TST: Skip more pyarrow csv tests that fail in pyarrow.csv.read_csv (#56098)

Author: mroeschke

pandas/tests/io/parser/common/test_read_errors.py

@@ -186,7 +186,8 @@ def test_error_bad_lines(all_parsers):
     msg = "Expected 1 fields in line 3, saw 3"
 
     if parser.engine == "pyarrow":
-        msg = "CSV parse error: Expected 1 columns, got 3: 1,2,3"
+        # "CSV parse error: Expected 1 columns, got 3: 1,2,3"
+        pytest.skip(reason="https://github.com/apache/arrow/issues/38676")
 
     with pytest.raises(ParserError, match=msg):
         parser.read_csv(StringIO(data), on_bad_lines="error")
@@ -222,7 +223,8 @@ def test_read_csv_wrong_num_columns(all_parsers):
     msg = "Expected 6 fields in line 3, saw 7"
 
     if parser.engine == "pyarrow":
-        msg = "Expected 6 columns, got 7: 6,7,8,9,10,11,12"
+        # Expected 6 columns, got 7: 6,7,8,9,10,11,12
+        pytest.skip(reason="https://github.com/apache/arrow/issues/38676")
 
     with pytest.raises(ParserError, match=msg):
         parser.read_csv(StringIO(data))
@@ -246,10 +248,9 @@ def test_null_byte_char(request, all_parsers):
         tm.assert_frame_equal(out, expected)
     else:
         if parser.engine == "pyarrow":
-            msg = (
-                "CSV parse error: Empty CSV file or block: "
-                "cannot infer number of columns"
-            )
+            # CSV parse error: Empty CSV file or block: "
+            # cannot infer number of columns"
+            pytest.skip(reason="https://github.com/apache/arrow/issues/38676")
         else:
             msg = "NULL byte detected"
         with pytest.raises(ParserError, match=msg):
@@ -299,7 +300,8 @@ def test_bad_header_uniform_error(all_parsers):
             "number of columns, but 3 left to parse."
         )
     elif parser.engine == "pyarrow":
-        msg = "CSV parse error: Expected 1 columns, got 4: col1,col2,col3,col4"
+        # "CSV parse error: Expected 1 columns, got 4: col1,col2,col3,col4"
+        pytest.skip(reason="https://github.com/apache/arrow/issues/38676")
 
     with pytest.raises(ParserError, match=msg):
         parser.read_csv(StringIO(data), index_col=0, on_bad_lines="error")

pandas/tests/io/parser/test_unsupported.py

@@ -178,8 +178,8 @@ def test_close_file_handle_on_invalid_usecols(all_parsers):
 
     error = ValueError
     if parser.engine == "pyarrow":
-        pyarrow = pytest.importorskip("pyarrow")
-        error = pyarrow.lib.ArrowKeyError
+        # Raises pyarrow.lib.ArrowKeyError
+        pytest.skip(reason="https://github.com/apache/arrow/issues/38676")
 
     with tm.ensure_clean("test.csv") as fname:
         Path(fname).write_text("col1,col2\na,b\n1,2", encoding="utf-8")

pandas/tests/io/parser/usecols/test_usecols_basic.py

@@ -477,11 +477,8 @@ def test_usecols_subset_names_mismatch_orig_columns(all_parsers, usecols, reques
             with pytest.raises(ValueError, match=_msg_pyarrow_requires_names):
                 parser.read_csv(StringIO(data), header=0, names=names, usecols=usecols)
             return
-        mark = pytest.mark.xfail(
-            reason="pyarrow.lib.ArrowKeyError: Column 'A' in include_columns "
-            "does not exist"
-        )
-        request.applymarker(mark)
+        # "pyarrow.lib.ArrowKeyError: Column 'A' in include_columns does not exist"
+        pytest.skip(reason="https://github.com/apache/arrow/issues/38676")
 
     result = parser.read_csv(StringIO(data), header=0, names=names, usecols=usecols)
     expected = DataFrame({"A": [1, 5], "C": [3, 7]})