Deprecate line_terminator in to_csv

Pandas 1.5 deprecated line_terminator in favour of lineterminator (to
align with read_csv), and removed it in 2.0. Align the cuDF API by
preferring lineterminator and providing a deprecation warning for use
of line_terminator. Closes rapidsai#12894.

python/cudf/cudf/_fuzz_testing/tests/fuzz_test_csv.py

@@ -1,4 +1,4 @@
-# Copyright (c) 2020-2021, NVIDIA CORPORATION.
+# Copyright (c) 2020-2023, NVIDIA CORPORATION.
 
 import sys
 from io import StringIO
@@ -54,12 +54,12 @@ def csv_writer_test(pdf):
         ],
         "columns": ALL_POSSIBLE_VALUES,
         "index": [True, False],
-        "line_terminator": ["\n", "\r", "\r\n"],
+        "lineterminator": ["\n", "\r", "\r\n"],
         "chunksize": ALL_POSSIBLE_VALUES,
     },
 )
 def csv_writer_test_params(
-    pdf, sep, header, na_rep, columns, index, line_terminator, chunksize
+    pdf, sep, header, na_rep, columns, index, lineterminator, chunksize
 ):
     gdf = cudf.from_pandas(pdf)
 
@@ -69,7 +69,7 @@ def csv_writer_test_params(
         na_rep=na_rep,
         columns=columns,
         index=index,
-        line_terminator=line_terminator,
+        lineterminator=lineterminator,
         chunksize=chunksize,
     )
     gd_buffer = gdf.to_csv(
@@ -78,7 +78,7 @@ def csv_writer_test_params(
         na_rep=na_rep,
         columns=columns,
         index=index,
-        line_terminator=line_terminator,
+        lineterminator=lineterminator,
         chunksize=chunksize,
     )
 
@@ -90,13 +90,13 @@ def csv_writer_test_params(
         StringIO(gd_buffer),
         delimiter=sep,
         na_values=na_rep,
-        lineterminator=line_terminator,
+        lineterminator=lineterminator,
     )
     expected = pd.read_csv(
         StringIO(pd_buffer),
         delimiter=sep,
         na_values=na_rep,
-        lineterminator=line_terminator,
+        lineterminator=lineterminator,
     )
     if not header:
         # TODO: Remove renaming columns once the following bug is fixed:

python/cudf/cudf/_lib/csv.pyx

@@ -472,7 +472,7 @@ def write_csv(
     object sep=",",
     object na_rep="",
     bool header=True,
-    object line_terminator="\n",
+    object lineterminator="\n",
     int rows_per_chunk=8,
     bool index=True,
 ):
@@ -488,7 +488,7 @@ def write_csv(
     )
     cdef bool include_header_c = header
     cdef char delim_c = ord(sep)
-    cdef string line_term_c = line_terminator.encode()
+    cdef string line_term_c = lineterminator.encode()
     cdef string na_c = na_rep.encode()
     cdef int rows_per_chunk_c = rows_per_chunk
     cdef vector[string] col_names

python/cudf/cudf/core/dataframe.py

@@ -6,6 +6,7 @@
 import inspect
 import itertools
 import numbers
+import os
 import pickle
 import re
 import sys
@@ -604,7 +605,6 @@ class DataFrame(IndexedFrame, Serializable, GetAttrGetItemMixin):
     def __init__(
         self, data=None, index=None, columns=None, dtype=None, nan_as_null=True
     ):
-
         super().__init__()
 
         if isinstance(columns, (Series, cudf.BaseIndex)):
@@ -918,7 +918,7 @@ def _init_from_dict_like(
 
         if len(data):
             self._data.multiindex = True
-            for (i, col_name) in enumerate(data):
+            for i, col_name in enumerate(data):
                 self._data.multiindex = self._data.multiindex and isinstance(
                     col_name, tuple
                 )
@@ -1199,7 +1199,6 @@ def __setitem__(self, arg, value):
                     if is_scalar(value):
                         self._data[col_name][scatter_map] = value
                     else:
-
                         self._data[col_name][scatter_map] = column.as_column(
                             value
                         )[scatter_map]
@@ -5445,7 +5444,6 @@ def interpolate(
         downcast=None,
         **kwargs,
     ):
-
         if all(dt == np.dtype("object") for dt in self.dtypes):
             raise TypeError(
                 "Cannot interpolate with all object-dtype "
@@ -6358,13 +6356,29 @@ def to_csv(
         index=True,
         encoding=None,
         compression=None,
-        line_terminator="\n",
+        lineterminator=None,
+        line_terminator=None,
         chunksize=None,
         storage_options=None,
     ):
         """{docstring}"""
         from cudf.io import csv
 
+        if line_terminator is not None:
+            warnings.warn(
+                "line_terminator is a deprecated keyword argument, "
+                "use lineterminator instead.",
+                FutureWarning,
+            )
+            if lineterminator is not None:
+                warnings.warn(
+                    f"Ignoring {line_terminator=} in favour "
+                    f"of {lineterminator=}"
+                )
+            else:
+                lineterminator = line_terminator
+        if lineterminator is None:
+            lineterminator = os.linesep
         return csv.to_csv(
             self,
             path_or_buf=path_or_buf,
@@ -6373,7 +6387,7 @@ def to_csv(
             columns=columns,
             header=header,
             index=index,
-            line_terminator=line_terminator,
+            lineterminator=line_terminator,
             chunksize=chunksize,
             encoding=encoding,
             compression=compression,
@@ -6738,7 +6752,6 @@ def append(
             current_cols = self._data.to_pandas_index()
             combined_columns = other.index.to_pandas()
             if len(current_cols):
-
                 if cudf.utils.dtypes.is_mixed_with_object_dtype(
                     current_cols, combined_columns
                 ):

python/cudf/cudf/io/csv.py

@@ -1,4 +1,4 @@
-# Copyright (c) 2018-2022, NVIDIA CORPORATION.
+# Copyright (c) 2018-2023, NVIDIA CORPORATION.
 
 from collections import abc
 from io import BytesIO, StringIO
@@ -155,7 +155,7 @@ def to_csv(
     index=True,
     encoding=None,
     compression=None,
-    line_terminator="\n",
+    lineterminator="\n",
     chunksize=None,
     storage_options=None,
 ):
@@ -233,7 +233,7 @@ def to_csv(
                 sep=sep,
                 na_rep=na_rep,
                 header=header,
-                line_terminator=line_terminator,
+                lineterminator=lineterminator,
                 rows_per_chunk=rows_per_chunk,
                 index=index,
             )
@@ -244,7 +244,7 @@ def to_csv(
             sep=sep,
             na_rep=na_rep,
             header=header,
-            line_terminator=line_terminator,
+            lineterminator=lineterminator,
             rows_per_chunk=rows_per_chunk,
             index=index,
         )