The London Perl and Raku Workshop takes place on 26th Oct 2024. If your company depends on Perl, please consider sponsoring and/or attending.


populate - Syncronize the development web server with the CVS tree


populate { -h | project_name } [ options... ]


The populate tool is the first stage of the three-tool release process. It is used to synchronize the development web server documents area with the CVS repository.


There are a wide range of options to this tool. Unless using the -h option, the command line must include at least one project name. Other options are:


Print a short usage summary and terse explanation of the available options. Exits immediately, even if passed with project names.

-t title

When updating the topic-list file, use title in lieu of the project name. The topic-list file is used by the release management system during other stages, in constructing headers for data transmission, etc.

-r symbol

Specificy a particular CVS/RCS tag for the CVS commands to retrieve during the update of files from the CVS repository. Without this, CVS will update using the most-recent (head) revisions of files.

-e addr

If any errors occur that are deemed critical enough to warrant e-mail, this will specify additional addresses. The person running the command, the listed "owner" of the project, and the Webmaster Team are already targeted in the e-mail, so this option need not reference any of those.

-cvsroot dir

Specify an alternate CVSROOT value. This is rarely used, as the CVS environment is very carefully managed by ATG.


Enable some debugging-related output. This is mostly to keep users abreast of what point within the process the tool is currently at.


Enable some additional informational messages within the tool. This differs from -debug in that the messages are more geared towards progress reporting than iterating command-line options before command execution, for example.


Suppress the majority of the "normal" output, resulting in less feedback from the tool than is usually generated.

-log file

Specify a different file name for logging output. The normal run of the release tools generates some amount of logging data, used mainly as a checklist in the case of problem diagnosis. If you wish the output to go to a specific location, use this option. Note that if the file exists, it must be writable by the user ID that the tools run as, not your own. This is generally idsweb. If the file does not exist, it will be created and owned by this UID.


Lastly, this option displays a brief summary of these options, then causes the tool to exit, even if there are any projects specified.

For any of the options besides -h, one or more project names must be specified. If more than one is passed, they are processed in the order given. At the end of execution, you will be informed if any were not able to successfully complete execution.


stage, release, cvs


Rather than always force redundant processing, the stage tool no longer runs populate unconditionally. It first checks that it is necessary to run this tool.


Randy J. Ray <>


Kevin Greene