dataflow-core 2.0.3__py3-none-any.whl → 2.0.6__py3-none-any.whl

authenticator/dataflowairflowauthenticator.py

@@ -1,34 +1,19 @@
-from .package.configuration import ConfigurationManager
 from .package.models.database import DatabaseManager
-from .package.models import (
-    user as m_user,
-    session as m_session
-    )
+from dataflow.dataflow import Dataflow
 
 from typing import Any, Callable
 from airflow.www.security import FabAirflowSecurityManagerOverride
+from airflow.configuration import conf
 
 class DataflowAirflowAuthenticator(FabAirflowSecurityManagerOverride):
     def __init__(self, wsgi_app: Callable) -> None:
         self.wsgi_app = wsgi_app
+        self.dataflow = Dataflow()
+        self.airflow_database_url = conf.get("database", "sql_alchemy_conn")
 
-        # Dataflow database configuration
-        self.dataflow_config = ConfigurationManager('/dataflow/app/config/dataflow.cfg')
-        self.dataflow_database_url = self.dataflow_config.get_config_value('database', 'database_url')
-
-        self.dataflow_db_instance = DatabaseManager(self.dataflow_database_url)
-        self.dataflow_db = next(self.dataflow_db_instance.get_session())
-
-        # Airflow database configuration
-        self.airflow_config = ConfigurationManager('airflow.cfg')
-        self.airflow_database_url = self.airflow_config.get_config_value('database', 'sql_alchemy_conn')
-        
         self.airflow_db_instance = DatabaseManager(self.airflow_database_url)
         self.airflow_db = next(self.airflow_db_instance.get_session())
 
-        m_user.Base.metadata.create_all(bind=self.dataflow_db_instance.get_engine())
-        m_session.Base.metadata.create_all(bind=self.dataflow_db_instance.get_engine())
-
     def __call__(self, environ: dict, start_response: Callable) -> Any:
 
         path = environ.get('PATH_INFO', '')
@@ -51,39 +36,28 @@ class DataflowAirflowAuthenticator(FabAirflowSecurityManagerOverride):
             user_session_id = user_session_id.split('=')[1]
 
             # Retrieving user details
-            user_data = self.find_dataflow_user(user_session_id)
+            user_data = self.dataflow.auth(user_session_id)
 
             if user_data is None:
                 raise Exception("No user found for the dataflow_session id")
             
-            user = self.find_user(user_data.user_name)
+            user = self.find_user(user_data["user_name"])
 
             if not user:
-                user_role = self.find_role(user_data.role.title())
-                user = self.add_user(username=user_data.user_name, first_name=self.not_none(user_data.first_name), last_name=self.not_none(user_data.last_name), email=self.not_none(user_data.email), role=user_role, password=self.not_none(user_data.password))
+                user_role = self.find_role(user_data["role"].title())
+                user = self.add_user(
+                    username=user_data["user_name"],
+                    first_name=user_data.get("first_name", ""),
+                    last_name=user_data.get("last_name", ""),
+                    email=user_data.get("email", ""),
+                    role=user_role
+                )
 
             environ['REMOTE_USER'] = user.username
-            self.write_user_id(user_data.user_id)
             return self.wsgi_app(environ, start_response)
 
         except Exception as e:
             return self.wsgi_app(environ, start_response)
-    
-    def not_none(self, value):
-        return value if value is not None else ""
-    
-    def find_dataflow_user(self, user_session_id):
-        """Find user by session_id in dataflow database."""
-        query = self.dataflow_db.query(m_session.Session_table)
-        session = query.filter(m_session.Session_table.session_id == user_session_id).first()
-        if session is None:
-            return None
-
-        user_data = self.dataflow_db.query(m_user.User).filter(m_user.User.user_id == session.user_id).first()
-        if user_data is None:
-            return None
-        
-        return user_data
 
     def find_user(self, username=None):
         """Find user by username or email."""
@@ -106,16 +80,4 @@ class DataflowAirflowAuthenticator(FabAirflowSecurityManagerOverride):
         self.airflow_db.add(user)
         self.airflow_db.commit()
         return user
-
-    def write_user_id(self, user_id):
-        """
-        Write the given user_id to a file named dataflow_user_id.txt.
-        
-        Args:
-            user_id (str): The user ID to be written to the file.
-        """
-        file_name = 'dataflow_user_id.txt'
-        with open(file_name, 'w') as file:
-            file.write(str(user_id))
-
         

authenticator/dataflowsupersetauthenticator.py

@@ -9,39 +9,30 @@ from .package.models import (
     user as m_user,
     session as m_session
     )
+from dataflow.dataflow import Dataflow
 
 class CustomAuthDBView(AuthDBView):
     def __init__(self):
-        self.dataflow_config = ConfigurationManager('/dataflow/app/config/dataflow.cfg')
-        self.dataflow_database_url = self.dataflow_config.get_config_value('database', 'database_url')
-        self.dataflow_db_instance = DatabaseManager(self.dataflow_database_url)
-        self.dataflow_db = next(self.dataflow_db_instance.get_session())
-        m_user.Base.metadata.create_all(bind=self.dataflow_db_instance.get_engine())
-        m_session.Base.metadata.create_all(bind=self.dataflow_db_instance.get_engine())
-
-    def _get_user_id_from_session(self, session_id):
-        query = self.dataflow_db.query(m_session.Session_table).filter(m_session.Session_table.session_id == session_id).first()
-        return query.user_id if query!=None else None
-
-    def _get_user_details_from_user_table(self, user_id):
-        user_details = self.dataflow_db.query(m_user.User).filter(m_user.User.user_id == user_id).first()
-        return user_details if user_details!=None else None
-    
-    def not_none(self, value):
-        return value if value is not None else ""
+        self.dataflow = Dataflow()
 
     @expose('/login/', methods=['GET'])
     def login(self):
         try:
             session_id = request.cookies.get('dataflow_session')
             
-            user_id = self._get_user_id_from_session(session_id)
-            user_details = self._get_user_details_from_user_table(user_id)
-            user = self.appbuilder.sm.find_user(username=user_details.user_name)
+            user_details = self.dataflow.auth(session_id)
+            user = self.appbuilder.sm.find_user(username=user_details['user_name'])
             if user:
                 login_user(user, remember=False)
             else:
-                user = self.appbuilder.sm.add_user(username=self.not_none(user_details.user_name), first_name=self.not_none(user_details.first_name), last_name=self.not_none(user_details.last_name), email=self.not_none(user_details.email), role=self.appbuilder.sm.find_role('Admin'), password=self.not_none(user_details.password))
+                user = self.appbuilder.sm.add_user(
+                    username=user_details['user_name'], 
+                    first_name=user_details.get("first_name", ""),
+                    last_name=user_details.get("last_name", ""), 
+                    email=user_details.get("email", ""), 
+                    role=self.appbuilder.sm.find_role('Admin'), 
+                    password=""
+                )
                 if user:
                     login_user(user, remember=False)
                     

dataflow/__init__.py

@@ -0,0 +1 @@
+from .dataflow import Dataflow

dataflow/dataflow.py

@@ -1,45 +1,40 @@
-import os
-from .configuration import ConfigurationManager
-from .models import (
-    session as m_session,
-    user as m_user,
-)
+import os, requests
 from .models.database import DatabaseManager
 from sqlalchemy.inspection import inspect
 from .utils.aws_secrets_manager import SecretsManagerClient
-import json
+import json, asyncio, pkg_resources
+from authenticator.package.configuration import ConfigurationManager
 
 
 class Dataflow:
     def __init__(self):
-        self.dataflow_config = ConfigurationManager('/dataflow/app/config/dataflow.cfg')
-        self.dataflow_database_url = self.dataflow_config.get_config_value('database', 'database_url')
-
-        self.dataflow_db_instance = DatabaseManager(self.dataflow_database_url)
-        self.dataflow_db = next(self.dataflow_db_instance.get_session())
-
         self.secrets_manager = SecretsManagerClient('us-east-1')
 
-        m_user.Base.metadata.create_all(bind=self.dataflow_db_instance.get_engine())
-        m_session.Base.metadata.create_all(bind=self.dataflow_db_instance.get_engine())
-
     def auth(self, session_id: str):
-        """Find user by session_id in dataflow database."""
+        """Retrieve user information from the auth API."""
         try:
-            query = self.dataflow_db.query(m_session.Session_table)
-            session = query.filter(m_session.Session_table.session_id == session_id).first()
-            if session is None:
-                return False
-
-            user_data = self.dataflow_db.query(m_user.User).filter(m_user.User.user_id == session.user_id).first()
-            if user_data is None:
-                return False
+            dataflow_config = ConfigurationManager('/dataflow/app/auth_config/dataflow_auth.cfg')
+            auth_api = dataflow_config.get_config_value('auth', 'ui_auth_api')
+            response = requests.get(
+                auth_api,
+                cookies={"dataflow_session": session_id, "jupyterhub-hub-login": ""}
+            )
             
-            user_dict = {"user_name": user_data.user_name, "name": f"{user_data.first_name} {user_data.last_name}", "email": user_data.email, "role": user_data.role}
+            if response.status_code != 200:
+                return response.json()
+            
+            user_data = response.json()
+            user_dict = {
+                "user_name": user_data["user_name"], 
+                "first_name": user_data["first_name"],
+                "last_name": user_data["last_name"] if user_data.get("last_name") else "",
+                "email": user_data["email"],
+                "role": user_data["role"]
+            }
             return user_dict
-        
+                  
         except Exception as e:
-            return False
+            return e
     
     def variable(self, variable_name: str):
         """Get variable value from secrets manager."""
@@ -91,3 +86,42 @@ class Dataflow:
         
         except Exception as e:
             return None
+        
+    async def create_env(self, env_name, py_version, py_requirements, status, env_version=None):
+        """
+        Creates a conda environment at the specified path and installs libraries in one command.
+        """
+        config = ConfigurationManager('/dataflow/app/config/dataflow.cfg')
+        status = status.lower()
+        if status == "published":
+            env_base_path = config.get_config_value('paths', 'published_env_path')
+            conda_env_path = os.path.join(env_base_path, env_name)
+        else:
+            env_base_path = config.get_config_value('paths', 'drafts_env_path')
+            conda_env_path = os.path.join(env_base_path, env_name, f"{env_name}_v{env_version}")
+        try:
+            if not os.path.exists(conda_env_path):
+                os.makedirs(conda_env_path, exist_ok=True)
+
+            py_requirements = ",".join(py_requirements) 
+
+            script_path = pkg_resources.resource_filename('dataflow', 'scripts/create_environment.sh')
+    
+            # Make the script executable
+            os.chmod(script_path, 0o755)
+            
+            # Prepare command with arguments
+            command = ["bash", script_path, py_requirements, conda_env_path, py_version]
+
+            process = await asyncio.create_subprocess_exec(
+                *command, 
+                stdout=asyncio.subprocess.PIPE, 
+                stderr=asyncio.subprocess.PIPE
+            )
+            
+            return process
+        except OSError as e:
+            print(f"OS error while creating {conda_env_path}: {e}")
+        except Exception as e:
+            print(f"Unexpected error while creating {conda_env_path}: {e}")
+            return {"error": str(e)}

dataflow/scripts/create_environment.sh

@@ -0,0 +1,34 @@
+#!/bin/bash
+set -e
+
+IFS=',' read -r -a  libraries <<< $1
+conda_env_path=$2
+py_version=$3
+
+# Use an isolated conda package cache to avoid concurrency issues
+export CONDA_PKGS_DIRS=$(mktemp -d)
+# to delete conda package cache after script finishes
+trap 'rm -rf "$CONDA_PKGS_DIRS"' EXIT
+
+# 1. Creating conda environment
+conda create --prefix ${conda_env_path} --yes python=${py_version}                         
+${conda_env_path}/bin/pip install --root-user-action ignore ${libraries[@]}      
+
+# 3. Install Dataflow Airflow to a separate path in environment 
+${conda_env_path}/bin/pip install \
+    --force-reinstall --root-user-action ignore \
+    --no-warn-conflicts dataflow-airflow==2.10.3 \
+    --target ${conda_env_path}/bin/airflow-libraries/ \
+    --constraint https://raw.githubusercontent.com//apache/airflow/constraints-2.10.3/constraints-${py_version}.txt || true 
+
+files=(
+    ${conda_env_path}/lib/python${py_version}/site-packages/dbt/config/profile.py 
+    ${conda_env_path}/lib/python${py_version}/site-packages/dbt/task/debug.py
+)
+for file in ${files[@]}
+do      
+    awk '{gsub("from dbt.clients.yaml_helper import load_yaml_text", "from dbt.dataflow_config.secrets_manager import load_yaml_text"); print}' $file > temp 
+    mv temp $file
+done
+
+echo "Environment Creation Successful"

{dataflow_core-2.0.3.dist-info → dataflow_core-2.0.6.dist-info}/METADATA

@@ -1,11 +1,18 @@
 Metadata-Version: 2.1
 Name: dataflow-core
-Version: 2.0.3
+Version: 2.0.6
 Summary: Dataflow core package
+Home-page: UNKNOWN
 Author: Dataflow
-Author-email: 
+Author-email: UNKNOWN
+License: UNKNOWN
+Platform: UNKNOWN
 Requires-Dist: sqlalchemy
 Requires-Dist: boto3
 Requires-Dist: psycopg2-binary
 Requires-Dist: pymysql
+Requires-Dist: requests
+
+UNKNOWN
+
 

{dataflow_core-2.0.3.dist-info → dataflow_core-2.0.6.dist-info}/RECORD

@@ -1,24 +1,25 @@
 authenticator/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-authenticator/dataflowairflowauthenticator.py,sha256=wzd807UAb5ojIVMO8FVHQnNi3Tc78NjkhbDSfzgo2Qg,4790
+authenticator/dataflowairflowauthenticator.py,sha256=PcGlL2cq5EA9RdtClyrOupvNXxk6h54UPQcQ-g4VQtA,3117
 authenticator/dataflowhubauthenticator.py,sha256=UgjZiXSP-hlbciyJVb-QIgUjS3_EgX0WIvbhD563rVw,2918
-authenticator/dataflowsupersetauthenticator.py,sha256=UNgem68eg8G-1uAok2yVc6NvUeXaJ-9MPmdpfSyr9Jg,2759
+authenticator/dataflowsupersetauthenticator.py,sha256=M2pNpIdmWwzJVuMQ6trXLWQT4HpcYT84bC93uOSm0fQ,1772
 authenticator/package/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 authenticator/package/configuration.py,sha256=7To6XwH1eESiYp39eqPcswXWwrdBUdPF6xN6WnazOF0,663
 authenticator/package/models/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 authenticator/package/models/database.py,sha256=y09pqnglsBBtDZlyhvqDAlpUSFovwAzBAi6jOYl_XNk,896
 authenticator/package/models/session.py,sha256=j6PhbrTMJxEkeDT4Vf5SqGtM_LI_vZy9O4vxn6LtIbc,495
 authenticator/package/models/user.py,sha256=IYogp_vt0yDBG5i936uNPjgTis77VYPzITn9XpQUIyw,788
-dataflow/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+dataflow/__init__.py,sha256=WTRg8HMpMWSgxYJ9ZGVldx4k07fAbta3mBmZ1hG9mWE,30
 dataflow/configuration.py,sha256=7To6XwH1eESiYp39eqPcswXWwrdBUdPF6xN6WnazOF0,663
-dataflow/dataflow.py,sha256=CAn47BkdUwmTWe-kTSFKfRZYnjx29MG0SrYsXpvSkuo,3726
+dataflow/dataflow.py,sha256=viEKxu1bVFtenbPx8etCb7Gf-Hbwv0KvkE7occVbX8I,5136
 dataflow/models/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 dataflow/models/database.py,sha256=y09pqnglsBBtDZlyhvqDAlpUSFovwAzBAi6jOYl_XNk,896
 dataflow/models/session.py,sha256=C9crPh6ZDFuL27hZ_zhUXDZZ0ZiIDE8ZD19O_4WPw-I,488
 dataflow/models/user.py,sha256=IYogp_vt0yDBG5i936uNPjgTis77VYPzITn9XpQUIyw,788
+dataflow/scripts/create_environment.sh,sha256=r4cKRTMf1G9pXx1fqMpzkvd-ROzUH0Qp16f9t84Nh5c,1300
 dataflow/utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 dataflow/utils/aws_secrets_manager.py,sha256=FqHm3YRynv580FpFsS0RfI1MSGY5aq-V7t4blpiYsS4,2588
-dataflow_core-2.0.3.dist-info/METADATA,sha256=N3tOAUkMWoSaOXpJn6UkzZ-p9Rpfb51xYVmAOEmJBX8,222
-dataflow_core-2.0.3.dist-info/WHEEL,sha256=Wyh-_nZ0DJYolHNn1_hMa4lM7uDedD_RGVwbmTjyItk,91
-dataflow_core-2.0.3.dist-info/entry_points.txt,sha256=ppj_EIbYrJJwCPg1kfdsZk5q1N-Ejfis1neYrnjhO8o,117
-dataflow_core-2.0.3.dist-info/top_level.txt,sha256=SZsUOpSCK9ntUy-3Tusxzf5A2e8ebwD8vouPb1dPt_8,23
-dataflow_core-2.0.3.dist-info/RECORD,,
+dataflow_core-2.0.6.dist-info/METADATA,sha256=BCXu55SZS71k7ZQ2ilfFn_nimogjDv0SDKtrDyje5iw,317
+dataflow_core-2.0.6.dist-info/WHEEL,sha256=tZoeGjtWxWRfdplE7E3d45VPlLNQnvbKiYnx7gwAy8A,92
+dataflow_core-2.0.6.dist-info/entry_points.txt,sha256=lDLG2MMWlKfkqsVWFghF7sx-yEvM2xqMmHE7rMTysE4,118
+dataflow_core-2.0.6.dist-info/top_level.txt,sha256=SZsUOpSCK9ntUy-3Tusxzf5A2e8ebwD8vouPb1dPt_8,23
+dataflow_core-2.0.6.dist-info/RECORD,,

{dataflow_core-2.0.3.dist-info → dataflow_core-2.0.6.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: setuptools (71.1.0)
+Generator: bdist_wheel (0.45.1)
 Root-Is-Purelib: true
 Tag: py3-none-any
 

{dataflow_core-2.0.3.dist-info → dataflow_core-2.0.6.dist-info}/entry_points.txt

@@ -1,2 +1,3 @@
 [jupyterhub.authenticators]
 dataflow_authenticator = authenticator.dataflowhubauthenticator:DataflowHubAuthenticator
+