Fix streaming TAR archive (#1)

- Fix streaming TAR archive (95b34272712a66e48e0535333f6cd2e2a6fb0171)
- Fix type hint (bdaf5520d4933615eb694fb9fe120428a4e1773a)


Co-authored-by: Albert Villanova <albertvillanova@users.noreply.huggingface.co>

spl_adr_200db.py

@@ -57,8 +57,7 @@ https://bionlp.nlm.nih.gov/tac2017adversereactions/
 import xml.etree.ElementTree as ET
 from collections import defaultdict
 from itertools import accumulate
-from pathlib import Path
-from typing import Dict, List, Tuple
+from typing import BinaryIO, Dict, Iterable, List, Tuple
 
 import datasets
 
@@ -228,15 +227,13 @@ class SplAdr200DBDataset(datasets.GeneratorBasedBuilder):
 
         urls = _URLS[_DATASETNAME][subset_name]
 
-        data_dir = dl_manager.download_and_extract(urls)
-
-        data_files = (Path(data_dir) / f"{subset_name}_xml").glob("*.xml")
+        data_dir = dl_manager.download(urls)
 
         return [
             datasets.SplitGenerator(
                 name=datasets.Split.TRAIN,
                 gen_kwargs={
-                    "filepaths": tuple(data_files),
+                    "filepaths": dl_manager.iter_archive(data_dir),
                 },
             ),
         ]
@@ -381,11 +378,11 @@ class SplAdr200DBDataset(datasets.GeneratorBasedBuilder):
             "coreferences": [],
         }
 
-    def _generate_examples(self, filepaths: Tuple[Path]) -> Tuple[int, Dict]:
+    def _generate_examples(self, filepaths: Iterable[Tuple[str, BinaryIO]]) -> Tuple[int, Dict]:
         """Yields examples as (key, example) tuples."""
 
-        for file_index, drug_filename in enumerate(filepaths):
-            element_tree = ET.parse(drug_filename)
+        for file_index, (drug_filename, drug_file) in enumerate(filepaths):
+            element_tree = ET.parse(drug_file)
 
             if self.config.schema == "source":
                 features = self._source_features_from_xml(