TST: better exception messages with na_values and pyarrow engine (#56090)

* TST: better exception messages with na_values and pyarrow engine * remove commented-out * remove commented-out

Author: jbrockmendel

pandas/io/parsers/arrow_parser_wrapper.py

@@ -250,6 +250,15 @@ def read(self) -> DataFrame:
  include = self.convert_options.get("include_columns", None)
  if include is not None:
  self._validate_usecols(include)
+
+ nulls = self.convert_options.get("null_values", set())
+ if not lib.is_list_like(nulls) or not all(
+ isinstance(x, str) for x in nulls
+ ):
+ raise TypeError(
+ "The 'pyarrow' engine requires all na_values to be strings"
+ )
+
  raise
 
  try:

pandas/io/parsers/readers.py

@@ -1716,7 +1716,10 @@ def _clean_options(
 
  # Converting values to NA
  keep_default_na = options["keep_default_na"]
- na_values, na_fvalues = _clean_na_values(na_values, keep_default_na)
+ floatify = engine != "pyarrow"
+ na_values, na_fvalues = _clean_na_values(
+ na_values, keep_default_na, floatify=floatify
+ )
 
  # handle skiprows; this is internally handled by the
  # c-engine, so only need for python and pyarrow parsers
@@ -1928,7 +1931,7 @@ def TextParser(*args, **kwds) -> TextFileReader:
  return TextFileReader(*args, **kwds)
 
 
-def _clean_na_values(na_values, keep_default_na: bool = True):
+def _clean_na_values(na_values, keep_default_na: bool = True, floatify: bool = True):
  na_fvalues: set | dict
  if na_values is None:
  if keep_default_na:
@@ -1956,7 +1959,7 @@ def _clean_na_values(na_values, keep_default_na: bool = True):
  else:
  if not is_list_like(na_values):
  na_values = [na_values]
- na_values = _stringify_na_values(na_values)
+ na_values = _stringify_na_values(na_values, floatify)
  if keep_default_na:
  na_values = na_values | STR_NA_VALUES
 
@@ -1978,7 +1981,7 @@ def _floatify_na_values(na_values):
  return result
 
 
-def _stringify_na_values(na_values):
+def _stringify_na_values(na_values, floatify: bool):
  """return a stringified and numeric for these values"""
  result: list[str | float] = []
  for x in na_values:
@@ -1993,13 +1996,15 @@ def _stringify_na_values(na_values):
  result.append(f"{v}.0")
  result.append(str(v))
 
- result.append(v)
- except (TypeError, ValueError, OverflowError):
- pass
- try:
- result.append(int(x))
+ if floatify:
+ result.append(v)
  except (TypeError, ValueError, OverflowError):
  pass
+ if floatify:
+ try:
+ result.append(int(x))
+ except (TypeError, ValueError, OverflowError):
+ pass
  return set(result)