Spaces:
Sleeping
Sleeping
File size: 15,731 Bytes
c69ecf4 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 | import ast
import os
from datetime import datetime
from urllib.parse import urlparse
import requests
from bs4 import BeautifulSoup
from ddgs import DDGS
from dotenv import load_dotenv
from langchain_core.tools import tool
from tavily import TavilyClient
from functools import lru_cache
from rag.rag_pipeline import RagPipeline
load_dotenv()
TIMEOUT = 15
HEADERS = {"User-Agent": "AgenticChatbot/1.0"}
# ==================== WEB SEARCH ====================
@tool
def web_search_tavily(query: str, max_results: int = 5) -> str:
"""Search the web using Tavily. Requires TAVILY_API_KEY."""
key = os.getenv("TAVILY_API_KEY")
if not key:
return "Add TAVILY_API_KEY to your .env file."
try:
data = TavilyClient(api_key=key).search(
query=query, max_results=max(1, min(max_results, 10)),
search_depth="basic", include_answer=True
)
results = [data["answer"]] if data.get("answer") else []
results += [
f"- {r.get('title', 'Untitled')}\nURL: {r.get('url', '')}\n{r.get('content', '')}"
for r in data.get("results", [])
]
return "\n\n".join(results) or "No results found."
except Exception as e:
return f"Web search failed: {e}"
@tool
def web_search_free(query: str, max_results: int = 5) -> str:
"""Search the web using DuckDuckGo. No API key required."""
try:
results = DDGS().text(query, max_results=max(1, min(max_results, 10)))
return "\n".join(
f"- {r.get('title', 'Untitled')}\nURL: {r.get('href', '')}\n{r.get('body', '')}"
for r in results
) or "No results found."
except Exception as e:
return f"Web search failed: {e}"
@tool
def wikipedia_search(query: str, max_results: int = 5) -> str:
"""Search Wikipedia pages using DuckDuckGo."""
try:
results = DDGS().text(
f"site:wikipedia.org {query}",
max_results=max(1, min(max_results, 10))
)
return "\n".join(
f"- {r.get('title', 'Untitled')}\nURL: {r.get('href', '')}\n{r.get('body', '')}"
for r in results
) or "No Wikipedia results found."
except Exception as e:
return f"Wikipedia search failed: {e}"
@tool
def fetch_webpage(url: str, max_chars: int = 8000) -> str:
"""Get readable text from a public webpage."""
parsed = urlparse(url)
if parsed.scheme not in ("http", "https") or not parsed.netloc or parsed.username:
return "Provide a valid public HTTP/HTTPS URL."
try:
response = requests.get(url, timeout=TIMEOUT, headers=HEADERS)
response.raise_for_status()
if "text/html" not in response.headers.get("Content-Type", "").lower():
return "The URL did not return an HTML page."
soup = BeautifulSoup(response.text, "html.parser")
for tag in soup(["script", "style", "nav", "footer", "header", "noscript"]):
tag.decompose()
text = " ".join(soup.get_text(" ", strip=True).split())
return text[:max(500, min(max_chars, 20000))] or "No readable text found."
except Exception as e:
return f"Could not fetch webpage: {e}"
# ==================== IMAGE SEARCH ====================
@tool
def image_search_unsplash(query: str, per_page: int = 5) -> str:
"""Search Unsplash photos. Requires UNSPLASH_ACCESS_KEY."""
key = os.getenv("UNSPLASH_ACCESS_KEY")
if not key:
return "Add UNSPLASH_ACCESS_KEY to your .env file."
try:
response = requests.get(
"https://api.unsplash.com/search/photos",
params={"query": query, "per_page": max(1, min(per_page, 10)),
"content_filter": "high"},
headers={**HEADERS, "Authorization": f"Client-ID {key}"},
timeout=TIMEOUT
)
response.raise_for_status()
photos = response.json().get("results", [])
return "\n\n".join(
f"- {p.get('alt_description') or p.get('description') or query}\n"
f"Image: {p.get('urls', {}).get('regular', '')}\n"
f"Photo page: {p.get('links', {}).get('html', '')}\n"
f"Photographer: {p.get('user', {}).get('name', 'Unknown')}"
for p in photos
) or "No Unsplash photos found."
except Exception as e:
return f"Image search failed: {e}"
@tool
def image_search_free(query: str, max_results: int = 5) -> str:
"""Search images using DuckDuckGo. Check usage rights before reuse."""
try:
results = DDGS().images(query, max_results=max(1, min(max_results, 10)))
return "\n".join(
f"- {r.get('title', 'Untitled')}\nImage: {r.get('image', '')}\nSource: {r.get('url', '')}"
for r in results
) or "No image results found."
except Exception as e:
return f"Image search failed: {e}"
# ==================== NEWS ====================
@tool
def search_news(query: str, page_size: int = 5, language: str = "en") -> str:
"""Search recent news. Requires NEWS_API_KEY."""
key = os.getenv("NEWS_API_KEY")
if not key:
return "Add NEWS_API_KEY to your .env file."
try:
response = requests.get(
"https://newsapi.org/v2/everything",
params={"q": query, "pageSize": max(1, min(page_size, 10)),
"language": language, "sortBy": "publishedAt"},
headers={**HEADERS, "X-Api-Key": key},
timeout=TIMEOUT
)
response.raise_for_status()
data = response.json()
if data.get("status") != "ok":
return f"NewsAPI error: {data.get('message', 'Unknown error')}"
return "\n\n".join(
f"- {a.get('title', 'Untitled')} ({(a.get('source') or {}).get('name', 'Unknown')})\n"
f"Published: {a.get('publishedAt', 'Unknown')}\nURL: {a.get('url', '')}\n"
f"{a.get('description') or ''}"
for a in data.get("articles", [])
) or "No news articles found."
except Exception as e:
return f"News search failed: {e}"
@tool
def get_top_headlines(country: str = "in", category: str = "technology",
page_size: int = 5) -> str:
"""Get top news headlines. Requires NEWS_API_KEY."""
key = os.getenv("NEWS_API_KEY")
if not key:
return "Add NEWS_API_KEY to your .env file."
try:
response = requests.get(
"https://newsapi.org/v2/top-headlines",
params={"country": country.lower(), "category": category.lower(),
"pageSize": max(1, min(page_size, 10))},
headers={**HEADERS, "X-Api-Key": key},
timeout=TIMEOUT
)
response.raise_for_status()
data = response.json()
if data.get("status") != "ok":
return f"NewsAPI error: {data.get('message', 'Unknown error')}"
return "\n\n".join(
f"- {a.get('title', 'Untitled')}\n"
f"Source: {(a.get('source') or {}).get('name', 'Unknown')}\n"
f"URL: {a.get('url', '')}\nPublished: {a.get('publishedAt', 'Unknown')}"
for a in data.get("articles", [])
) or "No headlines found."
except Exception as e:
return f"Headlines lookup failed: {e}"
# ==================== WEATHER ====================
def find_location(location: str):
"""Find a place's latitude and longitude with Open-Meteo."""
response = requests.get(
"https://geocoding-api.open-meteo.com/v1/search",
params={"name": location, "count": 1, "language": "en", "format": "json"},
timeout=TIMEOUT
)
response.raise_for_status()
places = response.json().get("results", [])
return places[0] if places else None
@tool
def get_current_weather(location: str) -> str:
"""Get current weather for a city or place."""
try:
place = find_location(location)
if not place:
return f"Location not found: {location}"
response = requests.get(
"https://api.open-meteo.com/v1/forecast",
params={
"latitude": place["latitude"],
"longitude": place["longitude"],
"current": "temperature_2m,relative_humidity_2m,apparent_temperature,precipitation,weather_code,wind_speed_10m",
"timezone": "auto"
},
timeout=TIMEOUT
)
response.raise_for_status()
current = response.json().get("current", {})
return (
f"Location: {place.get('name')}, {place.get('country', '')}\n"
f"Time: {current.get('time', 'Unknown')}\n"
f"Temperature: {current.get('temperature_2m', 'Unknown')} °C\n"
f"Feels like: {current.get('apparent_temperature', 'Unknown')} °C\n"
f"Humidity: {current.get('relative_humidity_2m', 'Unknown')}%\n"
f"Precipitation: {current.get('precipitation', 'Unknown')} mm\n"
f"Wind: {current.get('wind_speed_10m', 'Unknown')} km/h"
)
except Exception as e:
return f"Weather lookup failed: {e}"
@tool
def get_weather_forecast(location: str, days: int = 3) -> str:
"""Get a daily weather forecast for a city or place."""
try:
place = find_location(location)
if not place:
return f"Location not found: {location}"
days = max(1, min(days, 7))
response = requests.get(
"https://api.open-meteo.com/v1/forecast",
params={
"latitude": place["latitude"],
"longitude": place["longitude"],
"daily": "weather_code,temperature_2m_max,temperature_2m_min,precipitation_probability_max",
"forecast_days": days,
"timezone": "auto"
},
timeout=TIMEOUT
)
response.raise_for_status()
daily = response.json().get("daily", {})
return "\n".join(
f"- {date}: low {daily['temperature_2m_min'][i]} °C, "
f"high {daily['temperature_2m_max'][i]} °C, "
f"rain chance {daily['precipitation_probability_max'][i]}%"
for i, date in enumerate(daily.get("time", []))
) or "No forecast found."
except Exception as e:
return f"Forecast lookup failed: {e}"
# ==================== UTILITIES ====================
@tool
def calculator(expression: str) -> str:
"""Calculate basic arithmetic such as (25 * 4) + 10."""
def calculate(node):
if isinstance(node, ast.Constant) and type(node.value) in (int, float):
return node.value
if isinstance(node, ast.UnaryOp):
value = calculate(node.operand)
if isinstance(node.op, ast.USub):
return -value
if isinstance(node.op, ast.UAdd):
return value
if isinstance(node, ast.BinOp):
left, right = calculate(node.left), calculate(node.right)
if isinstance(node.op, ast.Add):
return left + right
if isinstance(node.op, ast.Sub):
return left - right
if isinstance(node.op, ast.Mult):
return left * right
if isinstance(node.op, ast.Div):
return left / right
if isinstance(node.op, ast.FloorDiv):
return left // right
if isinstance(node.op, ast.Mod):
return left % right
if isinstance(node.op, ast.Pow):
if abs(right) > 10:
raise ValueError("Exponent must be between -10 and 10.")
return left ** right
raise ValueError("Use numbers and basic arithmetic operators only.")
try:
if len(expression) > 200:
return "Expression too long."
result = calculate(ast.parse(expression, mode="eval").body)
return f"{expression} = {result}"
except Exception as e:
return f"Calculation failed: {e}"
@tool
def get_current_datetime() -> str:
"""Get the current date and time on the machine running the app."""
return datetime.now().astimezone().isoformat()
@tool
def convert_units(value: float, from_unit: str, to_unit: str) -> str:
"""Convert length, mass, or temperature units."""
source, target = from_unit.lower().strip(), to_unit.lower().strip()
length = {"m": 1, "km": 1000, "cm": 0.01, "mi": 1609.344, "ft": 0.3048}
mass = {"kg": 1, "g": 0.001, "lb": 0.45359237}
if source in length and target in length:
result = value * length[source] / length[target]
elif source in mass and target in mass:
result = value * mass[source] / mass[target]
elif source in ("c", "f") and target in ("c", "f"):
celsius = value if source == "c" else (value - 32) * 5 / 9
result = celsius if target == "c" else celsius * 9 / 5 + 32
else:
return "Unsupported units. Length: m/km/cm/mi/ft; mass: kg/g/lb; temperature: C/F."
return f"{value} {from_unit} = {result:.4f} {to_unit}"
# ==================== RAG ====================
@lru_cache(maxsize=1)
def get_rag_pipeline() -> RagPipeline:
"""
Lazily initialize the RAG pipeline once per process.
This avoids loading the embedding and reranker models when the application
starts if the current conversation never uses RAG.
"""
return RagPipeline()
def _search_knowledge_base(query: str, owner_id: str) -> dict:
if not query or not query.strip():
return {
"found": False,
"query": query,
"context": "",
"sources": [],
"reason": "Empty query.",
}
try:
rag = get_rag_pipeline()
result = rag.retrieve(query.strip(), owner_id=owner_id)
context = (result.get("context") or "").strip()
sources = result.get("sources") or []
found = bool(
result.get("success")
and result.get("has_documents")
and context
)
if not found:
return {
"found": False,
"query": query,
"context": "",
"sources": sources,
"reason": (
"No relevant information was found in the "
"knowledge base."
),
}
return {
"found": True,
"query": query,
"context": context,
"sources": sources,
"count": result.get("count", len(sources)),
}
except Exception as exc:
return {
"found": False,
"query": query,
"context": "",
"sources": [],
"reason": f"RAG execution failed: {exc}",
}
def make_search_knowledge_base_tool(owner_id: str):
"""Bind document search to the current browser profile, not model input."""
@tool("search_knowledge_base")
def search_knowledge_base(query: str) -> dict:
"""
Search the current user's uploaded PDFs for relevant information.
Returns whether relevant context was found, grounded document context,
and source/page metadata.
"""
return _search_knowledge_base(query, owner_id)
return search_knowledge_base |