Add Azure Blob Storage driver support

saberraz · claude · saberraz · commit 89402dfd65b0 · 2026-03-19T12:55:28.000+01:00
Implements AzureBlobDriver following the same pattern as MinioDriver,
allowing media files to be synced to Azure Blob Storage containers.

Changes:
- drivers.py: Add DriverType.AZURE enum value and AzureBlobDriver class
  using azure-storage-blob SDK; supports optional blob_path_prefix
- config.py: Add AZURE to supported driver check and validate azure_blob
  config block (account_name, account_key, container required)
- config.yaml.default: Add azure_blob config section with all options
- Pipfile: Add azure-storage-blob ~=12.0 dependency
- test/conftest.py: Add TEST_AZURE_STORAGE_* env vars and reset defaults
- test/test_sync.py: Add test_azure_blob_backend covering invalid config,
  basic upload verification, and blob_path_prefix behaviour

Co-Authored-By: Claude Sonnet 4.6 &lt;noreply@anthropic.com&gt;
diff --git a/Pipfile b/Pipfile
@@ -12,6 +12,7 @@ minio = "~=7.1"
 mergin-client = "==0.9.3"
 dynaconf = {extras = ["ini"],version = "~=3.1"}
 google-api-python-client = "==2.24"
+azure-storage-blob = "~=12.0"
 
 [requires]
 python_version = "3"
diff --git a/config.py b/config.py
@@ -33,6 +33,7 @@ def validate_config(config):
         config.driver == DriverType.LOCAL
         or config.driver == DriverType.MINIO
         or config.driver == DriverType.GOOGLE_DRIVE
+        or config.driver == DriverType.AZURE
     ):
         raise ConfigError("Config error: Unsupported driver")
 
@@ -78,6 +79,17 @@ def validate_config(config):
     ):
         raise ConfigError("Config error: Incorrect GoogleDrive driver settings")
 
+    if config.driver == DriverType.AZURE and not (
+        hasattr(config, "azure_blob")
+        and hasattr(config.azure_blob, "account_name")
+        and hasattr(config.azure_blob, "account_key")
+        and hasattr(config.azure_blob, "container")
+        and config.azure_blob.account_name
+        and config.azure_blob.account_key
+        and config.azure_blob.container
+    ):
+        raise ConfigError("Config error: Incorrect Azure Blob Storage driver settings")
+
 
 def update_config_path(
     path_param: str,
diff --git a/config.yaml.default b/config.yaml.default
@@ -25,10 +25,16 @@ minio:
   bucket_subpath:
 
 google_drive:
-  service_account_file: 
+  service_account_file:
   folder:
   share_with:
 
+azure_blob:
+  account_name:
+  account_key:
+  container:
+  blob_path_prefix:
+
 references:
   - file: survey.gpkg
     table: notes
diff --git a/drivers.py b/drivers.py
@@ -21,11 +21,15 @@
 from googleapiclient.discovery import build, Resource
 from googleapiclient.http import MediaFileUpload
 
+from azure.storage.blob import BlobServiceClient
+from azure.core.exceptions import AzureError
+
 
 class DriverType(enum.Enum):
     LOCAL = "local"
     MINIO = "minio"
     GOOGLE_DRIVE = "google_drive"
+    AZURE = "azure"
 
     def __eq__(self, value):
         if isinstance(value, str):
@@ -282,6 +286,46 @@ def _get_share_with(self, config_google_drive) -> typing.List[str]:
         return emails_to_share_with
 
 
+class AzureBlobDriver(Driver):
+    """Driver to handle connection to Azure Blob Storage"""
+
+    def __init__(self, config):
+        super(AzureBlobDriver, self).__init__(config)
+
+        try:
+            self.account_name = config.azure_blob.account_name
+            connection_string = (
+                f"DefaultEndpointsProtocol=https;"
+                f"AccountName={self.account_name};"
+                f"AccountKey={config.azure_blob.account_key};"
+                f"EndpointSuffix=core.windows.net"
+            )
+            service_client = BlobServiceClient.from_connection_string(connection_string)
+            self.container = config.azure_blob.container
+            container_client = service_client.get_container_client(self.container)
+            if not container_client.exists():
+                container_client.create_container()
+            self.client = container_client
+
+            self.blob_path_prefix = None
+            if hasattr(config.azure_blob, "blob_path_prefix"):
+                if config.azure_blob.blob_path_prefix:
+                    self.blob_path_prefix = config.azure_blob.blob_path_prefix
+
+        except AzureError as e:
+            raise DriverError("Azure Blob Storage driver init error: " + str(e))
+
+    def upload_file(self, src: str, obj_path: str) -> str:
+        if self.blob_path_prefix:
+            obj_path = f"{self.blob_path_prefix}/{obj_path}"
+        try:
+            with open(src, "rb") as data:
+                self.client.upload_blob(name=obj_path, data=data, overwrite=True)
+        except AzureError as e:
+            raise DriverError("Azure Blob Storage driver error: " + str(e))
+        return f"https://{self.account_name}.blob.core.windows.net/{self.container}/{obj_path}"
+
+
 def create_driver(config):
     """Create driver object based on type defined in config"""
     driver = None
@@ -291,4 +335,6 @@ def create_driver(config):
         driver = MinioDriver(config)
     elif config.driver == DriverType.GOOGLE_DRIVE:
         driver = GoogleDriveDriver(config)
+    elif config.driver == DriverType.AZURE:
+        driver = AzureBlobDriver(config)
     return driver
diff --git a/test/conftest.py b/test/conftest.py
@@ -20,6 +20,9 @@
 GOOGLE_DRIVE_SERVICE_ACCOUNT_FILE = os.environ.get(
     "TEST_GOOGLE_DRIVE_SERVICE_ACCOUNT_FILE"
 )
+AZURE_STORAGE_ACCOUNT_NAME = os.environ.get("TEST_AZURE_STORAGE_ACCOUNT_NAME")
+AZURE_STORAGE_ACCOUNT_KEY = os.environ.get("TEST_AZURE_STORAGE_ACCOUNT_KEY")
+AZURE_STORAGE_CONTAINER = os.environ.get("TEST_AZURE_STORAGE_CONTAINER")
 
 
 @pytest.fixture(scope="function")
@@ -49,6 +52,10 @@ def setup_config():
             "MINIO__BUCKET_SUBPATH": "",
             "MINIO__SECURE": False,
             "MINIO__REGION": "",
+            "AZURE_BLOB__ACCOUNT_NAME": "",
+            "AZURE_BLOB__ACCOUNT_KEY": "",
+            "AZURE_BLOB__CONTAINER": "",
+            "AZURE_BLOB__BLOB_PATH_PREFIX": "",
         }
     )
 
diff --git a/test/test_sync.py b/test/test_sync.py
@@ -11,7 +11,7 @@
 import shutil
 import sqlite3
 
-from drivers import MinioDriver, LocalDriver, GoogleDriveDriver
+from drivers import MinioDriver, LocalDriver, GoogleDriveDriver, AzureBlobDriver
 from media_sync import (
     main,
     config,
@@ -33,6 +33,9 @@
     MINIO_SECRET_KEY,
     GOOGLE_DRIVE_SERVICE_ACCOUNT_FILE,
     GOOGLE_DRIVE_FOLDER,
+    AZURE_STORAGE_ACCOUNT_NAME,
+    AZURE_STORAGE_ACCOUNT_KEY,
+    AZURE_STORAGE_CONTAINER,
     cleanup,
     prepare_mergin_project,
 )
@@ -634,3 +637,111 @@ def test_google_drive_backend(mc):
     # files in mergin project still exist (copy mode)
     assert os.path.exists(os.path.join(work_project_dir, "img1.png"))
     assert os.path.exists(os.path.join(work_project_dir, "images", "img2.jpg"))
+
+
+def test_azure_blob_backend(mc):
+    """Test media sync connected to Azure Blob Storage backend (needs valid Azure credentials)"""
+    project_name = "mediasync_test_azure"
+    full_project_name = WORKSPACE + "/" + project_name
+    work_project_dir = os.path.join(TMP_DIR, project_name + "_work")
+
+    cleanup(mc, full_project_name, [work_project_dir])
+    prepare_mergin_project(mc, full_project_name)
+
+    # invalid config - missing required fields
+    config.update(
+        {
+            "MERGIN__USERNAME": API_USER,
+            "MERGIN__PASSWORD": USER_PWD,
+            "MERGIN__URL": SERVER_URL,
+            "MERGIN__PROJECT_NAME": full_project_name,
+            "PROJECT_WORKING_DIR": work_project_dir,
+            "OPERATION_MODE": "copy",
+            "REFERENCES": [
+                {
+                    "file": None,
+                    "table": None,
+                    "local_path_column": None,
+                    "driver_path_column": None,
+                }
+            ],
+            "DRIVER": "azure",
+            "AZURE_BLOB__ACCOUNT_NAME": AZURE_STORAGE_ACCOUNT_NAME,
+            "AZURE_BLOB__ACCOUNT_KEY": "",
+            "AZURE_BLOB__CONTAINER": AZURE_STORAGE_CONTAINER,
+        }
+    )
+
+    with pytest.raises(ConfigError):
+        validate_config(config)
+
+    # patch config to fit testing purposes
+    config.update(
+        {
+            "MERGIN__USERNAME": API_USER,
+            "MERGIN__PASSWORD": USER_PWD,
+            "MERGIN__URL": SERVER_URL,
+            "MERGIN__PROJECT_NAME": full_project_name,
+            "PROJECT_WORKING_DIR": work_project_dir,
+            "OPERATION_MODE": "copy",
+            "REFERENCES": [
+                {
+                    "file": None,
+                    "table": None,
+                    "local_path_column": None,
+                    "driver_path_column": None,
+                }
+            ],
+            "DRIVER": "azure",
+            "AZURE_BLOB__ACCOUNT_NAME": AZURE_STORAGE_ACCOUNT_NAME,
+            "AZURE_BLOB__ACCOUNT_KEY": AZURE_STORAGE_ACCOUNT_KEY,
+            "AZURE_BLOB__CONTAINER": AZURE_STORAGE_CONTAINER,
+        }
+    )
+
+    main()
+
+    # verify files were uploaded to Azure Blob Storage
+    driver = AzureBlobDriver(config)
+    blob_names = [b.name for b in driver.client.list_blobs()]
+    assert "img1.png" in blob_names
+    assert "images/img2.jpg" in blob_names
+
+    # files in mergin project still exist (copy mode)
+    assert os.path.exists(os.path.join(work_project_dir, "img1.png"))
+    assert os.path.exists(os.path.join(work_project_dir, "images", "img2.jpg"))
+
+    # test with blob_path_prefix
+    cleanup(mc, full_project_name, [work_project_dir])
+    prepare_mergin_project(mc, full_project_name)
+
+    config.update(
+        {
+            "MERGIN__USERNAME": API_USER,
+            "MERGIN__PASSWORD": USER_PWD,
+            "MERGIN__URL": SERVER_URL,
+            "MERGIN__PROJECT_NAME": full_project_name,
+            "PROJECT_WORKING_DIR": work_project_dir,
+            "OPERATION_MODE": "copy",
+            "REFERENCES": [
+                {
+                    "file": None,
+                    "table": None,
+                    "local_path_column": None,
+                    "driver_path_column": None,
+                }
+            ],
+            "DRIVER": "azure",
+            "AZURE_BLOB__ACCOUNT_NAME": AZURE_STORAGE_ACCOUNT_NAME,
+            "AZURE_BLOB__ACCOUNT_KEY": AZURE_STORAGE_ACCOUNT_KEY,
+            "AZURE_BLOB__CONTAINER": AZURE_STORAGE_CONTAINER,
+            "AZURE_BLOB__BLOB_PATH_PREFIX": "subPath",
+        }
+    )
+
+    main()
+
+    driver = AzureBlobDriver(config)
+    blob_names = [b.name for b in driver.client.list_blobs()]
+    assert "subPath/img1.png" in blob_names
+    assert "subPath/images/img2.jpg" in blob_names

Original file line number	Diff line number	Diff line change
`@@ -20,6 +20,9 @@`
`20`	`20`	`GOOGLE_DRIVE_SERVICE_ACCOUNT_FILE = os.environ.get(`
`21`	`21`	`"TEST_GOOGLE_DRIVE_SERVICE_ACCOUNT_FILE"`
`22`	`22`	`)`
	`23`	`+AZURE_STORAGE_ACCOUNT_NAME = os.environ.get("TEST_AZURE_STORAGE_ACCOUNT_NAME")`
	`24`	`+AZURE_STORAGE_ACCOUNT_KEY = os.environ.get("TEST_AZURE_STORAGE_ACCOUNT_KEY")`
	`25`	`+AZURE_STORAGE_CONTAINER = os.environ.get("TEST_AZURE_STORAGE_CONTAINER")`
`23`	`26`
`24`	`27`
`25`	`28`	`@pytest.fixture(scope="function")`
`@@ -49,6 +52,10 @@ def setup_config():`
`49`	`52`	`"MINIO__BUCKET_SUBPATH": "",`
`50`	`53`	`"MINIO__SECURE": False,`
`51`	`54`	`"MINIO__REGION": "",`
	`55`	`+ "AZURE_BLOB__ACCOUNT_NAME": "",`
	`56`	`+ "AZURE_BLOB__ACCOUNT_KEY": "",`
	`57`	`+ "AZURE_BLOB__CONTAINER": "",`
	`58`	`+ "AZURE_BLOB__BLOB_PATH_PREFIX": "",`
`52`	`59`	`}`
`53`	`60`	`)`
`54`	`61`