Pete/internal/storage/db.go

package storage

import (
	"database/sql"
	"fmt"
	"log/slog"
	"os"
	"path/filepath"
	"strings"
	"sync"

	_ "modernc.org/sqlite"
)

var (
	mu       sync.RWMutex
	globalDB *sql.DB
)

// Init opens (or creates) the SQLite database and runs migrations.
func Init(dbPath string) error {
	mu.Lock()
	defer mu.Unlock()

	if globalDB != nil {
		return nil
	}

	dir := filepath.Dir(dbPath)
	if err := os.MkdirAll(dir, 0o755); err != nil {
		return fmt.Errorf("create data dir: %w", err)
	}

	d, err := sql.Open("sqlite", dbPath+"?_pragma=journal_mode(WAL)&_pragma=busy_timeout(5000)&_pragma=foreign_keys(ON)")
	if err != nil {
		return fmt.Errorf("open database: %w", err)
	}

	d.SetMaxOpenConns(1)

	if err := runMigrations(d); err != nil {
		return fmt.Errorf("run migrations: %w", err)
	}

	globalDB = d
	slog.Info("database initialized", "path", dbPath)
	return nil
}

// Get returns the global database handle. Panics if Init was not called.
func Get() *sql.DB {
	mu.RLock()
	db := globalDB
	mu.RUnlock()
	if db == nil {
		panic("storage.Get() called before storage.Init()")
	}
	return db
}

// Close closes the global database handle.
func Close() error {
	mu.Lock()
	defer mu.Unlock()
	if globalDB != nil {
		err := globalDB.Close()
		globalDB = nil
		return err
	}
	return nil
}

func runMigrations(d *sql.DB) error {
	if _, err := d.Exec(schema); err != nil {
		return fmt.Errorf("create schema: %w", err)
	}

	// Idempotent column adds for DBs created before the dedup columns existed.
	// SQLite errors with "duplicate column name" when the column is already there;
	// we swallow that specifically.
	addColumnIfMissing(d, "stories", "url_canonical", "TEXT")
	addColumnIfMissing(d, "stories", "headline_norm", "TEXT")
	addColumnIfMissing(d, "stories", "paywalled", "INTEGER NOT NULL DEFAULT 0")
	// content holds the full article text (feed content:encoded when present,
	// else the body scraped during paywall detection) for reader mode. Stories
	// ingested before this column existed simply have NULL and fall back to lede.
	addColumnIfMissing(d, "stories", "content", "TEXT")
	// content_chars caches the character count of content so the "N min read"
	// chip never has to LENGTH() the full body on the hot listing path. Filled at
	// insert time; the backfill below populates rows that predate the column.
	addColumnIfMissing(d, "stories", "content_chars", "INTEGER NOT NULL DEFAULT 0")
	backfillContentChars(d)
	addColumnIfMissing(d, "stories", "published_at", "INTEGER")
	addColumnIfMissing(d, "post_log", "url_canonical", "TEXT")
	addColumnIfMissing(d, "post_log", "forced", "INTEGER NOT NULL DEFAULT 0")
	addColumnIfMissing(d, "round_robin_state", "last_channel", "TEXT")

	// FTS5 virtual tables don't support IF NOT EXISTS reliably.
	// Check sqlite_master before creating.
	var ftsExists int
	if err := d.QueryRow(`SELECT COUNT(*) FROM sqlite_master WHERE type='table' AND name='stories_fts'`).Scan(&ftsExists); err != nil {
		return fmt.Errorf("probe FTS5 table: %w", err)
	}
	if ftsExists == 0 {
		if _, err := d.Exec(ftsSchema); err != nil {
			return fmt.Errorf("create FTS5 table: %w", err)
		}
		if _, err := d.Exec(ftsTriggers); err != nil {
			return fmt.Errorf("create FTS5 triggers: %w", err)
		}
		slog.Info("created FTS5 search index")
	}

	return nil
}

// RunMaintenance prunes stale data. Called periodically.
func RunMaintenance() {
	// Prune old stories (30 days) and their post logs / reactions
	storyCutoff := nowUnix() - int64(30*86400)
	exec("prune old stories",
		`DELETE FROM stories WHERE seen_at < ? AND classified = 1`, storyCutoff)
	exec("prune old post_log",
		`DELETE FROM post_log WHERE posted_at < ?`, storyCutoff)
	exec("prune old reactions",
		`DELETE FROM reactions WHERE reacted_at < ?`, storyCutoff)

	// Drop per-user read/bookmark rows whose story has been pruned above, so the
	// table can't accumulate dangling references as stories age out.
	exec("prune orphan user_story_state",
		`DELETE FROM user_story_state WHERE story_id NOT IN (SELECT id FROM stories)`)

	// Same for per-story view counts once their story has aged out.
	exec("prune orphan story_views",
		`DELETE FROM story_views WHERE story_id NOT IN (SELECT id FROM stories)`)

	// Daily unique tokens are only useful for the recent window; their salts are
	// long gone. page_views is kept forever (tiny aggregate, all-time totals).
	exec("prune old daily_visitors",
		`DELETE FROM daily_visitors WHERE day < ?`, unixDay()-30)

	exec("wal checkpoint", "PRAGMA wal_checkpoint(TRUNCATE)")
	exec("optimize", "PRAGMA optimize")
}

// exec is a fire-and-forget helper that logs errors. Several callers run it from
// background goroutines (metrics, view counts), which can outlive a Close() — so
// unlike Get() it must not panic on a nil handle: it simply skips the write.
func exec(label, query string, args ...any) {
	mu.RLock()
	db := globalDB
	mu.RUnlock()
	if db == nil {
		slog.Warn("db exec skipped: no database", "op", label)
		return
	}
	if _, err := db.Exec(query, args...); err != nil {
		slog.Error("db exec failed", "op", label, "err", err)
	}
}

// backfillContentChars populates content_chars for rows carrying a body but a
// zero count — i.e. stories ingested before the column existed. LENGTH() counts
// characters (code points) for TEXT, matching the utf8.RuneCountInString done at
// insert. After the first run this matches no rows (bodied stories are set,
// bodyless ones stay 0 and are filtered by content IS NOT NULL), so it's a cheap
// startup no-op thereafter.
func backfillContentChars(d *sql.DB) {
	if _, err := d.Exec(
		`UPDATE stories SET content_chars = LENGTH(content)
		 WHERE content_chars = 0 AND content IS NOT NULL AND content <> ''`); err != nil {
		slog.Error("backfill content_chars failed", "err", err)
	}
}

func addColumnIfMissing(d *sql.DB, table, column, columnType string) {
	q := fmt.Sprintf("ALTER TABLE %s ADD COLUMN %s %s", table, column, columnType)
	if _, err := d.Exec(q); err != nil {
		// SQLite returns "duplicate column name" when the column already exists.
		if !strings.Contains(err.Error(), "duplicate column name") {
			slog.Error("alter table failed", "table", table, "column", column, "err", err)
		}
	}
}