feat(curator): vendor the application backend as the scenario's tracked source

The curator Python backend (package, tests, systemd units, config templates, scripts) now lives under scenarios/curator/backend and is the single source of truth; the live checkout at the workspace path is a runtime copy. Exported from the app repo's tracked tree via git archive (no history, .pi/venv/caches excluded). 149 unit tests pass from the new location. profile.toml backend is now repo-relative (scenarios/curator/backend); verify-generated.sh resolves a relative backend against REPO_ROOT. verify-no-secrets ASSIGN heuristic now requires value entropy so vendored kwargs like token=extraction_token no longer false-positive. README documents the backend/ layout and the operator-owned app rollout step.
2026-08-30 18:49:10 -07:00
parent 14c97d88cf
commit 88b06d782f
57 changed files with 14236 additions and 9 deletions
@@ -0,0 +1,108 @@
+#!/usr/bin/env python3
+from __future__ import annotations
+
+import argparse
+import json
+import os
+import shutil
+import sqlite3
+from datetime import UTC, datetime
+from pathlib import Path
+
+from curator.epub import inspect_epub
+from curator.library import safe_component, safe_filename
+
+
+def main() -> None:
+    parser = argparse.ArgumentParser(description="Merge one Curator book work into another")
+    parser.add_argument("database", type=Path)
+    parser.add_argument("target_work_id", type=int)
+    parser.add_argument("source_work_id", type=int)
+    parser.add_argument("--apply", action="store_true")
+    args = parser.parse_args()
+
+    connection = sqlite3.connect(args.database)
+    connection.row_factory = sqlite3.Row
+    connection.execute("PRAGMA foreign_keys=ON")
+    target = connection.execute("SELECT * FROM works WHERE id=?", (args.target_work_id,)).fetchone()
+    source = connection.execute("SELECT * FROM works WHERE id=?", (args.source_work_id,)).fetchone()
+    if not target or not source or target["media_type"] != "book" or source["media_type"] != "book":
+        raise SystemExit("target and source must both be existing book works")
+    rows = connection.execute(
+        """SELECT a.*, e.id AS source_edition_id, e.language, e.variant, e.isbn
+           FROM assets a JOIN editions e ON e.id=a.edition_id WHERE e.work_id=?""",
+        (args.source_work_id,),
+    ).fetchall()
+    if not rows:
+        raise SystemExit("source work has no assets")
+
+    changes: list[dict[str, object]] = []
+    for row in rows:
+        old_path = Path(row["path"])
+        info = inspect_epub(old_path) if row["format"] == "epub" else None
+        language = info.language if info else row["language"]
+        edition_dir = safe_component(f"{language}-{row['variant']}", "und-original")
+        destination_dir = old_path.parents[3] / safe_component(target["author"], "Unknown Author") / safe_component(target["title"], "Untitled") / edition_dir
+        filename = safe_filename(row["filename"], f"book.{row['format']}")
+        destination = destination_dir / filename
+        metadata = json.loads(row["metadata_json"] or "{}")
+        if info:
+            metadata.update({
+                "display_title": info.display_title,
+                "title_aliases": list(info.title_aliases),
+                "declared_language": info.declared_language,
+                "identifiers": list(info.identifiers),
+                "source_identifiers": list(info.source_identifiers),
+            })
+        changes.append({
+            "asset_id": row["id"], "edition_id": row["source_edition_id"], "old": old_path,
+            "new": destination, "filename": filename, "language": language,
+            "isbn": info.isbn if info else row["isbn"], "metadata": metadata,
+        })
+
+    print(json.dumps([{**change, "old": str(change["old"]), "new": str(change["new"])} for change in changes], ensure_ascii=False, indent=2, default=str))
+    if not args.apply:
+        return
+
+    backup = args.database.with_name(f"{args.database.stem}-before-merge-{datetime.now(UTC).strftime('%Y%m%dT%H%M%SZ')}.sqlite3")
+    with sqlite3.connect(backup) as backup_connection:
+        connection.backup(backup_connection)
+    moved: list[tuple[Path, Path]] = []
+    try:
+        connection.execute("BEGIN IMMEDIATE")
+        for change in changes:
+            old_path = change["old"]
+            destination = change["new"]
+            assert isinstance(old_path, Path) and isinstance(destination, Path)
+            destination.parent.mkdir(parents=True, exist_ok=True)
+            if destination.exists() and destination != old_path:
+                raise FileExistsError(destination)
+            os.replace(old_path, destination)
+            moved.append((destination, old_path))
+            connection.execute(
+                "UPDATE editions SET work_id=?, language=?, isbn=? WHERE id=?",
+                (args.target_work_id, change["language"], change["isbn"], change["edition_id"]),
+            )
+            connection.execute(
+                "UPDATE assets SET filename=?, path=?, metadata_json=? WHERE id=?",
+                (change["filename"], str(destination), json.dumps(change["metadata"], ensure_ascii=False, sort_keys=True), change["asset_id"]),
+            )
+        connection.execute("DELETE FROM works WHERE id=?", (args.source_work_id,))
+        connection.execute(
+            "UPDATE works SET updated_at=? WHERE id=?",
+            (datetime.now(UTC).replace(microsecond=0).isoformat(), args.target_work_id),
+        )
+        connection.commit()
+    except Exception:
+        connection.rollback()
+        for current, original in reversed(moved):
+            original.parent.mkdir(parents=True, exist_ok=True)
+            os.replace(current, original)
+        raise
+    finally:
+        connection.close()
+    print(f"backup={backup}")
+
+
+if __name__ == "__main__":
+    main()