py-hutun-robot/xhs/rotate_ident.py

"""用于处理小红书旋转验证码"""

import logging
import os
import time
import base64
from io import BytesIO
import requests
from PIL import Image
from playwright.sync_api import Playwright, Page
from util import playwright_util

root_dir = os.path.dirname(os.path.abspath(__file__))


def pil_base64(img, coding='utf-8'):
    """
    PIL图片保存为base64编码
    """
    img_format = img.format
    if img_format is None:
        img_format = 'JPEG'

    format_str = 'JPEG'
    if 'png' == img_format.lower():
        format_str = 'PNG'
    if 'gif' == img_format.lower():
        format_str = 'gif'

    if img.mode == "P":
        img = img.convert('RGB')
    if img.mode == "RGBA":
        format_str = 'PNG'
        img_format = 'PNG'

    output_buffer = BytesIO()
    # img.save(output_buffer, format=format_str)
    img.save(output_buffer, quality=100, format=format_str)
    byte_data = output_buffer.getvalue()
    base64_str = 'data:image/' + img_format.lower() + ';base64,' + base64.b64encode(byte_data).decode(coding)
    return base64_str


def invoke_ident_api(img):
    """
    验证码识别接口
    """
    url = "http://www.detayun.cn/openapi/verify_code_identify/"
    data = {
        # 用户的key
        "key": "2XbUYAP0jeiaiBV8uAvg",
        # 验证码类型
        "verify_idf_id": "24",
        # 样例图片
        "img_base64": pil_base64(img),
        "img_byte": None,
        # 中文点选，空间语义类型验证码的文本描述（这里缺省为空字符串）
        "words": ""
    }
    header = {"Content-Type": "application/json"}
    # 发送请求调用接口
    response = requests.post(url=url, json=data, headers=header, timeout=30)
    logging.info('verify_code_identify response: %s', response.text)
    return response.json()



class RotateIdent:
    """
    旋转验证码处理类
    """

    def __init__(self, page: Page):
        self.page = page
        self.img_path = None

    def need_ident(self) -> bool:
        """
        是否需需要识别
        """
        return playwright_util.is_element_present(self.page, '//div[@id="red-captcha-rotate"]/img')

    def handle_rotate(self):
        """
        处理旋转验证码的核心方法
        """
        try_count = 0
        while self.need_ident() and try_count < 4:
            try_count += 1
            logging.info('开始处理旋转验证码，第 %s 次', try_count)
            tag2 = self.page.query_selector('//div[@class="red-captcha-slider"]')
            img = self.download_img()
            response = invoke_ident_api(img)
            if response['code'] != 200:
                logging.error(response['msg'])
            else:
                angle = response['data']['angle']
                # 使用鼠标操作进行点击并保持
                bbox = tag2.bounding_box()
                x_center = bbox['x'] + bbox['width'] / 2
                y_center = bbox['y'] + bbox['height'] / 2
                self.page.mouse.move(x_center, y_center)
                self.page.mouse.down()
                # 计算实际滑动距离 = 像素距离 + 前面空白距离
                move_x = angle * 0.79
                # 滑动滑块，且坐标多增加一个像素，添加误差
                self.page.mouse.move(x_center + move_x + 1, y_center + 5, steps=10)
                # 等待一段时间
                time.sleep(1)
                # 释放鼠标按钮
                self.page.mouse.up()
                time.sleep(5)
        if self.img_path is not None and os.path.exists(self.img_path):
            os.remove(self.img_path)

    def download_img(self):
        """
        下载待处理的验证码并保存
        """
        # 找到【旋转图像】元素
        tag1 = self.page.query_selector('//div[@id="red-captcha-rotate"]/img')
        # 获取图像链接
        img_url = tag1.get_attribute('src')
        logging.info('ident url: %s', img_url)
        header = {
            "Host": "picasso-static.xiaohongshu.com",
            "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:72.0) Gecko/20100101 Firefox/72.0",
            "Accept": "text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8",
            "Accept-Language": "zh-CN,zh;q=0.8,zh-TW;q=0.7,zh-HK;q=0.5,en-US;q=0.3,en;q=0.2",
            "Accept-Encoding": "gzip, deflate, br",
            "Connection": "keep-alive",
            "Cookie": "xsecappid=login; a1=1896916369fehn0yq7nomanvre3fghfkj0zubt7zx50000120287; webId=75af27905db67b6fcb29a4899d200062; web_session=030037a385d8a837e5e590cace234a6e266fd5; gid=yYjKjyK484VKyYjKjyKqK89WjidxI8vAWIl6uuC0IhFdq728ikxiTD888yJ8JYW84DySKW0Y; webBuild=2.17.8; websectiga=634d3ad75ffb42a2ade2c5e1705a73c845837578aeb31ba0e442d75c648da36a; sec_poison_id=41187a04-9f82-4fbc-8b98-d530606b7696",
            "Upgrade-Insecure-Requests": "1",
            "If-Modified-Since": "Thu, 06 Jul 2023 11:42:07 GMT",
            "If-None-Match": '"7e53c313a9f321775e8f5e190de21081"',
            "TE": "Trailers",
        }
        # 下载图片
        return self.do_download_img(img_url)

    def do_download_img(self, img_url):
        # 下载图片
        response = requests.get(url=img_url, timeout=20)
        img = Image.open(BytesIO(response.content))
        img_folder = os.path.join(root_dir, 'train_img')
        # 如果目标文件夹不存在，则创建
        if not os.path.exists(img_folder):
            os.makedirs(img_folder)
        # 构建图片路径并保存图片
        self.img_path = os.path.join(img_folder, f'{int(time.time() * 1000)}.jpg')
        img.convert('RGB').save(self.img_path)
        return img