Does RSS support only work with Feedbin and Feedly? about hpi HOT 6 CLOSED

Thanks for the pointers - I'll add a PR for a new module when I have time.

from hpi.

Thats correct. I have a personal module that uses newsboat, though that just tracks when I added/removed feeds not the contents of the RSS feed

But it sounds like youre talking about using rss as a source for data, not particularly tracking your feeds -- correct?

If you want to track hackernews and reddit, could you use their respective modules instead?

https://github.com/karlicoss/HPI/blob/master/my/hackernews/dogsheep.py using https://github.com/dogsheep/hacker-news-to-sqlite

https://github.com/karlicoss/HPI/tree/master/my/reddit using https://github.com/karlicoss/rexport, see setup

That'll also likely have more info than an RSS feed

from hpi.

Not saying some other functionality to use RSS as a data source couldn't be added -- but it would likely widely vary depending on the site, I'd assume

from hpi.

And just to expand a bit, HPI probably wouldnt connect to a live RSS feed -- goes against one of its core principles

Likely, we'd create some basic separate tool/repo, like export_rss_data, which does something like:

In [25]: import requests, xmltodict; data = xmltodict.parse(requests.get("https://xkcd.com/rss.xml").text)

In [26]: data['rss']['channel']['item'][0]
Out[26]: 
OrderedDict([('title', 'Consensus Time'),
             ('link', 'https://xkcd.com/2594/'),
             ('description',
              '<img src="https://imgs.xkcd.com/comics/consensus_time.png" title="Now, you may argue that the varying hour lengths and feedback effects would cause chaos. To which I say, yeah, and I\'m also curious to see how the weekday cycle interacts with it! So, you in?" alt="Now, you may argue that the varying hour lengths and feedback effects would cause chaos. To which I say, yeah, and I\'m also curious to see how the weekday cycle interacts with it! So, you in?" />'),
             ('pubDate', 'Wed, 16 Mar 2022 04:00:00 -0000'),
             ('guid', 'https://xkcd.com/2594/')])

saving rss feeds you give it to a file, say in ~/data/rss_data

You'd run some cli command like export_rss_data url1 url2 url3 --to ~/data/rss_data, which you could put in cron or run manually, to save stuff every couple hours/daily

and then a file would be created like my.feed.rss_data.py, which can parse those files, removing any duplicates because of overlapping backups

If thats something youre interested in, let me know -- otherwise can just use the hackernews and reddit modules

from hpi.

No, I'd like to use the hackernews and reddit modules. But what do you actually put in your promnesia/config.py file to use the my.hackernews.dogsheep HPI module?

from hpi.

I think dogsheep is a relatively new module, so it doesnt have a corresponding source in promnesia yet.

Given its structure, it shouldn't be hard to add one -- can just use the url and permalink attributes on the object

I think a src/promnesia/sources/dogsheep.py would look very similar to something like this -- where it calls the dogsheep.items function, returning a visit for the permalink, and then using the extract_urls helper to extract any URLs from the text.

I don't use hackernews, so is a bit hard for me to write one myself, but feel free to open an issue in promnesia or create a PR with something basic like this if you feel up to it.

from hpi.