From e056294d106b00720e9f1eb5444089efe087d17c Mon Sep 17 00:00:00 2001
From: Daiyi Yang <daiyi@dsensei.app>
Date: Tue, 19 Sep 2023 15:08:48 -0700
Subject: [PATCH] Update docker entry point and make bigquery credential not
 required when starting

---
 Dockerfile                                      |  2 +-
 .../app/data_source/bigquery/bigquery_source.py | 17 +++++++++++------
 backend/run.py                                  |  4 ++++
 3 files changed, 16 insertions(+), 7 deletions(-)
 create mode 100644 backend/run.py

diff --git a/Dockerfile b/Dockerfile
index 53f1057..9629e98 100644
--- a/Dockerfile
+++ b/Dockerfile
@@ -26,4 +26,4 @@ WORKDIR /app/backend
 EXPOSE 5001
 
 # Run Flask application
-CMD . /opt/venv/bin/activate && exec flask run -h 0.0.0.0 -p 5001
+CMD . /opt/venv/bin/activate && exec python run.py
diff --git a/backend/app/data_source/bigquery/bigquery_source.py b/backend/app/data_source/bigquery/bigquery_source.py
index c588c4e..c799433 100644
--- a/backend/app/data_source/bigquery/bigquery_source.py
+++ b/backend/app/data_source/bigquery/bigquery_source.py
@@ -10,15 +10,20 @@
 
 class BigquerySource:
 
-    def __init__(self) -> None:
-        self.client = bigquery.Client()
+    client: bigquery.Client = None
+
+    def get_client(self) -> bigquery.Client:
+        if self.client is None:
+            self.client = bigquery.Client()
+
+        return self.client
 
     @staticmethod
     def convert_field_type(bq_type: str) -> str:
         pass
 
     def get_schema(self, full_name: str) -> BigquerySchema:
-        table = self.client.get_table(full_name)
+        table = self.get_client().get_table(full_name)
 
         selections = ','.join(
             [f'APPROX_COUNT_DISTINCT({field.name}) as {field.name}' for field in table.schema if field.field_type != 'RECORD'])
@@ -59,7 +64,7 @@ def get_schema(self, full_name: str) -> BigquerySchema:
         return schema
 
     def list_dataset(self) -> list[Dataset]:
-        dataset_list_res = self.client.list_datasets()
+        dataset_list_res = self.get_client().list_datasets()
 
         return [Dataset(
             name=dataset.dataset_id,
@@ -68,7 +73,7 @@ def list_dataset(self) -> list[Dataset]:
             for dataset in dataset_list_res]
 
     def list_tables(self, dataset: Dataset = None) -> list[BigquerySchema]:
-        tables = self.client.list_tables(dataset)
+        tables = self.get_client().list_tables(dataset)
         schemas = []
         for row in tables:
             schema = self.get_schema(row.full_table_id)
@@ -85,5 +90,5 @@ def run_queries_in_parallel(self, queries) -> list[RowIterator]:
 
     def run_query(self, query) -> RowIterator:
         # Run the query and return the results
-        query_job = self.client.query(query)
+        query_job = self.get_client().query(query)
         return query_job.result()
diff --git a/backend/run.py b/backend/run.py
new file mode 100644
index 0000000..8c78ddf
--- /dev/null
+++ b/backend/run.py
@@ -0,0 +1,4 @@
+from app import app
+
+if __name__ == '__main__':
+    app.run(debug=True, port=5001, host="::", threaded=False, processes=4)