Add errors argument to to_csv() call to enable error handling for encoders

Author: roberthdevries

pandas/core/generic.py

@@ -3028,6 +3028,7 @@ def to_csv(
         index_label: Optional[Union[bool_t, str, Sequence[Label]]] = None,
         mode: str = "w",
         encoding: Optional[str] = None,
+        errors: str = "strict",
         compression: Optional[Union[str, Mapping[str, str]]] = "infer",
         quoting: Optional[int] = None,
         quotechar: str = '"',
@@ -3081,6 +3082,10 @@ def to_csv(
             for easier importing in R.
         mode : str
             Python write mode, default 'w'.
+        errors : str, default 'strict'
+            Specifies how encoding and decoding errors are to be handled.
+            See the errors argument for :func:`open` for a full list
+            of options.
         encoding : str, optional
             A string representing the encoding to use in the output file,
             defaults to 'utf-8'.
@@ -3161,6 +3166,7 @@ def to_csv(
             line_terminator=line_terminator,
             sep=sep,
             encoding=encoding,
+            errors=errors,
             compression=compression,
             quoting=quoting,
             na_rep=na_rep,

pandas/io/common.py

@@ -330,6 +330,7 @@ def get_handle(
     path_or_buf,
     mode: str,
     encoding=None,
+    errors=None,
     compression: Optional[Union[str, Mapping[str, Any]]] = None,
     memory_map: bool = False,
     is_text: bool = True,
@@ -441,7 +442,7 @@ def get_handle(
     elif is_path:
         if encoding:
             # Encoding
-            f = open(path_or_buf, mode, encoding=encoding, newline="")
+            f = open(path_or_buf, mode, encoding=encoding, errors=errors, newline="")
         elif is_text:
             # No explicit encoding
             f = open(path_or_buf, mode, errors="replace", newline="")
@@ -454,7 +455,7 @@ def get_handle(
     if is_text and (compression or isinstance(f, need_text_wrapping)):
         from io import TextIOWrapper
 
-        g = TextIOWrapper(f, encoding=encoding, newline="")
+        g = TextIOWrapper(f, encoding=encoding, errors=errors, newline="")
         if not isinstance(f, (BufferedIOBase, RawIOBase)):
             handles.append(g)
         f = g

pandas/io/formats/csvs.py

@@ -44,6 +44,7 @@ def __init__(
         index_label: Optional[Union[bool, Hashable, Sequence[Hashable]]] = None,
         mode: str = "w",
         encoding: Optional[str] = None,
+        errors: str = "strict",
         compression: Union[str, Mapping[str, str], None] = "infer",
         quoting: Optional[int] = None,
         line_terminator="\n",
@@ -77,6 +78,7 @@ def __init__(
         if encoding is None:
             encoding = "utf-8"
         self.encoding = encoding
+        self.errors = errors
         self.compression = infer_compression(self.path_or_buf, compression)
 
         if quoting is None:
@@ -185,6 +187,7 @@ def save(self) -> None:
                 self.path_or_buf,
                 self.mode,
                 encoding=self.encoding,
+                errors=self.errors,
                 compression=dict(self.compression_args, method=self.compression),
             )
             close = True
@@ -216,6 +219,7 @@ def save(self) -> None:
                         self.path_or_buf,
                         self.mode,
                         encoding=self.encoding,
+                        errors=self.errors,
                         compression=compression,
                     )
                     f.write(buf)