First thing to note is, it has to be able to read the email archive.  While it
can read Maildir and mbox mailboxes, it was designed for use with MLMMJ, and can
read MLMMJ's own archives directly if it is run by a user that can read those.

Its configuration is expected to be found in a directory named:
  /etc/ml-web-archiver

There's an optional config file: /etc/ml-web-archiver/ml-web-archiver.conf.php

There are only two config settings currently: $syslog = TRUE;  Which causes it
to send some informational messages and errors to the syslog.  Errors are sent
to STDERR either way, which will cause cron to send you an email, but if you use
systemd, you won't get an email because systemd doesn't work that way. *sigh*

And the other is: $verbose = TRUE;  Which causes it to show the informational
messages to STDOUT.

As this is a PHP file, it needs to look like:
  <?php
  $syslog = TRUE;
  $verbose = TRUE;

There is an optional list of lists at: /etc/ml-web-archiver/lists

The format for this a (PHP's implementation) INI file, and looks like:
[mlmmj-test@lists.example.com]
description = MLMMJ Test List
input = /var/spool/mlmmj/lists.example.com/mlmmj-test/archive
output = /var/www/ron/lists.example.com/www/archives/mlmmj-test
archiveurl = https://lists.example.com/archives/mlmmj-test
listurl = https://lists.example.com/mlmmj-test
searchurl = https://lists.example.com/search

The list address appears in the brackets, and the following lines describe that
list.
The archiveurl is necessary to make the RSS feed work.
The listurl is optional, and points to a Web page dedicated to the list.
The searchurl is optional, its presence enables the search functionality for that
list.

This file can contain comments, where the comment character is a semicolon: ;

To run the script, after populating the list of lists, simply run:
   ml-web-archiver

Or you can process a single list by running:
   ml-web-archiver --list listaddress

Where listaddress is the address of the list to process.

If you are using the searching functionality, you need to define the searchurl
value in the /etc/ml-web-archiver/lists file, and either copy search.php, or
symlink search.php into the filesystem location that corresponds to searchurl.

That's it.  Presuming it runs (it should work with PHP-CLI 7.3+), and correctly,
you'll get web pages in the output directory, along with zip files, and an RSS
feed.
