I have a pandas dataframe with some rows containing json data

I would like to pivot the table so that I have a time series of the different variables, like

CodePudding user response：

Convert values to Series with convert jsons to dicts, last get first non missing value per datetimes by GroupBy.first:

import ast


def f(x):
    try:
        return pd.Series(ast.literal_eval(x))
    except SyntaxError:
        return pd.Series()

df = df.set_index('Date')['message'].apply(f).groupby(level=0).first()
print (df)
            var1 var2
Date                 
2020-01-01   foo  foo
2020-01-02  None  bar

CodePudding user response：

Not sure what your data exactly looks like, but you could do the following

• Use pd.json_normalize() to parse some of the nested json content
• Then use groupby() to get for example the first result

Here is an example:

import pandas as pd

#Input data
data = {
        "Date": ["2020-01-01","2020-01-01","2020-01-01","2020-01-02","2020-01-02"],
        "message": [{'var1': 'foo'}, {'stupidkey':'stupidval'}, {'var2':'foo'}, "bla", {'var2':'bar'}]
        }
df = pd.DataFrame(data)

# Generate result
varsToKeep = ["var1", "var2"]
df_vars = pd.json_normalize(df["message"])[varsToKeep]
df = df.join(df_vars)
df_result = df.groupby("Date").first()[varsToKeep]