carry_video_rewriting/carry_video/nrfx_carry_video.py

import html
import json
import os
import random
import re
import time
import uuid
import requests
from datetime import datetime
from urllib.parse import urlparse, parse_qs
from loguru import logger
from common import Oss, Feishu, AliyunLogger, Material
from common.download_video import DownLoad
from common.ffmpeg import FFmpeg
from common.google_ai_studio import GoogleAI
from common.gpt4o_mini_help import GPT4oMini
from common.redis import in_carry_video_data
from common.sql_help import sqlCollect
from common.tag_video import Tag
from common.tts_help import TTS
from data_channel.piaoquan import PQ


class NrfxCarryViode:

    def get_text_dy_video(self,url):
        max_retries = 3
        retry_count = 0
        while retry_count < max_retries:
            try:
                if "&vid=" in url:
                    parsed_url = urlparse(url)
                    params = parse_qs(parsed_url.query)
                    video_id = params.get('vid', [None])[0]
                elif "?modal_id=" in url:
                    parsed_url = urlparse(url)
                    params = parse_qs(parsed_url.query)
                    video_id = params.get('modal_id', [None])[0]
                else:
                    headers = {
                        'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,image/apng,*/*;'
                                  'q=0.8,application/signed-exchange;v=b3;q=0.7',
                        'Accept-Language': 'zh-CN,zh;q=0.9',
                        'Cache-Control': 'no-cache',
                        'Pragma': 'no-cache',
                        'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) '
                                      'Chrome/127.0.0.0 Safari/537.36',
                    }
                    response = requests.request(url=url, method='GET', headers=headers, allow_redirects=False, timeout=30)
                    location = response.headers.get('Location', None)
                    video_id = re.search(r'/video/(\d+)/?', location.split('?')[0] if location else url).group(1)
                url = "http://8.217.192.46:8889/crawler/dou_yin/detail"
                if not video_id or not video_id.strip():
                    return None, None, None
                payload = json.dumps({
                    "content_id": str(video_id)
                })
                headers = {
                    'Content-Type': 'application/json'
                }

                response = requests.request("POST", url, headers=headers, data=payload, timeout= 60)
                response = response.json()
                code = response["code"]
                if code == 0:
                    data = response["data"]["data"]
                    video_url = data["video_url_list"][0]["video_url"]
                    original_title = data["title"]
                    return video_url, original_title, video_id
                if code == 22002:
                    if '抖音内容已被删除或无法访问' in response['msg']:
                        return "作品不存在", None, None
            except Exception as e:
                retry_count += 1
                logger.error(f"[+] 抖音{url}获取视频链接失败，失败信息{e}")
                time.sleep(1)
        return None, None, None

    def get_text_ks_video(self,url):
        try:
            headers = {
                'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,image/apng,*/*;'
                          'q=0.8,application/signed-exchange;v=b3;q=0.7',
                'Accept-Language': 'zh-CN,zh;q=0.9',
                'Cache-Control': 'no-cache',
                'Pragma': 'no-cache',
                'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) '
                              'Chrome/127.0.0.0 Safari/537.36',
            }
            response = requests.request(url=url, method='GET', headers=headers, allow_redirects=False, timeout= 30)
            location = response.headers.get('Location', None)
            video_id = re.search(r'/(f|photo|short-video|long-video)/(.*)/?',
                                 location.split('?')[0] if location else url).group(2)
            url = "http://8.217.192.46:8889/crawler/kuai_shou/detail"
            if not video_id or not video_id.strip():
                return None, None, None
            payload = json.dumps({
                "content_id": str(video_id)
            })
            headers = {
                'Content-Type': 'application/json'
            }
            time.sleep(random.uniform(10, 30))
            response = requests.request("POST", url, headers=headers, data=payload, timeout= 30)
            response = response.json()
            code = response["code"]
            if code == 0:
                data = response["data"]["data"]
                content_type = data['content_type']
                if content_type == 'note':
                    return "note","note"
                video_url = data["video_url_list"][0]["video_url"]
                original_title = data["title"]
                return video_url, original_title, video_id
            elif code == 27006:
                if "作品不存在" in response['msg'] or "不存在" in response['msg'] or "私密作品" in response['msg']:
                    return "作品不存在", None, None
            time.sleep(3)
        except Exception as e:
            logger.error(f"[+] 快手{url}获取视频链接失败，失败信息{e}")
            return None, None,None


    def insert_pq(self, REDIS_NAME, data, oss_object_key, title, tags, tag_transport_channel, channel_mark, n_ids, type):
        logger.info(f"[+] {REDIS_NAME}的{data}开始写入票圈")
        if ',' in n_ids:
            n_id_list = n_ids.split(',')
        else:
            n_id_list = [n_ids]
        pq_list = []
        for n_id in n_id_list:
            code = PQ.insert_piaoquantv(oss_object_key, title, n_id)
            if not code:
                logger.error(f"[+] {REDIS_NAME}的{data}写入票圈后台失败")
                AliyunLogger.logging(data["name"], type, tag_transport_channel, data["video_url"],
                                     "改造失败,写入票圈后台失败", "3003", str(data))
                continue

            pq_list.append(code)
            logger.info(f"[+] {REDIS_NAME}的{data}写入票圈成功，返回视频id{code}")
            tag_status = Tag.video_tag(code, str(tags))
            if tag_status == 0:
                logger.info(f"[+] {REDIS_NAME}的{data}写入标签成功，后台视频ID为{code}")
            try:
                current_time = datetime.now()
                formatted_time = current_time.strftime("%Y-%m-%d %H:%M:%S")
                sqlCollect.insert_machine_making_data(data["name"], type, tag_transport_channel,
                                                      data["video_url"], data["video_url"], data["pq_ids"],
                                                      data["title_category"],
                                                      code,
                                                      formatted_time, data["title_category"], oss_object_key)
                pq_url = f'https://admin.piaoquantv.com/cms/post-detail/{code}/detail'  # 站内视频链接
                values = [
                    [
                        str(code),
                        str(n_id),
                        formatted_time,
                        channel_mark,
                        data["name"],
                        data["pq_ids"],
                        data["pq_label"],
                        data["activate_data"],
                        data["video_url"],
                        data["title_category"],
                        tag_transport_channel,
                        data["tag_transport_scene"],
                        data["tag_transport_keyword"],
                        data["tag"],
                        data["transform_rule"],
                        data["video_share"],
                        data["trailer_share"],
                        data["trailer_share_audio"],
                        data["video_clipping"],
                        data["video_clipping_time"],
                        data["title_transform"],
                        pq_url
                    ]
                ]
                Feishu.insert_columns("R4dLsce8Jhz9oCtDMr9ccpFHnbI", 'Um1nWA', "ROWS", 1, 2)
                time.sleep(0.5)
                Feishu.update_values("R4dLsce8Jhz9oCtDMr9ccpFHnbI", 'Um1nWA', "A2:Z2", values)
                logger.info(f"[+] {REDIS_NAME}的{data}写入飞书成功")
            except Exception as e:
                logger.error(f"[+] {REDIS_NAME}的{data}写入飞书失败{e}")
                pass
        AliyunLogger.logging(data["name"], "内容分析", tag_transport_channel, data["video_url"],
                             "改造成功", "1000", str(data), str(pq_list))
        return



    def main(self, data, file_path, GEMINI_API_KEY):
        REDIS_NAME = 'task:carry_redis_nrfx'
        try:
            if data["transform_rule"] == '否':
                return
            url = data['video_url']
            if "&vid=" in url or "?modal_id=" in url:
                host = urlparse(url).netloc
            else:
                msg = html.unescape(url).split('?')[0]
                pattern = re.search(r'https?://[^\s<>"\'\u4e00-\u9fff]+', msg)
                if not pattern:
                    in_carry_video_data(REDIS_NAME, json.dumps(data, ensure_ascii=False, indent=4))
                    return
                url = pattern.group()
                host = urlparse(url).netloc
            if host in ['v.douyin.com', 'www.douyin.com', 'www.iesdouyin.com']:
                tag_transport_channel = "抖音"
                logger.info(f"[+] {url}开始获取抖音视频链接")
                url, original_title, video_id =  self.get_text_dy_video(url=url)
            elif host in ['v.kuaishou.com', 'www.kuaishou.com', 'v.m.chenzhongtech.com', 'creater.eozatvmq.com']:
                tag_transport_channel = "快手"
                logger.info(f"[+] {url}开始获取快手视频链接")
                url, original_title, video_id = self.get_text_ks_video(url=url)
            else:
                AliyunLogger.logging(data["name"], "内容分析", "", data["video_url"], "扫描到一条视频",
                                     "2001", str(data))
                logger.error(f"[+] {url}该链接不是抖/快 不做处理")
                AliyunLogger.logging(data["name"], "内容分析","", data["video_url"],
                                     "不是抖/快不做处理", "1001", str(data))
                return
            if url == "作品不存在":
                return
        except Exception as e:
            logger.info(f"[+] 获取视频链接异常{e}")
            in_carry_video_data(REDIS_NAME, json.dumps(data, ensure_ascii=False, indent=4))
            return
        AliyunLogger.logging(data["name"],"内容分析", tag_transport_channel, data["video_url"], "扫描到一条视频", "2001", str(data))
        AliyunLogger.logging(data["name"], "内容分析",tag_transport_channel, data["video_url"],  "符合规则等待改造", "2004", str(data))
        if not url:
            in_carry_video_data(REDIS_NAME, json.dumps(data, ensure_ascii=False, indent=4))
            logger.info(f"[+] {url}没有获取到视频链接,等待重新处理")
            AliyunLogger.logging(data["name"], "内容分析",tag_transport_channel, data["video_url"],
                                 "没有获取到视频链接，等待重新处理", "1002", str(data))
            return
        if url == "note":
            logger.info(f"[+] {url}是图文不做处理")
            AliyunLogger.logging(data["name"], "内容分析", tag_transport_channel, data["video_url"],
                                 "是图文不做处理", "1002", str(data))
            return

        logger.info(f"[+] {url}开始下载视频")
        video_path = DownLoad.download_video(url, file_path, tag_transport_channel, video_id)
        if not os.path.exists(video_path) or os.path.getsize(video_path) == 0:
            in_carry_video_data(REDIS_NAME, json.dumps(data, ensure_ascii=False, indent=4))
            logger.error(f"[+] {url}下载失败")
            AliyunLogger.logging(data["name"],"内容分析", tag_transport_channel, data["video_url"],
                                 "视频下载失败等待重新处理", "3002", str(data))
            return
        logger.info(f"[+] {url}开始视频下载成功")
        logger.info(f"[+] {url}开始处理标题")
        if data["title_category"] == "AI标题" or data["trailer_share"] == "AI标题":
            title = GPT4oMini.get_ai_mini_title(
                original_title if data["title_category"] == "AI标题" else data["title_category"])
        else:
            title = original_title if data["title_category"] == "原标题" else data["title_category"]
        if tag_transport_channel == "抖音":
            if "复制打开抖音" in data['video_url']:
                channel_mark = "APP"
            else:
                channel_mark = "PC"
        else:
            if "https://www.kuaishou.com/f" in data['video_url']:
                channel_mark = "PC"
            else:
                channel_mark = "APP"
        if data["transform_rule"] == "仅改造" or data["transform_rule"] == "是":
            width, height = FFmpeg.get_w_h_size(video_path)
            if width < height:  # 判断是否需要修改为竖屏
                video_path = FFmpeg.update_video_h_w(video_path, file_path)
            logger.info(f"[+] {REDIS_NAME}的{data}视频更改分辨率处理")
            video_path = FFmpeg.video_640(video_path, file_path)
            if not os.path.exists(video_path) or os.path.getsize(video_path) == 0:
                in_carry_video_data(REDIS_NAME, json.dumps(data, ensure_ascii=False, indent=4))
                logger.error(f"[+] {REDIS_NAME}的{data}视频更改分辨率失败")
                AliyunLogger.logging(data["name"], "内容分析", tag_transport_channel, data["video_url"],
                                     "改造失败,片尾拼接失败", "3001", str(data))
                return
            logger.info(f"[+] {REDIS_NAME}的{data}视频更改分辨率处理成功")
            if data["video_clipping"]:  # 判断是否需要裁剪
                video_path = FFmpeg.video_crop(video_path, file_path)
            if data["video_clipping_time"]:  # 判断是否需要指定视频时长
                video_path = FFmpeg.video_ggduration(video_path, file_path, data["video_clipping_time"])
            logger.info(f"[+] 内容分析-开始获取视频口播内容")
            video_text = GoogleAI.run(GEMINI_API_KEY, video_path)
            if not video_text:
                AliyunLogger.logging(data["name"], "内容分析", "", data["video_url"],
                                     "内容分析,获取口播文案失败", "3003", str(data))
                return
            logger.info(f"[+] 内容分析-开始获取AI片尾")
            pw_srt_text = GPT4oMini.get_content_understanding_pw(video_text)
            voice = data['trailer_share_audio']
            if voice:
                if ',' in voice:
                    voices = voice.split(',')
                else:
                    voices = [voice]
                voice = random.choice(voices)
            else:
                voice = "zhifeng_emo"
            pw_url = TTS.get_pw_zm(pw_srt_text, voice)
            if not pw_url:
                logger.error(f"[+] 内容分析-片尾获取失败")
                data["transform_rule"] = "仅改造"
                AliyunLogger.logging(data["name"], "内容分析", "", data["video_url"],
                                     "内容分析,片尾获取失败", "3003", str(data))

                return
            logger.info(f"[+] 内容分析-片尾获取成功")
            pw_srt = TTS.getSrt(pw_url)
            if not pw_srt:
                AliyunLogger.logging(data["name"], "内容分析", "", data["video_url"],
                                     "内容分析,片尾音频下载失败", "3003", str(data))
                return
            pw_mp3_path = TTS.download_mp3(pw_url, file_path)
            if not pw_mp3_path:
                AliyunLogger.logging(data["name"], "内容分析", "", data["video_url"],
                                     "内容分析,片尾音频下载失败", "3003", str(data))
                return
            logger.info(f"[+] 内容分析-片尾音频下载成功")
            logger.info(f"[+] 内容分析-片尾获取最后一帧成功")
            jpg_path = FFmpeg.video_png(video_path, file_path)  # 生成视频最后一帧jpg
            pw_path = FFmpeg.nrfx_pw_video(jpg_path, file_path, pw_mp3_path, pw_srt)  # 生成片尾视频
            if not os.path.exists(pw_path) or os.path.getsize(pw_path) == 0:
                logger.error(f"[+] 内容分析-片尾拼接失败")
                AliyunLogger.logging(data["name"], "内容分析", "", data["video_url"],
                                     "内容分析,片尾拼接失败", "3003", str(data))
                return
            logger.info(f"[+] 内容分析-合并开始拼接")
            video_path = FFmpeg.h_b_video(video_path, pw_path, file_path)
            single_video_path = FFmpeg.single_video(video_path, file_path, data["video_share"])
            if not os.path.exists(single_video_path) or os.path.getsize(single_video_path) == 0:
                data["transform_rule"] = "仅改造"
                in_carry_video_data(REDIS_NAME, json.dumps(data, ensure_ascii=False, indent=4))
                logger.error(f"[+] 内容分析-添加片中字幕失败")
                AliyunLogger.logging(data["name"], "内容分析", tag_transport_channel, data["video_url"],
                                     "内容分析,添加片中字幕失败", "3003", str(data))
                return
            logger.info(f"[+] 内容分析-添加片中字幕成功")
            logger.info(f"[+] 内容分析-开始发送oss")
            oss_object_key = Oss.stitching_sync_upload_oss(single_video_path, str(uuid.uuid4()))  # 视频发送OSS
            status = oss_object_key.get("status")
            if status != 200:
                logger.error(f"[+] 内容分析-发送oss失败")
                AliyunLogger.logging(data["name"], "内容分析", tag_transport_channel, data["video_url"],
                                     "内容分析,发送oss失败", "3003", str(data))
                return
            logger.info(f"[+] 内容分析-发送oss成功")
            oss_object_key = oss_object_key.get("oss_object_key")
            tags = 'lev-供给,rol-机器,#str-搬运改造内容理解引导语实验_60'
            self.insert_pq(REDIS_NAME, data, oss_object_key, title, tags, tag_transport_channel, channel_mark, "50322062", "内容分析")
            return


        return