将具有多个值的字符串转换为dict？

Question

我有一个数据帧，例如：

Path                          Attribute
/home/accessbile/wheelchair   type:threshold;width:42;gaurded:no
/home/accessible/wheelcahir   type:threshold;width:45
/home/accessbile/armchair     weight(lbs):100;width:30
/home/accessible/armchair     type:foldind;weight(lbs):100;width:30

我需要在属性中获取关于路径的值（type、width、weight、gaurded，比如..）的唯一计数

输出：

    PATH                          Attributes  count


/home/accessbile/wheelchair     type          2
/home/accessbile/wheelchair     width         2
/home/accessbile/wheelchair    gaurded        1
/home/accessbile/armchair      weight(lbs)    2
/home/accessbile/armchair      width          2
/home/accessbile/armchair      type           1 

尝试通过ast.eval将属性转换为dict，并将键作为唯一计数。。但不起作用

Answer 1

首先，检查Path中所有行的拼写，一旦修复，可以执行以下操作：

df.Attribute=df.Attribute.str.split(";")
df_new=pd.DataFrame([[x] + [z] for x, y in df.values for z in y],columns=df.columns)
df_new['Attribute']=df_new.Attribute.str.split(":").str[0]
df_new.groupby('Path')['Attribute'].value_counts().reset_index(0).\
               rename(columns={'Attribute':'Count'}).reset_index()

     Attribute                         Path  Count
0         type  /home/accessbile/wheelchair      2
1        width  /home/accessbile/wheelchair      2
2      gaurded  /home/accessbile/wheelchair      1
3  weight(lbs)    /home/accessible/armchair      2
4        width    /home/accessible/armchair      2
5         type    /home/accessible/armchair      1