rov-offline/alg_growth_common.py

#! /usr/bin/env python
# -*- coding: utf-8 -*-
# vim:fenc=utf-8
#
# Copyright © 2024 StrayWarrior <i@straywarrior.com>

"""
Common functions for growth jobs
"""
import pandas as pd

from my_utils import request_post

UNSAFE_VIDEO_IDS = [14403867, 13696461, 13671819, 13587868, 13680796, 14050873,
                    26348326, 28623786]
AUDIT_APPROVED_GH_IDS = ['gh_b63b9dde3f4b','gh_330ef0db846d','gh_330ef0db846d','gh_e2318164f869','gh_620af8e24fb9',
'gh_620af8e24fb9','gh_133c36b99b14','gh_133c36b99b14','gh_5ac72e2b9130','gh_ef8ade0fad92',
'gh_1e03b6de22bf','gh_5538fe297e59','gh_8c6fffcbaac1','gh_8c6fffcbaac1','gh_d0e830b7547e',
'gh_fb234f4e32a5','gh_84c5d01a61e7','gh_87c4b8ae885e','gh_29d8a63d5e5e','gh_b144210318e5',
'gh_b144210318e5','gh_1f9bf4cfa788','gh_4f47d12bbe04','gh_8c6af276df98','gh_1f16bc6ac60d',
'gh_4920bc4c5720','gh_5177a8c57917','gh_5177a8c57917','gh_5e3e6cd5e35c','gh_5e3e6cd5e35c',
'gh_d2c72bcc05c9','gh_d2c72bcc05c9','gh_5f2400da935c','gh_5f2400da935c','gh_669555ebea28',
'gh_28ce883486c3','gh_28ce883486c3','gh_7057ef30222b','gh_7057ef30222b','gh_b0048adc0b46',
'gh_6e61a2d5db85','gh_01cd19465b39','gh_01cd19465b39','gh_126c99b39cea','gh_4a1174e36ceb',
'gh_f81c27eb8c48','gh_f81c27eb8c48','gh_3170dc15e246','gh_1ccfb5620605','gh_315be76a746d',
'gh_4f47d12bbe04','gh_4f47d12bbe04','gh_4f47d12bbe04']

def check_unsafe_video(df, force_replace=True):
    unsafe_video_condition = ','.join([str(x) for x in UNSAFE_VIDEO_IDS])
    unsafe_rows = df.query(f'video_id in ({unsafe_video_condition})')
    if len(unsafe_rows) > 0:
        print(unsafe_rows)
        if not force_replace:
            raise Exception("video unsafe")
        df.loc[unsafe_rows.index, 'video_id'] = 20463342


def filter_unsafe_video(df):
    unsafe_video_condition = ','.join([str(x) for x in UNSAFE_VIDEO_IDS])
    df = df.query(f'video_id not in ({unsafe_video_condition})')
    return df




def filter_audit_failed_video(df):
    video_id_list = df['video_id'].tolist()
    chunk_size = 20
    result = [video_id_list[i:i + chunk_size] for i in range(0, len(video_id_list), chunk_size)]
    video_audit_failure = []
    for chunk in result:
        json_structure = {
            "videoIdList": chunk
        }
        result = request_post('https://longvideoapi.piaoquantv.com/longvideoapi/openapi/video/batchSelectVideoInfo',
                              json_structure)
        if result is None:
            continue
        if result['code'] != 0:
            continue
        for item in result['data']:
            if item['auditStatus'] != 5 or item['appAuditStatus'] != 5:
                video_audit_failure.append(item['id'])
    condition = (df['video_id'].isin(video_audit_failure)) & (df['gh_id'].isin(AUDIT_APPROVED_GH_IDS))
    filtered_df = df[~condition]
    return filtered_df


if __name__ == '__main__':
    # 定义包含 ID 的字符串
    id_str = '20463342,12794884,13788955,13586800,4780859,33332362,19175397,4555247,14403867,12117356,14050873,14142458,17638023,14945734,13680796,13042177,10587660,14552795,12418493,12700411,13671819,13825547,12166346,13587868,19096953,14095344,13817005,1275943,13437896,12492103'

    # 将字符串按逗号分割成列表
    id_list = id_str.split(',')

    # 将列表中的元素转换为整数（如果需要）
    id_list = [int(id) for id in id_list]

    # 创建 DataFrame 并将 ID 列表放入 'video_id' 列
    df = pd.DataFrame({'video_id': id_list})
    video_audit_failure = filter_audit_failed_video(df)
    print(df)
    print(video_audit_failure)
    filtered_df = df[~df['video_id'].isin(video_audit_failure)]
    print(filtered_df)