OfflineRSS

A rewrite of the offlinerss ruby gem in Go.

Downloads RSS feed URLs and split each entry to an XML file and write them to ~/rss/INBOX.

Offlinerss will not write the entry to the inbox directory if it exists in any subdirectory in ~/rss. This means if you read an article and you want to move it to another directory you can move it to ~/rss/archive for example and it won't be created again when you run offlinerss.

Using Go toolchain:

$ go install github.com/emad-elsaid/offlinerss@master

Copy config.example.json to ~/rss/config.json and fill it with your RSS feeds URLs

Run offlinerss when you want to download your RSS feeds.

• offlinerss will read ~/rss/config.json
• Will read each URL in parallel
• For each entry it will write its XML to a file in ~/rss/INBOX/ with a name in this format sha1(feed url)-sha1(entry content).rss
• If the file sha1(feed url)-sha1(entry content).rss exist under any subdirectory in ~/rss it will not be written again.
• The rest of the RSS feed is written to ~/rss/.meta to a file with a name in this format sha1(feed url).rss

• No dependencies. Just Go! itself
• Sync in parallel as the process is mostly IO bound.
• No sub packages.
• No unnedeeded abstractions

• Any application can read it
• You can search with any file search tool (grep, ag, ripgrep..etc)
• You can move the files as you wish to any directory
• You can version it with Git or sync it with rsync
• You can replace offlinerss with another implementation and the data doesn't need any transformation, as offlinerss doesn't do any transformation at all, just splits it to file.
• You can view the files with any simple application that reads the file and renders it to HTML or any other format

I wrote about this script for the first time couple days ago in my blog describing why I was set to create it

The Go implementation has a small difference in the rss item file name. Instead of getting the item id/guid/link and hashing it with SHA1 I used the item itself, this was simpler to do as it doesn't require parsing the feed or the item at all.

So if you used the ruby implementation and you ran the Go implementation on the same rss directory it will write existing rss item again to INBOX as the filename is different.

So to upgrade, make sure you have read and moved all your RSS items from INBOX then run the go implementation which will write again all rss items from your feeds to INBOX. After that move all items in INBOX to another directory say "Duplicates" or "Trash".

Another difference is the config file. now is in JSON format instead of YAML