
Apache::RedirectDBI - Redirect requests to different directories based on the existence of a user in one or more database tables

PerlModule Apache::DBI Apache::RedirectDBI
<Location /path/to/virtual/directory> SetHandler perl-script PerlHandler Apache::RedirectDBI PerlAuthenHandler Apache::AuthenDBI
PerlSetVar Auth_DBI_data_source dbi:Oracle:CERT
# :
# and other Auth_DBI_* variables
# :
PerlSetVar RedirectDBI_data_source dbi:Oracle:CERT
PerlSetVar RedirectDBI_username nobody
PerlSetVar RedirectDBI_password nobody
PerlSetVar RedirectDBI_location /path/to/virtual/directory
PerlSetVar RedirectDBI_default /path/to/virtual/directory.1
PerlSetVar RedirectDBI_table2uri "t1 /directory.2 t2 /directory.3"
PerlSetVar RedirectDBI_field DB_field_name_containing_username
PerlSetVar RedirectDBI_url DB_field_name_containing_redirect_url
PerlSetVar RedirectDBI_external External_REDIRECT_is_issued
AuthName "Realm"
AuthType Basic
Require valid-user
</Location>

Apache::RedirectDBI allows you to create a virtual path in your document hierarchy.
All requests for access to this virtual path should require a username and password to access. When the user attempts to access this virtual path their username is looked up in one or more database tables. The table in which the username is found in determines the physical path from which files are served.
For internal redirects the files are served to the user without the URL changing, so they never know that they have been redirected. For external redirects the web browser is sent a redirected to the desired URL.
The user is redirected to a default location if they are not in any of the database tables.

First, define the virtual location that a user will see. You must also create other directories from which files will be served.
For example, specify /dir as the virtual directory, and have $DOCROOT/dir.1, $DOCROOT/dir.2 and $DOCROOT/dir.3 as three possible directories that files will be served from, depending on the table that lists the user.
<Location /dir>
...
</Location>
The different configuration directives in httpd.conf have the following meanings;
A DBI identifier for the data source that contains the tables that will be used to determine which directory to send the user to.
The username to use when connecting to the data source.
The password to use when connecting to the data source.
The same path as used in the <Location ...> section of this configuration.
Path (relative to the document root) from which files will be served if the user does not exist in any of the database files.
A sting that provides the field name that contains the login username.
A string that provides the field name containing the URL name to use for the user logged in. If this directive is not provided, an URI is obtianed from the second subparameter from the RedirectDBI_table2uri directive.
If the directive is set to any value, an external REDIRECT is presented to the client. If it is not set, an internal redirect is issued.
A string containing white space seperated elements. If the RedirectDBI_url directive is not set, each element is forms a pair where the first element is the name of the user table and the second element is the redirect location. If the RedirectDBI_url is set, the list is just the user tables to scan. The URI directory is relative to the document root from which files will be redireted if the user is in this table.

The tables listed in the RedirectDBI_table2uri string must contain one or more columns. One of these columns must be contain the username, which is specified by the RedirectDBI_field directive.
These tables do not necessarily have to be real tables. If the backend database supports it then they could be views. This allows for a lot of flexibility in specifying the criteria for the inclusion of a user in the table.


perl(1), Apache(3), Apache::DBI(3)

Original Apache module
Original CGI scripts which this replaces, and this documentation.