Update wikipedia.py

wikipedia.py

@@ -909,6 +909,16 @@ class WikipediaConfig(datasets.BuilderConfig):
 _DATE = "20220301"
 
 
+PROCESSED_CONFIG_TO_NUM_SHARDS = {
+    "20220301.de": 18,
+    "20220301.en": 41,
+    "20220301.fr": 15,
+    "20220301.frr": 1,
+    "20220301.it": 10,
+    "20220301.simple": 1,
+}
+
+
 class Wikipedia(datasets.ArrowBasedBuilder):
     """Wikipedia dataset."""
 
@@ -951,13 +961,16 @@ class Wikipedia(datasets.ArrowBasedBuilder):
         )
 
     def _split_generators(self, dl_manager):
-        processed_data_dir = os.path.join(dl_manager._base_path, "data", self.config.name)
-        is_processed = os.path.exists(processed_data_dir)
-        if is_processed:
-            parquet_urls = sorted(os.path.join(processed_data_dir, parquet_file) for parquet_file in os.listdir(processed_data_dir))
+        if self.config.name in PROCESSED_CONFIG_TO_NUM_SHARDS:
+            num_shards = PROCESSED_CONFIG_TO_NUM_SHARDS[self.config.name]
+            parquet_urls = [
+                os.path.join("data", self.config.name, f"train-{i:05d}-{num_shards:05d}.parquet")
+                for i in range(num_shards)
+            ]
             # Use dictionary since testing mock always returns the same result.
             downloaded_files = dl_manager.download({"parquet": parquet_urls})
             files = downloaded_files["parquet"]
+            is_processed = True
         else:
             def _base_url(lang):
                 return _BASE_URL_TMPL.format(lang=lang.replace("-", "_"), date=self.config.date)
@@ -989,6 +1002,7 @@ class Wikipedia(datasets.ArrowBasedBuilder):
             # Use dictionary since testing mock always returns the same result.
             downloaded_files = dl_manager.download({"xml": xml_urls})
             files = downloaded_files["xml"]
+            is_processed = False
 
         return [
             datasets.SplitGenerator(  # pylint:disable=g-complex-comprehension