Add the original info URL to the release object stored for Newznab/Torznab search results

[juandjara]

Hi team! I wanted to contribute this feature I have created while working on my alternative React-based client for Medusa.

Motivation

When picking between candidates in a manual search, users often want to open a release on the indexer's site to check the file list or see bundled subtitles. Today Medusa exposes only the download URL on cached results, so a frontend can let you snatch a release but not preview it.

This information is already available to use: Newznab and Torznab XML responses include a <comments> element pointing to the release's info page.

I considered recovering the URL on the frontend by rewriting the download URL or querying the configured Prowlarr server at /api/v1/search but the solution would be sub-optimal and end up as a half-assed fix rather than a good feature. Plumbing <comments> through the API to the DB is the smallest change with the broadest reach: every Newznab/Torznab provider gets a real per-release info link with no extra requests.

Changes

A new info_url follows the same path url does today:

• SearchResult (medusa/classes.py) gains an info_url attribute, populated from the parser on live results and from the cache row on reload.
• GenericProvider._get_info_url reads item['info_url'] by default — any provider whose parse() adds that key gets piped through automatically.
• Newznab and Torznab parsers extract <comments> into the item dict.
• Provider cache table (medusa/tv/cache.py) gains an info_url TEXT column. New installs get it from CREATE TABLE; existing tables pick it up via the same auto-add-column loop
  already handling release_group, proper_tags, etc. — no DB migration version bump. INSERT/UPDATE persist it alongside url.
• /api/v2/providers/<id>/results exposes the field as infoUrl, returning null for rows written before the column existed so consumers fall back gracefully.

Test plan

• Confirmed the column is added on new installs and via the auto-migration loop on existing ones.
• Manual searches against a Newznab and a Prowlarr-fronted indexer both populate info_url; the Prowlarr-fronted case points to the upstream tracker domain, not the Prowlarr host.
• Pre-migration cache rows return infoUrl: null without errors.

No existing issue tracking this but happy to file one if the maintainers would prefer a reference

• PR is based on the DEVELOP branch
• Don't send big changes all at once. Split up big PRs into multiple smaller PRs that are easier to manage and review
• Read the contribution guide