offline mode for firewalled envs (part 2) (#10569)

* more readable test

* add all the missing places

* one more nltk

* better exception check

* revert

src/transformers/feature_extraction_utils.py

@@ -35,6 +35,7 @@ from .file_utils import (
     cached_path,
     hf_bucket_url,
     is_flax_available,
+    is_offline_mode,
     is_remote_url,
     is_tf_available,
     is_torch_available,
@@ -342,6 +343,10 @@ class PreTrainedFeatureExtractor:
         local_files_only = kwargs.pop("local_files_only", False)
         revision = kwargs.pop("revision", None)
 
+        if is_offline_mode() and not local_files_only:
+            logger.info("Offline mode: forcing local_files_only=True")
+            local_files_only = True
+
         pretrained_model_name_or_path = str(pretrained_model_name_or_path)
         if os.path.isdir(pretrained_model_name_or_path):
             feature_extractor_file = os.path.join(pretrained_model_name_or_path, FEATURE_EXTRACTOR_NAME)

src/transformers/file_utils.py

@@ -1105,6 +1105,10 @@ def cached_path(
     if isinstance(cache_dir, Path):
         cache_dir = str(cache_dir)
 
+    if is_offline_mode() and not local_files_only:
+        logger.info("Offline mode: forcing local_files_only=True")
+        local_files_only = True
+
     if is_remote_url(url_or_filename):
         # URL, so get it from the cache (downloading if necessary)
         output_path = get_from_cache(

src/transformers/modeling_flax_utils.py

@@ -28,7 +28,7 @@ from flax.traverse_util import flatten_dict, unflatten_dict
 from jax.random import PRNGKey
 
 from .configuration_utils import PretrainedConfig
-from .file_utils import FLAX_WEIGHTS_NAME, WEIGHTS_NAME, cached_path, hf_bucket_url, is_remote_url
+from .file_utils import FLAX_WEIGHTS_NAME, WEIGHTS_NAME, cached_path, hf_bucket_url, is_offline_mode, is_remote_url
 from .utils import logging
 
 
@@ -229,6 +229,10 @@ class FlaxPreTrainedModel(ABC):
         use_auth_token = kwargs.pop("use_auth_token", None)
         revision = kwargs.pop("revision", None)
 
+        if is_offline_mode() and not local_files_only:
+            logger.info("Offline mode: forcing local_files_only=True")
+            local_files_only = True
+
         # Load config if we don't provide a configuration
         if not isinstance(config, PretrainedConfig):
             config_path = config if config is not None else pretrained_model_name_or_path

src/transformers/modeling_tf_utils.py

@@ -36,6 +36,7 @@ from .file_utils import (
     ModelOutput,
     cached_path,
     hf_bucket_url,
+    is_offline_mode,
     is_remote_url,
 )
 from .generation_tf_utils import TFGenerationMixin
@@ -1151,6 +1152,10 @@ class TFPreTrainedModel(tf.keras.Model, TFModelUtilsMixin, TFGenerationMixin):
         revision = kwargs.pop("revision", None)
         mirror = kwargs.pop("mirror", None)
 
+        if is_offline_mode() and not local_files_only:
+            logger.info("Offline mode: forcing local_files_only=True")
+            local_files_only = True
+
         # Load config if we don't provide a configuration
         if not isinstance(config, PretrainedConfig):
             config_path = config if config is not None else pretrained_model_name_or_path