Just now I finished writing a program that clocks in at under 60 lines of ruby code. What it does is parse through all of my gaim logs for the current day. Then it takes any URLs it finds in those logs and creates an RSS feed out of them after removing duplicates. This is very handy for me because the links go into live bookmarks in firefox. It is handy for other people who want to see what sites me and my contacts are talking about.
Right now the program is pretty scruffy. All it does is parse the logs and create an RSS file. That is why it is version 0.1. I mean, it’s so ghetto I have it updating the feed on the webserver on a half-hourly basis using cron. In the future I will add all sorts of better functionality. First I will find a way for it to fill in the item.title with something besides the URL itself. Then maybe I’ll get it to give ranks to links, so if I send something to all my friends it will recognize that and have some sort of metadata which points out how often the link was sent/recieved. Maybe I’ll even get it to recognize which ones were sent as opposed to received, or both. For safety’s sake I am going to make it filter out https and other non-http links. Maybe handle some special cases like ftp. Definitely want to make it update automatically through a gaim plugin or something similar instead of cron uploading a file every so often. Also it would be nice if I could take out some of the lines of code specific to myself like file paths and such. In the future I’ll probably just make those into constants at the top of the file and have the script itself be its own configuration file. This would be a lot easier if my gaim logs were on my webserver because then turning this script into cgi would be a joke. Any other ideas on how to make this better are of course appreciated.
Oh, I’m sure you all want to see what links me and my friends are sending each other. Well you can’t, since this project is dead. Just visit my del.icio.us instead.
UPDATE: I have finished version .3 of glues. It now does all sorts of other nifty stuff. Because of this update it now stores the current rss feed in a new location, I have update the above link accordingly. Also, it stores archives of previous day’s RSS feeds. Where date is the day of the RSS feed in the format YYYY-MM-DD. Oh, and for people looking for the source, I am going to open it up, but right now I’m too ashamed of many things in it being low quality to let anyone look at it. I want to do as much on it as I can to make it nice before other people start crowding in.
UPDATE2: Project discontinued.
