I have a dataframe like so:
| Store | matches |
|---|---|
| Murphy’s | [{‘domain’: ‘murphyscolumbus.com’, ‘location’: ‘Columbus, OH’}, {‘domain’: ‘murphystampa.com’, ‘location’: ‘Tampa, FL’}] |
| Bill’s | [{‘domain’: ‘billsdallas.com’, ‘location’: ‘Dallas, TX’}, {‘domain’: ‘billsorlando.com’, ‘location’: ‘Orlando, FL’}] |
What I want is a dataframe like so:
| Store | domains |
|---|---|
| Murphy’s | [‘murphyscolumbus.com’, ‘murphystampa.com’] |
| Bill’s | [‘billsdallas.com’,’billsorlando.com’] |
I’m hoping for something less computationally expensive than something like a for loop that steps through as the dataframe is quite large.
>Solution :
Try:
from ast import literal_eval
# apply literal_eval if necessary
df["matches"] = df["matches"].apply(literal_eval)
df["domains"] = df.pop("matches").apply(lambda x: [d["domain"] for d in x])
print(df)
Prints:
Store domains
0 Murphy's [murphyscolumbus.com, murphystampa.com]
1 Bill's [billsdallas.com, billsorlando.com]