feat(api)(rag): add text extraction api (#137)

This PR adds a text extraction API to the RAG service.

Part of #108

nesis/api/alembic/versions/ea840c8d9e58_add_extract_management_fields_to_.py

@@ -0,0 +1,107 @@
+"""add extract management fields to document
+
+Revision ID: ea840c8d9e58
+Revises: 7cfa662dff86
+Create Date: 2024-07-01 12:29:36.310411
+
+"""
+
+from typing import Sequence, Union
+
+from alembic import op
+import sqlalchemy as sa
+from sqlalchemy.dialects import postgresql
+
+# revision identifiers, used by Alembic.
+revision: str = "ea840c8d9e58"
+down_revision: Union[str, None] = "7cfa662dff86"
+branch_labels: Union[str, Sequence[str], None] = None
+depends_on: Union[str, Sequence[str], None] = None
+
+document_status = postgresql.ENUM(
+    "SUCCESS", "PROCESSING", "ERROR", name="document_status"
+)
+
+
+def upgrade() -> None:
+    # ### commands auto generated by Alembic - please adjust! ###
+    document_status.create(op.get_bind())
+
+    op.add_column(
+        "document",
+        sa.Column(
+            "extract_metadata", postgresql.JSONB(astext_type=sa.Text()), nullable=True
+        ),
+    )
+    op.add_column(
+        "document", sa.Column("datasource_id", sa.Unicode(length=255), nullable=True)
+    )
+    op.add_column(
+        "document",
+        sa.Column(
+            "status",
+            sa.Enum("SUCCESS", "PROCESSING", "ERROR", name="document_status"),
+            nullable=False,
+        ),
+    )
+    op.add_column("document", sa.Column("last_modified", sa.DateTime(), nullable=False))
+    op.add_column(
+        "document", sa.Column("last_processed", sa.DateTime(), nullable=False)
+    )
+    op.add_column(
+        "document", sa.Column("last_processed_message", sa.Text(), nullable=True)
+    )
+    op.alter_column(
+        "document",
+        "base_uri",
+        existing_type=sa.VARCHAR(length=255),
+        type_=sa.Unicode(length=4096),
+        existing_nullable=False,
+    )
+    op.alter_column(
+        "document",
+        "filename",
+        existing_type=sa.VARCHAR(length=255),
+        type_=sa.Unicode(length=4096),
+        existing_nullable=False,
+    )
+    op.alter_column(
+        "document",
+        "rag_metadata",
+        existing_type=postgresql.JSONB(astext_type=sa.Text()),
+        nullable=True,
+    )
+    # ### end Alembic commands ###
+
+
+def downgrade() -> None:
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.alter_column(
+        "document",
+        "rag_metadata",
+        existing_type=postgresql.JSONB(astext_type=sa.Text()),
+        nullable=False,
+    )
+    op.alter_column(
+        "document",
+        "filename",
+        existing_type=sa.Unicode(length=4096),
+        type_=sa.VARCHAR(length=255),
+        existing_nullable=False,
+    )
+    op.alter_column(
+        "document",
+        "base_uri",
+        existing_type=sa.Unicode(length=4096),
+        type_=sa.VARCHAR(length=255),
+        existing_nullable=False,
+    )
+    op.drop_column("document", "last_processed_message")
+    op.drop_column("document", "last_processed")
+    op.drop_column("document", "last_modified")
+    op.drop_column("document", "status")
+    op.drop_column("document", "datasource_id")
+    op.drop_column("document", "extract_metadata")
+
+    document_status.drop(op.get_bind())
+    # ### end Alembic commands ###

nesis/api/core/controllers/predictions.py

@@ -2,7 +2,7 @@
 from . import GET, POST, DELETE, PUT
 from .api import app, error_message
 
-from nesis.api.core.models.entities import Module
+from nesis.api.core.models.objects import Module
 import logging
 import nesis.api.core.services as services
 import nesis.api.core.services.util as util