feat(documents): importer for organized S3/filesystem buckets

One-shot script that walks an existing organized bucket tree, builds
matching document_folders rows mirroring the path, then inserts
documents + files rows pointing at the existing storage keys verbatim
— no path rewrite. For migrating from a legacy MinIO bucket whose
folder structure is already the source of truth.

Idempotency:
  • Folders: sibling-name unique index swallows duplicate creates;
    we reuse the row on ConflictError.
  • Documents: skipped when (port_id, fileStoragePath) already exists.

Adds StorageBackend.listByPrefix (recursive readdir on filesystem;
listObjectsV2 stream-drain on s3) — the first one-shot caller, not
a hot path. Pure parseImportPath helper extracted to its own module
and unit-tested for trailing slashes, empty intermediate segments,
prefix mismatch, and special-character folder names (8 tests).

Audit log per imported doc carries source='organized-bucket-importer',
storageKey, and folderSegments so the documents inspector can filter
on imports later.

CLI:
  pnpm tsx scripts/import-organized-documents.ts \\
      --port-slug <slug> \\
      --bucket-prefix "legacy-imports/" \\
      (--dry-run | --apply) [--uploaded-by <userId>]

Folds in Prettier post-hook drift on documents.service.ts +
download handler — same lint-staged formatting the earlier commits
already absorbed.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

src/lib/services/document-import.ts

@@ -0,0 +1,48 @@
+/**
+ * Pure helpers for the organized-bucket document importer
+ * (`scripts/import-organized-documents.ts`).
+ *
+ * The script walks an existing storage prefix that already represents real
+ * organisation (e.g. `legacy-imports/Deals 2026/Q1/contract.pdf`) and
+ * materialises matching `document_folders` + `documents` rows in the CRM
+ * without rewriting the storage keys. Splitting these helpers out of the
+ * script body makes the path-parser unit-testable in isolation.
+ */
+
+export interface ParsedImportPath {
+  /** Folder names from outermost to innermost; empty when the file is at the prefix root. */
+  folderSegments: string[];
+  /** Filename only, never empty. */
+  filename: string;
+}
+
+/**
+ * Decompose a storage key into folder segments + filename relative to the
+ * importer prefix. Both `prefix` and `key` use POSIX separators (the
+ * filesystem backend's `listByPrefix` already normalises Windows paths).
+ *
+ * Edge cases:
+ *   - Trailing slashes on prefix are tolerated (`legacy/` ≡ `legacy`).
+ *   - Empty intermediate segments (`a//b`) collapse to `[a, b]`.
+ *   - Leading-prefix mismatch throws — the caller should never feed in keys
+ *     outside the prefix it asked the backend to list.
+ *   - A key that ends in `/` (directory placeholder) yields an empty
+ *     filename — the caller must filter those out before invoking.
+ */
+export function parseImportPath(prefix: string, key: string): ParsedImportPath {
+  const normalizedPrefix = prefix.replace(/\/+$/, '');
+  let relative = key;
+  if (normalizedPrefix.length > 0) {
+    if (!key.startsWith(`${normalizedPrefix}/`)) {
+      throw new Error(`Key "${key}" is not under prefix "${prefix}"`);
+    }
+    relative = key.slice(normalizedPrefix.length + 1);
+  }
+
+  const parts = relative.split('/').filter((segment) => segment.length > 0);
+  if (parts.length === 0) {
+    throw new Error(`Key "${key}" has no filename after stripping prefix`);
+  }
+  const filename = parts.pop()!;
+  return { folderSegments: parts, filename };
+}

src/lib/services/documents.service.ts

@@ -307,10 +307,7 @@ async function hydrateDocumentsWithDownloadUrl(
     const filename = row.fileId ? (filenameById.get(row.fileId) ?? null) : null;
     return {
       ...row,
-      downloadUrl: buildDocumentDownloadUrl(
-        { id: row.id, folderId: row.folderId, filename },
-        tree,
-      ),
+      downloadUrl: buildDocumentDownloadUrl({ id: row.id, folderId: row.folderId, filename }, tree),
     };
   });
 }