ENH: skiprows after header in read_csv

[davetapley]

Feature Type

• Adding new functionality to pandas

• Changing existing functionality in pandas

• Removing existing functionality in pandas

Problem Description

There isn't a way to tell read_csv to skiprows after the specified header row.

Feature Description

To avoid breaking existing logic, I think a new argument would be required, perhaps skiprows_after?

So one could, parse, for example:

ignore,me
foo,bar
ignore,me
123,1
456,2

with read_csv(f, header=1, skiprows_after=1) and get:

   foo  bar
0  123    1
1  456    2

Alternative Solutions

Workaround:

header = read_csv(f, skiprows=1, nrows=1)
names = header.columns.to_list()
read_csv(f, skiprows=4, names=names)

Additional Context

No response

[Aloqeely]

Thanks for the suggestion! CSV reading is already too complicated in pandas with many options for skipping certain rows (nrows, skipfooter, etc.) and so introducing another parameter will complicate read_csv even further.

However, I think you can implement this in a prettier way using iterator=True, e.g.

import pandas as pd

chunker = pd.read_csv("test.csv", header=1, iterator=True)
chunker.get_chunk(1)  # skip 1 row after header
df = chunker.get_chunk()  # the rest of the file

[mroeschke]

Thanks for the report but agreed that the snippet above would be a sufficient workaround and not worth the complexity to add another keyword argument to read_csv so closing