SERPent / app.py
Game4all's picture
Extract field, background and description
cf1c265
raw
history blame
7.79 kB
from contextlib import asynccontextmanager
from typing import Optional
import expiringdict
from fastapi import APIRouter, FastAPI
from fastapi.routing import APIRouter as Router
import httpx
from pydantic import BaseModel, Field
from playwright.async_api import async_playwright, Browser, BrowserContext, Page
import logging
import uvicorn
from scrap import scrap_patent_async, scrap_patent_bulk_async
from search import APISearchResults, APIPatentResults, query_bing_search, query_brave_search, query_ddg_search, query_google_patents
logging.basicConfig(
level=logging.INFO,
format='[%(asctime)s][%(levelname)s][%(filename)s:%(lineno)d]: %(message)s',
datefmt='%Y-%m-%d %H:%M:%S'
)
# playwright global context
playwright = None
pw_browser: Optional[Browser] = None
# httpx client
httpx_client = httpx.AsyncClient(timeout=30, limits=httpx.Limits(
max_connections=30, max_keepalive_connections=20))
@asynccontextmanager
async def api_lifespan(app: FastAPI):
global playwright, pw_browser
playwright = await async_playwright().start()
pw_browser = await playwright.chromium.launch(headless=True)
yield
await pw_browser.close()
await playwright.stop()
app = FastAPI(lifespan=api_lifespan)
backend_status = expiringdict.ExpiringDict(max_len=5, max_age_seconds=15*60)
# Router for scrapping related endpoints
scrap_router = APIRouter(prefix="/scrap", tags=["scrapping"])
@app.get('/')
async def status():
backend_keys = [k[0] for k in backend_status.items()]
backend_status_dict = {}
for k in backend_keys:
backend_status_dict[k] = backend_status.get(k)
return {"status": "running", "backend_status": backend_status_dict}
# ===================== Search endpoints =====================
class APISearchParams(BaseModel):
queries: list[str] = Field(...,
description="The list of queries to search for")
n_results: int = Field(
10, description="Number of results to return for each query. Valid values are 10, 25, 50 and 100")
@app.post("/search_scholar")
async def query_google_scholar(params: APISearchParams):
"""Queries google scholar for the specified query"""
return {"error": "Unimplemented"}
@app.post("/search_patents")
async def search_patents(params: APISearchParams) -> APIPatentResults:
"""Searches google patents for the specified queries and returns the found documents."""
results = []
for q in params.queries:
logging.info(f"Searching Google Patents with query `{q}`")
try:
res = await query_google_patents(pw_browser, q, params.n_results)
results.extend(res)
except Exception as e:
backend_status["gpatents"] = "rate-limited"
logging.error(
f"Failed to query Google Patents with query `{q}`: {e}")
return APIPatentResults(results=results, error=None)
@app.post("/search_brave")
async def search_brave(params: APISearchParams) -> APISearchResults:
"""Searches brave search for the specified queries and returns the found documents."""
results = []
last_exception: Optional[Exception] = None
for q in params.queries:
logging.info(f"Searching Brave search with query `{q}`")
try:
res = await query_brave_search(pw_browser, q, params.n_results)
results.extend(res)
except Exception as e:
last_exception = e
backend_status["brave"] = "rate-limited"
logging.error(
f"Failed to query Brave search with query `{q}`: {e}")
return APISearchResults(results=results, error=str(last_exception) if len(results) == 0 and last_exception else None)
@app.post("/search_bing")
async def search_bing(params: APISearchParams) -> APISearchResults:
"""Searches Bing search for the specified queries and returns the found documents."""
results = []
last_exception: Optional[Exception] = None
for q in params.queries:
logging.info(f"Searching Bing search with query `{q}`")
try:
res = await query_brave_search(pw_browser, q, params.n_results)
results.extend(res)
except Exception as e:
last_exception = e
backend_status["bing"] = "rate-limited"
logging.error(
f"Failed to query Bing search with query `{q}`: {e}")
return APISearchResults(results=results, error=str(last_exception) if len(results) == 0 and last_exception else None)
@app.post("/search_duck")
async def search_duck(params: APISearchParams) -> APISearchResults:
"""Searches duckduckgo for the specified queries and returns the found documents"""
results = []
last_exception: Optional[Exception] = None
for q in params.queries:
logging.info(f"Querying DDG with query: `{q}`")
try:
res = await query_ddg_search(q, params.n_results)
results.extend(res)
except Exception as e:
last_exception = e
backend_status["duckduckgo"] = "rate-limited"
logging.error(f"Failed to query DDG with query `{q}`: {e}")
return APISearchResults(results=results, error=str(last_exception) if len(results) == 0 and last_exception else None)
@app.post("/search")
async def search(params: APISearchParams):
"""Attempts to search the specified queries using ALL backends"""
results = []
for q in params.queries:
try:
logging.info(f"Querying DDG with query: `{q}`")
res = await query_ddg_search(q, params.n_results)
results.extend(res)
continue
except Exception as e:
logging.error(f"Failed to query DDG with query `{q}`: {e}")
logging.info("Trying with next browser backend.")
try:
logging.info(f"Querying Brave Search with query: `{q}`")
res = await query_brave_search(pw_browser, q, params.n_results)
results.extend(res)
continue
except Exception as e:
logging.error(
f"Failed to query Brave Search with query `{q}`: {e}")
logging.info("Trying with next browser backend.")
try:
logging.info(f"Querying Bing with query: `{q}`")
res = await query_bing_search(pw_browser, q, params.n_results)
results.extend(res)
continue
except Exception as e:
logging.error(f"Failed to query Bing search with query `{q}`: {e}")
logging.info("Trying with next browser backend.")
if len(results) == 0:
return APISearchResults(results=[], error="All backends are rate-limited.")
return APISearchResults(results=results, error=None)
# =========================== Scrapping endpoints ===========================
@scrap_router.get("/scrap_patent/{patent_id}")
async def scrap_patent(patent_id: str):
"""Scraps the specified patent from Google Patents."""
patent = await scrap_patent_async(httpx_client, f"https://patents.google.com/patent/{patent_id}/en")
return patent
class ScrapPatentsRequest(BaseModel):
"""Request model for scrapping multiple patents."""
patent_ids: list[str] = Field(...,
description="List of patent IDs to scrap")
@scrap_router.post("/scrap_patents_bulk")
async def scrap_patents(params: ScrapPatentsRequest):
"""Scraps multiple patents from Google Patents."""
patents = await scrap_patent_bulk_async(httpx_client, [
f"https://patents.google.com/patent/{pid}/en" for pid in params.patent_ids])
return patents
# ===============================================================================
app.include_router(scrap_router)
uvicorn.run(app, host="0.0.0.0", port=7860)