from aiohttp import web

app = web.Application()
app.router.add_view('/render.html', RenderHtmlView)
app.router.add_view('/render.png', RenderPngView)
app.router.add_view('/render.jpeg', RenderJpegView)
app.router.add_view('/render.json', RenderJsonView)

c = LaunchChrome()
app.on_startup.append(c.on_startup_tasks)
app.on_cleanup.append(c.on_cleanup_tasks)

async def on_startup_tasks(self, app: web.Application) -> None:
		page_count = 4
		await asyncio.create_task(self._launch())
		app["browser"] = self.browser
		tasks = [asyncio.create_task(self.launch_tab()) for _ in range(page_count-1)]
		await asyncio.gather(*tasks)
		queue = asyncio.Queue(maxsize=page_count+1)
		for i in await self.browser.pages():
				await queue.put(i)
		app["pages_queue"] = queue
		app["screenshot_lock"] = asyncio.Lock()

async def on_cleanup_tasks(self, app: web.Application) -> None:
		await self.browser.close()

async def _goto(self, page: Optional[Page], options: AjaxPostData) -> Dict:
		try:
				await page.goto(options.url, 
						waitUntil=options.wait_util, timeout=options.timeout*1000)
		except PPTimeoutError:
				#await page.evaluate('() => window.stop()')
				await page._client.send("Page.stopLoading")
		finally:
				page.remove_all_listeners("request")

class HtmlPostData(BaseModel):
    url: str
    timeout: float = 30
    wait_util: str = "domcontentloaded"
    wait: float = 0   
    js_name: str = "" 
    filters: List[str] = [] 
    images: bool = 0  
    forbidden_content_types: List[str] = ["image", "media"]
    cache: bool = 1 
    cookie: bool = 0 
    text: bool = 1 
		headers: bool = 1

class PngPostData(HtmlPostData):
    render_all: int = 0
    text: bool = 0
    images: bool = 1
    forbidden_content_types: List[str] = []
    wait_util: str = "networkidle2"

import sys
import asyncio
import aiohttp

if sys.platform == 'win32':
    asyncio.set_event_loop_policy(asyncio.WindowsSelectorEventLoopPolicy())

async def get_sign(session, delay):
    url = f"http://www.httpbin.org/delay/{delay}"
    api = f'http://127.0.0.1:8080/render.html?url={url}'
    async with session.get(api) as resp:
        data = await resp.json()
        print(url, data.get("status"))
        return data

async def main():
    headers = {
        "Content-Type": "application/json",
        'user-agent': 'Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/86.0.4240.198 Safari/537.36',
    }
    loop = asyncio.get_event_loop()
    t = loop.time()
    async with aiohttp.ClientSession(headers=headers) as session:
        tasks = [asyncio.create_task(get_sign(session, i)) for i in range(1, 5)]
        await asyncio.gather(*tasks)
    print("耗時(shí): ", loop.time()-t)

        
if __name__ == "__main__":
    asyncio.run(main())

import json
import sys
import asyncio
import aiohttp

if sys.platform == 'win32':
    asyncio.set_event_loop_policy(asyncio.WindowsSelectorEventLoopPolicy())

async def get_sign(session, url):
    api = f'http://127.0.0.1:8080/render.json'
    data = {
        "url": url,
        "xhr": "/api/", # 攔截接口包含/api/的響應(yīng)并返回
        "cache": 0,
        "filters": [".png", ".jpg"]
    }
    async with session.post(api, data=json.dumps(data)) as resp:
        data = await resp.json()
        print(url, data)
        return data

async def main():
    urls = ["https://spa1.scrape.center/"]
    headers = {
        "Content-Type": "application/json",
        'user-agent': 'Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/86.0.4240.198 Safari/537.36',
    }
    loop = asyncio.get_event_loop()
    t = loop.time()
    async with aiohttp.ClientSession(headers=headers) as session:
        tasks = [asyncio.create_task(get_sign(session, url)) for url in urls]
        await asyncio.gather(*tasks)
    print(loop.time()-t)

        
if __name__ == "__main__":
    asyncio.run(main())

import json
import sys
import asyncio
import base64
import aiohttp

if sys.platform == 'win32':
    asyncio.set_event_loop_policy(asyncio.WindowsSelectorEventLoopPolicy())

async def get_sign(session, url, name):
    api = f'http://127.0.0.1:8080/render.png'
    data = {
        "url": url,
        #"render_all": 1,
        "images": 1,
        "cache": 1,
        "wait": 1 
    }
    async with session.post(api, data=json.dumps(data)) as resp:
        data = await resp.json()
        if data.get('image'):
            image_bytes = base64.b64decode(data["image"])
            with open(name, 'wb') as f:
                f.write(image_bytes)
            print(url, name, len(image_bytes))
        return data

async def main():
    urls = [
        "https://www.baidu.com/s?ie=utf-8&f=8&rsv_bp=1&tn=44004473_102_oem_dg&wd=%E5%9B%BE%E7%89%87&rn=50",
        "https://www.toutiao.com/article/7145668657396564518/",
        "https://new.qq.com/rain/a/NEW2022092100053400",
        "https://new.qq.com/rain/a/DSG2022092100053300"
    ]
    headers = {
        "Content-Type": "application/json",
        'user-agent': 'Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/86.0.4240.198 Safari/537.36',
    }
    loop = asyncio.get_event_loop()
    t = loop.time()
    async with aiohttp.ClientSession(headers=headers) as session:
        tasks = [asyncio.create_task(get_sign(session, url, f"{n}.png")) for n,url in enumerate(urls)]
        await asyncio.gather(*tasks)
    print(loop.time()-t)


if __name__ == "__main__":
    asyncio.run(main())

import json
import logging
from scrapy.exceptions import NotConfigured

logger = logging.getLogger(__name__)

class BrowserMiddleware(object):
    def __init__(self, browser_base_url: str):
        self.browser_base_url = browser_base_url
        self.logger = logger
        
    @classmethod
    def from_crawler(cls, crawler):
        s = crawler.settings
        browser_base_url = s.get('PYPPETEER_CLUSTER_URL')
        if not browser_base_url:
            raise NotConfigured
        o = cls(browser_base_url)
        return o
    
    def process_request(self, request, spider):
        if "browser_options" not in request.meta or request.method != "GET":
            return
        browser_options = request.meta["browser_options"]
        url = request.url
        browser_options["url"] = url
        uri = browser_options.get('browser_uri', "/render.html")
        browser_url = self.browser_base_url.rstrip('/') + '/' + uri.lstrip('/')
        new_request = request.replace(
            url=browser_url,
            method='POST',
            body=json.dumps(browser_options)
        )
        new_request.meta["ori_url"] = url
        return new_request

    def process_response(self, request, response, spider):
        if "browser_options" not in request.meta or "ori_url" not in request.meta:
            return response
        try:
            datas = json.loads(response.text)
        except json.decoder.JSONDecodeError:
            return response.replace(url=url, status=500)
        datas = self.deal_datas(datas)
        url = request.meta["ori_url"]
        new_response = response.replace(url=url, **datas)
        return new_response
    
    def deal_datas(self, datas: dict) -> dict:
        status = datas["status"]
        text: str = datas.get('text') or datas.get('content')
        headers = datas.get('headers')
        response = {
            "status": status,
            "headers": headers,
            "body": text.encode()
        }
        return response