From 8f0c4d2ca6856cc345917d62c3f989ada00617c0 Mon Sep 17 00:00:00 2001
From: jbrockmendel <jbrockmendel@gmail.com>
Date: Sun, 19 Nov 2023 11:26:45 -0800
Subject: [PATCH] TST: de-xfail pyarrow parser tests (#56056)

* TST: un-xfail pyarrow thousands test

* TST: de-xfail memory_map tests

* remove unused

* revert no-longer-needed
---
 pandas/tests/io/parser/test_encoding.py    | 31 ++++++++++++++++------
 pandas/tests/io/parser/test_parse_dates.py | 17 ++++++++----
 2 files changed, 35 insertions(+), 13 deletions(-)

diff --git a/pandas/tests/io/parser/test_encoding.py b/pandas/tests/io/parser/test_encoding.py
index 36d7a19cf6781..cbd3917ba9c04 100644
--- a/pandas/tests/io/parser/test_encoding.py
+++ b/pandas/tests/io/parser/test_encoding.py
@@ -23,7 +23,6 @@
     "ignore:Passing a BlockManager to DataFrame:DeprecationWarning"
 )
 
-xfail_pyarrow = pytest.mark.usefixtures("pyarrow_xfail")
 skip_pyarrow = pytest.mark.usefixtures("pyarrow_skip")
 
 
@@ -238,7 +237,6 @@ def test_parse_encoded_special_characters(encoding):
     tm.assert_frame_equal(result, expected)
 
 
-@xfail_pyarrow  # ValueError: The 'memory_map' option is not supported
 @pytest.mark.parametrize("encoding", ["utf-8", None, "utf-16", "cp1255", "latin-1"])
 def test_encoding_memory_map(all_parsers, encoding):
     # GH40986
@@ -252,11 +250,17 @@ def test_encoding_memory_map(all_parsers, encoding):
     )
     with tm.ensure_clean() as file:
         expected.to_csv(file, index=False, encoding=encoding)
+
+        if parser.engine == "pyarrow":
+            msg = "The 'memory_map' option is not supported with the 'pyarrow' engine"
+            with pytest.raises(ValueError, match=msg):
+                parser.read_csv(file, encoding=encoding, memory_map=True)
+            return
+
         df = parser.read_csv(file, encoding=encoding, memory_map=True)
     tm.assert_frame_equal(df, expected)
 
 
-@xfail_pyarrow  # ValueError: The 'memory_map' option is not supported
 def test_chunk_splits_multibyte_char(all_parsers):
     """
     Chunk splits a multibyte character with memory_map=True
@@ -272,11 +276,17 @@ def test_chunk_splits_multibyte_char(all_parsers):
     df.iloc[2047] = "a" * 127 + "ą"
     with tm.ensure_clean("bug-gh43540.csv") as fname:
         df.to_csv(fname, index=False, header=False, encoding="utf-8")
-        dfr = parser.read_csv(fname, header=None, memory_map=True, engine="c")
+
+        if parser.engine == "pyarrow":
+            msg = "The 'memory_map' option is not supported with the 'pyarrow' engine"
+            with pytest.raises(ValueError, match=msg):
+                parser.read_csv(fname, header=None, memory_map=True)
+            return
+
+        dfr = parser.read_csv(fname, header=None, memory_map=True)
     tm.assert_frame_equal(dfr, df)
 
 
-@xfail_pyarrow  # ValueError: The 'memory_map' option is not supported
 def test_readcsv_memmap_utf8(all_parsers):
     """
     GH 43787
@@ -300,9 +310,14 @@ def test_readcsv_memmap_utf8(all_parsers):
     df = DataFrame(lines)
     with tm.ensure_clean("utf8test.csv") as fname:
         df.to_csv(fname, index=False, header=False, encoding="utf-8")
-        dfr = parser.read_csv(
-            fname, header=None, memory_map=True, engine="c", encoding="utf-8"
-        )
+
+        if parser.engine == "pyarrow":
+            msg = "The 'memory_map' option is not supported with the 'pyarrow' engine"
+            with pytest.raises(ValueError, match=msg):
+                parser.read_csv(fname, header=None, memory_map=True, encoding="utf-8")
+            return
+
+        dfr = parser.read_csv(fname, header=None, memory_map=True, encoding="utf-8")
     tm.assert_frame_equal(df, dfr)
 
 
diff --git a/pandas/tests/io/parser/test_parse_dates.py b/pandas/tests/io/parser/test_parse_dates.py
index 30580423a3099..ee19a20a155aa 100644
--- a/pandas/tests/io/parser/test_parse_dates.py
+++ b/pandas/tests/io/parser/test_parse_dates.py
@@ -1979,8 +1979,6 @@ def test_date_parser_multiindex_columns_combine_cols(all_parsers, parse_spec, co
     tm.assert_frame_equal(result, expected)
 
 
-# ValueError: The 'thousands' option is not supported with the 'pyarrow' engine
-@xfail_pyarrow
 def test_date_parser_usecols_thousands(all_parsers):
     # GH#39365
     data = """A,B,C
@@ -1989,12 +1987,21 @@ def test_date_parser_usecols_thousands(all_parsers):
     """
 
     parser = all_parsers
-    warn = UserWarning
+
     if parser.engine == "pyarrow":
         # DeprecationWarning for passing a Manager object
-        warn = (UserWarning, DeprecationWarning)
+        msg = "The 'thousands' option is not supported with the 'pyarrow' engine"
+        with pytest.raises(ValueError, match=msg):
+            parser.read_csv(
+                StringIO(data),
+                parse_dates=[1],
+                usecols=[1, 2],
+                thousands="-",
+            )
+        return
+
     result = parser.read_csv_check_warnings(
-        warn,
+        UserWarning,
         "Could not infer format",
         StringIO(data),
         parse_dates=[1],