ENH: Add on_bad_lines for pyarrow

Author: amithkk

pandas/io/parsers/arrow_parser_wrapper.py

@@ -1,11 +1,14 @@
 from __future__ import annotations
 
 from typing import TYPE_CHECKING
+import warnings
 
 from pandas._config import using_pyarrow_string_dtype
 
 from pandas._libs import lib
+from pandas._libs.parsers import ParserWarning
 from pandas.compat._optional import import_optional_dependency
+from pandas.util._exceptions import find_stack_level
 
 from pandas.core.dtypes.inference import is_integer
 
@@ -85,6 +88,29 @@ def _get_pyarrow_options(self) -> None:
             and option_name
             in ("delimiter", "quote_char", "escape_char", "ignore_empty_lines")
         }
+
+        if "on_bad_lines" in self.kwds:
+            if callable(self.kwds["on_bad_lines"]):
+                self.parse_options["invalid_row_handler"] = self.kwds["on_bad_lines"]
+            elif self.kwds["on_bad_lines"] == ParserBase.BadLineHandleMethod.ERROR:
+                self.parse_options[
+                    "invalid_row_handler"
+                ] = None  # PyArrow raises an exception by default
+            elif self.kwds["on_bad_lines"] == ParserBase.BadLineHandleMethod.WARN:
+
+                def handle_warning(invalid_row):
+                    warnings.warn(
+                        f"Expected {invalid_row.expected_columns} columns, but found "
+                        f"{invalid_row.actual_columns}: {invalid_row.text}",
+                        ParserWarning,
+                        stacklevel=find_stack_level(),
+                    )
+                    return "skip"
+
+                self.parse_options["invalid_row_handler"] = handle_warning
+            elif self.kwds["on_bad_lines"] == ParserBase.BadLineHandleMethod.SKIP:
+                self.parse_options["invalid_row_handler"] = lambda _: "skip"
+
         self.convert_options = {
             option_name: option_value
             for option_name, option_value in self.kwds.items()

pandas/io/parsers/readers.py

@@ -390,6 +390,13 @@
           expected, a ``ParserWarning`` will be emitted while dropping extra elements.
           Only supported when ``engine='python'``
 
+    .. versionchanged:: 1.4.1
+
+        - Callable, function with signature
+         as described in `pyarrow documentation
+         <https://arrow.apache.org/docs/python/generated/pyarrow.csv.ParseOptions.html
+         #pyarrow.csv.ParseOptions.invalid_row_handler>_` when ``engine='pyarrow'``
+
 delim_whitespace : bool, default False
     Specifies whether or not whitespace (e.g. ``' '`` or ``'\\t'``) will be
     used as the ``sep`` delimiter. Equivalent to setting ``sep='\\s+'``. If this option
@@ -483,7 +490,6 @@ class _Fwf_Defaults(TypedDict):
     "thousands",
     "memory_map",
     "dialect",
-    "on_bad_lines",
     "delim_whitespace",
     "quoting",
     "lineterminator",
@@ -2038,9 +2044,10 @@ def _refine_defaults_read(
     elif on_bad_lines == "skip":
         kwds["on_bad_lines"] = ParserBase.BadLineHandleMethod.SKIP
     elif callable(on_bad_lines):
-        if engine != "python":
+        if engine not in ["python", "pyarrow"]:
             raise ValueError(
-                "on_bad_line can only be a callable function if engine='python'"
+                "on_bad_line can only be a callable function "
+                "if engine='python' or 'pyarrow'"
             )
         kwds["on_bad_lines"] = on_bad_lines
     else: