Fix reading from interleaved files in the case there is one file

mc4_nl_cleaned.py

@@ -195,20 +195,19 @@ class Mc4(datasets.GeneratorBasedBuilder):
         args = [iter(iterable)] * n
         return zip_longest(*args, fillvalue=fillvalue)
 
+    @staticmethod
+    def gzip_open(filepath):
+        if filepath:
+            return gzip.open(open(filepath, "rb"), "rt", encoding="utf-8")
+
     def _generate_examples(self, filepaths):
         """This function returns the examples in the raw (text) form by iterating on all the files."""
         id_ = 0
-        for filepath1, filepath2 in self.grouper(filepaths, 2, None):
-            logger.info(f"Generating examples from {filepath1} and {filepath2}")
-            with gzip.open(
-                open(filepath1, "rb"), "rt", encoding="utf-8"
-            ) as f1, gzip.open(open(filepath2, "rb"), "rt", encoding="utf-8") as f2:
-                for line1, line2 in zip(f1, f2):
-                    if line1:
-                        example = json.loads(line1)
-                        yield id_, example
-                        id_ += 1
-                    if line2:
-                        example = json.loads(line2)
-                        yield id_, example
-                        id_ += 1
+        for files in self.grouper(filepaths, 2, None):
+            logger.info(f"Generating examples from {files}")
+            gzip_iters = [self.gzip_open(file) for file in files if file is not None]
+            for lines in zip(*gzip_iters):
+                for line in lines:
+                    example = json.loads(line)
+                    yield id_, example
+                    id_ += 1

test_mc4_nl_cleaned.py

@@ -0,0 +1,57 @@
+from itertools import chain
+
+import pytest
+from datasets import load_dataset
+
+
+def test_streaming_dataset():
+    datasets = load_dataset("./mc4_nl_cleaned.py", "tiny", streaming=True)
+    assert list(datasets.keys()) == ["train", "validation"]
+
+    train_ds = datasets["train"]
+    val_ds = datasets["validation"]
+
+    i = iter(train_ds)
+    e = next(i)
+    assert len(e["text"]) == 1264
+    e = next(i)
+    assert len(e["text"]) == 1091
+
+    assert sum(1 for _ in iter(val_ds)) == 16189
+
+
+def test_batch_dataset():
+    datasets = load_dataset(
+        "./mc4_nl_cleaned.py",
+        "micro",
+    )
+    assert list(datasets.keys()) == ["train", "validation"]
+
+    train_ds = datasets["train"]
+    val_ds = datasets["validation"]
+
+    e = next(iter(train_ds))
+    assert len(e["text"]) == 1264
+
+    assert train_ds.num_rows == 125938
+    assert val_ds.num_rows == 16189
+
+
+def test_nl_en():
+    datasets = load_dataset(
+        "./mc4_nl_cleaned.py",
+        "micro_en_nl",
+    )
+    assert list(datasets.keys()) == ["train", "validation"]
+
+    train_ds = datasets["train"]
+    val_ds = datasets["validation"]
+
+    i = iter(train_ds)
+    e = next(i)
+    assert len(e["text"]) == 1264  ## Length of 'Japanse bedrijven zijn ...'
+    e = next(i)
+    assert len(e["text"]) == 747  ## Length of 'Beginners BBQ class ...'
+
+    assert train_ds.num_rows == 251900
+    assert val_ds.num_rows == 32378