From a44c3dcf9cd39691689c2c51a1ee37660ff4a7a3 Mon Sep 17 00:00:00 2001
From: Kevin Yan <kevnya@amazon.com>
Date: Fri, 10 Jan 2025 01:28:25 +0000
Subject: [PATCH] Fix unexpected reader_type kwarg error

---
 .../tests/utils/data/test_file.parquet.gz     | Bin 0 -> 630 bytes
 deltacat/tests/utils/test_pyarrow.py          |  23 ++++++++++++++++++
 deltacat/utils/pyarrow.py                     |   4 +--
 3 files changed, 25 insertions(+), 2 deletions(-)
 create mode 100644 deltacat/tests/utils/data/test_file.parquet.gz

diff --git a/deltacat/tests/utils/data/test_file.parquet.gz b/deltacat/tests/utils/data/test_file.parquet.gz
new file mode 100644
index 0000000000000000000000000000000000000000..673dc0146e0a009b550e0ff9d0a64909fa582db0
GIT binary patch
literal 630
zcmV-+0*U<}iwFoJ#_(hS19W9`bYEs^Y-KKRVRCVGWpn_wRLySFKoDO4#5h$dMOAjS
zmV)F%YI;aRT2w$)inLA>1rbV!KyA4olUQyf#|gGecmkfGPr$K9)JNciIB=*?kDNGi
zKpZ%*W5<cwnnR^_q<Ck)ot^o<*-<*IyM|?~<5HV1XbGWwO+>ND(-kAUgh?4IIM07d
zk*diFRg+kPR#H%ISsgZtAV{)tDipD-ZRC-nQN?@h+D?E)qL9x5`6?JKL%m9Btqg>*
z2=ww8V7}kPkJ_9g=(s&rIo5Qo;lOi73imsf?~g)>d)p(2-zs1_7j+-4C~*Ll_%RP!
zD8<Z0sf3I#auMee>XtVcT5d%BSRkJS<CkO{NtaTjGOXfC!jQ@`$_lm2R8W83?sLqj
z+MlUU`T#w-j9w&tax1;x-vx3e7#|hmEOV9Ysd{#BH_>zF2&A+|eqb+h$Mgqb=-PS^
zT7Hn_K;8Dj!PxHGDZS4`*XZK8OBeN=E~G!gw@A;=tnpX<f37t4T5SVSCFmX4Lkr6s
zSQe8;A)JFvTxA)xi_n*xR@TZlJ;v8GYveb-W2oeVRDdcsKMdFXJmgp+z>1tm1rcdQ
znq%GWv)#@3x#%C^F4zMG_MFA`gs}-T9oA&bVRa|MOp~#v;M5s=90Bu{#je?y?RQ@A
zqz}2%XvW|Kz700BeE<*M4Bc>lcM_F0r|10h66*5$E#Q$H#@47erX*7G9qz5acVFED
znhx8JNcE<nV9j|r+uA}@A6x$0LpxmUjmN7n1rz<9?FWOAr?0J7s+Fo%=JN_8`-69~
Q!~f&|0sQTPm*@om0ET@y2><{9

literal 0
HcmV?d00001

diff --git a/deltacat/tests/utils/test_pyarrow.py b/deltacat/tests/utils/test_pyarrow.py
index 90a2a838..58afe5d5 100644
--- a/deltacat/tests/utils/test_pyarrow.py
+++ b/deltacat/tests/utils/test_pyarrow.py
@@ -16,6 +16,7 @@
 import pyarrow as pa
 
 PARQUET_FILE_PATH = "deltacat/tests/utils/data/test_file.parquet"
+PARQUET_GZIP_COMPRESSED_FILE_PATH = "deltacat/tests/utils/data/test_file.parquet.gz"
 EMPTY_UTSV_PATH = "deltacat/tests/utils/data/empty.csv"
 NON_EMPTY_VALID_UTSV_PATH = "deltacat/tests/utils/data/non_empty_valid.csv"
 OVERFLOWING_DECIMAL_PRECISION_UTSV_PATH = (
@@ -789,3 +790,25 @@ def test_s3_file_to_table_when_parquet_schema_overridden(self):
             self.assertEqual(field.name, schema.field(index).name)
 
         self.assertEqual(result.schema.field(1).type, "string")
+
+    def test_s3_file_to_table_when_parquet_gzip(self):
+
+        pa_kwargs_provider = lambda content_type, kwargs: {
+            "reader_type": "pyarrow",
+            **kwargs,
+        }
+
+        result = s3_file_to_table(
+            PARQUET_GZIP_COMPRESSED_FILE_PATH,
+            ContentType.PARQUET.value,
+            ContentEncoding.GZIP.value,
+            ["n_legs", "animal"],
+            ["n_legs"],
+            pa_read_func_kwargs_provider=pa_kwargs_provider,
+        )
+
+        self.assertEqual(len(result), 6)
+        self.assertEqual(len(result.column_names), 1)
+        schema = result.schema
+        schema_index = schema.get_field_index("n_legs")
+        self.assertEqual(schema.field(schema_index).type, "int64")
diff --git a/deltacat/utils/pyarrow.py b/deltacat/utils/pyarrow.py
index 653ac614..31167cc6 100644
--- a/deltacat/utils/pyarrow.py
+++ b/deltacat/utils/pyarrow.py
@@ -569,8 +569,8 @@ def s3_file_to_table(
                 **s3_client_kwargs,
             )
 
-        if READER_TYPE_KWARG in kwargs:
-            kwargs.pop(READER_TYPE_KWARG)
+    if READER_TYPE_KWARG in kwargs:
+        kwargs.pop(READER_TYPE_KWARG, None)
 
     filesystem = io
     if s3_url.startswith("s3://"):