Inserting row of text before and after when a column's text changes in a Pandas Dataframe, in python

Question

Python beginner here. I have a Pandas Dataframe where I would like to insert a row with text in the first column based off values from another column. My example of data is below:

Current Dataframe:

value	type	section
0	brown	sect1
1	green	sect1
0	blue	sect2
0	brown	sect3
0	black	sect4
1	yellow	sect4
2	green	sect4
0	blue	sect5

Desired updated Dataframe:

value	type	section
START	brown	sect1
1	green	sect1
END
START	blue	sect2
END
START	brown	sect3
END
START	black	sect4
END
START	black	sect4
1	yellow	sect4
2	green	sect4
END
START	blue	sect5
END

Notice I would like to remove the beginning values of '0' and replace with 'START'. The 'START' and 'END' rows need to occur when the sect value changes to a new value.

I have my current Dataframe set up as seen above, but I truly am not sure how to correctly update the Dataframe to do this. Any suggestions on how to handle this problem?

Answer 1

Using boolean indexing and repeat:

# identify changes of section
m = df['section'].ne(df['section'].shift(-1))

# repeat last row per group
idx = df.index.repeat(m.add(1))
out = df.loc[idx]

# clear END
# assign END to last value
# assign START to first value
m2 = idx.to_series().duplicated()
out[m2] = ''
out.loc[m2, 'value'] = 'END'
out.loc[m2.shift(fill_value=True), 'value'] = 'START'

Output:

   value    type section
0  START   brown   sect1
1      1   green   sect1
1    END                
2  START    blue   sect2
2    END                
3  START   brown   sect3
3    END                
4  START   black   sect4
5      1  yellow   sect4
6      2   green   sect4
6    END                
7  START    blue   sect5
7    END