From 79a7e8a603e6666732954755553dadbe54370e31 Mon Sep 17 00:00:00 2001
From: Albert Villanova del Moral
 <8515462+albertvillanova@users.noreply.github.com>
Date: Tue, 30 Aug 2022 11:45:18 +0200
Subject: [PATCH 1/2] Support streaming swda dataset

---
 datasets/swda/swda.py | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

diff --git a/datasets/swda/swda.py b/datasets/swda/swda.py
index 4ad179b55e3..9c0b573e732 100644
--- a/datasets/swda/swda.py
+++ b/datasets/swda/swda.py
@@ -435,10 +435,8 @@ def _split_generators(self, dl_manager):
         dl_dir = dl_manager.download_and_extract(_URL)
         # Use swda/ folder.
         data_dir = os.path.join(dl_dir, "swda")
-        # Handle partitions files.
-        urls_to_download = self._URLS
-        # Download extract and return paths of split files.
-        downloaded_files = dl_manager.download_and_extract(urls_to_download)
+        # Handle partitions files: download extract and return paths of split files.
+        downloaded_files = dl_manager.download(self._URLS)
 
         return [
             # Return whole data path and train splits file downloaded path.
@@ -476,7 +474,8 @@ def _generate_examples(self, data_dir, split_file):
         """
 
         # Read in the split file.
-        split_file = io.open(file=split_file, mode="r", encoding="utf-8").read().splitlines()
+        with open(file=split_file, mode="r", encoding="utf-8") as f:
+            split_file = f.read().splitlines()
         # Read in corpus data using split files.
         corpus = CorpusReader(src_dirname=data_dir, split_file=split_file)
         # Generate examples.

From d2b63f6e399cb5d25c3928dfa5fa1b50a24c7eee Mon Sep 17 00:00:00 2001
From: Albert Villanova del Moral
 <8515462+albertvillanova@users.noreply.github.com>
Date: Tue, 30 Aug 2022 12:41:53 +0200
Subject: [PATCH 2/2] Remove unused import

---
 datasets/swda/swda.py | 1 -
 1 file changed, 1 deletion(-)

diff --git a/datasets/swda/swda.py b/datasets/swda/swda.py
index 9c0b573e732..7323f871832 100644
--- a/datasets/swda/swda.py
+++ b/datasets/swda/swda.py
@@ -26,7 +26,6 @@
 import csv
 import datetime
 import glob
-import io
 import os
 import re