Skip to content

PERF: read_sas #43333

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 1 commit into from
Sep 1, 2021
Merged

PERF: read_sas #43333

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
11 changes: 6 additions & 5 deletions pandas/io/sas/sas7bdat.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -789,7 +789,7 @@ def _chunk_to_dataframe(self) -> DataFrame:
n = self._current_row_in_chunk_index
m = self._current_row_in_file_index
ix = range(m - n, m)
rslt = DataFrame(index=ix)
rslt = {}

js, jb = 0, 0
for j in range(self.column_count):
Expand All @@ -798,25 +798,26 @@ def _chunk_to_dataframe(self) -> DataFrame:

if self._column_types[j] == b"d":
rslt[name] = self._byte_chunk[jb, :].view(dtype=self.byte_order + "d")
rslt[name] = np.asarray(rslt[name], dtype=np.float64)
rslt[name] = pd.Series(rslt[name], dtype=np.float64, index=ix)
if self.convert_dates:
if self.column_formats[j] in const.sas_date_formats:
rslt[name] = _convert_datetimes(rslt[name], "d")
elif self.column_formats[j] in const.sas_datetime_formats:
rslt[name] = _convert_datetimes(rslt[name], "s")
jb += 1
elif self._column_types[j] == b"s":
rslt[name] = self._string_chunk[js, :]
rslt[name] = pd.Series(self._string_chunk[js, :], index=ix)
if self.convert_text and (self.encoding is not None):
rslt[name] = rslt[name].str.decode(
self.encoding or self.default_encoding
)
if self.blank_missing:
ii = rslt[name].str.len() == 0
rslt.loc[ii, name] = np.nan
rslt[name][ii] = np.nan
js += 1
else:
self.close()
raise ValueError(f"unknown column type {self._column_types[j]}")

return rslt
df = DataFrame(rslt, columns=self.column_names, index=ix, copy=False)
return df
15 changes: 4 additions & 11 deletions pandas/tests/io/sas/test_sas7bdat.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,10 +7,7 @@
import numpy as np
import pytest

from pandas.errors import (
EmptyDataError,
PerformanceWarning,
)
from pandas.errors import EmptyDataError
import pandas.util._test_decorators as td

import pandas as pd
Expand Down Expand Up @@ -202,15 +199,11 @@ def test_compact_numerical_values(datapath):
tm.assert_series_equal(result, expected, check_exact=True)


def test_many_columns(datapath, using_array_manager):
def test_many_columns(datapath):
# Test for looking for column information in more places (PR #22628)
fname = datapath("io", "sas", "data", "many_columns.sas7bdat")
expected_warning = None
if not using_array_manager:
expected_warning = PerformanceWarning
with tm.assert_produces_warning(expected_warning):
# Many DataFrame.insert calls
df = pd.read_sas(fname, encoding="latin-1")

df = pd.read_sas(fname, encoding="latin-1")

fname = datapath("io", "sas", "data", "many_columns.csv")
df0 = pd.read_csv(fname, encoding="latin-1")
Expand Down