use 64-bit counts of seconds with nanosecond subcounts, and can represent all the times in the known lifetime of the universe to a resolution of 1 nanosecond.

On most systems, these counts ignore the presence of leap seconds. For example, on most systems `@915148799' represents 1998-12-31 23:59:59 UTC, `@915148800' represents 1999-01-01 00:00:00 UTC, and there is no way to represent the intervening leap second 1998-12-31 23:59:60 UTC.

A.6.9 Specifying time zone rules

Normally, dates are interpreted using the rules of the current time zone, which in turn are specified by the TZ environment variable, or by a system default if TZ is not set. To specify a different set of default time zone rules that apply just to one date, start the date with a string of the form `TZ="rule"'. The two quote characters (`"') must be present in the date, and any quotes or backslashes within rule must be escaped by a backslash.

For example, with the GNU date command you can answer the question "What time is it in New York when a Paris clock shows 6:30am on October 31, 2004?" by using a date beginning with `TZ="Europe/Paris"' as shown in the following shell transcript:

$ export TZ="America/New_York"
$ date --date='TZ="Europe/Paris" 2004-10-31 06:30'
Sun Oct 31 01:30:00 EDT 2004

In this example, the `--date' operand begins with its own TZ setting, so the rest of that operand is processed according to `Europe/Paris' rules, treating the string `2004-10-31 06:30' as if it were in Paris. However, since the output of the date command is processed according to the overall time zone rules, it uses New York time. (Paris was normally six hours ahead of New York in 2004, but this example refers to a brief Halloween period when the gap was five hours.)

A TZ value is a rule that typically names a location in the `tz' database. A recent catalog of location names appears in the TWiki Date and Time Gateway. A few non-GNU hosts require a colon before a location name in a TZ setting, e.g., `TZ=":America/New_York"'.

The `tz' database includes a wide variety of locations ranging from `Arctic/Longyearbyen' to `Antarctica/South_Pole', but if you are at sea and have your own private time zone, or if you are using a non-GNU host that does not support the `tz' database, you may need to use a POSIX rule instead. Simple POSIX rules like `UTC0' specify a time zone without daylight saving time; other rules can specify simple daylight saving regimes. See (libc)TZ Variable section `Specifying the Time Zone with TZ' in The GNU C Library.

A.6.10 Authors of get_date

get_date was originally implemented by Steven M. Bellovin (smb@research.att.com) while at the University of North Carolina at Chapel Hill. The code was later tweaked by a couple of people on Usenet, then completely overhauled by Rich $alz (rsalz@bbn.com) and Jim Berets (jberets@bbn.com) in August, 1990. Various revisions for the GNU system were made by David MacKenzie, Jim Meyering, Paul Eggert and others.

This chapter was originally produced by François Pinard (pinard@iro.umontreal.ca) from the `getdate.y' source code, and then edited by K. Berry (kb@cs.umb.edu).

A.7 admin--Administration

• Requires: repository, working directory.
• Changes: repository.
• Synonym: rcs

This is the CVS interface to assorted administrative facilities. Some of them have questionable usefulness for CVS but exist for historical purposes. Some of the questionable options are likely to disappear in the future. This command does work recursively, so extreme care should be used.

On unix, if there is a group named cvsadmin, only members of that group can run cvs admin commands, except for those specified using the UserAdminOptions configuration option in the `CVSROOT/config' file. Options specified using UserAdminOptions can be run by any user. See The CVSROOT/config configuration file for more on UserAdminOptions.

The cvsadmin group should exist on the server, or any system running the non-client/server CVS. To disallow cvs admin for all users, create a group with no users in it. On NT, the cvsadmin feature does not exist and all users can run cvs admin.

A.7.1 admin options

Some of these options have questionable usefulness for CVS but exist for historical purposes. Some even make it impossible to use CVS until you undo the effect!

-Aoldfile

Might not work together with CVS. Append the access list of oldfile to the access list of the RCS file.

-alogins

Might not work together with CVS. Append the login names appearing in the comma-separated list logins to the access list of the RCS file.

-b[rev]

Set the default branch to rev. In CVS, you normally do not manipulate default branches; sticky tags (see section Sticky tags) are a better way to decide which branch you want to work on. There is one reason to run cvs admin -b: to revert to the vendor's version when using vendor branches (see section Reverting to the latest vendor release). There can be no space between `-b' and its argument.

-cstring

Sets the comment leader to string. The comment leader is not used by current versions of CVS or RCS 5.7. Therefore, you can almost surely not worry about it. See section Keyword substitution.

-e[logins]

Might not work together with CVS. Erase the login names appearing in the comma-separated list logins from the access list of the RCS file. If logins is omitted, erase the entire access list. There can be no space between `-e' and its argument.

-I

Run interactively, even if the standard input is not a terminal. This option does not work with the client/server CVS and is likely to disappear in a future release of CVS.

-i

Useless with CVS. This creates and initializes a new RCS file, without depositing a revision. With CVS, add files with the cvs add command (see section Adding files to a directory).

-ksubst

Set the default keyword substitution to subst. See section Keyword substitution. Giving an explicit `-k' option to cvs update, cvs export, or cvs checkout overrides this default.

-l[rev]

Lock the revision with number rev. If a branch is given, lock the latest revision on that branch. If rev is omitted, lock the latest revision on the default branch. There can be no space between `-l' and its argument.

This can be used in conjunction with the `rcslock.pl' script in the `contrib' directory of the CVS source distribution to provide reserved checkouts (where only one user can be editing a given file at a time). See the comments in that file for details (and see the `README' file in that directory for disclaimers about the unsupported nature of contrib). According to comments in that file, locking must set to strict (which is the default).

-L

Set locking to strict. Strict locking means that the owner of an RCS file is not exempt from locking for checkin. For use with CVS, strict locking must be set; see the discussion under the `-l' option above.

-mrev:msg

Replace the log message of revision rev with msg.

-Nname[:[rev]]

Act like `-n', except override any previous assignment of name. For use with magic branches, see Magic branch numbers.

-nname[:[rev]]

Associate the symbolic name name with the branch or revision rev. It is normally better to use `cvs tag' or `cvs rtag' instead. Delete the symbolic name if both `:' and rev are omitted; otherwise, print an error message if name is already associated with another number. If rev is symbolic, it is expanded before association. A rev consisting of a branch number followed by a `.' stands for the current latest revision in the branch. A `:' with an empty rev stands for the current latest revision on the default branch, normally the trunk. For example, `cvs admin -nname:' associates name with the current latest revision of all the RCS files; this contrasts with `cvs admin -nname:$' which associates name with the revision numbers extracted from keyword strings in the corresponding working files.

-orange

Deletes (outdates) the revisions given by range.

Note that this command can be quite dangerous unless you know exactly what you are doing (for example see the warnings below about how the rev1:rev2 syntax is confusing).

If you are short on disc this option might help you. But think twice before using it--there is no way short of restoring the latest backup to undo this command! If you delete different revisions than you planned, either due to carelessness or (heaven forbid) a CVS bug, there is no opportunity to correct the error before the revisions are deleted. It probably would be a good idea to experiment on a copy of the repository first.

Specify range in one of the following ways:

rev1::rev2

Collapse all revisions between rev1 and rev2, so that CVS only stores the differences associated with going from rev1 to rev2, not intermediate steps. For example, after `-o 1.3::1.5' one can retrieve revision 1.3, revision 1.5, or the differences to get from 1.3 to 1.5, but not the revision 1.4, or the differences between 1.3 and 1.4. Other examples: `-o 1.3::1.4' and `-o 1.3::1.3' have no effect, because there are no intermediate revisions to remove.

::rev

Collapse revisions between the beginning of the branch containing rev and rev itself. The branchpoint and rev are left intact. For example, `-o ::1.3.2.6' deletes revision 1.3.2.1, revision 1.3.2.5, and everything in between, but leaves 1.3 and 1.3.2.6 intact.

rev::

Collapse revisions between rev and the end of the branch containing rev. Revision rev is left intact but the head revision is deleted.

rev

Delete the revision rev. For example, `-o 1.3' is equivalent to `-o 1.2::1.4'.

rev1:rev2

Delete the revisions from rev1 to rev2, inclusive, on the same branch. One will not be able to retrieve rev1 or rev2 or any of the revisions in between. For example, the command `cvs admin -oR_1_01:R_1_02 .' is rarely useful. It means to delete revisions up to, and including, the tag R_1_02. But beware! If there are files that have not changed between R_1_02 and R_1_03 the file will have the same numerical revision number assigned to the tags R_1_02 and R_1_03. So not only will it be impossible to retrieve R_1_02; R_1_03 will also have to be restored from the tapes! In most cases you want to specify rev1::rev2 instead.

:rev

Delete revisions from the beginning of the branch containing rev up to and including rev.

rev:

Delete revisions from revision rev, including rev itself, to the end of the branch containing rev.

None of the revisions to be deleted may have branches or locks.

If any of the revisions to be deleted have symbolic names, and one specifies one of the `::' syntaxes, then CVS will give an error and not delete any revisions. If you really want to delete both the symbolic names and the revisions, first delete the symbolic names with cvs tag -d, then run cvs admin -o. If one specifies the non-`::' syntaxes, then CVS will delete the revisions but leave the symbolic names pointing to nonexistent revisions. This behavior is preserved for compatibility with previous versions of CVS, but because it isn't very useful, in the future it may change to be like the `::' case.

Due to the way CVS handles branches rev cannot be specified symbolically if it is a branch. See section Magic branch numbers, for an explanation.

Make sure that no-one has checked out a copy of the revision you outdate. Strange things will happen if he starts to edit it and tries to check it back in. For this reason, this option is not a good way to take back a bogus commit; commit a new revision undoing the bogus change instead (see section Merging differences between any two revisions).

-q

Run quietly; do not print diagnostics.

-sstate[:rev]

Useful with CVS. Set the state attribute of the revision rev to state. If rev is a branch number, assume the latest revision on that branch. If rev is omitted, assume the latest revision on the default branch. Any identifier is acceptable for state. A useful set of states is `Exp' (for experimental), `Stab' (for stable), and `Rel' (for released). By default, the state of a new revision is set to `Exp' when it is created. The state is visible in the output from cvs log (see section log--Print out log information for files), and in the `$Log$' and `$State$' keywords (see section Keyword substitution). Note that CVS uses the dead state for its own purposes; to take a file to or from the dead state use commands like cvs remove and cvs add, not cvs admin -s.

-t[file]

Useful with CVS. Write descriptive text from the contents of the named file into the RCS file, deleting the existing text. The file pathname may not begin with `-'. The descriptive text can be seen in the output from `cvs log' (see section log--Print out log information for files). There can be no space between `-t' and its argument.

If file is omitted, obtain the text from standard input, terminated by end-of-file or by a line containing `.' by itself. Prompt for the text if interaction is possible; see `-I'.

-t-string

Similar to `-tfile'. Write descriptive text from the string into the RCS file, deleting the existing text. There can be no space between `-t' and its argument.

-U

Set locking to non-strict. Non-strict locking means that the owner of a file need not lock a revision for checkin. For use with CVS, strict locking must be set; see the discussion under the `-l' option above.

-u[rev]

See the option `-l' above, for a discussion of using this option with CVS. Unlock the revision with number rev. If a branch is given, unlock the latest revision on that branch. If rev is omitted, remove the latest lock held by the caller. Normally, only the locker of a revision may unlock it; somebody else unlocking a revision breaks the lock. This causes the original locker to be sent a commit notification (see section Telling CVS to notify you). There can be no space between `-u' and its argument.

-Vn

In previous versions of CVS, this option meant to write an RCS file which would be acceptable to RCS version n, but it is now obsolete and specifying it will produce an error.

-xsuffixes

In previous versions of CVS, this was documented as a way of specifying the names of the RCS files. However, CVS has always required that the RCS files used by CVS end in `,v', so this option has never done anything useful.

A.8 annotate--What revision modified each line of a file?

• Synopsis: annotate [options] files…
• Requires: repository.
• Changes: nothing.

For each file in files, print the head revision of the trunk, together with information on the last modification for each line.

A.8.1 annotate options

These standard options are supported by annotate (see section Common command options, for a complete description of them):

-l

Local directory only, no recursion.

-R

Process directories recursively.

-f

Use head revision if tag/date not found.

-F

Annotate binary files.

-r revision

Annotate file as of specified revision/tag.

-D date

Annotate file as of specified date.

A.8.2 annotate example

For example:

$ cvs annotate ssfile
Annotations for ssfile
***************
1.1          (mary     27-Mar-96): ssfile line 1
1.2          (joe      28-Mar-96): ssfile line 2

The file `ssfile' currently contains two lines. The ssfile line 1 line was checked in by mary on March 27. Then, on March 28, joe added a line ssfile line 2, without modifying the ssfile line 1 line. This report doesn't tell you anything about lines which have been deleted or replaced; you need to use cvs diff for that (see section diff--Show differences between revisions).

The options to cvs annotate are listed in Quick reference to CVS commands, and can be used to select the files and revisions to annotate. The options are described in more detail there and in Common command options.

A.9 checkout--Check out sources for editing

• Synopsis: checkout [options] modules…
• Requires: repository.
• Changes: working directory.
• Synonyms: co, get

Create or update a working directory containing copies of the source files specified by modules. You must execute checkout before using most of the other CVS commands, since most of them operate on your working directory.

The modules are either symbolic names for some collection of source directories and files, or paths to directories or files in the repository. The symbolic names are defined in the `modules' file. See section The modules file.

Depending on the modules you specify, checkout may recursively create directories and populate them with the appropriate source files. You can then edit these source files at any time (regardless of whether other software developers are editing their own copies of the sources); update them to include new changes applied by others to the source repository; or commit your work as a permanent change to the source repository.

Note that checkout is used to create directories. The top-level directory created is always added to the directory where checkout is invoked, and usually has the same name as the specified module. In the case of a module alias, the created sub-directory may have a different name, but you can be sure that it will be a sub-directory, and that checkout will show the relative path leading to each file as it is extracted into your private work area (unless you specify the `-Q' global option).

The files created by checkout are created read-write, unless the `-r' option to CVS (see section Global options) is specified, the CVSREAD environment variable is specified (see section All environment variables which affect CVS), or a watch is in effect for that file (see section Mechanisms to track who is editing files).

Note that running checkout on a directory that was already built by a prior checkout is also permitted. This is similar to specifying the `-d' option to the update command in the sense that new directories that have been created in the repository will appear in your work area. However, checkout takes a module name whereas update takes a directory name. Also to use checkout this way it must be run from the top level directory (where you originally ran checkout from), so before you run checkout to update an existing directory, don't forget to change your directory to the top level directory.

For the output produced by the checkout command see update output.

A.9.1 checkout options

These standard options are supported by checkout (see section Common command options, for a complete description of them):

-D date

Use the most recent revision no later than date. This option is sticky, and implies `-P'. See Sticky tags, for more information on sticky tags/dates.

-f

Only useful with the `-D date' or `-r tag' flags. If no matching revision is found, retrieve the most recent revision (instead of ignoring the file).

-k kflag

Process keywords according to kflag. See Keyword substitution. This option is sticky; future updates of this file in this working directory will use the same kflag. The status command can be viewed to see the sticky options. See Quick reference to CVS commands, for more information on the status command.

-l

Local; run only in current working directory.

-n

Do not run any checkout program (as specified with the `-o' option in the modules file; see section The modules file).

-P

Prune empty directories. See Moving and renaming directories.

-p

Pipe files to the standard output.

-R

Checkout directories recursively. This option is on by default.

-r tag

Use revision tag. This option is sticky, and implies `-P'. See Sticky tags, for more information on sticky tags/dates.

In addition to those, you can use these special command options with checkout:

-A

Reset any sticky tags, dates, or `-k' options. See Sticky tags, for more information on sticky tags/dates.

-c

Copy the module file, sorted, to the standard output, instead of creating or modifying any files or directories in your working directory.

-d dir

Create a directory called dir for the working files, instead of using the module name. In general, using this flag is equivalent to using `mkdir dir; cd dir' followed by the checkout command without the `-d' flag.

There is an important exception, however. It is very convenient when checking out a single item to have the output appear in a directory that doesn't contain empty intermediate directories. In this case only, CVS tries to "shorten" pathnames to avoid those empty directories.

For example, given a module `foo' that contains the file `bar.c', the command `cvs co -d dir foo' will create directory `dir' and place `bar.c' inside. Similarly, given a module `bar' which has subdirectory `baz' wherein there is a file `quux.c', the command `cvs co -d dir bar/baz' will create directory `dir' and place `quux.c' inside.

Using the `-N' flag will defeat this behavior. Given the same module definitions above, `cvs co -N -d dir foo' will create directories `dir/foo' and place `bar.c' inside, while `cvs co -N -d dir bar/baz' will create directories `dir/bar/baz' and place `quux.c' inside.

-j tag

With two `-j' options, merge changes from the revision specified with the first `-j' option to the revision specified with the second `j' option, into the working directory.

With one `-j' option, merge changes from the ancestor revision to the revision specified with the `-j' option, into the working directory. The ancestor revision is the common ancestor of the revision which the working directory is based on, and the revision specified in the `-j' option.

In addition, each -j option can contain an optional date specification which, when used with branches, can limit the chosen revision to one within a specific date. An optional date is specified by adding a colon (:) to the tag: `-jSymbolic_Tag:Date_Specifier'.

See section Branching and merging.

-N

Only useful together with `-d dir'. With this option, CVS will not "shorten" module paths in your working directory when you check out a single module. See the `-d' flag for examples and a discussion.

-s

Like `-c', but include the status of all modules, and sort it by the status string. See section The modules file, for info about the `-s' option that is used inside the modules file to set the module status.

A.9.2 checkout examples

Get a copy of the module `tc':

$ cvs checkout tc

Get a copy of the module `tc' as it looked one day ago:

$ cvs checkout -D yesterday tc

A.10 commit--Check files into the repository

• Synopsis: commit [-lnRf] [-m 'log_message' | -F file] [-r revision] [files…]
• Requires: working directory, repository.
• Changes: repository.
• Synonym: ci

Use commit when you want to incorporate changes from your working source files into the source repository.

If you don't specify particular files to commit, all of the files in your working current directory are examined. commit is careful to change in the repository only those files that you have really changed. By default (or if you explicitly specify the `-R' option), files in subdirectories are also examined and committed if they have changed; you can use the `-l' option to limit commit to the current directory only.

commit verifies that the selected files are up to date with the current revisions in the source repository; it will notify you, and exit without committing, if any of the specified files must be made current first with update (see section update--Bring work tree in sync with repository). commit does not call the update command for you, but rather leaves that for you to do when the time is right.

When all is well, an editor is invoked to allow you to enter a log message that will be written to one or more logging programs (see section The modules file, and see section Loginfo) and placed in the RCS file inside the repository. This log message can be retrieved with the log command; see log--Print out log information for files. You can specify the log message on the command line with the `-m message' option, and thus avoid the editor invocation, or use the `-F file' option to specify that the argument file contains the log message.

A.10.1 commit options

These standard options are supported by commit (see section Common command options, for a complete description of them):

-l

Local; run only in current working directory.

-R

Commit directories recursively. This is on by default.

-r revision

Commit to revision. revision must be either a branch, or a revision on the main trunk that is higher than any existing revision number (see section Assigning revisions). You cannot commit to a specific revision on a branch.

commit also supports these options:

-c

Refuse to commit files unless the user has registered a valid edit on the file via cvs edit. This is most useful when `commit -c' and `edit -c' have been placed in all `.cvsrc' files. A commit can be forced anyways by either regestering an edit retroactively via cvs edit (no changes to the file will be lost) or using the -f option to commit. Support for commit -c requires both client and a server versions 1.12.10 or greater.

-F file

Read the log message from file, instead of invoking an editor.

-f

Note that this is not the standard behavior of the `-f' option as defined in Common command options.

Force CVS to commit a new revision even if you haven't made any changes to the file. As of CVS version 1.12.10, it also causes the -c option to be ignored. If the current revision of file is 1.7, then the following two commands are equivalent:

$ cvs commit -f file $ cvs commit -r 1.8 file

The `-f' option disables recursion (i.e., it implies `-l'). To force CVS to commit a new revision for all files in all subdirectories, you must use `-f -R'.

-m message

Use message as the log message, instead of invoking an editor.

A.10.2 commit examples

A.10.2.1 Committing to a branch

You can commit to a branch revision (one that has an even number of dots) with the `-r' option. To create a branch revision, use the `-b' option of the rtag or tag commands (see section Branching and merging). Then, either checkout or update can be used to base your sources on the newly created branch. From that point on, all commit changes made within these working sources will be automatically added to a branch revision, thereby not disturbing main-line development in any way. For example, if you had to create a patch to the 1.2 version of the product, even though the 2.0 version is already under development, you might do:

$ cvs rtag -b -r FCS1_2 FCS1_2_Patch product_module
$ cvs checkout -r FCS1_2_Patch product_module
$ cd product_module
[[ hack away ]]
$ cvs commit

This works automatically since the `-r' option is sticky.

A.10.2.2 Creating the branch after editing

Say you have been working on some extremely experimental software, based on whatever revision you happened to checkout last week. If others in your group would like to work on this software with you, but without disturbing main-line development, you could commit your change to a new branch. Others can then checkout your experimental stuff and utilize the full benefit of CVS conflict resolution. The scenario might look like:

[[ hacked sources are present ]]
$ cvs tag -b EXPR1
$ cvs update -r EXPR1
$ cvs commit

The update command will make the `-r EXPR1' option sticky on all files. Note that your changes to the files will never be removed by the update command. The commit will automatically commit to the correct branch, because the `-r' is sticky. You could also do like this:

[[ hacked sources are present ]]
$ cvs tag -b EXPR1
$ cvs commit -r EXPR1

but then, only those files that were changed by you will have the `-r EXPR1' sticky flag. If you hack away, and commit without specifying the `-r EXPR1' flag, some files may accidentally end up on the main trunk.

To work with you on the experimental change, others would simply do

$ cvs checkout -r EXPR1 whatever_module