Skip to content

BUG: read_csv adding additional columns as integers instead of strings #47137

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
May 27, 2022
Merged

BUG: read_csv adding additional columns as integers instead of strings #47137

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
ef898b4
BUG: read_csv adding additional columns as integers instead of strings
phofl May 27, 2022
6559309
Reword whatsnew
phofl May 27, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions doc/source/whatsnew/v1.5.0.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -809,6 +809,7 @@ I/O
- Bug in :func:`read_parquet` when ``engine="pyarrow"`` which caused partial write to disk when column of unsupported datatype was passed (:issue:`44914`)
- Bug in :func:`DataFrame.to_excel` and :class:`ExcelWriter` would raise when writing an empty DataFrame to a ``.ods`` file (:issue:`45793`)
- Bug in :func:`read_html` where elements surrounding ``<br>`` were joined without a space between them (:issue:`29528`)
- Bug in :func:`read_csv` when data is longer than header leading to issues with callables in ``usecols`` expecting strings (:issue:`46997`)
- Bug in Parquet roundtrip for Interval dtype with ``datetime64[ns]`` subtype (:issue:`45881`)
- Bug in :func:`read_excel` when reading a ``.ods`` file with newlines between xml elements (:issue:`45598`)
- Bug in :func:`read_parquet` when ``engine="fastparquet"`` where the file was not closed on error (:issue:`46555`)
Expand Down
4 changes: 3 additions & 1 deletion pandas/_libs/parsers.pyx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1304,8 +1304,10 @@ cdef class TextReader:
if self.header is not None:
j = i - self.leading_cols
# generate extra (bogus) headers if there are more columns than headers
# These should be strings, not integers, because otherwise we might get
# issues with callables as usecols GH#46997
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

if these are truely bogus, why are they even passed to the usecols callable? There are no circumstances where these "auto-generated" column names are/should be in the result?

Copy link
Member Author

@phofl phofl May 27, 2022 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

see #47138, this is the bogus case

You can access them when using an index-based usecols setting, but the column name is shifted afterwards.

Edit: If we want to change this (I think I would be in favor of that), we have to deprecate first

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

but the column name is shifted afterwards.

right, so if they don't currently make their way into the result, we are not at risk of changing existing behavior with this PR?

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

No.

At least I can not see a way in how this would impact anything.

You can not select these columns by name via usecols. If you select them by position, an available name is used, not the bogus name.

The only impact this should have is consistensy when feeding them into the usecols callable

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

At least I can not see a way in how this would impact anything.

cool.

if j >= len(self.header[0]):
return j
return str(j)
elif self.has_mi_columns:
return tuple(header_row[j] for header_row in self.header)
else:
Expand Down
18 changes: 18 additions & 0 deletions pandas/tests/io/parser/usecols/test_usecols_basic.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -416,3 +416,21 @@ def test_usecols_indices_out_of_bounds(all_parsers, names):
if names is None and parser.engine == "python":
expected = DataFrame({"a": [1]})
tm.assert_frame_equal(result, expected)


def test_usecols_additional_columns(all_parsers):
# GH#46997
parser = all_parsers
usecols = lambda header: header.strip() in ["a", "b", "c"]
result = parser.read_csv(StringIO("a,b\nx,y,z"), index_col=False, usecols=usecols)
expected = DataFrame({"a": ["x"], "b": "y"})
tm.assert_frame_equal(result, expected)


def test_usecols_additional_columns_integer_columns(all_parsers):
# GH#46997
parser = all_parsers
usecols = lambda header: header.strip() in ["0", "1"]
result = parser.read_csv(StringIO("0,1\nx,y,z"), index_col=False, usecols=usecols)
expected = DataFrame({"0": ["x"], "1": "y"})
tm.assert_frame_equal(result, expected)