Simplify AI memory pipeline

2026-04-30 16:22:55 +08:00
parent 7617ea902c
commit 3234396254
35 changed files with 1002 additions and 579 deletions
--- a/api/alembic/versions/0016_memory_pipeline_status.py
+++ b/api/alembic/versions/0016_memory_pipeline_status.py
@@ -0,0 +1,109 @@
+"""Memory pipeline status columns.
+
+Revision ID: 0016_memory_pipeline_status
+Revises: 0015_memory_single_chain
+"""
+
+from typing import Sequence, Union
+
+import sqlalchemy as sa
+
+from alembic import op
+
+revision: str = "0016_memory_pipeline_status"
+down_revision: Union[str, None] = "0015_memory_single_chain"
+branch_labels: Union[str, Sequence[str], None] = None
+depends_on: Union[str, Sequence[str], None] = None
+
+
+def _has_column(table: str, column: str) -> bool:
+    bind = op.get_bind()
+    return any(c["name"] == column for c in sa.inspect(bind).get_columns(table))
+
+
+def upgrade() -> None:
+    if not _has_column("memory_sources", "embedding_status"):
+        op.add_column(
+            "memory_sources",
+            sa.Column("embedding_status", sa.String(), nullable=True),
+        )
+    if not _has_column("memory_sources", "embedding_error"):
+        op.add_column(
+            "memory_sources",
+            sa.Column("embedding_error", sa.Text(), nullable=True),
+        )
+    if not _has_column("memory_sources", "enrichment_status"):
+        op.add_column(
+            "memory_sources",
+            sa.Column("enrichment_status", sa.String(), nullable=True),
+        )
+    if not _has_column("memory_sources", "enrichment_error"):
+        op.add_column(
+            "memory_sources",
+            sa.Column("enrichment_error", sa.Text(), nullable=True),
+        )
+    if not _has_column("memory_chunks", "embedding_status"):
+        op.add_column(
+            "memory_chunks",
+            sa.Column("embedding_status", sa.String(), nullable=True),
+        )
+    if not _has_column("memory_chunks", "embedding_error"):
+        op.add_column(
+            "memory_chunks",
+            sa.Column("embedding_error", sa.Text(), nullable=True),
+        )
+
+    op.execute(
+        """
+        UPDATE memory_sources
+        SET
+          embedding_status = COALESCE(embedding_status, 'success'),
+          enrichment_status = COALESCE(enrichment_status, 'pending')
+        """
+    )
+    op.execute(
+        """
+        UPDATE memory_chunks
+        SET embedding_status = COALESCE(
+          embedding_status,
+          CASE WHEN embedding IS NULL THEN 'pending' ELSE 'success' END
+        )
+        """
+    )
+    op.create_index(
+        "ix_memory_sources_user_embedding_status",
+        "memory_sources",
+        ["user_id", "embedding_status"],
+        unique=False,
+    )
+    op.create_index(
+        "ix_memory_sources_user_enrichment_status",
+        "memory_sources",
+        ["user_id", "enrichment_status"],
+        unique=False,
+    )
+    op.create_index(
+        "ix_memory_chunks_user_embedding_status",
+        "memory_chunks",
+        ["user_id", "embedding_status"],
+        unique=False,
+    )
+
+
+def downgrade() -> None:
+    op.drop_index("ix_memory_chunks_user_embedding_status", table_name="memory_chunks")
+    op.drop_index(
+        "ix_memory_sources_user_enrichment_status", table_name="memory_sources"
+    )
+    op.drop_index("ix_memory_sources_user_embedding_status", table_name="memory_sources")
+    for column in ("embedding_error", "embedding_status"):
+        if _has_column("memory_chunks", column):
+            op.drop_column("memory_chunks", column)
+    for column in (
+        "enrichment_error",
+        "enrichment_status",
+        "embedding_error",
+        "embedding_status",
+    ):
+        if _has_column("memory_sources", column):
+            op.drop_column("memory_sources", column)