ripster/docs/architecture/database.md

Datenbank

Ripster verwendet SQLite (backend/data/ripster.db).

---

Tabellen

settings_schema
settings_values
jobs
pipeline_state
scripts
script_chains
script_chain_steps
user_presets
cron_jobs
cron_run_logs

---

jobs

Speichert Pipeline-Lifecycle und Artefakte pro Job.

Zentrale Felder:

• Metadaten: title, year, imdb_id, poster_url, omdb_json, selected_from_omdb
• Laufzeit: start_time, end_time, status, last_state
• Pfade: raw_path, output_path, encode_input_path
• Tool-Ausgaben: makemkv_info_json, handbrake_info_json, mediainfo_info_json, encode_plan_json
• Kontrolle: encode_review_confirmed, rip_successful, error_message
• Audit: created_at, updated_at

---

pipeline_state

Singleton-Tabelle (id = 1) für aktiven Snapshot:

• state
• active_job_id
• progress
• eta
• status_text
• context_json
• updated_at

---

settings_schema + settings_values

• settings_schema: Definition (Typ, Default, Validation, Reihenfolge)
• settings_values: aktueller Wert pro Key

---

scripts, script_chains, script_chain_steps

• scripts: Shell-Skripte (name, script_body, order_index)
• script_chains: Ketten (name, order_index)
• script_chain_steps: Schritte je Kette
  • step_type: script oder wait
  • script_id oder wait_seconds

---

user_presets

Benannte HandBrake-Preset-Sets:

• name
• media_type (bluray|dvd|other|all)
• handbrake_preset
• extra_args
• description

---

cron_jobs + cron_run_logs

• cron_jobs: Zeitplan + Status
• cron_run_logs: einzelne Läufe
  • status: running|success|error
  • output
  • error_message

---

Migration/Recovery

Beim Start werden Schema und Settings-Metadaten automatisch abgeglichen.

Bei korruptem SQLite-File:

1. Datei wird nach backend/data/corrupt-backups/ verschoben
2. neue DB wird initialisiert
3. Schema wird neu aufgebaut

---

Direkte Inspektion

sqlite3 backend/data/ripster.db

.mode table
SELECT id, status, title, created_at FROM jobs ORDER BY created_at DESC;
SELECT key, value FROM settings_values ORDER BY key;