我是Python编程的初学者。我正在探索正则表达式。我正在尝试从“描述”列中提取一个单词（数据库名称）。我无法给出多个正则表达式模式。

请参阅下面的描述和代码。

描述

Summary: AD1: Low free DATA space in database AD1ADS: 10.00% Date: 06/28/2017 Severity: Warning Res
Summary: Database SV1V1CH has used log space: 90.00% Date: 02/06/2017 Severity: Warning ResourceId: s
Summary: SAP SolMan Sys=SM1Tempdb,MO=AGEEPM49,Alert=Database Host Status,Desc=A database hos
*** Clearing Event Received *** SNG01AMMSOL04_age SAP SolMan Sys=SM1_SNG01AMMSOL04,MO=AGEEQM46,Alert

提取的数据库名称的预期输出

AD1ADS
SV1V1CH
SM1Tempdb
SNG01AMMSOL04

尝试过的代码

sentence = df['Description']
frame = pd.DataFrame({'logs': sentence})

import re
pattern = re.compile(r'[dD]atabase (\w+)|Sys=(\w+)')

for _, line in frame.iterrows():
    name = pattern.findall(line['logs'])
    if name:
        line['names'] = name[0]
    else:
        line['names'] = 'Miscellaneous'

谁能告诉我，我在这里犯了什么错误。

我现在得到的输出

(u'AD1ADS', u'')
(u'SV1V1CH', u'')
(u'', u'CM1_CHE01AMMSOL04')
Miscellaneous

您可以使用str.extract http://pandas.pydata.org/pandas-docs/stable/generated/pandas.Series.str.extract.html with fillna http://pandas.pydata.org/pandas-docs/stable/generated/pandas.Series.fillna.html:

p = r'[dD]atabase (\w+)|Sys=(\w+)'
s = df['logs'].str.extract(p, expand=True)
print (s)
         0                  1
0   AD1ADS                NaN
1  SV1V1CH                NaN
2      NaN          SM1Tempdb
3      NaN  SM1_SNG01AMMSOL04

df['db'] = s[0].fillna(s[1]).fillna('Miscellaneous')
#alternatively 
#df['db'] = s[0].combine_first(s[1]).fillna('Miscellaneous')
print (df)
                                                logs                 db
0  Summary: AD1: Low free DATA space in database ...             AD1ADS
1  Summary: Database SV1V1CH has used log space: ...            SV1V1CH
2  Summary: SAP SolMan Sys=SM1Tempdb,MO=AGEEPM49,...          SM1Tempdb
3  *** Clearing Event Received *** SNG01AMMSOL04_...  SM1_SNG01AMMSOL04

如果想提取所有可能的值，请使用extractall http://pandas.pydata.org/pandas-docs/stable/generated/pandas.Series.str.extractall.html进而join如有必要，他们：

p = r'[dD]atabase (\w+)|Sys=(\w+)'
s = df['logs'].str.extractall(p)
print (s)
               0                  1
  match                            
0 0       AD1ADS                NaN
1 0      SV1V1CH                NaN
2 0          NaN          SM1Tempdb
  1         Host                NaN
  2          hos                NaN
3 0          NaN  SM1_SNG01AMMSOL04

df['db'] = s[0].fillna(s[1]).groupby(level=0).apply(', '.join)
df['db'] = df['db'].fillna('Miscellaneous')
print (df)
                                                logs                    db
0  Summary: AD1: Low free DATA space in database ...                AD1ADS
1  Summary: Database SV1V1CH has used log space: ...               SV1V1CH
2  Summary: SAP SolMan Sys=SM1Tempdb,MO=AGEEPM49,...  SM1Tempdb, Host, hos
3  *** Clearing Event Received *** SNG01AMMSOL04_...     SM1_SNG01AMMSOL04