feat: migrate doc-service to use storage-service for file I/O (Phase 2)

- storage.py: replace aiofiles filesystem ops with httpx calls to storage-service PUT/GET/DELETE /objects/documents/{key} - Document model: rename file_path → storage_key (plain object key, no path prefix) - Migration 0008: ALTER COLUMN + data migration strips /data/documents/ prefix - documents.py: update upload, delete, download endpoints; _extract_pdf_text now takes bytes (pdfplumber.open(BytesIO)) instead of a filesystem path - file_watcher.py: store storage_key instead of file_path on ingestion - doc-service config: add STORAGE_SERVICE_URL env var Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-20 15:57:29 +02:00
parent 5349f21752
commit 2f3efb9bf9
6 changed files with 128 additions and 36 deletions
@@ -0,0 +1,56 @@
+"""rename file_path to storage_key and strip filesystem prefix from existing rows
+
+Revision ID: 0008
+Revises: 0007
+Create Date: 2026-04-20
+
+Renames the documents.file_path column to storage_key.
+Existing rows have paths like '/data/documents/{user_id}/{doc_id}.pdf' or
+'/data/documents/watch/{doc_id}.pdf'.  The migration strips the leading
+'/data/documents/' prefix so the value becomes a plain storage key
+(e.g. '{user_id}/{doc_id}.pdf') that the storage-service uses as the object key.
+"""
+from typing import Sequence, Union
+
+from alembic import op
+import sqlalchemy as sa
+
+revision: str = "0008"
+down_revision: Union[str, None] = "0007"
+branch_labels: Union[str, Sequence[str], None] = None
+depends_on: Union[str, Sequence[str], None] = None
+
+
+def upgrade() -> None:
+    with op.batch_alter_table("documents") as batch_op:
+        batch_op.alter_column(
+            "file_path",
+            new_column_name="storage_key",
+            existing_type=sa.String(),
+            existing_nullable=False,
+        )
+
+    # Strip the '/data/documents/' filesystem prefix from pre-migration rows.
+    op.execute(
+        sa.text(
+            "UPDATE documents SET storage_key = REPLACE(storage_key, '/data/documents/', '')"
+            " WHERE storage_key LIKE '/data/documents/%'"
+        )
+    )
+
+
+def downgrade() -> None:
+    # Restore the filesystem prefix so old code can still find the files.
+    op.execute(
+        sa.text(
+            "UPDATE documents SET storage_key = '/data/documents/' || storage_key"
+            " WHERE storage_key NOT LIKE '/data/documents/%'"
+        )
+    )
+    with op.batch_alter_table("documents") as batch_op:
+        batch_op.alter_column(
+            "storage_key",
+            new_column_name="file_path",
+            existing_type=sa.String(),
+            existing_nullable=False,
+        )