"""Add knowledge base tables with pgvector support

Revision ID: e5f7g9h1cd45
Revises: d4f6g8h0ab23
Create Date: 2026-05-23 12:00:00.000000

"""
from typing import Sequence, Union

import sqlalchemy as sa
from alembic import op
from sqlalchemy.dialects import postgresql

# revision identifiers, used by Alembic.
revision: str = "e5f7g9h1cd45"
down_revision: Union[str, None] = "d4f6g8h0ab23"
branch_labels: Union[str, Sequence[str], None] = None
depends_on: Union[str, Sequence[str], None] = None


def upgrade() -> None:
    # ------------------------------------------------------------------ #
    # 1. Enable pgvector extension
    # ------------------------------------------------------------------ #
    op.execute("CREATE EXTENSION IF NOT EXISTS vector")

    # ------------------------------------------------------------------ #
    # 2. knowledge_bases
    # ------------------------------------------------------------------ #
    op.create_table(
        "knowledge_bases",
        sa.Column(
            "id",
            postgresql.UUID(as_uuid=True),
            primary_key=True,
            nullable=False,
        ),
        sa.Column(
            "organization_id",
            postgresql.UUID(as_uuid=True),
            sa.ForeignKey("organizations.id", ondelete="CASCADE"),
            nullable=False,
        ),
        sa.Column("name", sa.String(200), nullable=False),
        sa.Column("type", sa.String(20), nullable=False),
        sa.Column("description", sa.Text, nullable=True),
        sa.Column("document_count", sa.Integer, server_default="0", nullable=False),
        sa.Column("status", sa.String(20), server_default="active", nullable=False),
        sa.Column(
            "created_by",
            postgresql.UUID(as_uuid=True),
            sa.ForeignKey("users.id", ondelete="SET NULL"),
            nullable=True,
        ),
        sa.Column(
            "created_at",
            sa.DateTime(timezone=True),
            server_default=sa.text("now()"),
            nullable=False,
        ),
        sa.Column(
            "updated_at",
            sa.DateTime(timezone=True),
            server_default=sa.text("now()"),
            nullable=False,
        ),
    )
    op.create_index("idx_knowledge_bases_organization_id", "knowledge_bases", ["organization_id"])
    op.create_index("idx_knowledge_bases_type", "knowledge_bases", ["type"])
    op.create_index("idx_knowledge_bases_status", "knowledge_bases", ["status"])

    # ------------------------------------------------------------------ #
    # 3. knowledge_documents
    # ------------------------------------------------------------------ #
    op.create_table(
        "knowledge_documents",
        sa.Column(
            "id",
            postgresql.UUID(as_uuid=True),
            primary_key=True,
            nullable=False,
        ),
        sa.Column(
            "knowledge_base_id",
            postgresql.UUID(as_uuid=True),
            sa.ForeignKey("knowledge_bases.id", ondelete="CASCADE"),
            nullable=False,
        ),
        sa.Column("title", sa.String(500), nullable=False),
        sa.Column("source_type", sa.String(20), nullable=False),
        sa.Column("source_url", sa.String(2000), nullable=True),
        sa.Column("content", sa.Text, nullable=False),
        sa.Column("content_hash", sa.String(64), nullable=False),
        sa.Column("chunk_count", sa.Integer, server_default="0", nullable=False),
        sa.Column("status", sa.String(20), server_default="processing", nullable=False),
        sa.Column("error_message", sa.Text, nullable=True),
        sa.Column("metadata", postgresql.JSONB(astext_type=sa.Text()), nullable=True),
        sa.Column(
            "created_at",
            sa.DateTime(timezone=True),
            server_default=sa.text("now()"),
            nullable=False,
        ),
        sa.Column(
            "updated_at",
            sa.DateTime(timezone=True),
            server_default=sa.text("now()"),
            nullable=False,
        ),
    )
    op.create_index(
        "idx_knowledge_documents_knowledge_base_id",
        "knowledge_documents",
        ["knowledge_base_id"],
    )
    op.create_index("idx_knowledge_documents_status", "knowledge_documents", ["status"])
    op.create_index(
        "idx_knowledge_documents_content_hash", "knowledge_documents", ["content_hash"]
    )

    # ------------------------------------------------------------------ #
    # 4. knowledge_chunks  (embedding column via raw SQL for vector type)
    # ------------------------------------------------------------------ #
    op.create_table(
        "knowledge_chunks",
        sa.Column(
            "id",
            postgresql.UUID(as_uuid=True),
            primary_key=True,
            nullable=False,
        ),
        sa.Column(
            "document_id",
            postgresql.UUID(as_uuid=True),
            sa.ForeignKey("knowledge_documents.id", ondelete="CASCADE"),
            nullable=False,
        ),
        sa.Column("content", sa.Text, nullable=False),
        sa.Column("chunk_index", sa.Integer, nullable=False),
        sa.Column("token_count", sa.Integer, server_default="0", nullable=False),
        sa.Column("metadata", postgresql.JSONB(astext_type=sa.Text()), nullable=True),
        sa.Column(
            "created_at",
            sa.DateTime(timezone=True),
            server_default=sa.text("now()"),
            nullable=False,
        ),
    )

    # Add vector embedding column via raw SQL (pgvector type not in SA dialect)
    # Dimension 1536 matches OpenAI text-embedding-3-small
    op.execute(
        "ALTER TABLE knowledge_chunks ADD COLUMN embedding vector(1536)"
    )

    op.create_index("idx_knowledge_chunks_document_id", "knowledge_chunks", ["document_id"])
    op.create_index(
        "idx_knowledge_chunks_chunk_index",
        "knowledge_chunks",
        ["document_id", "chunk_index"],
    )

    # HNSW index for approximate nearest-neighbor cosine similarity search
    op.execute(
        "CREATE INDEX ix_knowledge_chunks_embedding "
        "ON knowledge_chunks USING hnsw (embedding vector_cosine_ops)"
    )

    # ------------------------------------------------------------------ #
    # 5. knowledge_search_logs
    # ------------------------------------------------------------------ #
    op.create_table(
        "knowledge_search_logs",
        sa.Column(
            "id",
            postgresql.UUID(as_uuid=True),
            primary_key=True,
            nullable=False,
        ),
        sa.Column(
            "organization_id",
            postgresql.UUID(as_uuid=True),
            sa.ForeignKey("organizations.id", ondelete="CASCADE"),
            nullable=False,
        ),
        sa.Column(
            "user_id",
            postgresql.UUID(as_uuid=True),
            sa.ForeignKey("users.id", ondelete="SET NULL"),
            nullable=True,
        ),
        sa.Column("query", sa.Text, nullable=False),
        sa.Column("knowledge_base_ids", postgresql.JSONB(astext_type=sa.Text()), nullable=True),
        sa.Column("results_count", sa.Integer, server_default="0", nullable=False),
        sa.Column("latency_ms", sa.Integer, server_default="0", nullable=False),
        sa.Column(
            "created_at",
            sa.DateTime(timezone=True),
            server_default=sa.text("now()"),
            nullable=False,
        ),
    )
    op.create_index(
        "idx_knowledge_search_logs_organization_id",
        "knowledge_search_logs",
        ["organization_id"],
    )
    op.create_index(
        "idx_knowledge_search_logs_user_id", "knowledge_search_logs", ["user_id"]
    )
    op.create_index(
        "idx_knowledge_search_logs_created_at", "knowledge_search_logs", ["created_at"]
    )


def downgrade() -> None:
    # Drop tables in reverse dependency order
    op.drop_table("knowledge_search_logs")
    op.execute("DROP INDEX IF EXISTS ix_knowledge_chunks_embedding")
    op.drop_table("knowledge_chunks")
    op.drop_table("knowledge_documents")
    op.drop_table("knowledge_bases")
    # Note: we do NOT drop the vector extension as other tables might rely on it