Rename kaggle.py to kaggle_gcp.py which will then be imported into the kaggle namespace by the Worker.

Author: vimota

Dockerfile

@@ -505,7 +505,7 @@ RUN pip install --upgrade dask && \
 
 # Add BigQuery client proxy settings
 ENV PYTHONUSERBASE "/root/.local"
-ADD patches/kaggle.py /root/.local/lib/python3.6/site-packages/kaggle.py
+ADD patches/kaggle_gcp.py /root/.local/lib/python3.6/site-packages/kaggle_gcp.py
 ADD patches/sitecustomize.py /root/.local/lib/python3.6/site-packages/sitecustomize.py
 
 # Set backend for matplotlib

patches/kaggle.py patches/kaggle_gcp.pypatches/kaggle.py renamed to patches/kaggle_gcp.py

@@ -38,10 +38,7 @@ class PublicBigqueryClient(bigquery.client.Client):
     client = PublicBigqueryClient()
     """
 
-    def __init__(self, project=None):
-        if project:
-            raise Exception("In order to query a private BigQuery project, please connect a GCP account. "
-                            "Otherwise do not specify a project to use Kaggle's public dataset BigQuery integration.")
+    def __init__(self):
         data_proxy_project = os.getenv("KAGGLE_DATA_PROXY_PROJECT")
         anon_credentials = credentials.AnonymousCredentials()
         anon_credentials.refresh = lambda *args: None

patches/sitecustomize.py

@@ -6,7 +6,7 @@
     from google.auth import credentials
     from google.cloud import bigquery
     from google.cloud.bigquery._http import Connection
-    from kaggle import PublicBigqueryClient
+    from kaggle_gcp import PublicBigqueryClient
 
     def monkeypatch_bq(bq_client, *args, **kwargs):
         data_proxy_project = os.getenv("KAGGLE_DATA_PROXY_PROJECT")

tests/test_bigquery.py

@@ -8,7 +8,7 @@
 
 from google.cloud import bigquery
 from google.auth.exceptions import DefaultCredentialsError
-from kaggle import KaggleKernelCredentials, PublicBigqueryClient
+from kaggle_gcp import KaggleKernelCredentials, PublicBigqueryClient
 
 
 class TestBigQuery(unittest.TestCase):