Web Server Utility Sources
(This page needs updating. Please update it.)
To maintain the GNU project web server, we use a lot of different free
software. Most of them are already available on the Internet, but we also
have some software which has been developed by GNU volunteers and has
never been released as GNU software, because they are somewhat specific to
our needs. Still, we feel that those programs should also be made
available so that others can make use of them. This page is a list of
We recommend storing the scripts in the usual www
CVS tree. This allows easier sharing between GNU machines that maintain
a full or partial image of the www.gnu.org web. See the
Savannah CVS page,
the “Webpages repository” information.
LINC stands for “LINC Is Not Checklink”. It scans our
web site (or any site) and makes a report of broken links on our web site. It runs
weekly on fencepost from user gnun's cron.
This Perl script fetches and saves N feeds from planet.gnu.org, using the RSS feed
http://planet.gnu.org/rss20.xml, in html format. It is then included
in the homepage. Each feed is truncated to M characters.
This program generates the sitemap of
our web site, based on its directory structure on the file
system. It runs on fencepost in user gnun's cron.
The goal is to cope with the inability of CVS to deal
with symbolic files. By adding special files (".symlinks") into
the CVS tree that are interpreted as specifications to build
symbolic links. The "symlinks" script used to be run immediately
after a "cvs update" to fix the symbolic links according to the
specifications included in the ".symlinks" files.
This script is no longer used on the live web site. Instead, the
.symlinks files are transformed into RewriteRule directives.
More information about .symlinks files can be found in the Symbolic
Links section of the GNU Webmaster Guidelines.
This script can be ran from the webroot to traverse through English html
files replacing the old "Verbatim coping" text with the new CC license.
This script generates the list of recent releases (from
archives). It runs on fencepost in user gnun's cron.
- Author: Tim Bedding
- Current version: 1.0
This is a program made in Emacs LISP that is used to check references
in HTML documents. It takes the file input.txt as input which contains
a list of HTML documents and outputs te file output.txt. The output
file contains information of documents which can not be reached at all
and documents which can only be reached by following more than two links.
We use this to make sure that all important documents can be reached
fairly easy from the main page.