Tomas Kral - Developer Tools and Builds

Tool

A player for an archive that never had one

Bassdrive Archive is a directory listing of recorded drum & bass radio shows. It is complete and well kept, but browsing it means walking folders and downloading 160 MB files one at a time — there is no way to see what landed recently across every show, and no way to listen without leaving the page.

This indexes the archive on a schedule and gives it two views: a timeline of the most recent episodes across all shows, and every episode grouped under the show it belongs to. Audio streams directly from the archive’s own servers — nothing is copied, re-hosted, or proxied here, and the index holds only show names, dates, sizes, and links.

Most recently broadcast, across every show

Loading the index…

How the index is built

Cloudflare Pages has no cron triggers, so a separate scheduled Worker does the crawling and writes to D1, which this page reads. A full pass over the archive is only about 135 requests, but a Worker invocation is capped at 50 subrequests — so each run walks a single weekday section and advances a rotation cursor, completing a full sweep roughly once a day. Requests are spaced at least 500 ms apart and carry a User-Agent that identifies this tool and links back here.

A few things about the archive make it more awkward to index than it first appears: listing link text is truncated, so the full filename only exists in the title attribute; broadcast dates live in the filename while the listing’s own date column is the upload time; around a hundred entries are 60-byte placeholders rather than audio; and roughly half the show folders are empty. One show keeps its pre-2023 history in year subdirectories, so the crawler recurses rather than assuming a fixed depth.

HTTP API

# Most recent episodes across every show (keyset paginated)
curl 'https://tomaskral.eu/api/v1/bassdrive/episodes?limit=20'

# {"episodes":[{"id":1,"showSlug":"...","title":"...","url":"https://bassdrivearchive.com/...",
#   "airdate":"2026-08-19","sizeBytes":172836177}],"nextCursor":"2026-08-19_1"}

# Every show that has episodes
curl 'https://tomaskral.eu/api/v1/bassdrive/shows'

# One show, newest episodes first
curl 'https://tomaskral.eu/api/v1/bassdrive/shows/<slug>?limit=20'

# Subscribe to every episode of one show as a podcast
curl 'https://tomaskral.eu/api/v1/bassdrive/shows/<slug>/feed.xml'

Pass the nextCursor from a response back as ?cursor= to page further. Paging is keyset rather than offset-based, so a crawl inserting episodes while you page cannot make you skip or repeat rows.