df=spark.sql("select key, name, subjects from table")
df in from above select statement :
key name subjects
12 x,y,z 1,2,3
20 a,b 8,7
df out :
12 x 1
12 y 2
12 z 3
20 a 8
20 b 7
tried converting to list , explode. Still throwing error. pls help the efficient way to achieve this ?
One way using pandas.DataFrame.apply:
# df["name"] = df["name"].str.split(",")
# df["subjects"] = df["subjects"].str.split(",")
# If not already split
new_df = df.apply(pd.Series.explode)
print(new_df)
Output:
key name subjects
0 12 x 1
0 12 y 2
0 12 z 3
1 20 a 8
1 20 b 7
Thanks chris. It is getting exploded. Still facing the error - Cannot reindex from a duplicate axis. Concat with ignore index is not working .Is it possible to generate temp unique indexes as key is duplicated during explode. pandasversion -1.0.5
df["name"] = df["name"].str.split(",")
df["subjects"] = df["subjects"].str.split(",")
new_df= df.apply(pd.Series.explode).reindex()