如何将逗号分隔的字典字符串拆分为Pandas数据框

Question

I have a string in this format:

{apple:"34253453",oranges:"Sweet",x:"COOL"},{apple:"34222453",oranges:"Dry",x:"WARM"},{apple:"31113453",oranges:"Bitter",x:"HOT"},{apple:"38883453",oranges:"Sweet",x:"COOL"}

And looking to create a dataframe with the column labels as 'apple','oranges','x' and the values placed in their respective rows.

I've tried to use this solution: Python convert comma separated list to pandas dataframe as well as ast.literal_eval to convert it into a list before transforming it into a dataframe but to no luck.

Answer 1

Your string is invalid json , so necessary some replace first:

import ast

s = '{apple:"34253453",oranges:"Sweet",x:"COOL"},{apple:"34222453",oranges:"Dry",x:"WARM"},{apple:"31113453",oranges:"Bitter",x:"HOT"},{apple:"38883453",oranges:"Sweet",x:"COOL"}'

ss = '[' + s.replace('{', '{"').replace(':"','":"').replace('",', '","') + ']'
print (ss)

[{"apple":"34253453","oranges":"Sweet","x":"COOL"},
 {"apple":"34222453","oranges":"Dry","x":"WARM"},
 {"apple":"31113453","oranges":"Bitter","x":"HOT"},
 {"apple":"38883453","oranges":"Sweet","x":"COOL"}]

---

df = pd.DataFrame(ast.literal_eval(ss))
print (df)
      apple oranges     x
0  34253453   Sweet  COOL
1  34222453     Dry  WARM
2  31113453  Bitter   HOT
3  38883453   Sweet  COOL

---

df = pd.DataFrame(pd.io.json.loads(ss))
print (df)
      apple oranges     x
0  34253453   Sweet  COOL
1  34222453     Dry  WARM
2  31113453  Bitter   HOT
3  38883453   Sweet  COOL