Merge pull request #11 from GSK-Biostatistics/feature/hive-loader

Feature/hive loader

data_loaders/__init__.py

@@ -1,3 +1,4 @@
 from data_loaders.file_data_loader import FileDataLoader
 from data_loaders.sql_server_data_loader import SQLServerDataLoader
 from data_loaders.azure_data_loader import AzureDataLoader
+from data_loaders.hive_data_loader import HiveDataLoader

data_loaders/hive_data_loader.py

@@ -0,0 +1,43 @@
+import neointerface
+import os
+import jaydebeapi
+import pandas as pd
+
+
+class HiveDataLoader:
+    def __init__(self,
+                 db_host=os.environ.get("HIVE_SERVER"),
+                 schema=os.environ.get("HIVE_SCHEMA"),
+                 user=os.environ.get("HIVE_USER_ID"),
+                 password=os.environ.get("HIVE_PASSWORD"),
+                 driver_path=os.environ.get("HIVE_JDBC_DRIVER_PATH"),
+                 driver_class=os.environ.get("HIVE_JDBC_DRIVER"),
+                 *args, **kwargs):
+        self._connection = None
+        self._schema = schema
+        self._driver_class = driver_class
+        self._db_host = db_host
+        self._user = user
+        self._password = password
+        self._driver_path = driver_path
+
+    def open(self):
+        self._connection = jaydebeapi.connect(self._driver_class, f'jdbc:hive2://{self._db_host}/{self._schema}',
+                                              {'UID': self._user, 'PWD': self._password, 'AuthMech': "3"}, self._driver_path)
+
+    def query(self, statement: str):
+        cursor = self._connection.cursor()
+        cursor.execute(statement)
+        results = cursor.fetchall()
+        names = [item[0] for item in cursor.description]
+        df = pd.DataFrame(results)
+        df.columns = names
+        cursor.close()
+        return df
+
+    def get_tables(self):
+        tables_df = self.query(f'show tables from {self._schema}')
+        return tables_df
+
+    def close(self):
+        self._connection.close()

requirements-local.txt

@@ -10,3 +10,4 @@ azure-identity==1.6.0
 datacompy==0.7.2
 colorlog==6.6.0
 PyGithub==1.55
+jaydebeapi==1.2.3

requirements.txt

@@ -10,3 +10,4 @@ azure-identity==1.6.0
 datacompy==0.7.2
 colorlog==6.6.0
 PyGithub==1.55
+jaydebeapi==1.2.3

setup.py

@@ -30,7 +30,7 @@ def read_text(file_name: str):
 
 setuptools.setup(
     name="tab2neo",                           # This is the name of the package
-    version="1.0.1.0",
+    version="1.1.0.0",                      # Release.Major Feature.Minor Feature.Bug Fix
     author="Alexey Kuznetsov",              # Full name of the author
     description="Clinical Linked Data: High-level Python classes to load, model and reshape tabular data imported into Neo4j database",
     long_description=long_description,      # Long description read from the the readme file