//! airplay: collects what Danish radio stations play, for modelling when artists get played again. //! //! airplay collect [--db data/airplay.db] [--days 7] [--all-bauer] [--only dr|bauer|hitlisten|deezer|musicbrainz|spotify] [--backfill] //! airplay stats [--db data/airplay.db] //! airplay check [--db data/airplay.db] gaps, duplicates, DR cross-check, chart completeness //! //! --backfill fetches all history at once (every DR artist page, the whole Hitlisten archive); //! without it a run takes a bounded slice, which is what the daily timer does. mod bauer; mod check; mod db; mod deezer; mod dr; mod hitlisten; mod http; mod musicbrainz; mod spotify; use anyhow::Result; use chrono::Utc; use rusqlite::params; fn arg(args: &[String], name: &str) -> Option { args.iter().position(|a| a == name).and_then(|i| args.get(i + 1)).cloned() } fn main() -> Result<()> { let args: Vec = std::env::args().collect(); let db_path = arg(&args, "--db").unwrap_or_else(|| "data/airplay.db".into()); if let Some(dir) = std::path::Path::new(&db_path).parent() { std::fs::create_dir_all(dir)?; } let db = db::Db::open(&db_path)?; match args.get(1).map(String::as_str) { Some("collect") => collect(&db, &args, &db_path), Some("stats") => stats(&db), Some("check") => check::run(&db), _ => { eprintln!("usage: airplay collect [--db FILE] [--days 7] [--all-bauer] [--only dr|bauer|hitlisten|deezer|musicbrainz|spotify] [--backfill]\n airplay stats [--db FILE]\n airplay check [--db FILE]"); std::process::exit(2); } } } fn collect(db: &db::Db, args: &[String], db_path: &str) -> Result<()> { let days: i64 = arg(args, "--days").and_then(|d| d.parse().ok()).unwrap_or(7); let only = arg(args, "--only"); let wants = |source: &str| only.as_deref().map_or(true, |o| o == source); let backfill = args.iter().any(|a| a == "--backfill"); let http = http::Http::new(400)?; let started = Utc::now().to_rfc3339(); db.conn.execute("INSERT INTO runs (started_at) VALUES (?1)", params![started])?; let run = db.conn.last_insert_rowid(); let mut log: Vec = Vec::new(); let mut new = 0; let step = |name: &str, r: Result, log: &mut Vec| match r { Ok(msg) => { println!("{name}: {msg}"); } Err(e) => { eprintln!("{name}: {e}"); log.push(format!("{name}: {e}")); } }; if wants("dr") { let r = dr::collect_plays(db, &http, days, &mut log); if let Ok(n) = &r { new += n; } step("dr plays", r.map(|n| format!("{n} new")), &mut log); let dir = std::path::Path::new(db_path).parent().unwrap_or(std::path::Path::new(".")).join("rotations"); let r = dr::collect_rotation_pdf(db, &http, &dir); step("dr rotation list", r.map(|f| f.map(|f| format!("saved {f}")).unwrap_or_else(|| "unchanged".into())), &mut log); let r = dr::collect_artists(db, &http, 6, if backfill { usize::MAX } else { 400 }, &mut log); step("dr artist pages", r.map(|n| format!("{n} fetched")), &mut log); } if wants("bauer") { match bauer::stations(db, &http, args.iter().any(|a| a == "--all-bauer")) { Ok(stations) => { for (code, name) in stations { let r = bauer::collect(db, &http, &code, days, &mut log); if let Ok(n) = &r { new += n; } step(&format!("bauer {code} ({name})"), r.map(|n| format!("{n} new")), &mut log); } } Err(e) => step("bauer stations", Err(e), &mut log), } } if wants("deezer") { // Deezer allows 50 requests per 5 seconds; its own client keeps well under that. let deezer_http = http::Http::new(150)?; let r = deezer::collect(db, &deezer_http, arg(args, "--limit").and_then(|l| l.parse().ok()).unwrap_or(if backfill { usize::MAX } else { 3000 }), &mut log); step("deezer song metadata", r.map(|n| format!("{n} songs looked up")), &mut log); } if wants("musicbrainz") { // MusicBrainz allows one request per second. let mb_http = http::Http::new(1100)?; let r = musicbrainz::collect(db, &mb_http, arg(args, "--limit").and_then(|l| l.parse().ok()).unwrap_or(if backfill { usize::MAX } else { 2000 }), &mut log); step("musicbrainz release dates", r.map(|n| format!("{n} ISRCs looked up")), &mut log); } if wants("spotify") { // kworb.net is a one-person site: one request a second. let kworb_http = http::Http::new(1000)?; let r = spotify::collect(db, &kworb_http, arg(args, "--limit").and_then(|l| l.parse().ok()).unwrap_or(if backfill { usize::MAX } else { 400 }), &mut log); step("spotify chart history (kworb)", r.map(|(l, p)| format!("{l} songs listed, {p} histories read")), &mut log); } if wants("hitlisten") { let r = hitlisten::collect(db, &http, if backfill { usize::MAX } else { 60 }, &mut log); step("hitlisten airplay chart", r.map(|n| format!("{n} weeks fetched")), &mut log); } db.conn.execute( "UPDATE runs SET finished_at = ?1, new_plays = ?2, errors = ?3, log = ?4 WHERE id = ?5", params![Utc::now().to_rfc3339(), new as i64, log.len() as i64, log.join("\n"), run], )?; println!("{new} new plays, {} problem(s)", log.len()); for l in log.iter().take(20) { eprintln!(" {l}"); } Ok(()) } fn stats(db: &db::Db) -> Result<()> { let mut st = db.conn.prepare( "SELECT p.station, coalesce(s.name, p.station), count(*), min(played_at), max(played_at), count(DISTINCT artist) FROM plays p LEFT JOIN stations s ON s.code = p.station GROUP BY p.station ORDER BY count(*) DESC")?; println!("{:<12} {:<22} {:>7} {:>8} {:<20} {}", "station", "name", "plays", "artists", "from", "to"); let rows = st.query_map([], |r| Ok((r.get::<_, String>(0)?, r.get::<_, String>(1)?, r.get::<_, i64>(2)?, r.get::<_, String>(3)?, r.get::<_, String>(4)?, r.get::<_, i64>(5)?)))?; for r in rows { let (code, name, n, from, to, artists) = r?; println!("{code:<12} {name:<22} {n:>7} {artists:>8} {from:<20} {to}"); } let weeks: i64 = db.conn.query_row("SELECT count(*) FROM dr_artist_weeks", [], |r| r.get(0))?; let artists: i64 = db.conn.query_row("SELECT count(*) FROM dr_artists", [], |r| r.get(0))?; let pdfs: i64 = db.conn.query_row("SELECT count(*) FROM dr_rotation_pdfs", [], |r| r.get(0))?; println!("DR artist history: {artists} artists, {weeks} artist-week-channel rows; rotation PDFs: {pdfs}"); hitlisten::create_tables(db)?; let (n, first, last): (i64, Option, Option) = db.conn.query_row( "SELECT count(DISTINCT year * 100 + week), min(printf('%d-W%02d', year, week)), max(printf('%d-W%02d', year, week)) FROM hitlisten_airplay", [], |r| Ok((r.get(0)?, r.get(1)?, r.get(2)?)))?; println!("Hitlisten airplay chart: {n} weeks ({} to {})", first.unwrap_or_default(), last.unwrap_or_default()); deezer::create_tables(db)?; let (searched, matched, tracks, albums, labels): (i64, i64, i64, i64, i64) = db.conn.query_row( "SELECT (SELECT count(*) FROM deezer_searches), (SELECT count(*) FROM deezer_searches WHERE best_track_id IS NOT NULL), (SELECT count(*) FROM deezer_tracks), (SELECT count(*) FROM deezer_albums), (SELECT count(*) FROM deezer_albums WHERE coalesce(label, '') != '')", [], |r| Ok((r.get(0)?, r.get(1)?, r.get(2)?, r.get(3)?, r.get(4)?)))?; spotify::create_tables(db)?; let (sp_tracks, sp_read, sp_rows, sp_from): (i64, i64, i64, Option) = db.conn.query_row( "SELECT (SELECT count(*) FROM spotify_dk_tracks), (SELECT count(*) FROM spotify_dk_tracks WHERE history_fetched_at IS NOT NULL), (SELECT count(*) FROM spotify_chart_history WHERE country = 'DK'), (SELECT min(date) FROM spotify_chart_history WHERE country = 'DK' AND chart = 'daily')", [], |r| Ok((r.get(0)?, r.get(1)?, r.get(2)?, r.get(3)?)))?; println!("Spotify Denmark: {sp_tracks} songs that charted, {sp_read} histories read, {sp_rows} DK chart rows (daily from {})", sp_from.unwrap_or_default()); musicbrainz::create_tables(db)?; let (mb, mb_known): (i64, i64) = db.conn.query_row( "SELECT count(*), count(*) FILTER (WHERE recordings > 0) FROM mb_isrc", [], |r| Ok((r.get(0)?, r.get(1)?)))?; println!("MusicBrainz: {mb} ISRCs looked up, {mb_known} known (original release dates)"); println!("Deezer: {searched} songs looked up, {matched} matched; {tracks} tracks, {albums} releases ({labels} with a label)"); Ok(()) }