SUMMARY

RSS::ArchiveReader is a Perl framework that accesses archived web content and presents its entries as an RSS feed.

An "archive" in the sense of this framework is simply a set of HTML pages, each connected to the next by a link of some kind (typically a "next" hyperlink). Each time the framework is run, it accesses the page it most recently visited and follows the link to the next page (or starts on a given page if it hasn't been run before), renders that page into the body of an RSS item, follows the link to the next page, and so on, until it has rendered a fixed number of pages.

EXAMPLE

I recently became aware of the webcomic Spacetrawler which has been around for a few years. I wanted to start reading it from the beginning, but it's kind of a hassle to read a few strips, bookmark a stopping place, return to the bookmark later, delete it, and repeat until I'm all caught up. Instead, I'd rather have a fixed number of strips automatically served up in my RSS reader every day. I created the following subclass of RSS::ArchiveReader:

package SpacetrawlerArchive;

use parent qw(RSS::ArchiveReader);
use strict;

use constant {
    FEED_TITLE => 'Spacetrawler Archive',
    RSS_FILE   => "$ENV{HOME}/www/rss/spacetrawler.xml",
    FIRST_PAGE => 'http://spacetrawler.com/2010/01/01/spacetrawler-4/',
    RENDER     => '//div[contains(@class,"comicpane")]//img',
    NEXT_PAGE  => '//a[contains(@class,"navi-next")]/@href',
};

1;

Now, whenever I run the command

perl -MSpacetrawlerArchive -e 'SpacetrawlerArchive->new->run'

...another comic is appended to the spacetrawler.xml file. I can run this command in a daily cron job to get a steady stream of comics.

The base class RSS::ArchiveReader provides a number of overridable methods, but (as in the example) no overriding may be needed if an archive's content is simple enough:

• If the content to be rendered into the RSS feed is single HTML element accessible by an XPath expression, the default render method does so using the RENDER parameter.
• If each page is reached from the previous one via an attribute node (most commonly, if not always, by an <a> element's href attribute) accessible by an XPath expression, the default next_page method does so using the NEXT_PAGE parameter.
• The default title method titles each RSS item using the source page's HTML title.

See the documentation of the RSS::ArchiveReader class for more complete information. More extensive examples can be found in the examples directory.