Fix dataset download problem (#715)

Also add help for the `--timeout` option.

Author: stevemessick

CHANGELOG.md

@@ -1,6 +1,11 @@
 Changelog
 ====
 
+### 1.7.4.1
+
+* Fix a dataset download problem. Datasets that had a license were failing to download.
+* Update the documentation to include code competition submit.
+
 ### 1.7.4
 
 Version 1.7.3 was never released. There were errors in versioning on

kaggle/api/kaggle_api_extended.py

@@ -1737,13 +1737,13 @@ def dataset_download_cli(self,
         raise Exception(response.error_message)
       metadata = response.info
 
-    if 'info' in metadata and 'licenses' in metadata['info']:
+    if metadata and metadata.licenses:
       # license_objs format is like: [{ 'name': 'CC0-1.0' }]
-      license_objs = metadata['info']['licenses']
+      license_objs = metadata.licenses
       licenses = [
-          license_obj['name']
+          license_obj.name
           for license_obj in license_objs
-          if 'name' in license_obj
+          if license_obj.name
       ]
     else:
       licenses = [
@@ -2440,6 +2440,7 @@ def kernels_push(self, folder, timeout=None) -> ApiSaveKernelResponse:
             Parameters
             ==========
             folder: the path of the folder
+            timeout: maximum run time iin seconds
         """
     if not os.path.isdir(folder):
       raise ValueError('Invalid folder: ' + folder)
@@ -2561,6 +2562,7 @@ def kernels_push_cli(self, folder, timeout):
         Parameters
             ==========
             folder: the path of the folder
+            timeout: maximum run time in seconds
         """
     folder = folder or os.getcwd()
     result = self.kernels_push(folder, timeout)

kaggle/cli.py

@@ -759,7 +759,8 @@ def parse_kernels(subparsers):
       '-t',
       '--timeout',
       type=int,
-      dest='timeout')
+      dest='timeout',
+      help=Help.param_kernel_timeout)
   parser_kernels_push._action_groups.append(parser_kernels_push_optional)
   parser_kernels_push.set_defaults(func=api.kernels_push_cli)
 
@@ -1526,6 +1527,9 @@ class Help(object):
   param_kernel_competition = 'Find kernels for a given competition slug'
   param_kernel_dataset = ('Find kernels for a given dataset slug. Format is '
                           '{username/dataset-slug}')
+  param_kernel_timeout = ('Limit the run time of a kernel to the given number '
+                          'of seconds. The global maximum time will not be '
+                          'exceeded.')
   param_kernel_user = 'Find kernels created by a given username'
   # TODO(b/129357583): Pull these from the same spot as the api impl
   param_kernel_language = (

kagglesdk/kaggle_object.py

@@ -275,7 +275,10 @@ def __repr__(self):
     return KaggleObject.to_json(self, ignore_defaults=False)
 
   def __contains__(self, field_name):
-    field = self._get_field(field_name)
+    try:
+      field = self._get_field(field_name)
+    except ValueError:
+      return False
     value = getattr(self, field.private_field_name)
     if field.optional:
       return value is not None

src/kaggle/api/kaggle_api_extended.py

@@ -1737,13 +1737,13 @@ def dataset_download_cli(self,
         raise Exception(response.error_message)
       metadata = response.info
 
-    if 'info' in metadata and 'licenses' in metadata['info']:
+    if metadata and metadata.licenses:
       # license_objs format is like: [{ 'name': 'CC0-1.0' }]
-      license_objs = metadata['info']['licenses']
+      license_objs = metadata.licenses
       licenses = [
-          license_obj['name']
+          license_obj.name
           for license_obj in license_objs
-          if 'name' in license_obj
+          if license_obj.name
       ]
     else:
       licenses = [
@@ -2440,6 +2440,7 @@ def kernels_push(self, folder, timeout=None) -> ApiSaveKernelResponse:
             Parameters
             ==========
             folder: the path of the folder
+            timeout: maximum run time iin seconds
         """
     if not os.path.isdir(folder):
       raise ValueError('Invalid folder: ' + folder)
@@ -2561,6 +2562,7 @@ def kernels_push_cli(self, folder, timeout):
         Parameters
             ==========
             folder: the path of the folder
+            timeout: maximum run time in seconds
         """
     folder = folder or os.getcwd()
     result = self.kernels_push(folder, timeout)

src/kaggle/cli.py

@@ -759,7 +759,8 @@ def parse_kernels(subparsers):
       '-t',
       '--timeout',
       type=int,
-      dest='timeout')
+      dest='timeout',
+      help=Help.param_kernel_timeout)
   parser_kernels_push._action_groups.append(parser_kernels_push_optional)
   parser_kernels_push.set_defaults(func=api.kernels_push_cli)
 
@@ -1526,6 +1527,9 @@ class Help(object):
   param_kernel_competition = 'Find kernels for a given competition slug'
   param_kernel_dataset = ('Find kernels for a given dataset slug. Format is '
                           '{username/dataset-slug}')
+  param_kernel_timeout = ('Limit the run time of a kernel to the given number '
+                          'of seconds. The global maximum time will not be '
+                          'exceeded.')
   param_kernel_user = 'Find kernels created by a given username'
   # TODO(b/129357583): Pull these from the same spot as the api impl
   param_kernel_language = (

src/kagglesdk/kaggle_object.py

@@ -275,7 +275,10 @@ def __repr__(self):
     return KaggleObject.to_json(self, ignore_defaults=False)
 
   def __contains__(self, field_name):
-    field = self._get_field(field_name)
+    try:
+      field = self._get_field(field_name)
+    except ValueError:
+      return False
     value = getattr(self, field.private_field_name)
     if field.optional:
       return value is not None

tests/test_commands.sh

@@ -61,6 +61,7 @@ sed -i s/INSERT_TITLE_HERE/TitleHere/ tests/dataset/dataset-metadata.json
 sed -i s/INSERT_SLUG_HERE/$SLUG/ tests/dataset/dataset-metadata.json
 kaggle d create -p tests/dataset --public -q -t -r skip
 echo "kaggle datasets download"
+kaggle datasets download -d willianoliveiragibin/pixar-films # this one breaks in some environments
 kaggle d download goefft/public-datasets-with-file-types-and-columns -p tmp --unzip -o -q
 kaggle d download goefft/public-datasets-with-file-types-and-columns -f dataset_results.csv -w -q -o
 echo "kaggle datasets version"