Why?

For over 15 years and until Not long ago the _rss=1 query string instructed any ebay /sch to respond with an RSS feed. For a while an altenative endpoint /dsc seemed to still be supporting the _rss=1 parameter, yet this support too has been removed by now, which leaves us with no official RSS feed for ebay search queries.

What?

This is really just a Newsboat filter that translates ebay search results into an RSS feed

But why is this Newboat specific, why not something more generic?

You're right to ask that, my initial thought was to write a lightweight translation service (an http proxy) that takes an ebay URL and seemlessly returns a feed. But since I have multiple ebay URLs on Newsboat, such service would have to handle concurrency and throttling related issues which Newsboat already does, and I rather keep it that way, for now at least. And with that in mind, by using a filter (rather than exec script), I let Snowboat be in charge of request handling.

Usage

$ pip install parsel
$ git clone https://github.com/almog/newsboat-ebay2rss-filter/blob/main/ebay2rss_filter.py

Simple search query:

"filter:~/<eaby2rss-path>/ebay2rss_filter.py:https://www.ebay.com/sch/i.html?&_nkw=test"

Migrating existing newsboat urls file

./migrate_urls <input_file> <filter_path> transforms any ebay.com urls file to work with the ebay2rss filter (output is written to stdio)

./migrate_urls ~/.newsboat/urls ~/proj/newsboat-ebay2rss-filter/ebay2rss_filter.py > ./transformed_urls

Reads a urls file, a filter path and writes a new file ./transformed_url.

Logging

Note that Newsboat filters error log is not written to Newsboat error log, the filter however logs all errors to ~/.newsboat/ebay2rss.log directly.