2024-05-06 04:27:46 +00:00
|
|
|
import json
|
|
|
|
import logging
|
|
|
|
|
|
|
|
import requests
|
|
|
|
|
|
|
|
from apps.rag.search.main import SearchResult
|
2024-06-02 02:57:00 +00:00
|
|
|
from config import SRC_LOG_LEVELS
|
2024-05-06 04:27:46 +00:00
|
|
|
|
|
|
|
log = logging.getLogger(__name__)
|
|
|
|
log.setLevel(SRC_LOG_LEVELS["RAG"])
|
|
|
|
|
|
|
|
|
2024-06-02 02:57:00 +00:00
|
|
|
def search_serper(api_key: str, query: str, count: int) -> list[SearchResult]:
|
2024-05-06 04:27:46 +00:00
|
|
|
"""Search using serper.dev's API and return the results as a list of SearchResult objects.
|
|
|
|
|
|
|
|
Args:
|
|
|
|
api_key (str): A serper.dev API key
|
|
|
|
query (str): The query to search for
|
|
|
|
"""
|
|
|
|
url = "https://google.serper.dev/search"
|
|
|
|
|
|
|
|
payload = json.dumps({"q": query})
|
|
|
|
headers = {"X-API-KEY": api_key, "Content-Type": "application/json"}
|
|
|
|
|
|
|
|
response = requests.request("POST", url, headers=headers, data=payload)
|
|
|
|
response.raise_for_status()
|
|
|
|
|
|
|
|
json_response = response.json()
|
|
|
|
results = sorted(
|
|
|
|
json_response.get("organic", []), key=lambda x: x.get("position", 0)
|
|
|
|
)
|
|
|
|
return [
|
|
|
|
SearchResult(
|
2024-05-06 08:39:25 +00:00
|
|
|
link=result["link"],
|
2024-05-06 04:27:46 +00:00
|
|
|
title=result.get("title"),
|
|
|
|
snippet=result.get("description"),
|
|
|
|
)
|
2024-06-02 02:57:00 +00:00
|
|
|
for result in results[:count]
|
2024-05-06 04:27:46 +00:00
|
|
|
]
|