Extract Text From Pandas DataFrame Based On Filters of a different column - python-3.x

Question

I have a df. In one column is "State" and in another column is "Text". I want to make a new column called "my_new_col" that extracts the word "Lime" from the "Text" column, only when the State Column = "Idaho"
df = {'State': ["Idaho", "Washington","Oregon","Idaho","Oregon"], 'Text': ["Lime Light","New Egg","Lime Inc","Monteray","NovaDing"]}
df = pd.DataFrame(df)
df
Output:
State Text
0 Idaho Lime Light
1 Washington New Egg
2 Oregon Lime Inc
3 Idaho Monteray
4 Oregon NovaDing
How do I get a dataframe that shows the following
State Text my_new_col
0 Idaho Lime Light Lime
1 Washington New Egg None
2 Oregon Lime Inc None
3 Idaho Monteray None
4 Oregon NovaDing None
Another example could be to pull out text that matches regex into a new column
df = {'State': ["Idaho", "Washington","Oregon","Idaho","Oregon"], 'Text': ["1,234 Light","New Egg","Lime Inc","1223 Ring","NovaDing"]}
df = pd.DataFrame(df)
df
Output:
State Text
0 Idaho 1,234 Light
1 Washington New Egg
2 Oregon Lime Inc
3 Idaho 1223 Ring
4 Oregon NovaDing
How do I get a dataframe that shows the following. The regex would be \d,\d\d\d
State Text my_new_col
0 Idaho 1,234 Light 1,234
1 Washington New Egg None
2 Oregon Lime Inc None
3 Idaho 1223 Ring None
4 Oregon NovaDing None

Answer 1

If it's case-sensitive:
df['my_new_col'] = None
df.loc[(df['State']=='Idaho') & (df['Text'].str.contains("Lime")), 'my_new_col'] = 'Lime'
print(df)
State Text my_new_col
0 Idaho Lime Light Lime
1 Washington New Egg None
2 Oregon Lime Inc None
3 Idaho Monteray None
4 Oregon NovaDing None
If case-insensitive:
df.loc[(df['State']=='Idaho') & (df['Text'].str.contains("Lime", case=False)), 'my_new_col'] = 'Lime'
...based on the update to the question, from the second example dataframe:
df.loc[(df['State']=='Idaho'), 'my_new_col'] = df['Text'].str.extract(r"(\d,\d\d\d)")[0]
That puts NaN values in the column instead of None. If that matters:
df['my_new_col'] = None
df.loc[(df['State']=='Idaho'), 'my_new_col'] = df['Text'].str.extract(r"(\d,\d\d\d)")[0]
df.loc[df['my_new_col'].isnull(), 'my_new_col'] = None