rss_context.yaml

# Example context for the RSS/Atom data source (`type: rss`).
# Reference: docs/rss.md
#
# Try it — registration performs no network I/O and `feeds` is a pure state
# read, so this works offline and fetches nothing:
#
#   cargo run -p skardi-cli --features rss -- \
#     query --ctx docs/sample_data/rss_context.yaml \
#     --sql "SELECT name, last_status FROM news.main.feeds"
#
# Every row comes back `never`: no `items` scan has run, so no feed has been
# attempted. The hosts below are under the reserved `.invalid` TLD and are
# deliberately unresolvable — swap in real feed URLs before scanning `items`.

kind: context

metadata:
  name: news-context
  version: 1.0.0

spec:
  data_sources:
    - name: news
      type: rss
      hierarchy_level: catalog
      access_mode: read_only
      description: "Subscribed RSS/Atom feeds: health in main.feeds, the live window in main.items"
      rss:
        feeds:
          - url: https://blog.example.invalid/feed.xml
            name: example-blog          # optional; defaults to the URL
          - url: https://weekly.example.invalid/rss.xml
            name: example-weekly
          - url: https://notes.example.invalid/atom.xml
        # or: opml: subscriptions.opml  # mutually exclusive with feeds:
        ttl_seconds: 900                # 0 = always live
        max_concurrent: 6               # in-flight fetches for THIS source; not per-host, not per-process
        request_timeout_seconds: 10     # per-request timeout
        scan_timeout_seconds: 60        # deadline for one full scan across all feeds
        max_response_bytes: 5242880     # cap on one decoded response body (5 MiB)
        # user_agent defaults to "skardi-rss/<crate version> (+https://github.com/SkardiLabs/skardi)";
        # override it only to add your own contact URL. Feed hosts ban anonymous clients.
        user_agent: "skardi-rss (+https://example.invalid/contact)"