video_rewriting/video_rewriting/video_prep.py

import configparser
import os
import random
import re
import sys
import threading
import time

from datetime import datetime
import concurrent.futures

from common.tts_help import TTS

sys.path.append(os.getcwd())

from common import Material, Feishu, Common, Oss
from common.ffmpeg import FFmpeg
from common.gpt4o_help import GPT4o
from data_channel.douyin import DY
from data_channel.kuaishou import KS
from data_channel.kuaishouchuangzuozhe import KsFeedVideo
from data_channel.piaoquan import PQ
from common.sql_help import sqlCollect
from data_channel.shipinhao import SPH

config = configparser.ConfigParser()
config.read('./config.ini')


class getVideo:

    """
    根据标示+任务标示创建目录
    """
    @classmethod
    def create_folders(cls, mark, task_mark):
        video_path_url = config['PATHS']['VIDEO_PATH'] + mark + "/" + task_mark + "/"
        if not os.path.exists(video_path_url):
            os.makedirs(video_path_url)
        return video_path_url

    """
    随机生成ID
    """
    @classmethod
    def random_id(cls):
        now = datetime.now()
        rand_num = random.randint(10000, 99999)
        oss_id = "{}{}".format(now.strftime("%Y%m%d%H%M%S"), rand_num)
        return oss_id


    """
    删除文件
    """
    @classmethod
    def remove_files(cls, video_path_url):
        if os.path.exists(video_path_url) and os.path.isdir(video_path_url):
            for root, dirs, files in os.walk(video_path_url):
                for file in files:
                    file_path = os.path.join(root, file)
                    os.remove(file_path)
                for dir in dirs:
                    dir_path = os.path.join(root, dir)
                    os.rmdir(dir_path)

    """
    飞书数据处理
    """
    @classmethod
    def video_task(cls, data):
        mark = data["mark"]
        name = data["name"]
        feishu_id = data["feishu_id"]
        feishu_sheet = data["feishu_sheet"]
        cookie_sheet = data["cookie_sheet"]
        pz_sheet = '500Oe0'
        pw_sheet = 'DgX7vC'
        task_data = Material.get_task_data(feishu_id, feishu_sheet)
        # task_data = task_data[::-1]
        if len(task_data) == 0:
            Feishu.bot(mark, '机器自动改造消息通知', f'今日任务为空，请关注', name)
            return mark
        lock = threading.Lock()
        def process_task(task):
            task_mark = task["task_mark"]  # 任务标示
            channel_id = str(task["channel_id"])
            channel_urls = str(task["channel_url"])
            piaoquan_id = str(task["piaoquan_id"])
            number = task["number"]  # 指定条数
            title = task["title"]
            video_share = task["video_share"]
            video_ending = task["video_ending"]
            crop_total = task["crop_total"]
            gg_duration_total = task["gg_duration_total"]
            video_path_url = cls.create_folders(mark, str(task_mark))  # 创建目录
            zm = Material.get_pzsrt_data("summary", pz_sheet, video_share)  # 获取srt
            if zm == '':
                Feishu.bot(mark, '机器自动改造消息通知', f'{task_mark}任务下片中标示填写错误，请关注！！！！', name)
            if ',' in channel_urls:
                channel_url = channel_urls.split(',')
            else:
                channel_url = [channel_urls]
            for url in channel_url:
                Common.logger("log").info(f"{name}的{task_mark}下的用户:{channel_url}开始获取视频")
                if channel_id == "抖音":
                    data_list = DY.get_dy_url(task_mark, url, number, mark, feishu_id, cookie_sheet, channel_id, name)
                elif channel_id == "票圈":
                    data_list = PQ.get_pq_url(task_mark, url, number, mark)
                elif channel_id == "视频号":
                    data_list = SPH.get_sph_url(task_mark, url, number, mark)
                elif channel_id == "快手":
                    data_list = KS.get_ks_url(task_mark, url, number, mark, feishu_id, cookie_sheet, channel_id, name)
                elif channel_id == "快手创作者版":
                    data_list = KsFeedVideo.get_data(number)
                    if len(data_list) == 0:
                        Common.logger("log").info(f"{name}的{task_mark}下的视频ID{url} 已经改造过了")
                        Feishu.bot("快手创作者版", '机器自动改造消息通知', f'本轮没有获取到改造的视频链接', "王雪珂")
                        cls.remove_files(video_path_url)
                        continue
                if len(data_list) == 0:
                    Common.logger("log").info(f"{name}的{task_mark}下的视频ID{url} 已经改造过了")
                    Feishu.bot(mark, '机器自动改造消息通知', f'{task_mark}任务下的用户ID{url},没有已经改造的视频了', name)
                    cls.remove_files(video_path_url)
                    continue
                Common.logger("log").info(f"{name}的{task_mark}下的ID{url} 获取视频完成，共{len(data_list)}条")

                try:
                    for video in data_list:
                        new_title = video['old_title'].strip().replace("\n", "") \
                            .replace("/", "").replace("\\", "").replace("\r", "") \
                            .replace(":", "").replace("*", "").replace("？", "") \
                            .replace("?", "").replace('"', "").replace("<", "") \
                            .replace(">", "").replace("|", "").replace(" ", "") \
                            .replace("&NBSP", "").replace(".", "。").replace(" ", "") \
                            .replace("'", "").replace("#", "").replace("Merge", "")
                        if title == "原标题":
                            if new_title == "" or len(new_title) == 0 or new_title == None:
                                new_title = '⭕分享给大家一个视频！值得细品！'
                        elif title == "AI标题":
                            if new_title == "" or len(new_title) == 0 or new_title == None:
                                new_title = '⭕分享给大家一个视频！值得细品！'
                            else:
                                new_title = GPT4o.get_ai_title(new_title)
                        else:
                            if '/' in title:
                                titles = title.split('/')
                            else:
                                titles = [title]
                            new_title = random.choice(titles)
                        v_id = video["video_id"]
                        cover = video["cover"]
                        video_url = video["video_url"]
                        rule = video['rule']
                        old_title = video['old_title']
                        if old_title == "" or len(old_title) == 0 or old_title == None:
                            old_title = '⭕分享给大家一个视频！值得细品！'
                        time.sleep(1)
                        pw_random_id = cls.random_id()
                        if channel_id == "票圈" or channel_id == "快手创作者版":
                            new_video_path = PQ.download_video(video_url, video_path_url, v_id)  # 下载视频地址
                        else:
                            new_video_path = Oss.download_video_oss(video_url, video_path_url, v_id)  # 下载视频地址
                            Common.logger("log").info(f"{name}的{task_mark}下的视频{url},{new_video_path}视频下载成功")
                        if not os.path.isfile(new_video_path):
                            Common.logger("log").info(f"{name}的{task_mark}下的视频{url},{new_video_path}视频下载失败")
                            cls.remove_files(video_path_url)
                            continue
                        Common.logger("log").info(f"{name}的{task_mark}下的视频{url},{new_video_path}视频下载成功")
                        if crop_total and crop_total != 'None':  # 判断是否需要裁剪
                            new_video_path = FFmpeg.video_crop(new_video_path, video_path_url, pw_random_id)
                        if gg_duration_total and gg_duration_total != 'None':  # 判断是否需要指定视频时长
                            new_video_path = FFmpeg.video_ggduration(new_video_path, video_path_url, pw_random_id, gg_duration_total)
                        width, height = FFmpeg.get_w_h_size(new_video_path)
                        if width < height:
                            new_video_path = FFmpeg.update_video_h_w(new_video_path, video_path_url, pw_random_id)
                            time.sleep(1)
                            if os.path.isfile(new_video_path):
                                if new_title != "" or len(new_title) != 0 or new_title != None:
                                    new_title_re = re.sub(r'[^\w\s\u4e00-\u9fff，。！？]', '', new_title)
                                    if len(new_title_re) > 12:
                                        new_text = '\n'.join(
                                            [new_title_re[i:i + 12] for i in range(0, len(new_title_re), 12)])
                                    new_video_path = FFmpeg.add_video_zm(new_video_path, video_path_url, pw_random_id, new_text)
                        if video_ending and video_ending != 'None':
                            if video_ending == "AI片尾引导":
                                pw_srt_text = GPT4o.get_ai_pw(old_title)
                                if pw_srt_text:
                                    pw_url = TTS.get_pw_zm(pw_srt_text)
                                    Common.logger("log").info(f"{name}的{task_mark}下的视频{url},获取AI片尾音频成功")
                                    if pw_url:
                                        pw_mp3_path = TTS.download_mp3(pw_url, video_path_url, pw_random_id)
                                        pw_url_sec = FFmpeg.get_video_duration(pw_mp3_path)  # 获取片尾秒数
                                        Common.logger("log").info(f"{name}的{task_mark}下的视频{url},获取AI片尾秒数成功{pw_url_sec}")
                                        pw_srt = TTS.getSrt(pw_url)
                                        Common.logger("log").info(f"{name}的{task_mark}下的视频{url},获取AI片尾srt成功")
                                    else:
                                        Feishu.bot(mark, 'TTS获取失败提示', f'无法获取到片尾音频，及时更换token', "张勇")
                                        Common.logger("log").info(f"{name}的{task_mark}下的视频{url},获取AI片尾失败")
                                        continue
                                else:
                                    Common.logger("log").info(f"{name}的{task_mark}下的视频{url},获取AI片尾失败")
                                    continue
                            else:
                                if ',' in video_ending:
                                    video_ending_list = video_ending.split(',')
                                else:
                                    video_ending_list = [video_ending]
                                ending = random.choice(video_ending_list)
                                pw_list = Material.get_pwsrt_data("summary", pw_sheet, ending)  # 获取srt
                                if pw_list:
                                    pw_id = pw_list["pw_id"]
                                    pw_srt = pw_list["pw_srt"]
                                    pw_url = PQ.get_pw_url(pw_id)
                                    pw_mp3_path = FFmpeg.get_video_mp3(pw_url, video_path_url, pw_random_id)
                                else:
                                    Feishu.bot(mark, '机器自动改造消息通知', f'{task_mark}任务下片尾标示错误，请关注！！！！', name)
                            for attempt in range(3):
                                jpg_path = FFmpeg.video_png(new_video_path, video_path_url, pw_random_id)  # 生成视频最后一帧jpg
                                if os.path.isfile(jpg_path):
                                    Common.logger("log").info(f"{name}的{task_mark}下的视频{url},生成视频最后一帧成功")
                                    break
                                time.sleep(1)
                            if not os.path.isfile(jpg_path):
                                Feishu.bot(mark, '机器自动改造消息通知', f'{task_mark}任务用户{url}下的视频{v_id},获取视频最后一帧失败，请关注', name)
                                cls.remove_files(video_path_url)
                                continue
                            for attempt in range(3):
                                Common.logger("log").info(f"{name}的{task_mark}下的视频{url},获取mp3成功")
                                pw_path = FFmpeg.pw_video(jpg_path, video_path_url, pw_mp3_path, pw_srt, pw_random_id, pw_mp3_path)  # 生成片尾视频
                                if os.path.isfile(pw_path):
                                    Common.logger("log").info(f"{task_mark}下的视频{url},生成片尾视频成功")
                                    break
                                time.sleep(1)
                            if not os.path.isfile(pw_path):
                                Feishu.bot(mark, '机器自动改造消息通知', f'{task_mark}任务用户{url}下的视频{v_id},生成片尾视频失败，请关注',
                                           name)
                                cls.remove_files(video_path_url)
                                continue
                            pw_video_list = [new_video_path, pw_path]
                            Common.logger("log").info(f"{task_mark}下的视频{url},视频与片尾开始拼接")
                            video_path = FFmpeg.concatenate_videos(pw_video_list, video_path_url)  # 视频与片尾拼接到一起
                            Common.logger("log").info(f"{name}的{task_mark}下的视频{url},视频与片尾拼接成功")
                            time.sleep(1)
                            if video_share and video_share != 'None':
                                new_video_path = FFmpeg.single_video(video_path, video_path_url, zm)
                            else:
                                new_video_path = video_path
                        else:
                            if video_share and video_share != 'None':
                                new_video_path = FFmpeg.single_video(new_video_path, video_path_url, zm)
                        time.sleep(1)
                        oss_id = cls.random_id()
                        Common.logger("log").info(f"{name}的{task_mark}下的视频{url},开始发送oss")
                        oss_object_key = Oss.stitching_sync_upload_oss(new_video_path, oss_id)  # 视频发送OSS
                        Common.logger("log").info(f"{name}的{task_mark}下的视频{url},发送oss成功{oss_object_key}")
                        status = oss_object_key.get("status")
                        if status == 200:
                            oss_object_key = oss_object_key.get("oss_object_key")
                            time.sleep(1)
                            jpg_path = PQ.download_video_jpg(cover, video_path_url, v_id)  # 下载视频封面
                            if os.path.isfile(jpg_path):
                                oss_jpg_key = Oss.stitching_fm_upload_oss(jpg_path, oss_id)  # 封面发送OSS
                                status = oss_jpg_key.get("status")
                                if status == 200:
                                    jpg = oss_jpg_key.get("oss_object_key")
                                else:
                                    jpg = None
                            else:
                                jpg = None
                            code = PQ.insert_piaoquantv(oss_object_key, new_title, jpg, piaoquan_id)
                            if code:
                                Common.logger("log").info(f"{name}的{task_mark}下的视频ID{v_id}发送成功")
                                sqlCollect.insert_task(task_mark, v_id, mark, channel_id)  # 插入数据库
                                current_time = datetime.now()
                                formatted_time = current_time.strftime("%Y-%m-%d %H:%M:%S")
                                sqlCollect.insert_machine_making_data(name, task_mark, channel_id, url, v_id, piaoquan_id, new_title, code, formatted_time, old_title)
                                if title == "原标题" or title == "AI标题":
                                    values = [[name, task_mark, channel_id, url, str(v_id), piaoquan_id, old_title, title,new_title, str(code), formatted_time, str(rule)]]
                                else:
                                    values = [[name, task_mark, channel_id, url, str(v_id), piaoquan_id, old_title, "",new_title, str(code), formatted_time, str(rule)]]
                                # 使用锁保护表格插入操作
                                with lock:
                                    if name == "王雪珂":
                                        sheet = "vfhHwj"
                                    elif name == "抖音品类账号-1":
                                        sheet = "61kvW7"
                                    elif name == "鲁涛":
                                        sheet = "FhewlS"
                                    elif name == "范军":
                                        sheet = "B6dCfS"
                                    elif name == "余海涛":
                                        sheet = "mfBrNT"
                                    elif name == "罗情":
                                        sheet = "2J3PwN"
                                    elif name == "王玉婷":
                                        sheet = "bBHFwC"
                                    elif name == "刘诗雨":
                                        sheet = "fBdxIQ"
                                    elif name == "信欣":
                                        sheet = "lPe1eT"
                                    elif name == "快手创作者版品类推荐流":
                                        sheet = "k7l7nQ"
                                    elif name == "抖音品类账号":
                                        sheet = "Bsg5UR"
                                    Feishu.insert_columns("ILb4sa0LahddRktnRipcu2vQnLb", sheet, "ROWS", 1, 2)
                                    time.sleep(0.5)
                                    # time.sleep(random.randint(1, 2))
                                    Feishu.update_values("ILb4sa0LahddRktnRipcu2vQnLb", sheet, "A2:Z2", values)
                            else:
                                Common.logger("warning").info(f"{name}的{task_mark}下的{url}视频{v_id} 视频发送票圈失败 {code}")
                            cls.remove_files(video_path_url)
                        else:
                            cls.remove_files(video_path_url)
                            Common.logger("warning").info(f"{name}的{task_mark}下的{url}视频{v_id} 视频发送OSS失败 ")
                    Feishu.bot(mark, '机器自动改造消息通知', f'{task_mark}任务改造完成，请关注', name)
                except Exception as e:
                    cls.remove_files(video_path_url)
                    Common.logger("warning").warning(f"{name}的{task_mark}任务处理失败:{e}\n")

        batch_size = 1
        with concurrent.futures.ThreadPoolExecutor(max_workers=batch_size) as executor:
            index = 0
            while index < len(task_data):
                # 计算剩余的任务数量
                remaining_tasks = len(task_data) - index
                # 当前批次大小为剩余任务数量和批次大小中较小的一个
                current_batch_size = min(batch_size, remaining_tasks)
                # 获取当前批次的任务
                current_batch = task_data[index:index + batch_size]
                futures = {executor.submit(process_task, task): task for task in current_batch}
                for future in concurrent.futures.as_completed(futures):
                    task = futures[future]
                    try:
                        future.result()
                        print(f"Task {task['task_mark']} 完成")
                    except Exception as exc:
                        Common.logger("warning").warning(f"{name}的{task['task_mark']}任务处理失败:{exc}\n")
                        print(f"Task {task['task_mark']} 异常信息: {exc}")
                # 移动到下一批任务
                index += current_batch_size

        Feishu.bot(mark, '机器自动改造消息通知', f'你的任务全部完成，请关注！！！！！', name)
        return mark