Olostep + LangChain 集成

pip install langchain-olostep

poetry add langchain-olostep

from langchain_olostep import scrape_website
import asyncio

# 抓取一个网站
content = asyncio.run(scrape_website.ainvoke({
    "url": "https://example.com",
    "format": "markdown"
}))

print(content)

# 等待动态内容
content = asyncio.run(scrape_website.ainvoke({
    "url": "https://example.com",
    "format": "markdown",
    "wait_before_scraping": 2000
}))

# 使用专门的解析器
content = asyncio.run(scrape_website.ainvoke({
    "url": "https://www.amazon.com/dp/PRODUCT_ID",
    "parser": "@olostep/amazon-product",
    "format": "json"
}))

from langchain_olostep import scrape_batch
import asyncio

# 抓取多个 URL
result = asyncio.run(scrape_batch.ainvoke({
    "urls": [
        "https://example1.com",
        "https://example2.com",
        "https://example3.com"
    ],
    "format": "markdown"
}))

print(result)
# 返回: {"batch_id": "batch_xxx", "status": "in_progress", ...}

from langchain_olostep import answer_question
import asyncio

# 提问一个简单问题
result = asyncio.run(answer_question.ainvoke({
    "task": "What is the capital of France?"
}))

print(result)
# 返回: {"answer": {"result": "Paris"}, "sources": [...]}

# 使用 JSON 架构获取结构化数据
result = asyncio.run(answer_question.ainvoke({
    "task": "What is the latest book by J.K. Rowling?",
    "json_schema": {
        "book_title": "",
        "author": "",
        "release_date": ""
    }
}))

print(result)
# 返回带有来源的结构化答案

# 丰富公司数据
result = asyncio.run(answer_question.ainvoke({
    "task": "Find the CEO and headquarters of Stripe",
    "json_schema": {
        "ceo_name": "",
        "headquarters": "",
        "founded_year": ""
    }
}))

# 使用 "NOT_FOUND" 值处理不确定性

from langchain_olostep import extract_urls
import asyncio

# 从网站获取所有 URL
result = asyncio.run(extract_urls.ainvoke({
    "url": "https://example.com",
    "top_n": 100
}))

print(result)
# 返回: {"urls": [...], "total_urls": 100, ...}

# 仅获取博客 URL
result = asyncio.run(extract_urls.ainvoke({
    "url": "https://example.com",
    "include_urls": ["/blog/**"],
    "exclude_urls": ["/admin/**", "/private/**"],
    "top_n": 50
}))

from langchain_olostep import crawl_website
import asyncio

# 爬取整个文档网站
result = asyncio.run(crawl_website.ainvoke({
    "start_url": "https://docs.example.com",
    "max_pages": 100
}))

print(result)
# 返回: {"crawl_id": "crawl_xxx", "status": "in_progress", ...}

# 使用 URL 过滤器进行爬取
result = asyncio.run(crawl_website.ainvoke({
    "start_url": "https://example.com",
    "max_pages": 200,
    "include_urls": ["/**"],
    "exclude_urls": ["/admin/**", "/private/**"],
    "max_depth": 3
}))

from langchain.agents import initialize_agent, AgentType
from langchain_openai import ChatOpenAI
from langchain_olostep import (
    scrape_website,
    answer_question,
    extract_urls
)

# 使用 Olostep 工具创建代理
tools = [scrape_website, answer_question, extract_urls]
llm = ChatOpenAI(model="gpt-4o-mini")

agent = initialize_agent(
    tools=tools,
    llm=llm,
    agent=AgentType.ZERO_SHOT_REACT_DESCRIPTION,
    verbose=True
)

# 使用代理
result = agent.run("""
Research the company at https://company.com:
1. Scrape their about page
2. Search for their latest funding round
3. Extract all their product pages
""")

print(result)

from langgraph.graph import StateGraph, END
from langchain_olostep import (
    scrape_website,
    scrape_batch,
    answer_question,
    extract_urls
)
from langchain_openai import ChatOpenAI
import json

def create_research_agent():
    workflow = StateGraph(dict)
    
    def discover_pages(state):
        # 从目标站点提取所有 URL
        result = extract_urls.invoke({
            "url": state["target_url"],
            "include_urls": ["/product/**"],
            "top_n": 50
        })
        state["urls"] = json.loads(result)["urls"]
        return state
    
    def scrape_pages(state):
        # 批量抓取发现的页面
        result = scrape_batch.invoke({
            "urls": state["urls"],
            "format": "markdown"
        })
        state["batch_id"] = json.loads(result)["batch_id"]
        return state
    
    def answer_questions(state):
        # 使用 AI 回答有关数据的问题
        result = answer_question.invoke({
            "task": state["research_question"],
            "json_schema": state["desired_format"]
        })
        state["answer"] = json.loads(result)["answer"]
        return state
    
    workflow.add_node("discover", discover_pages)
    workflow.add_node("scrape", scrape_pages)
    workflow.add_node("analyze", answer_questions)
    
    workflow.set_entry_point("discover")
    workflow.add_edge("discover", "scrape")
    workflow.add_edge("scrape", "analyze")
    workflow.add_edge("analyze", END)
    
    return workflow.compile()

# 使用代理
agent = create_research_agent()
result = agent.invoke({
    "target_url": "https://store.com",
    "research_question": "What are the top 5 most expensive products?",
    "desired_format": {
        "products": [{"name": "", "price": "", "url": ""}]
    }
})