fix tests to use new exceptions

tests/integration/io/parquet/test_reads_public_data.py

@@ -7,6 +7,7 @@
 from pyarrow import parquet as pq
 
 import daft
+from daft.exceptions import ConnectTimeoutError, ReadTimeoutError
 from daft.filesystem import get_filesystem, get_protocol_from_path
 from daft.table import MicroPartition, Table
 
@@ -413,7 +414,7 @@ def test_connect_timeout(multithreaded_io):
         )
     )
 
-    with pytest.raises(ValueError, match="HTTP connect timeout"):
+    with pytest.raises((ReadTimeoutError, ConnectTimeoutError), match=f"timed out when trying to connect to {url}"):
         MicroPartition.read_parquet(url, io_config=connect_timeout_config, multithreaded_io=multithreaded_io).to_arrow()
 
 
@@ -434,5 +435,5 @@ def test_read_timeout(multithreaded_io):
         )
     )
 
-    with pytest.raises(ValueError, match="HTTP read timeout"):
+    with pytest.raises((ReadTimeoutError, ConnectTimeoutError), match=f"timed out when trying to connect to {url}"):
         MicroPartition.read_parquet(url, io_config=read_timeout_config, multithreaded_io=multithreaded_io).to_arrow()

tests/integration/io/test_url_download_public_aws_s3.py

@@ -3,6 +3,7 @@
 import pytest
 
 import daft
+from daft.exceptions import ConnectTimeoutError, ReadTimeoutError
 
 
 @pytest.mark.integration()
@@ -81,7 +82,7 @@ def test_url_download_aws_s3_public_bucket_native_downloader_with_connect_timeou
         )
     )
 
-    with pytest.raises(ValueError, match="HTTP connect timeout"):
+    with pytest.raises((ReadTimeoutError, ConnectTimeoutError), match="timed out when trying to connect to"):
         df = df.with_column(
             "data", df["urls"].url.download(io_config=connect_timeout_config, use_native_downloader=True)
         ).collect()
@@ -101,7 +102,7 @@ def test_url_download_aws_s3_public_bucket_native_downloader_with_read_timeout(s
         )
     )
 
-    with pytest.raises(ValueError, match="HTTP read timeout"):
+    with pytest.raises((ReadTimeoutError, ConnectTimeoutError), match="timed out when trying to connect to"):
         df = df.with_column(
             "data", df["urls"].url.download(io_config=read_timeout_config, use_native_downloader=True)
         ).collect()