nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

rss.xml

nipper

OK, got a site that is DCR driven. Setting up an rss feed.

Figured out the XML form, so I can build this XML file, sorted by update time. Easy enough.

The big question is what the best way to update it ? My current approach is to regenerate it every time a page changes. Not the most scalable model in the world. Making things worse is that the clients CMS server is significantly less powerful than the laptop I carry around. So it takes 4-5 minutes to traverse templatedata, find RSS files, sort by mod date and build the file.

I could write something that will just push new pages at the top of the file, but still I need to worry about removing duplicates and old data.

How have other people handled this ?

Find more posts tagged with

Comments

Jamik

well, if you have a database being used on the web server side of things you could always just create a service that looks at the necessary fields in the database to generate the rss feed. Then that gets you away from using the slow CMS machine to generate the feed. Alternatively, you could have a batch job run daily or hourly on the webserver that looks for modified files and updates the rss feed(s) as needed. Other than that it's gonna have to be done either the way you're doing it now or through workflow. I suppose you could call a service on another machine to do the dirty work for you so that you take some of the load off the CMS machine.

One side question i have is does the client have SitePublisher license? We have a contract company in house currently for redesigning our external website and they've been pushing SitePublisher hard since evidently it has full fledged RSS capabilities. Have you looked into that maybe too? I'm not sure of the capabilities from SitePublishers side but our consultants swear by it.

Gregg Faus

When I generate RSS feeds, I parse a summary XML file (derived from each DCR) which picks out the fields that I use in the RSS feed (date, title, body, etc). However, the summary file is generated every time a DCR is submitted via workflow.

So in the case when I've had a large data directory to traverse/parse, instead of iterating through all DCR's and creating a new summary file every time, I use the XML:Twig library to update the summary file as needed. Then I generate the RSS feed from that.

I've got some data directories with over 3,000 records and by using the "update" approach it has saved well over 3-4 minutes of processing time per submittal.