Merge pull request #106 from rafsaf/add-azure-upload-provider

add azure upload provider

README.md

@@ -21,6 +21,7 @@ A tool for performing scheduled database backups and transferring encrypted data
 
 - Google Cloud Storage bucket
 - AWS S3 bucket
+- Azure Blob Storage
 - Debug (local)
 
 ## Notifications

backuper/config.py

@@ -27,6 +27,7 @@ class UploadProviderEnum(StrEnum):
     LOCAL_FILES_DEBUG = "debug"
     GOOGLE_CLOUD_STORAGE = "gcs"
     AWS_S3 = "aws"
+    AZURE = "azure"
 
 
 class BackupTargetEnum(StrEnum):

backuper/models/upload_provider_models.py

@@ -35,3 +35,8 @@ class AWSProviderModel(ProviderModel):
     key_secret: SecretStr
     region: str
     max_bandwidth: int | None = None
+
+
+class AzureProviderModel(ProviderModel):
+    container_name: str
+    connect_string: SecretStr

backuper/upload_providers/__init__.py

@@ -1,11 +1,13 @@
+from .aws_s3 import UploadProviderAWS
+from .azure import UploadProviderAzure
 from .base_provider import BaseUploadProvider
-from .google_cloud_storage import UploadProviderGCS
 from .debug import UploadProviderLocalDebug
-from .aws_s3 import UploadProviderAWS
+from .google_cloud_storage import UploadProviderGCS
 
 __all__ = [
     "BaseUploadProvider",
     "UploadProviderAWS",
     "UploadProviderGCS",
     "UploadProviderLocalDebug",
+    "UploadProviderAzure",
 ]

backuper/upload_providers/azure.py

@@ -0,0 +1,92 @@
+import logging
+from pathlib import Path
+
+from azure.storage.blob import BlobServiceClient
+from pydantic import SecretStr
+
+from backuper import config, core
+from backuper.upload_providers.base_provider import BaseUploadProvider
+
+log = logging.getLogger(__name__)
+
+
+class UploadProviderAzure(
+    BaseUploadProvider,
+    name=config.UploadProviderEnum.AZURE,
+):
+    """Azure blob storage for storing backups"""
+
+    def __init__(
+        self,
+        container_name: str,
+        connect_string: SecretStr,
+        **kwargs: str,
+    ) -> None:
+        self.container_name = container_name
+        self.connect_str = connect_string
+
+        blob_service_client = BlobServiceClient.from_connection_string(
+            connect_string.get_secret_value()
+        )
+        self.container_client = blob_service_client.get_container_client(
+            container=self.container_name
+        )
+
+    def _post_save(self, backup_file: Path) -> str:
+        zip_backup_file = core.run_create_zip_archive(backup_file=backup_file)
+
+        backup_dest_in_azure_container = "{}/{}".format(
+            zip_backup_file.parent.name,
+            zip_backup_file.name,
+        )
+        blob_client = self.container_client.get_blob_client(
+            blob=backup_dest_in_azure_container
+        )
+
+        log.info(
+            "start uploading %s to %s", zip_backup_file, backup_dest_in_azure_container
+        )
+
+        with open(file=zip_backup_file, mode="rb") as data:
+            blob_client.upload_blob(data=data)
+
+        log.info(
+            "uploaded %s to %s in %s",
+            zip_backup_file,
+            backup_dest_in_azure_container,
+            self.container_name,
+        )
+        return backup_dest_in_azure_container
+
+    def _clean(
+        self, backup_file: Path, max_backups: int, min_retention_days: int
+    ) -> None:
+        for backup_path in backup_file.parent.iterdir():
+            core.remove_path(backup_path)
+            log.info("removed %s from local disk", backup_path)
+
+        backup_list_cloud: list[str] = []
+        for blob in self.container_client.list_blobs(
+            name_starts_with=backup_file.parent.name
+        ):
+            backup_list_cloud.append(blob.name)
+
+        # remove oldest
+        backup_list_cloud.sort(reverse=True)
+
+        while len(backup_list_cloud) > max_backups:
+            backup_to_remove = backup_list_cloud.pop()
+            file_name = backup_to_remove.split("/")[-1]
+            if core.file_before_retention_period_ends(
+                backup_name=file_name, min_retention_days=min_retention_days
+            ):
+                log.info(
+                    "there are more backups than max_backups (%s/%s), "
+                    "but oldest cannot be removed due to min retention days",
+                    len(backup_list_cloud),
+                    max_backups,
+                )
+                break
+
+            self.container_client.delete_blob(blob=backup_to_remove)
+            log.info("deleted backup %s from azure blob storage", backup_to_remove)

docs/index.md

@@ -21,6 +21,7 @@ A tool for performing scheduled database backups and transferring encrypted data
 
 - Google Cloud Storage bucket
 - AWS S3 bucket
+- Azure Blob Storage
 - Debug (local)
 
 ## Notifications

docs/providers/azure.md

@@ -0,0 +1,46 @@
+---
+hide:
+  - toc
+---
+
+# Azure Blob Storage
+
+## Environment variable
+
+```bash
+BACKUP_PROVIDER="name=azure container_name=my-backuper-instance connect_string=DefaultEndpointsProtocol=https;AccountName=accname;AccountKey=secret;EndpointSuffix=core.windows.net"
+```
+
+Uses Azure Blob Storage for storing backups.
+
+!!! note
+    _There can be only one upload provider defined per app, using **BACKUP_PROVIDER** environemnt variable_. It's type is guessed by using `name`, in this case `name=azure`. Params must be included in value, splited by single space for example "value1=1 value2=foo".
+
+## Params
+
+| Name           | Type                 | Description                                                                                      | Default |
+| :------------- | :------------------- | :----------------------------------------------------------------------------------------------- | :------ |
+| name           | string[**requried**] | Must be set literaly to string `azure` to use Google Cloud Storage.                              | -       |
+| container_name | string[**requried**] | Storage account container name. It must be already created, backuper won't create new container. | -       |
+| connect_string | string[**requried**] | Connection string copied from your storage account "Access keys" section.                        | -       |
+
+## Examples
+
+```bash
+# 1. Storage account accname and container name my-backuper-instance
+BACKUP_PROVIDER="name=azure container_name=my-backuper-instance connect_string=DefaultEndpointsProtocol=https;AccountName=accname;AccountKey=secret;EndpointSuffix=core.windows.net"
+
+# 2. Storage account birds and container name birds
+BACKUP_PROVIDER="name=azure container_name=birds connect_string=DefaultEndpointsProtocol=https;AccountName=birds;AccountKey=secret;EndpointSuffix=core.windows.net"
+```
+
+
+## Resources
+
+#### Creating azure storage account
+
+[https://learn.microsoft.com/en-us/azure/storage/common/storage-account-create?tabs=azure-portal](https://learn.microsoft.com/en-us/azure/storage/common/storage-account-create?tabs=azure-portal)
+
+
+<br>
+<br>

mkdocs.yml

@@ -24,6 +24,7 @@ nav:
   - Upload providers:
       - providers/google_cloud_storage.md
       - providers/aws_s3.md
+      - providers/azure.md
       - providers/debug.md
   - Backup targets:
       - backup_targets/postgresql.md