read_csv skiprows vs. read_excel

[ragesz]

As I see pandas.read_csv() and pandas.read_excel() handle differently the skiprows argument. I have the same data in a CSV file and in an Excel file:

abc  def
  1   10
  2   11
  3   12
  4   13
  5   14

I want to use different column names when I read the data, so I specify the desired column names in names argument and I skip the first (header) row of my CSV & Excel files:
pd.read_excel('test.xlsx', skiprows=0, names=['foo', 'bar']) returns with my expected result:

   foo  bar
0    1   10
1    2   11
2    3   12
3    4   13
4    5   14

I get the same expected result with pd.read_csv('test.csv', skiprows=1, names=['foo', 'bar']). But pd.read_csv('test.csv', skiprows=0, names=['foo', 'bar']) keeps the first (header) row of the input file:

   foo  bar
0  abc  def
1    1   10
2    2   11
3    3   12
4    4   13
5    5   14

Is this the expected behavior of skiprows or something is wrong at pandas.read_csv()?

[TomAugspurger]

I think I would do this using the header argument.

In [17]: pd.read_excel("foo.xlsx", names=['foo', 'bar'], header=0)
Out[17]:
   foo  bar
0    1   10
1    2   11
2    3   12
3    4   13
4    5   14

In [18]: pd.read_csv("foo.csv", names=['foo', 'bar'], header=0)
Out[18]:
   foo  bar
0    1   10
1    2   11
2    3   12
3    4   13
4    5   14

This is a symptom of #11889

[TomAugspurger]

Since you actually hit that API inconsistency, your input would be appreciated in #11889