#!/usr/bin/env -S uv run --script --quiet
# /// script
# requires-python = ">=3.12"
# dependencies = ["httpx", "pydantic-settings"]
# ///
"""Check vector index and embeddings status."""

import os
import httpx
from pydantic_settings import BaseSettings, SettingsConfigDict


class Settings(BaseSettings):
    model_config = SettingsConfigDict(
        env_file=os.environ.get("ENV_FILE", ".env"), extra="ignore"
    )
    turso_url: str
    turso_token: str

    @property
    def turso_host(self) -> str:
        url = self.turso_url
        if url.startswith("libsql://"):
            url = url[len("libsql://"):]
        return url


def query(settings, sql):
    response = httpx.post(
        f"https://{settings.turso_host}/v2/pipeline",
        headers={
            "Authorization": f"Bearer {settings.turso_token}",
            "Content-Type": "application/json",
        },
        json={
            "requests": [
                {"type": "execute", "stmt": {"sql": sql}},
                {"type": "close"},
            ]
        },
        timeout=30,
    )
    response.raise_for_status()
    return response.json()


settings = Settings()  # type: ignore

# Check embeddings count
print("Checking embeddings...")
result = query(settings, "SELECT COUNT(*) as total, SUM(CASE WHEN embedding IS NOT NULL THEN 1 ELSE 0 END) as with_embeddings FROM documents")
data = result["results"][0]["response"]["result"]["rows"][0]
total = data[0]["value"] if isinstance(data[0], dict) else data[0]
with_emb = data[1]["value"] if isinstance(data[1], dict) else data[1]
print(f"  Total documents: {total}")
print(f"  With embeddings: {with_emb}")

# Check if vector index exists
print("\nChecking for vector index...")
result = query(settings, "SELECT name FROM sqlite_master WHERE type='table' AND name LIKE '%embedding%'")
rows = result["results"][0]["response"]["result"]["rows"]
for row in rows:
    name = row[0]["value"] if isinstance(row[0], dict) else row[0]
    print(f"  Found table: {name}")

# Try to use the index directly
print("\nTrying vector search...")
try:
    result = query(settings, """
        SELECT d.uri, d.title
        FROM vector_top_k('documents_embedding_idx',
          (SELECT embedding FROM documents LIMIT 1), 3) AS v
        JOIN documents d ON d.rowid = v.id
        LIMIT 3
    """)
    print(f"  Result: {result['results'][0]}")
except Exception as e:
    print(f"  Error: {e}")