Datasets:
fix script to display dataset preview
Browse files- euscrawl.py +16 -16
euscrawl.py
CHANGED
|
@@ -74,26 +74,26 @@ class EusCrawl(datasets.GeneratorBasedBuilder):
|
|
| 74 |
)
|
| 75 |
|
| 76 |
def _split_generators(self, dl_manager):
|
| 77 |
-
path = dl_manager.
|
| 78 |
-
filepath = f"{path}/{_FILEPATH}"
|
| 79 |
return [
|
| 80 |
datasets.SplitGenerator(
|
| 81 |
name=datasets.Split.TRAIN,
|
| 82 |
-
gen_kwargs={"
|
| 83 |
)
|
| 84 |
]
|
| 85 |
|
| 86 |
-
def _generate_examples(self,
|
| 87 |
-
|
| 88 |
-
|
| 89 |
-
|
|
|
|
| 90 |
|
| 91 |
-
|
| 92 |
-
|
| 93 |
-
|
| 94 |
-
|
| 95 |
-
|
| 96 |
-
|
| 97 |
-
|
| 98 |
-
|
| 99 |
-
|
|
|
|
| 74 |
)
|
| 75 |
|
| 76 |
def _split_generators(self, dl_manager):
|
| 77 |
+
path = dl_manager.download(_URL)
|
|
|
|
| 78 |
return [
|
| 79 |
datasets.SplitGenerator(
|
| 80 |
name=datasets.Split.TRAIN,
|
| 81 |
+
gen_kwargs={"filepaths": dl_manager.iter_archive(path)},
|
| 82 |
)
|
| 83 |
]
|
| 84 |
|
| 85 |
+
def _generate_examples(self, filepaths):
|
| 86 |
+
for filepath, file in filepaths:
|
| 87 |
+
if filepath == _FILEPATH:
|
| 88 |
+
for id, line in enumerate(file):
|
| 89 |
+
data = json.loads(line)
|
| 90 |
|
| 91 |
+
# defaut to empty string if field is missing
|
| 92 |
+
yield id, {
|
| 93 |
+
"id": id,
|
| 94 |
+
"title": data.get("title", ""),
|
| 95 |
+
"text": data.get("text", ""),
|
| 96 |
+
"source": data.get("source", ""),
|
| 97 |
+
"license": data.get("license", ""),
|
| 98 |
+
"url": data.get("url", ""),
|
| 99 |
+
}
|