python pandas模糊匹配 读取Excel后 获取指定指标的操作

import pandas as pd
#1.读取数据
df = pd.read_excel(r'./data/pfee.xlsx')
print(df)

cols = list(df.columns)
print(cols)

#2.获取含有装修 和 配置 字段的数据
zx_lists=[]
pz_lists=[]
for name in cols:
 if '装修' in name:
  zx_lists.append(name)
 elif '配置' in name:
  pz_lists.append(name)
print(zx_lists)
print(pz_lists)

#3.对装修和配置项费用进行求和计算
df['装修-求和'] =df[zx_lists].apply(lambda x:x.sum(),axis=1)
df['配置-求和'] = df[pz_lists].apply(lambda x:x.sum(),axis=1)
print(df)

df_obj[df_obj['title'].str.contains(r'.*?n.*')] #使用正则表达式进行模糊匹配,*匹配0或无限次,?匹配0或1次

df1=df.filter("uri rlike 
 'com.tencent.tmgp.sgame|%E8%80%85%E8%8D%A3%E8%80%80_|android.ugc.live|\
 %e7%88f%e8%a7%86%e9%a2%91|%E7%%8F%E8%A7%86%E9%A2%91'").groupBy("uri").\
 count().sort("count", ascending=False)

In [5]: df.filter("name rlike '%'").show()
+---+------+-----+
|age|height| name|
+---+------+-----+
| 4| 140|A%l%i|
| 6| 180| i%ce|
+---+------+-----+

spark.sql("select uri from t where uri like '%com.tencent.tmgp.sgame%' or uri like 'douyu'").show(5)

mysql> select count(*) from url_parse where uri regexp 'android.ugc.live|com.tencent.tmgp.sgame';
+----------+
| count(*) |
+----------+
|  9768 |
+----------+
1 row in set (0.52 sec)

In [9]: spark.sql('select * from t where name regexp "%l|t|_"').show()
+---+------+------+
|age|height| name|
+---+------+------+
| 1| 150|Al_ice|
| 4| 140| A%l%i|
+---+------+------+