256 lines
8.9 KiB
Python
256 lines
8.9 KiB
Python
|
#!/usr/bin/env python3
|
||
|
|
"""
|
||
|
|
Unified Search - Perplexity primary, SearXNG fallback
|
||
|
|
|
||
|
|
Usage:
|
||
|
|
search "your query" # Perplexity primary, SearXNG fallback
|
||
|
|
search p "your query" # Perplexity only
|
||
|
|
search perplexity "your query" # Perplexity only (alias)
|
||
|
|
search local "your query" # SearXNG only
|
||
|
|
search searxng "your query" # SearXNG only (alias)
|
||
|
|
search --citations "query" # Include citations (Perplexity)
|
||
|
|
search --model sonar-pro "query" # Use specific Perplexity model
|
||
|
|
"""
|
||
|
|
|
||
|
|
import json
|
||
|
|
import sys
|
||
|
|
import urllib.request
|
||
|
|
import urllib.parse
|
||
|
|
from pathlib import Path
|
||
|
|
|
||
|
|
# Configuration
|
||
|
|
PERPLEXITY_CONFIG = Path(__file__).parent.parent / "config.json"
|
||
|
|
SEARXNG_URL = "http://10.0.0.8:8888"
|
||
|
|
|
||
|
|
def load_perplexity_config():
|
||
|
|
"""Load Perplexity API configuration"""
|
||
|
|
try:
|
||
|
|
with open(PERPLEXITY_CONFIG) as f:
|
||
|
|
return json.load(f)
|
||
|
|
except Exception as e:
|
||
|
|
print(f"Error loading Perplexity config: {e}", file=sys.stderr)
|
||
|
|
return None
|
||
|
|
|
||
|
|
def search_perplexity(query, model="sonar", max_tokens=1000, include_citations=False, search_context="low"):
|
||
|
|
"""Search using Perplexity API"""
|
||
|
|
config = load_perplexity_config()
|
||
|
|
if not config:
|
||
|
|
return {"error": "Perplexity not configured", "fallback_needed": True}
|
||
|
|
|
||
|
|
api_key = config.get("api_key")
|
||
|
|
base_url = config.get("base_url", "https://api.perplexity.ai")
|
||
|
|
|
||
|
|
if not api_key:
|
||
|
|
return {"error": "Perplexity API key not set", "fallback_needed": True}
|
||
|
|
|
||
|
|
payload = {
|
||
|
|
"model": model,
|
||
|
|
"messages": [
|
||
|
|
{"role": "system", "content": "Be precise and concise."},
|
||
|
|
{"role": "user", "content": query}
|
||
|
|
],
|
||
|
|
"max_tokens": max_tokens,
|
||
|
|
"search_context_size": search_context
|
||
|
|
}
|
||
|
|
|
||
|
|
data = json.dumps(payload).encode()
|
||
|
|
|
||
|
|
req = urllib.request.Request(
|
||
|
|
f"{base_url}/chat/completions",
|
||
|
|
data=data,
|
||
|
|
headers={
|
||
|
|
"Authorization": f"Bearer {api_key}",
|
||
|
|
"Content-Type": "application/json"
|
||
|
|
}
|
||
|
|
)
|
||
|
|
|
||
|
|
try:
|
||
|
|
with urllib.request.urlopen(req, timeout=60) as response:
|
||
|
|
result = json.loads(response.read().decode())
|
||
|
|
|
||
|
|
output = {
|
||
|
|
"source": "perplexity",
|
||
|
|
"text": result["choices"][0]["message"]["content"],
|
||
|
|
"model": result.get("model"),
|
||
|
|
"usage": result.get("usage", {}),
|
||
|
|
"citations": result.get("citations", []),
|
||
|
|
"search_results": result.get("search_results", [])
|
||
|
|
}
|
||
|
|
|
||
|
|
return output
|
||
|
|
|
||
|
|
except urllib.error.HTTPError as e:
|
||
|
|
error_body = e.read().decode()
|
||
|
|
if e.code == 429: # Rate limit
|
||
|
|
return {"error": f"Perplexity rate limited: {error_body}", "fallback_needed": True}
|
||
|
|
return {"error": f"Perplexity HTTP {e.code}: {error_body}", "fallback_needed": True}
|
||
|
|
except Exception as e:
|
||
|
|
return {"error": f"Perplexity error: {str(e)}", "fallback_needed": True}
|
||
|
|
|
||
|
|
def search_searxng(query, limit=10):
|
||
|
|
"""Search using local SearXNG"""
|
||
|
|
try:
|
||
|
|
encoded_query = urllib.parse.quote(query)
|
||
|
|
url = f"{SEARXNG_URL}/search?q={encoded_query}&format=json"
|
||
|
|
|
||
|
|
req = urllib.request.Request(url)
|
||
|
|
with urllib.request.urlopen(req, timeout=30) as response:
|
||
|
|
result = json.loads(response.read().decode())
|
||
|
|
|
||
|
|
results = result.get("results", [])[:limit]
|
||
|
|
formatted_results = []
|
||
|
|
|
||
|
|
for r in results:
|
||
|
|
formatted_results.append({
|
||
|
|
"title": r.get("title", ""),
|
||
|
|
"url": r.get("url", ""),
|
||
|
|
"content": r.get("content", "")[:200] + "..." if len(r.get("content", "")) > 200 else r.get("content", "")
|
||
|
|
})
|
||
|
|
|
||
|
|
# Format as readable text
|
||
|
|
text_output = f"Search results for: {query}\n\n"
|
||
|
|
for i, r in enumerate(formatted_results, 1):
|
||
|
|
text_output += f"[{i}] {r['title']}\n{r['url']}\n{r['content']}\n\n"
|
||
|
|
|
||
|
|
return {
|
||
|
|
"source": "searxng",
|
||
|
|
"text": text_output.strip(),
|
||
|
|
"results": formatted_results,
|
||
|
|
"query": query
|
||
|
|
}
|
||
|
|
|
||
|
|
except Exception as e:
|
||
|
|
return {"error": f"SearXNG error: {str(e)}", "fallback_needed": False}
|
||
|
|
|
||
|
|
def unified_search(query, mode="default", model="sonar", include_citations=False, max_tokens=1000, search_context="low"):
|
||
|
|
"""
|
||
|
|
Unified search with Perplexity primary, SearXNG fallback
|
||
|
|
|
||
|
|
Modes:
|
||
|
|
default: Perplexity primary, SearXNG fallback
|
||
|
|
perplexity: Perplexity only
|
||
|
|
local/searxng: SearXNG only
|
||
|
|
"""
|
||
|
|
|
||
|
|
if mode in ["perplexity", "p"]:
|
||
|
|
# Perplexity only
|
||
|
|
result = search_perplexity(query, model, max_tokens, include_citations, search_context)
|
||
|
|
return result
|
||
|
|
|
||
|
|
elif mode in ["local", "searxng", "s"]:
|
||
|
|
# SearXNG only
|
||
|
|
result = search_searxng(query)
|
||
|
|
return result
|
||
|
|
|
||
|
|
else:
|
||
|
|
# Default: Perplexity primary, SearXNG fallback
|
||
|
|
result = search_perplexity(query, model, max_tokens, include_citations, search_context)
|
||
|
|
|
||
|
|
if result.get("fallback_needed") or result.get("error"):
|
||
|
|
print(f"⚠️ Perplexity failed: {result.get('error', 'Unknown error')}", file=sys.stderr)
|
||
|
|
print("🔄 Falling back to SearXNG...\n", file=sys.stderr)
|
||
|
|
|
||
|
|
fallback = search_searxng(query)
|
||
|
|
if not fallback.get("error"):
|
||
|
|
return fallback
|
||
|
|
else:
|
||
|
|
return {"error": f"Both Perplexity and SearXNG failed. Perplexity: {result.get('error')}, SearXNG: {fallback.get('error')}"}
|
||
|
|
|
||
|
|
return result
|
||
|
|
|
||
|
|
def main():
|
||
|
|
import argparse
|
||
|
|
|
||
|
|
parser = argparse.ArgumentParser(
|
||
|
|
description="Unified search: Perplexity primary, SearXNG fallback",
|
||
|
|
formatter_class=argparse.RawDescriptionHelpFormatter,
|
||
|
|
epilog="""
|
||
|
|
Examples:
|
||
|
|
search "latest AI news" # Perplexity primary, SearXNG fallback
|
||
|
|
search p "quantum computing explained" # Perplexity only
|
||
|
|
search local "ip address lookup" # SearXNG only
|
||
|
|
search --citations "who invented Python" # Include citations
|
||
|
|
search --model sonar-pro "coding help" # Use Pro model
|
||
|
|
"""
|
||
|
|
)
|
||
|
|
|
||
|
|
parser.add_argument("args", nargs="*", help="[mode] query (mode: p/perplexity/local/searxng)")
|
||
|
|
parser.add_argument("--citations", action="store_true",
|
||
|
|
help="Include citations (Perplexity only)")
|
||
|
|
parser.add_argument("--model", default="sonar",
|
||
|
|
choices=["sonar", "sonar-pro", "sonar-reasoning", "sonar-deep-research"],
|
||
|
|
help="Perplexity model to use")
|
||
|
|
parser.add_argument("--max-tokens", type=int, default=1000,
|
||
|
|
help="Maximum tokens in response (Perplexity)")
|
||
|
|
parser.add_argument("--search-context", default="low",
|
||
|
|
choices=["low", "medium", "high"],
|
||
|
|
help="Search context size (Perplexity)")
|
||
|
|
|
||
|
|
args = parser.parse_args()
|
||
|
|
|
||
|
|
# Parse positional arguments
|
||
|
|
mode = "default"
|
||
|
|
query_parts = []
|
||
|
|
|
||
|
|
if not args.args:
|
||
|
|
print("Error: No query provided", file=sys.stderr)
|
||
|
|
parser.print_help()
|
||
|
|
sys.exit(1)
|
||
|
|
|
||
|
|
# Check if first arg is a mode indicator
|
||
|
|
if args.args[0] in ["p", "perplexity", "local", "searxng", "s"]:
|
||
|
|
mode = args.args[0]
|
||
|
|
if mode == "p":
|
||
|
|
mode = "perplexity"
|
||
|
|
elif mode == "s":
|
||
|
|
mode = "searxng"
|
||
|
|
query_parts = args.args[1:]
|
||
|
|
else:
|
||
|
|
query_parts = args.args
|
||
|
|
|
||
|
|
query = " ".join(query_parts)
|
||
|
|
|
||
|
|
if not query:
|
||
|
|
print("Error: No query provided", file=sys.stderr)
|
||
|
|
parser.print_help()
|
||
|
|
sys.exit(1)
|
||
|
|
|
||
|
|
result = unified_search(
|
||
|
|
query,
|
||
|
|
mode=mode,
|
||
|
|
model=args.model,
|
||
|
|
include_citations=args.citations,
|
||
|
|
max_tokens=args.max_tokens,
|
||
|
|
search_context=args.search_context
|
||
|
|
)
|
||
|
|
|
||
|
|
if "error" in result:
|
||
|
|
print(f"Error: {result['error']}", file=sys.stderr)
|
||
|
|
sys.exit(1)
|
||
|
|
|
||
|
|
# Print result
|
||
|
|
if result.get("source") == "perplexity":
|
||
|
|
print(f"🔍 Perplexity ({result.get('model', 'unknown')})")
|
||
|
|
if result.get("usage"):
|
||
|
|
cost = result["usage"].get("cost", {})
|
||
|
|
total = cost.get("total_cost", "unknown")
|
||
|
|
print(f"💰 Cost: ${total}")
|
||
|
|
print()
|
||
|
|
print(result["text"])
|
||
|
|
|
||
|
|
if args.citations and result.get("citations"):
|
||
|
|
print("\n--- Sources ---")
|
||
|
|
for i, citation in enumerate(result["citations"][:5], 1):
|
||
|
|
print(f"[{i}] {citation}")
|
||
|
|
|
||
|
|
elif result.get("source") == "searxng":
|
||
|
|
print(f"🔍 SearXNG (local)")
|
||
|
|
print()
|
||
|
|
print(result["text"])
|
||
|
|
|
||
|
|
else:
|
||
|
|
print(result.get("text", "No results"))
|
||
|
|
|
||
|
|
if __name__ == "__main__":
|
||
|
|
main()
|