411 lines
23 KiB
Python
411 lines
23 KiB
Python
import pandas as pd
|
||
import os
|
||
import sqlite3
|
||
from mycode.base import BASE_DIR
|
||
import re
|
||
from openpyxl import load_workbook
|
||
from urllib.parse import urlparse
|
||
from datetime import datetime
|
||
import numpy as np
|
||
|
||
wechat_dir = os.path.join(BASE_DIR, 'article')
|
||
web_dir = os.path.join(BASE_DIR, 'web_dir')
|
||
output_dir = os.path.join(BASE_DIR, 'summary')
|
||
df_s = pd.read_excel(os.path.join(BASE_DIR, 'biao.xlsx'), sheet_name='筛查内容')
|
||
|
||
def fix_url_scheme(url, default_scheme='http'):
|
||
# 检查URL是否包含方案
|
||
if not url.startswith('http://') and not url.startswith('https://'):
|
||
# 如果没有方案，添加默认方案
|
||
url = f'{default_scheme}://{url}'
|
||
return url
|
||
|
||
def trans_to_json():
|
||
json_str = df_s.to_json(orient='records', force_ascii=False)
|
||
with open('biao.json', 'w', encoding='utf-8') as f:
|
||
f.write(json_str)
|
||
|
||
def make_simple_csv_from_db(now: datetime):
|
||
# 只查找当前月份更新的公众号数据
|
||
now_month_str = now.strftime('%Y-%m-%d 00:00:00')
|
||
conn = sqlite3.connect(os.path.join(BASE_DIR, 'db_folder/test.db'))
|
||
query = f"""
|
||
SELECT
|
||
id,
|
||
g.nickname,
|
||
a.title,
|
||
a.content_url,
|
||
datetime( a.p_date, 'unixepoch', 'localtime' ) AS pub_date,
|
||
datetime( g.updated_at, 'unixepoch', 'localtime' ) AS g_updated_at
|
||
FROM
|
||
articles a
|
||
LEFT JOIN gzhs g ON g.biz = a.biz
|
||
WHERE
|
||
g_updated_at > '{now_month_str}'
|
||
"""
|
||
df = pd.read_sql_query(query, conn)
|
||
# 关闭数据库连接
|
||
conn.close()
|
||
# 将数据写入CSV文件
|
||
df.to_csv(os.path.join(wechat_dir, 'articles.csv'), index=False)
|
||
|
||
def float_to_int(value):
|
||
try:
|
||
return int(value)
|
||
except:
|
||
return value
|
||
|
||
def get_cbma_info_from_db_and_ana(year: str = '2023'):
|
||
# 全年统计数据
|
||
zybiz = "MzIzMDU4Njg3MA=="
|
||
df_fx = pd.DataFrame({"单位": [ "中国建材总院",
|
||
"瑞泰科技", "国检集团", "中材高新", "哈玻院", "中国新材院", "秦皇岛院", "西安墙材院", "咸阳陶瓷院", "钟表所", "总院北分", "中岩科技", "水泥新材院", "中建材科创院", "科建苑", "办公室（董事会办公室）", "党委组织部/人力资源部", "财务部", "科技部", "投资部", "企业管理部、安全环保部", "党群部/宣传统战部",
|
||
"党风办/巡察办、纪委综合室", "监督执纪室", "审计办公室"],
|
||
"公众号Biz": [zybiz, "MzU0MzgwMzg1NA==", "MzI1MjYzNDQ3NA==", "MzA5MDkzNDA0NQ==", "Mzg2MDg0NjkwNw==", "MzI3MTY5NTExNA==", "MzI1MzY1Njg5MQ==", "MzIxOTQwNjE2MQ==",
|
||
"Mzg3OTI0NTYzMA==", "MzA3NTU5NjM2MA==", "", "Mzg2NDgyMDM3OA==","","MzA5NTQ5MjY4Nw==", "", "", "", "", "", "", "", "", "", "", "", ],
|
||
# "供总院稿数": [], "供总院专稿数": [], "供总院组稿数": [], "供总院阅读10000及以上数": [], "供总院阅读5000及以上数": [], "供总院阅读1000及以上数": [],
|
||
# "1月发布数": [], "1月最高点击文章": [],
|
||
# "2月发布数": [], "2月最高点击文章": [],
|
||
# "3月发布数": [], "3月最高点击文章": [],
|
||
# "4月发布数": [], "4月最高点击文章": [],
|
||
# "5月发布数": [], "5月最高点击文章": [],
|
||
# "6月发布数": [], "6月最高点击文章": [],
|
||
# "7月发布数": [], "7月最高点击文章": [],
|
||
# "8月发布数": [], "8月最高点击文章": [],
|
||
# "9月发布数": [], "9月最高点击文章": [],
|
||
# "10月发布数": [], "10月最高点击文章": [],
|
||
# "11月发布数": [], "11月最高点击文章": [],
|
||
# "12月发布数": [], "12月最高点击文章": [],
|
||
# "半年发布数": [], "半年最高点击文章": [],
|
||
# "全年发布数": [], "全年最高点击文章": []
|
||
})
|
||
# 查询所有指定公众号的文章并按年/月排序
|
||
conn = sqlite3.connect(os.path.join(BASE_DIR, 'db_folder/test.db'))
|
||
query_gzhs = f'''
|
||
SELECT
|
||
id,
|
||
g.biz as gbiz,
|
||
strftime('%Y', datetime(a.p_date, 'unixepoch', 'localtime')) as pub_year,
|
||
strftime('%m', datetime(a.p_date, 'unixepoch', 'localtime')) as pub_month,
|
||
strftime('%d', datetime(a.p_date, 'unixepoch', 'localtime')) as pub_day,
|
||
g.nickname,
|
||
a.title,
|
||
a.content_url,
|
||
a.read_num
|
||
FROM
|
||
articles a
|
||
LEFT JOIN
|
||
gzhs g ON g.biz = a.biz
|
||
WHERE
|
||
pub_year = '{year}'
|
||
AND
|
||
g.biz in ({', '.join([f"'{biz}'" for biz in df_fx["公众号Biz"].tolist()])})
|
||
ORDER BY
|
||
pub_year, pub_month, pub_day;
|
||
'''
|
||
df = pd.read_sql_query(query_gzhs, conn)
|
||
conn.close
|
||
|
||
# 追加总院数据来源
|
||
|
||
for ind, row in df.iterrows():
|
||
if row['gbiz'] == zybiz:
|
||
full_path = os.path.join(wechat_dir, row['nickname'], row['id'] + '.md')
|
||
try:
|
||
with open(full_path, encoding='utf-8') as f:
|
||
content = f.read()
|
||
# 从content中获取来源
|
||
a_match = re.findall('来源丨(.*?)\n', content)
|
||
if a_match:
|
||
# a = a_match[0].replace('\xa0', '、').replace(' ', '、')
|
||
# a = re.sub(r'、+', '、', a)
|
||
a = re.sub(r'[\xa0\s]+', '、', a_match[0].strip())
|
||
df.at[ind, 'source'] = a
|
||
except FileNotFoundError:
|
||
print(full_path + '---不存在')
|
||
df['source'] = df['source'].fillna('')
|
||
# df_fx['供总院稿数'] = df_fx['单位'].apply(lambda unit: (df['source'].str.contains(unit)).sum())
|
||
# df_fx['供总院专稿数'] = df_fx['单位'].apply(lambda unit: (df['source'] == unit).sum())
|
||
# df_fx['供总院组稿数'] = df_fx['单位'].apply(lambda unit: ((df['source'].str.contains(unit)&(df['source']!=unit))).sum())
|
||
# df_fx['供总院阅读10000及以上数'] = df_fx['单位'].apply(lambda unit: ((df['read_num']>=10000)&(df['source'].str.contains(unit))).sum())
|
||
# df_fx['供总院阅读5000及以上数'] = df_fx['单位'].apply(lambda unit: ((df['read_num']>=5000)&(df['read_num']<10000)&(df['source'].str.contains(unit))).sum())
|
||
# df_fx['供总院阅读1000及以上数'] = df_fx['单位'].apply(lambda unit: ((df['read_num']>=1000)&(df['read_num']<5000)&(df['source'].str.contains(unit))).sum())
|
||
# df_fx['1月发布数'] = df_fx['公众号Biz'].apply(lambda unit: ((df['pub_year']==year)&(df['pub_month']=='01')&(df['gbiz']==unit)).sum() if unit else '')
|
||
|
||
# 更详细的分析
|
||
for ind, row in df_fx.iterrows():
|
||
dw = row['单位']
|
||
gbiz = row['公众号Biz']
|
||
|
||
# 全年对总院供给统计
|
||
# if '、' in dw: # 针对这种同一部门的
|
||
# cons = (df['gbiz']==zybiz)
|
||
# cons_dw_1 = pd.Series(False, index=df.index)
|
||
# for item in dw.split('、'):
|
||
# cons_dw_1 = (df['source'].str.contains(item))|cons_dw_1
|
||
# df_fx.at[ind, '供总院全年稿数'] = ((cons_dw_1)&(cons)).sum()
|
||
# else:
|
||
# df_fx.at[ind, '供总院全年稿数'] = ((df['source'].str.contains(dw))&(df['gbiz']==zybiz)).sum()
|
||
df_fx.at[ind, '供总院全年专稿数'] = ((df['source'] == dw)&(df['gbiz']==zybiz)).sum()
|
||
df_fx.at[ind, '供总院全年组稿数'] = ((df['source'].str.contains(dw)&(df['source']!=dw))&(df['gbiz']==zybiz)).sum()
|
||
df_fx.at[ind, '供总院全年阅读10000及以上数'] = ((df['read_num']>=10000)&(df['source'].str.contains(dw))&(df['gbiz']==zybiz)).sum()
|
||
df_fx.at[ind, '供总院全年阅读5000及以上数'] = ((df['read_num']>=5000)&(df['read_num']<10000)&(df['source'].str.contains(dw))&(df['gbiz']==zybiz)).sum()
|
||
df_fx.at[ind, '供总院全年阅读1000及以上数'] = ((df['read_num']>=1000)&(df['read_num']<5000)&(df['source'].str.contains(dw))&(df['gbiz']==zybiz)).sum()
|
||
|
||
for i in ['1月', '2月', '3月', '4月', '5月', '6月', '7月', '8月', '9月', '10月', '11月', '12月', '上半年', '下半年', '全年']:
|
||
if '月' in i:
|
||
i_str = i.replace('月', '').zfill(2)
|
||
cons_y_m = (df['pub_month']==str(i_str))
|
||
elif i == '上半年':
|
||
cons_y_m = (df['pub_month'] =='01')|(df['pub_month'] =='02')|(df['pub_month'] =='03')|(df['pub_month'] =='04')|(df['pub_month'] =='05')|(df['pub_month'] =='06')
|
||
elif i == '下半年':
|
||
cons_y_m = (df['pub_month'] =='07')|(df['pub_month'] =='08')|(df['pub_month'] =='09')|(df['pub_month'] =='10')|(df['pub_month'] =='11')|(df['pub_month'] =='12')
|
||
elif i == '全年':
|
||
cons_y_m = pd.Series(True, index=df.index)
|
||
|
||
if '、' in dw: # 针对这种同一部门的
|
||
cons_dw_1 = pd.Series(False, index=df.index)
|
||
for item in dw.split('、'):
|
||
cons_dw_1 = (df['source'].str.contains(item))|cons_dw_1
|
||
df_fx.at[ind, f'供总院{i}稿数'] = ((cons_dw_1)&(cons_y_m)&(df['gbiz']==zybiz)).sum()
|
||
else:
|
||
df_fx.at[ind, f'供总院{i}稿数'] = (df['source'].str.contains(dw)&(cons_y_m)&(df['gbiz']==zybiz)).sum()
|
||
df_fx[f'供总院{i}稿数'] = df_fx[f'供总院{i}稿数'].fillna(0)
|
||
df_fx[f'供总院{i}稿数'] = df_fx[f'供总院{i}稿数'].astype(int)
|
||
|
||
if gbiz:
|
||
# 进行查询
|
||
# 条件
|
||
cons = (cons_y_m)&(df['gbiz']==gbiz)
|
||
cons_sum = (cons).sum()
|
||
df_fx.at[ind, f'{i}发布数'] = cons_sum
|
||
df_fx[f'{i}发布数'] = df_fx[f'{i}发布数'].fillna(0)
|
||
df_fx[f'{i}发布数'] = df_fx[f'{i}发布数'].astype(int)
|
||
df_fx.at[ind, f'{i}最高点击文章'] = ''
|
||
if cons_sum:
|
||
max_read_row = df[cons].loc[df[cons]['read_num'].idxmax()]
|
||
max_read_row_list = [max_read_row['id'], max_read_row['title'], str(max_read_row['read_num']), f'{max_read_row["pub_year"]}-{max_read_row["pub_month"]}-{max_read_row["pub_day"]}', max_read_row['source']]
|
||
df_fx.at[ind, f'{i}最高点击文章'] = '***'.join(max_read_row_list)
|
||
|
||
# 矫正数据类型
|
||
df_fx = df_fx.applymap(float_to_int)
|
||
# 先输出原始统计数据
|
||
origin_path = os.path.join(BASE_DIR, f'summary/{year}_fx.xlsx')
|
||
df_fx.to_excel(origin_path, index=True)
|
||
|
||
# 追加到总院年分析表中
|
||
template_path = os.path.join(BASE_DIR, 'summary/template_cbma.xlsx')
|
||
workbook = load_workbook(template_path)
|
||
sheet = workbook['公众号更新数']
|
||
sheet.cell(row=1, column=1, value=f'关于{year}年度中国建材总院新媒体更新情况明细表\n（官微）')
|
||
ind_zy = 0
|
||
for ind, row in df.iterrows():
|
||
if row['gbiz'] == zybiz:
|
||
sheet.cell(row=ind_zy+3, column=1, value=str(ind_zy+1))
|
||
sheet.cell(row=ind_zy+3, column=2, value=f'{row["pub_year"]}-{row["pub_month"]}-{row["pub_day"]}')
|
||
sheet.cell(row=ind_zy+3, column=3, value=row['title'])
|
||
sheet.cell(row=ind_zy+3, column=4, value=row['source'])
|
||
sheet.cell(row=ind_zy+3, column=6, value=row['read_num'])
|
||
sheet.cell(row=ind_zy+3, column=7, value=row['id'])
|
||
sheet.cell(row=ind_zy+3, column=8, value=row['content_url'])
|
||
ind_zy = ind_zy + 1
|
||
cbma_path = os.path.join(BASE_DIR, f'summary/{year}年_cbma.xlsx')
|
||
workbook.save(cbma_path)
|
||
print(f'总院{year}年文章表生成完毕!')
|
||
|
||
template_cal_path = os.path.join(BASE_DIR, 'summary/tempalte_cbma_cal.xlsx')
|
||
workbook2 = load_workbook(template_cal_path)
|
||
need_df_list = [ "瑞泰科技", "国检集团", "中材高新", "哈玻院", "中国新材院", "秦皇岛院", "西安墙材院", "咸阳陶瓷院", "钟表所", "总院北分", "中岩科技", "水泥新材院", "中建材科创院", "科建苑"]
|
||
sheet2= workbook2['打分表']
|
||
sheet2.cell(row=1, column=1, value=f'中国建材总院宣传工作计分表（{year}年度）')
|
||
for ind, val in enumerate(need_df_list):
|
||
row_ind_df_fx = df_fx['单位'].to_list().index(val)
|
||
sheet2.cell(row=6, column=5+2*ind, value=df_fx.at[row_ind_df_fx, '供总院全年专稿数'])
|
||
sheet2.cell(row=10, column=5+2*ind, value=df_fx.at[row_ind_df_fx, '供总院全年组稿数'])
|
||
sheet2.cell(row=12, column=5+2*ind, value=df_fx.at[row_ind_df_fx, '供总院全年阅读10000及以上数'])
|
||
sheet2.cell(row=13, column=5+2*ind, value=df_fx.at[row_ind_df_fx, '供总院全年阅读5000及以上数'])
|
||
sheet2.cell(row=14, column=5+2*ind, value=df_fx.at[row_ind_df_fx, '供总院全年阅读1000及以上数'])
|
||
cbma_cal_path = os.path.join(BASE_DIR, f'summary/{year}年_cbma_cal.xlsx')
|
||
workbook2.save(cbma_cal_path)
|
||
print(f'总院{year}年打分表生成完毕!')
|
||
|
||
# need_df_list_full = [ "瑞泰科技股份有限公司", "中国国检测试控股集团股份有限公司", "中材高新材料股份有限公司", "哈尔滨玻璃钢研究院有限公司", "中国新型建材设计研究院有限公司", "秦皇岛玻璃工业研究设计院有限公司", "西安墙体材料研究设计院有限公司", "咸阳陶瓷研究设计院有限公司", "西安轻工业钟表研究所有限公司", "中国建材总院北京分公司", "中建材中岩科技有限公司", "水泥科学与新型建筑材料研究院（中研益）", "中建材科创新技术研究院（山东）有限公司", "北京科建苑物业管理有限公司"]
|
||
dw_list = df_fx['单位'].to_list()
|
||
template_month_path = os.path.join(BASE_DIR, 'summary/template_month.xlsx')
|
||
workbook3 = load_workbook(template_month_path)
|
||
for i in ['1月', '2月', '3月', '4月', '5月', '6月', '7月', '8月', '9月', '10月', '11月', '12月', '上半年', '下半年', '全年']:
|
||
try:
|
||
sheet= workbook3[i]
|
||
except KeyError:
|
||
sheet = workbook3.copy_worksheet(workbook3['1月'])
|
||
sheet.title = i
|
||
sheet.cell(row=1, column=1, value=f'关于{year}年度中国建材总院各企业新媒体更新情况统计表\n（{i}）')
|
||
# 开始总院填充数据
|
||
sheet.cell(row=4, column=3, value=df_fx.at[0, f'{i}发布数'])
|
||
max_read_row = df_fx.at[dw_list.index('中国建材总院'), f'{i}最高点击文章']
|
||
if max_read_row:
|
||
_, title, read_num, pub_date, source = max_read_row.split('***')
|
||
sheet.cell(row=7, column=2, value=title)
|
||
sheet.cell(row=7, column=4, value=read_num)
|
||
sheet.cell(row=7, column=5, value=pub_date)
|
||
sheet.cell(row=7, column=6, value=source)
|
||
# 开始填充各单位数据
|
||
sheet.cell(row=14, column=3, value=df_fx.at[dw_list.index('瑞泰科技'), f'{i}发布数'])
|
||
sheet.cell(row=14, column=6, value=df_fx.at[dw_list.index('瑞泰科技'), f'供总院{i}稿数'])
|
||
|
||
sheet.cell(row=15, column=3, value=df_fx.at[dw_list.index('国检集团'), f'{i}发布数'])
|
||
sheet.cell(row=15, column=6, value=df_fx.at[dw_list.index('国检集团'), f'供总院{i}稿数'])
|
||
|
||
sheet.cell(row=16, column=3, value=df_fx.at[dw_list.index('中材高新'), f'{i}发布数'])
|
||
sheet.cell(row=16, column=6, value=df_fx.at[dw_list.index('中材高新'), f'供总院{i}稿数'])
|
||
|
||
sheet.cell(row=17, column=3, value=df_fx.at[dw_list.index('哈玻院'), f'{i}发布数'])
|
||
sheet.cell(row=17, column=6, value=df_fx.at[dw_list.index('哈玻院'), f'供总院{i}稿数'])
|
||
|
||
sheet.cell(row=18, column=3, value=df_fx.at[dw_list.index('中国新材院'), f'{i}发布数'])
|
||
sheet.cell(row=18, column=6, value=df_fx.at[dw_list.index('中国新材院'), f'供总院{i}稿数'])
|
||
|
||
sheet.cell(row=19, column=3, value=df_fx.at[dw_list.index('秦皇岛院'), f'{i}发布数'])
|
||
sheet.cell(row=19, column=6, value=df_fx.at[dw_list.index('秦皇岛院'), f'供总院{i}稿数'])
|
||
|
||
sheet.cell(row=20, column=3, value=df_fx.at[dw_list.index('西安墙材院'), f'{i}发布数'])
|
||
sheet.cell(row=20, column=6, value=df_fx.at[dw_list.index('西安墙材院'), f'供总院{i}稿数'])
|
||
|
||
sheet.cell(row=21, column=3, value=df_fx.at[dw_list.index('咸阳陶瓷院'), f'{i}发布数'])
|
||
sheet.cell(row=21, column=6, value=df_fx.at[dw_list.index('咸阳陶瓷院'), f'供总院{i}稿数'])
|
||
|
||
sheet.cell(row=22, column=3, value=df_fx.at[dw_list.index('钟表所'), f'{i}发布数'])
|
||
sheet.cell(row=22, column=6, value=df_fx.at[dw_list.index('钟表所'), f'供总院{i}稿数'])
|
||
|
||
# sheet.cell(row=23, column=3, value=df_fx.at[dw_list.index('总院北分'), f'{i}发布数'])
|
||
sheet.cell(row=23, column=6, value=df_fx.at[dw_list.index('总院北分'), f'供总院{i}稿数'])
|
||
|
||
sheet.cell(row=24, column=3, value=df_fx.at[dw_list.index('中岩科技'), f'{i}发布数'])
|
||
sheet.cell(row=24, column=6, value=df_fx.at[dw_list.index('中岩科技'), f'供总院{i}稿数'])
|
||
|
||
# sheet.cell(row=25, column=3, value=df_fx.at[dw_list.index('水泥新材院'), f'{i}发布数'])
|
||
sheet.cell(row=25, column=6, value=df_fx.at[dw_list.index('水泥新材院'), f'供总院{i}稿数'])
|
||
|
||
# sheet.cell(row=26, column=3, value=df_fx.at[dw_list.index('中建材科创院'), f'{i}发布数'])
|
||
sheet.cell(row=26, column=6, value=df_fx.at[dw_list.index('中建材科创院'), f'供总院{i}稿数'])
|
||
|
||
# sheet.cell(row=27, column=3, value=df_fx.at[dw_list.index('科建苑'), f'{i}发布数'])
|
||
sheet.cell(row=27, column=6, value=df_fx.at[dw_list.index('科建苑'), f'供总院{i}稿数'])
|
||
|
||
sheet.cell(row=29, column=2, value=df_fx.at[dw_list.index('办公室（董事会办公室）'), f'供总院{i}稿数'])
|
||
sheet.cell(row=30, column=2, value=df_fx.at[dw_list.index('党委组织部/人力资源部'), f'供总院{i}稿数'])
|
||
sheet.cell(row=31, column=2, value=df_fx.at[dw_list.index('财务部'), f'供总院{i}稿数'])
|
||
sheet.cell(row=32, column=2, value=df_fx.at[dw_list.index('科技部'), f'供总院{i}稿数'])
|
||
sheet.cell(row=33, column=2, value=df_fx.at[dw_list.index('投资部'), f'供总院{i}稿数'])
|
||
sheet.cell(row=29, column=7, value=df_fx.at[dw_list.index('企业管理部、安全环保部'), f'供总院{i}稿数'])
|
||
sheet.cell(row=30, column=7, value=df_fx.at[dw_list.index('党群部/宣传统战部'), f'供总院{i}稿数'])
|
||
sheet.cell(row=31, column=7, value=df_fx.at[dw_list.index('党风办/巡察办、纪委综合室'), f'供总院{i}稿数'])
|
||
sheet.cell(row=32, column=7, value=df_fx.at[dw_list.index('监督执纪室'), f'供总院{i}稿数'])
|
||
sheet.cell(row=33, column=7, value=df_fx.at[dw_list.index('审计办公室'), f'供总院{i}稿数'])
|
||
|
||
|
||
cbma_month_path = os.path.join(BASE_DIR, f'summary/{year}年_cbma_month.xlsx')
|
||
workbook3.save(cbma_month_path)
|
||
print(f'总院{year}年月度表生成完毕!')
|
||
|
||
return origin_path, cbma_path, cbma_cal_path, cbma_month_path
|
||
|
||
|
||
def make_wechat_articles_full():
|
||
df = pd.read_csv(os.path.join(wechat_dir, 'articles.csv'))
|
||
df['content'] = ''
|
||
for ind, row in df.iterrows():
|
||
full_path = os.path.join(wechat_dir, row['nickname'], row['id'] + '.md')
|
||
try:
|
||
with open(full_path, encoding='utf-8') as f:
|
||
df.at[ind, 'content'] = f.read()
|
||
except FileNotFoundError:
|
||
print(full_path + '---不存在')
|
||
output_path = os.path.join(wechat_dir, 'articles_full.csv')
|
||
df.to_csv(output_path)
|
||
|
||
def ana_wechat():
|
||
articles_full_path = os.path.join(wechat_dir, 'articles_full.csv')
|
||
if not os.path.exists(articles_full_path):
|
||
make_wechat_articles_full()
|
||
|
||
df = pd.read_csv(articles_full_path)
|
||
df['content'] = df['content'].fillna('')
|
||
|
||
output_data = []
|
||
index = 1
|
||
|
||
for ind, row in df_s.iterrows():
|
||
mask = df['content'].str.contains(row['错误表述'])
|
||
result = df[mask]
|
||
|
||
if not result.empty:
|
||
for ind2, row2 in result.iterrows():
|
||
if row['错误表述'] == '“两学一做”学习' and '“两学一做”学习教育' in row2['content']:
|
||
continue
|
||
if row['错误表述'] == '20大':
|
||
continue
|
||
output_row = [
|
||
index,
|
||
row2['nickname'],
|
||
row2['title'],
|
||
row['错误表述'],
|
||
row['建议修改词语'],
|
||
row['错误分类'],
|
||
row2['content_url']
|
||
]
|
||
output_data.append(output_row)
|
||
index += 1
|
||
print(f'找到公众号问题{index}---{row2["nickname"]}')
|
||
# output_data.insert(0, ['序号', '信源名称', '文章标题', '错误表述', '建议修改词语', '错误分类', '原文链接'])
|
||
|
||
return output_data
|
||
|
||
|
||
def ana_web():
|
||
output_data = []
|
||
index = 1
|
||
# for file in os.listdir(web_dir):
|
||
# full_path = os.path.join(web_dir, file)
|
||
# if '$' in full_path:
|
||
# continue
|
||
# print(full_path)
|
||
# 只分析websites中的
|
||
df = pd.read_excel('web_sites.xlsx', sheet_name='Sheet1')
|
||
for ind, row in df.iterrows():
|
||
group = row['单位']
|
||
name = row['主办']
|
||
url = fix_url_scheme(row['地址'].strip())
|
||
domain = urlparse(url).netloc.replace('www.', '')
|
||
full_path = os.path.join(BASE_DIR, f'web_dir/{name}_{domain}.xlsx')
|
||
if os.path.exists(full_path) and os.path.getsize(full_path) > 0:
|
||
df = pd.read_excel(os.path.join(full_path), engine='openpyxl')
|
||
for ind, row in df_s.iterrows():
|
||
mask = df['text'].str.contains(row['错误表述'], na=False)
|
||
result = df[mask]
|
||
if not result.empty:
|
||
for ind2, row2 in result.iterrows():
|
||
if row['错误表述'] == '“两学一做”学习' and '“两学一做”学习教育' in row2['text']:
|
||
continue
|
||
if row['错误表述'] == '20大':
|
||
continue
|
||
output_row = [
|
||
index,
|
||
row2['name'],
|
||
"/",
|
||
row['错误表述'],
|
||
row['建议修改词语'],
|
||
row['错误分类'],
|
||
row2['url']
|
||
]
|
||
output_data.append(output_row)
|
||
index += 1
|
||
print(f'找到官网问题{index}---{row2["name"]}')
|
||
# output_data.insert(0, ['序号', '信源名称', '文章标题', '错误表述', '建议修改词语', '错误分类', '原文链接'])
|
||
|
||
return output_data
|
||
|
||
if __name__ == "__main__":
|
||
get_cbma_info_from_db_and_ana()
|
||
|