Renamed a new arg to fingerprint

Author: simonreise

src/datasets/arrow_dataset.py

@@ -1047,7 +1047,7 @@ def from_generator(
         gen_kwargs: Optional[dict] = None,
         num_proc: Optional[int] = None,
         split: NamedSplit = Split.TRAIN,
-        dataset_id_suffix: Optional[str] = None,
+        fingerprint: Optional[str] = None,
         **kwargs,
     ):
         """Create a Dataset from a generator.
@@ -1074,9 +1074,9 @@ def from_generator(
                 Split name to be assigned to the dataset.
 
                 <Added version="2.21.0"/>
-            dataset_id_suffix (`str`, *optional*):
-                Suffix that will be used to generate dataset ID.
-                By default `dataset_id_suffix` is generated by hashing all the args which can be slow in case of a large dataset.
+            fingerprint (`str`, *optional*):
+                Fingerprint that will be used to generate dataset ID.
+                By default `fingerprint` is generated by hashing all the args which can be slow in case of a large dataset.
 
                 <Added version="3.6.0"/>
             **kwargs (additional keyword arguments):
@@ -1116,7 +1116,7 @@ def from_generator(
             gen_kwargs=gen_kwargs,
             num_proc=num_proc,
             split=split,
-            dataset_id_suffix=dataset_id_suffix,
+            fingerprint=fingerprint,
             **kwargs,
         ).read()
 

src/datasets/builder.py

@@ -141,6 +141,7 @@ def create_config_id(
         self,
         config_kwargs: dict,
         custom_features: Optional[Features] = None,
+        fingerprint: Optional[str] = None,
     ) -> str:
         """
         The config id is used to build the cache directory.
@@ -155,8 +156,9 @@ def create_config_id(
         """
         # Possibly add a suffix to the name to handle custom features/data_files/config_kwargs
         suffix: Optional[str] = None
-        if "dataset_id_suffix" in config_kwargs and config_kwargs["dataset_id_suffix"] is not None:
-            suffix = config_kwargs["dataset_id_suffix"]
+
+        if fingerprint is not None:
+           suffix = fingerprint
         else:
             config_kwargs_to_add_to_suffix = config_kwargs.copy()
             # name and version are already used to build the cache directory
@@ -316,6 +318,7 @@ def __init__(
         data_dir: Optional[str] = None,
         storage_options: Optional[dict] = None,
         writer_batch_size: Optional[int] = None,
+        fingerprint: Optional[str] = None,
         **config_kwargs,
     ):
         # DatasetBuilder name
@@ -346,6 +349,7 @@ def __init__(
         self.config, self.config_id = self._create_builder_config(
             config_name=config_name,
             custom_features=features,
+            fingerprint=fingerprint,
             **config_kwargs,
         )
 
@@ -536,7 +540,7 @@ def get_exported_dataset_info(self) -> DatasetInfo:
         return self.get_all_exported_dataset_infos().get(self.config.name, DatasetInfo())
 
     def _create_builder_config(
-        self, config_name=None, custom_features=None, **config_kwargs
+        self, config_name=None, custom_features=None, fingerprint=None, **config_kwargs
     ) -> tuple[BuilderConfig, str]:
         """Create and validate BuilderConfig object as well as a unique config id for this config.
         Raises ValueError if there are multiple builder configs and config_name and DEFAULT_CONFIG_NAME are None.
@@ -607,6 +611,7 @@ def _create_builder_config(
         config_id = builder_config.create_config_id(
             config_kwargs,
             custom_features=custom_features,
+            fingerprint=fingerprint,
         )
         is_custom = (config_id not in self.builder_configs) and config_id != "default"
         if is_custom:

src/datasets/io/generator.py

@@ -16,7 +16,7 @@ def __init__(
         gen_kwargs: Optional[dict] = None,
         num_proc: Optional[int] = None,
         split: NamedSplit = Split.TRAIN,
-        dataset_id_suffix: Optional[str] = None,
+        fingerprint: Optional[str] = None,
         **kwargs,
     ):
         super().__init__(
@@ -33,7 +33,7 @@ def __init__(
             generator=generator,
             gen_kwargs=gen_kwargs,
             split=split,
-            dataset_id_suffix=dataset_id_suffix,
+            fingerprint=fingerprint,
             **kwargs,
         )
 

src/datasets/packaged_modules/generator/generator.py

@@ -10,7 +10,7 @@ class GeneratorConfig(datasets.BuilderConfig):
     gen_kwargs: Optional[dict] = None
     features: Optional[datasets.Features] = None
     split: datasets.NamedSplit = datasets.Split.TRAIN
-    dataset_id_suffix: Optional[str] = None
+    fingerprint: Optional[str] = None
 
     def __post_init__(self):
         super().__post_init__()