The aeimport command is used to create a new project, and populate it by importing a foreign repository (such as RCS or CVS) without loss of project history.
Please note: unless you specify a version (see the
below) this command will default to creating branches to support version
1.0. If you discovered this too late, all is not lost: you can use the
aenbru(1) command to get rid of the branches you didn't want.
The project directory, under which the project baseline and history and
state and change data are kept, will be created at this time. If the
-DIRectory option is not given, the project directory will be
created in the directory specified by the default_project_directory
aeuconf(5), or if not set in current user's home directory;
in either case with the same name as the project.
The project is created with the current user and group as the owning user and group. The current user is an administrator for the project. The project has no other administrators (use aena(1) to add more).
The project will have all user names found in the history files (see blow) installed as developers, reviewers and integrators. This is probably too broad, but fairly accurately reproduces the wide-open permissions found in most repositories, and you will want to use aerd(1), aerrv(1) and aeri(1) as appropriate to winnow this list.
If only one name is found, the project will be set to
developers_may_review = true;” otherwise it will be false
aepattr(5) for more information). Use aepa(1) to change
this if you want a different setting.
The project's umask is derived from the current user's umask, but modified to guarantee that group members will have access and that only the project owner will have write access. In general, it's best of the project is not owned by an account with any other role, as this prevents a whole class of “oops, I thought I was somewhere else” errors.
The project's history commands (see
aepconf(5) for more information)
are set to those suitable for RCS. The build command is set to “exit
0”; you need to set it to something suitable. The symbolic link farm
is turned on.
The project pointer will be added to the first element of the search
path, or /var/lib/aegis if no path is set. If this is inappropriate,
-LIBrary option to explicitly set the desired location.
-LIBrary option for more information.
Alternatively, unset the AEGIS_PATH environment variable to add the project to the global project list.
You may specify the project version in two ways:
The version number may be implicit in the project name, in which case the version numbers will be stripped off. For example, “aeimport -p example.1.2” will create a project called “example” with branch number 1 created, and sub-branch 2 of branch 1 created.
The version number may be stated explicitly, in which case it will be subdivided for branch numbers. For example, “aeimport -p example -version 1.2” will create a project called “example” with branch number 1 created, and sub-branch 2 of branch 1 created.
In each case, these branches may be named wherever a project name may be given, such as “-p example.1” and “-p example-1.2”. The actual punctuation character is unimportant.
You may have any depth of version numbers you like. Both methods
of specifying version numbers may be used, and they will be combined.
If you want no version numbers at all, use
-version with a single
dash as the argument, as in “
If no version number is given, either explicitly or implicitly, version 1.0 is used.
Please Note: Aegis also consults the underlying file system, to
determine its notion of maximum file size. Where the file system's
maximum file size is less than
filesystem wins. This can happen, for example, when you are using the
Linux UMSDOS file system, or when you have an NFS mounted an ancient
V7 filesystem. Setting
maximum_filename_length to 255 in these
cases does not alter the fact that the underlying file systems limits
are far smaller (12 and 14, respectively).
If your development directories (or your whole project) is on filesystems with filename limitations, or a portion of the heterogeneous builds take place in such an environment, it helps to tell Aegis what they are (using the project config file's fields) so that you don't run into the situation where the project builds on the more permissive environments, but fails with mysterious errors in the more limited environments.
If your development directories are routinely on a Linux
UMSDOS filesystem, you would probably be better off setting
dos_filename_required = true,
and also changing the
Heterogeneous development with various Windows environments may
also require this.
Most file version systems do not operate using change sets. In order to import such repositories into Aegis it is necessary to “discover” these change sets. The following steps are taken:
The directory (dirpath) given on the command line, and all
directories below it, are scanned for appropriate files (for example,
RCS and CVS use files with a “
These files are read to obtain the file's history.
If you have been using a non-standard file suffix, aeimport won't be able to find the files.
If you have more than one module in your CVS repository, aeimport doesn't (yet) understand the CVSROOT/modules file. Pointing aeimport at your whole CVSROOT may produce an unexpectedly large result.
The history files discovered in the previous step are copied into the location used by Aegis. Unlike some other tools, Aegis has a repository per project, rather than all projects sharing the same repository.
This also means that Aegis will not modify the original history files. In particular, if the import produces unexpected results, simply remove the project (see aermpr(1) for more information) and start again.
It is not possible to leave all your history files under, say, $CVSROOT and have Aegis point to them.
For each user mentioned in the various file histories, the time stamps are examined to find groups of files which were committed at around the same time. Files changed within 1 minute of each other are considered a group.
Files change within one minute, but by different users, are not considered a group. This does not usually present a problem as developers mostly work alone. In rare cases where developers work together, only one of them does the commit.
In some cases the time window may be too large, and several very small changes may be seen as one larger change set. In practice, this isn't very common.
Groups of files are stored into the Aegis database as completed changes (i.e. as if aeipass(1) has already run). The description of the change is the concatenation of all the unique comments found attached to the relevant file versions. The time stamp used for the change is the latest time stamp of any file in the group.
There are times when small typographical errors between file comments result in longer-than-expected change descriptions. This can be corrected with aeca(1) or tkaeca(1) if desired. There are also times when the reverse is true: some files have no comments at all, and the resulting description is less than useful.
Tags are turned into delta names by transferring delta names from the files they are attached to, to the change sets they are attached to. When a tag would appear to be attached to more than one change, it is attached only to the latest change.
In common usage, the tags serve a similar purpose as Aegis' delta numbers. They are all (typically) applied in a single CVS command, in order that a particular release may be recreated later. However, because each file will be at a different version, and each will have had its latest version included in various random change sets.
Tags are used for other things too. The method given here is simply a guess, but it's one which works reasonably well.
Once aeimport has completed importing a project, you will be able to examine the results using the ael project_history and ael change_details commands. (See ael(1) for more information.)
The aeimport program is far from perfect. There are a number of known limitations.
At this time, there is no support for branching. (As soon as I figure out how to discern the root of a branch across loosely coupled files, I'll implement it. Ideas and/or code contributions welcome.)
Only RCS and SCCS formats are understood at present. It should be straight forward to add support for additional formats in the future. Only step 1 of the above process requires attention, the rest is file format neutral.
There is no support for CVS modules, and there needs to be.
You can't specify the time window size used to determine change sets. Time will tell whether this is necessary, but it begs the question: how will you know what window size you need in order to use the option at all.
You can't import a CVS repository into an existing project. You may only create a new project from a CVS repository.
You can't import a remote CVS repository.
The following options are understood:
This option may be used to specify which directory is to be used. It is an error if the current user does not have appropriate permissions to create the directory path given. This must be an absolute path.
Caution: If you are using an automounter do not use `pwd` to make an absolute path, it usually gives the wrong answer.
This option may be use to specify which history format is being imported. The following formats are understood:
Release Control System format has been around for quite a while.
It is the format underlying CVS (Concurrent Version System).
This is the default if no format name is specified.
Note: you must have RCS installed before you run aeimport
if you use this format, because RCS commands will be run during the import
process. The import will fail if RCS is not installed. You can find
a freeware implementation at
ftp.gnu.org, or a local mirror.
Source Code Control System is one of the earliest Unix version systems.
(I'm told this is the format underlying BitKeeper.)
Note: you must have SCCS installed before you run aeimport
if you use this format, because SCCS commands will be run during the
import process. The import will fail if SCCS is not installed.
The GNU Compatibly Stupid Source Control (CSSC) is a
freeware implementation of SCCS, and it may be found at
This option may be used to
specify a directory to be searched for global state files and
user state files.
for more information.)
Several library options may be present on the command line,
and are search in the order given.
Appended to this explicit search path are the directories specified
environment variable (colon separated),
is always searched.
All paths specified,
either on the command line or in the
must be absolute.
This option may be used to obtain a list of suitable subjects for this command. The list may be more general than expected.
This option may be used to select the project of interest.
option is specified,
environment variable is consulted.
If that does not exist,
file is examined for a default project field (see
for more information).
If that does not exist,
when the user is only working on changes within a single project,
the project name defaults to that project.
it is an error.
This option may be used to obtain more information about how to use the aegis program.
This option may be used to specify the version number
for the project.
Version numbers are implemented as branches.
Use a single dash (“
-”) as the argument
if you want no version branches created.
See also aegis(1) for options common to all aegis commands.
All options may be abbreviated; the abbreviation is documented as the upper case letters, all lower case letters and underscores (_) are optional. You must use consecutive sequences of optional letters.
All options are case insensitive, you may type them in upper case or lower case or a combination of both, case is not important.
the arguments "-project, "-PROJ" and "-p" are
all interpreted to mean the
The argument "-prj" will not be understood,
because consecutive optional characters were not supplied.
Options and other command line arguments may be mixed arbitrarily on the command line, after the function selectors.
The GNU long option names are understood.
Since all option names for
this means ignoring the extra leading '-'.
option=value" convention is also understood.
The aeimport command will exit with a status of 1 on any error. The aeimport command will only exit with a status of 0 if there are no errors.
See aegis(1) for a list of environment variables which may affect
aepconf(5) for the project configuration file's
project_specific field for how to set environment variables for
all commands executed by Aegis.
aeimport version 4.22 Copyright (C) 1991-2006 Peter Miller; All rights reserved.
The aeimport program comes with ABSOLUTELY NO WARRANTY; for details use the 'aeimport -VERSion License' command. This is free software and you are welcome to redistribute it under certain conditions; for details use the 'aeimport -VERSion License' command.