.env.example

@@ -1,15 +1,7 @@
-# (Required) Absolute path to the local folder to back up
+SOURCE_FOLDER=
-SOURCE_FOLDER=""
+TARGET_BUCKET=
-# (Required) Name of the remote bucket or container
-TARGET_BUCKET=""
 
-# (Required) Cloud provider (gcloud, aws, azure)
+# Cloud provider "gcs, aws, azure"
-SERVICE_NAME=""
+SERVICE_NAME=
-# (Required) Cloud provider access string or path to key file
+# Path to service account key file
-SERVICE_KEY=""
+SERVICE_KEY=
-
-# -----------------------------------------------------------
-
-# (Optional) Path to log file and level
-LOG_FILE=""
-LOG_LEVEL="WARNING"

README.md

@@ -1,91 +1,43 @@
 # Cloud Backup
 Backup and archive ordinary files and folders to Google Cloud, AWS or Azure.
 
-## Why this exists in a world with Rclone (and similar)
-
-This script was created to solve the specific task of archiving a folder to the cloud without read access to the bucket/container and its objects.
-Cloud Backup keeps an internal database of second-level files and folders which were sucessfully uploaded to the cloud.
-
-If a file or folder changes on disk, that specific file or folder is compressed, uploaded to the cloud, and the database gets updated. What happens to the object in the bucket is invisible to the program.
-
 ## Get started
 This program requires Python 3.6 or newer with PIP.
 
-Cloud Backups supports uploading to Google Cloud Storage, Azure Blob Storage and AWS S3.
-
 1. **Clone this repo**
-```bash
+```
 git clone https://github.com/VictorWesterlund/cloud-backup
 ```
 
 2. **Install dependencies**
-```bash
+```
-# Install dependencies for your cloud provider "gcloud", "aws" or "azure". Leaving it empty will install everything
-bash install.sh aws
-
-# Or install everything directly with PIP
 python3 -m pip install -r requirements.txt
 ```
 
 3. **Copy environment variables file**
-```bash
+```
 cp .env.example .env
 ```
 
-4. **Edit environment variables in `.env`**
+4. **Edit environment variables**
+Open `.env` with your text editor of choice and fill out these required variables
 ```bash
-# Remember to double-slash escape paths on Windows 'E:\\path\\to\\something'
+# Path to the local folder to back up
+SOURCE_FOLDER=
+# Name of the remote bucket (destination)
+TARGET_BUCKET=
 
-# Absolute path to folder whose contents should be backed up
+# Cloud provider (gcs, s3, azure)
-SOURCE_FOLDER="/home/me/backup/"
+SERVICE_NAME=
-# Name of bucket (or "container" in Azure)
+# Path to service account key file
-TARGET_BUCKET="home_backup"
+SERVICE_KEY=
-
-# Cloud provider. "gloud", "aws" or "azure"
-SERVICE_NAME="aws"
-# IAM authentication
-# GCS: Path to keyfile or string (GOOGLE_APPLICATION_CREDENTIALS)
-# Azure: "Connection string" from the Access Key to the container
-# AWS: Access key ID and secret seperated by a ";"
-SERVICE_KEY="SDJSBADYUAD;asid7sad123ebasdhasnk3dnsai"
 ```
 
 5. **Run backup script**
-```bash
+```
 python3 backup.py
 ```
 
 Second-level files and folders should now start uploading to your destination bucket as zip archives.
 Subsequent runs of the `backup.py` script will only upload changed files and folders.
 In-fact; modified state is cached locally and doesn't request anything from your cloud provider.
-
-----
-
-You can also run `backup.py` on a schedule with CRON or equivalent for your system. No requests will be sent to the cloud unless a file or folder has actually changed
-
-## More stuff
-
-Here are some additional settings and commands you can try
-
-### Back up a second-level file
-```bash
-python3 backup.py file 'relative/path/from/.env'
-```
-
-### Resolve CRC32 to path or vice versa
-```bash
-python3 resolve.py '587374759'
-# output: 'hello_world.txt'
-
-python3 resolve.py 'hello_world.txt'
-# output: '587374759'
-```
-
-### Optional flags in `.env`
-```bash
-# The following intbool flags can be added to .env to override default behavior
-# Their value in this demo is the "default" state
-
-# Archive files and folders before uploading
-COMPRESS="1"
-```

backup.py

@@ -1,31 +1,5 @@
 import sys
 
-from src import Backup, file_exists
+from src import Backup
 
-# Back up a single file by path
+Backup().backup_all()
-def file_backup():
-	path = sys.argv[2]
-
-	if len(sys.argv) < 3:
-		return print("Invalid argument length: Expected path to file or folder")
-	
-	if not file_exists(path):
-		return print(f"File or folder at '{path}' does not exist")
-
-	return Backup().backup_item(path, False)
-
-def run(method):
-	methods = {
-		"file": file_backup,
-		"all" : Backup().backup_all
-	}
-
-	if method not in methods:
-		return print(f"Invalid argument '{method}'")
-
-	return methods[method]()
-
-if len(sys.argv) > 1:
-	run(sys.argv[1])
-else:
-	run("all")

install.sh

@@ -1,28 +0,0 @@
-install () {
-	python3 -m pip install $1
-}
-
-install python-dotenv
-
-# Install Python libraries for cloud provider
-case $1 in
-	"gcloud")
-		install google-cloud-storage 
-		;;
-
-	"azure")
-		install azure-storage-blob 
-		;;
-	
-	"aws")
-		install boto3 
-		;;
-	
-	*) ;;
-esac
-
-# Create .env file if it doesn't exist
-if [ ! -f ".env" ]; then
-	cp .env.example .env
-	sed -i "s/SERVICE_NAME=\"\"/SERVICE_NAME=\"$1\"" .env
-fi

requirements.txt

@@ -1,4 +1,2 @@
 python-dotenv
 google-cloud-storage
-azure-storage-blob
-boto3

resolve.py

@@ -1,23 +0,0 @@
-import sys
-
-from src import Database
-
-class Resolve(Database):
-	def __init__(self):
-		super().__init__()
-
-		if self.item_exists(sys.argv[1]):
-			self.path_to_chksum()
-		else:
-			self.chksum_to_path()
-
-	def path_to_chksum(self):
-		print("Something")
-
-	def chksum_to_path(self):
-		print("Something else")
-
-if len(sys.argv) > 2:
-	Resolve()
-else:
-	print("Invalid argument length: Need at least two")

src/__init__.py

@@ -1,24 +1,9 @@
-import os
 from dotenv import load_dotenv
-
 from .db import Database, dbname
 from .fs import FileSystem, file_exists
 from .backup import Backup
 
-# Required environment variables
-required_vars = (
-	"SOURCE_FOLDER",
-	"TARGET_BUCKET",
-	"SERVICE_NAME",
-	"SERVICE_KEY",
-	"LOG_LEVEL"
-)
-
 if not file_exists(".env"):
 	raise FileNotFoundError("Environment variable file does not exist. Copy '.env.example' to '.env'")
 
 load_dotenv()
-
-# Check that required environment variables are set
-if not all(map(lambda var: os.getenv(var), required_vars)):
-	raise SystemExit("One or more required environment variables in '.env' have not been set")

src/backup.py

@@ -1,6 +1,3 @@
-import os
-import logging
-from logging.handlers import RotatingFileHandler
 from typing import Union
 
 from .cloud import Storage as StorageClient
@@ -10,7 +7,6 @@ from . import dbname
 class Backup(FileSystem):
     def __init__(self):
         super().__init__()
-        self.enable_logging()
 
         self.has_change = False
 
@@ -19,39 +15,8 @@ class Backup(FileSystem):
 
         self.compress = self.db.get_flag("COMPRESS")
 
-    # Configure logging
-    def enable_logging(self):
-        self.log = logging.getLogger(__name__)
-        self.log.debug("Start console logging")
-        log_format = logging.Formatter("[%(asctime)s][%(levelname)s]: %(name)s: %(message)s")
-
-        # Log to console
-        log_console = logging.StreamHandler()
-        log_console.setLevel(logging.INFO)
-        log_console.setFormatter(log_format)
-
-        self.log.addHandler(log_console)
-
-        # Log to file
-        log_file_path = os.getenv("LOG_FILE")
-        if log_file_path:
-            self.log.debug("Start file logging")
-            log_file = RotatingFileHandler(
-                log_file_path,
-                mode        = "a",
-                maxBytes    = 50 * 1024 * 1024,
-                backupCount = 5,
-                encoding    = None,
-                delay       = False
-            )
-
-            log_file.setLevel(os.getenv("LOG_LEVEL"))
-            log_file.setFormatter(log_format)
-
-            self.log.addHandler(log_file)
-
     # Backup a file or folder
-    def backup_item(self, item: Union[list, str], silent: bool = True) -> bool:
+    def backup_item(self, item: Union[list, str]) -> bool:
         if isinstance(item, str):
             item = self.get_item(item)
 
@@ -67,36 +32,26 @@ class Backup(FileSystem):
 
         self.has_change = True
 
-        self.log.info(f"'{item[0]}': Uploading")
+        print(f"Uploading: '{item[0]}' ... ", end="")
-        print(f"⏳ | Uploading: '{item[0]}'", end="\r")
 
         blob = item
         # Upload as zip archive
         if self.compress:
-            self.log.debug(f"'{item[0]}': Compressing")
             blob = FileSystem.zip(blob)
 
         # Upload to cloud
         if self.cloud.upload(blob):
-            self.log.debug(f"'{item[0]}': Uploaded")
-            print(f"✅ | Upload successful: '{item[0]}'")
             # Update local database
-            if not self.db.set_item(item):
+            if self.db.set_item(item):
-                self.log.warn(f"'{item[0]}': Failed to update database")
+                print("OK")
-                print("⚠️ | Failed to update database")
+            else:
+                print("OK, but failed to update database")
         else:
-            self.log.error(f"'{item[0]}': {self.cloud.error}")
+            print("FAILED")
-            print(f"❌ | Upload failed: '{item[0]}'")
 
         # Remove temp zip
         if self.compress:
             FileSystem.delete(blob)
-
-        # Deprecated: Run when a single item is backed up directly
-        if not silent and not self.has_change:
-            self.log.info("No changes found")
-            print("✅ | Up to date. No changes found")
-        
         return
 
     # Scan TARGET_FOLDER for files and folders to back up
@@ -106,5 +61,4 @@ class Backup(FileSystem):
             self.backup_item(item)
         
         if not self.has_change:
-            self.log.info("No changes found")
+            print("Up to date. No changes found")
-            print("✅ | Up to date. No changes found")

src/cloud/__init__.py

@@ -8,8 +8,6 @@ class Storage:
         self._service = None
         self.service = os.getenv("SERVICE_NAME")
 
-        self.error = None
-
     @property
     def service(self):
         return self._service
@@ -29,7 +27,4 @@ class Storage:
         return values
 
     def upload(self, *argv):
-        upload = self.service.upload(*argv)
+        return self.service.upload(*argv)
-        self.error = self.service.error
-        
-        return upload

src/cloud/aws.py

@@ -1,40 +0,0 @@
-import os
-import boto3
-from botocore.exceptions import ClientError
-
-from ..fs.utils import get_file
-
-class StorageClient:
-	def __init__(self):
-		self.set_access_key()
-		self.client = boto3.client("s3")
-
-		self._error = None
-
-	@property
-	def error(self):
-		return self._error
-
-	@error.setter
-	def error(self, state):
-		self._error = state
-
-	# Get IAM user access key and ID
-	def set_access_key(self):
-		key = os.getenv("SERVICE_KEY").split(";")
-		if len(key) != 2:
-			self.error = "Invalid AWS service key"
-			return False
-
-		os.environ["aws_access_key_id"] = key[0]
-		os.environ["aws_secret_access_key"] = key[1]
-
-	def upload(self, path: str) -> bool:
-		name = get_file(path)
-
-		try:
-			resp = self.client.upload_file(path, os.getenv("TARGET_BUCKET"), name)
-		except ClientError as e:
-			self.error = e
-			return False
-		return True

src/cloud/azure.py

@@ -1,31 +0,0 @@
-import os
-from azure.storage.blob import BlobServiceClient
-
-from ..fs.utils import get_file
-
-class StorageClient:
-	def __init__(self):
-		self.client = BlobServiceClient.from_connection_string(os.getenv("SERVICE_KEY"))
-
-		self._error = None
-
-	@property
-	def error(self):
-		return self._error
-
-	@error.setter
-	def error(self, state):
-		self._error = state
-
-	def upload(self, path: str) -> bool:
-		name = get_file(path)
-		blob = self.client.get_blob_client(container=os.getenv("TARGET_BUCKET"), blob=name)
-
-		try:
-			with open(path, "rb") as f:
-				blob.upload_blob(f,overwrite=True)
-			return True
-		except Exception as e:
-			if e.response.status_code == 403:
-				self.error = "Azure: Access key invalid or lacking required permissions"
-			return False

src/cloud/gcs.py

@@ -11,21 +11,10 @@ class StorageClient:
         client = storage.Client()
         self.bucket = client.bucket(self.get_bucket())
 
-        self._error = None
-
-    @property
-    def error(self):
-        return self._error
-
-    @error.setter
-    def error(self, state):
-        self._error = state
-
     def get_bucket(self):
         return os.getenv("TARGET_BUCKET")
 
     def upload(self, path: str) -> bool:
-        self.error = None
         name = get_file(path)
         blob = self.bucket.blob(name)
 
@@ -33,7 +22,5 @@ class StorageClient:
             with open(path, "rb") as f:
                 blob.upload_from_file(f)
             return True
-        except Exception as e:
+        except:
-            if e.response.status_code == 403:
-                self.error = "GCS: Forbidden: Account lacks 'storage.objects.create' role on target bucket"
             return False

src/db/database.py

@@ -27,18 +27,15 @@ class Database(SQLite):
 
     # Check if item exists in the database
     def item_exists(self, item: Union[list, tuple]) -> bool:
-        sql = "SELECT anchor FROM manifest WHERE anchor = ?"
+        sql = f"SELECT anchor FROM manifest WHERE anchor = '{item[0]}'"
-        res = self.query(sql, (item[0],))
+        res = self.query(sql)
 
         return res
 
     # Check if item should be backed up by comparing mtime and checksum
     def check_item(self, item: Union[list, tuple]) -> bool:
-        if os.getenv("FORCE_UPLOAD"):
+        sql = f"SELECT {self.columns} FROM manifest WHERE anchor = '{item[0]}'"
-            return True
+        db_item = self.query(sql)
-
-        sql = f"SELECT {self.columns} FROM manifest WHERE anchor = ?"
-        db_item = self.query(sql, (item[0],))
 
         # New item or item changed, so back it up
         if not db_item or (item != db_item[0]):
@@ -47,12 +44,10 @@ class Database(SQLite):
 
     # Insert or update item in database
     def set_item(self, item: Union[list, tuple]) -> bool:
-        values = [item[0], item[1], item[0]]
+        sql = f"UPDATE manifest SET anchor = '{item[0]}', chksum = {item[1]} WHERE anchor = '{item[0]}'"
-        sql = "UPDATE manifest SET anchor = ?, chksum = ? WHERE anchor = ?"
 
         if not self.item_exists(item):
-            sql = f"INSERT INTO manifest ({self.columns}) VALUES (?, ?)"
+            sql = f"INSERT INTO manifest ({self.columns}) VALUES ('{item[0]}', {item[1]})"
-            values.pop()
+        self.query(sql)
-        self.query(sql, values)
 
         return True

src/db/flags.py

@@ -0,0 +1,16 @@
+from .sqlite import SQLite
+
+class Flags(SQLite):
+    def __init__(self):
+        super().__init__()
+
+        self._columns = ["k", "v"]
+
+    @property
+    def columns(self):
+        return ",".join(self._columns)
+
+    @columns.setter
+    def columns(self, columns: list):
+        self._columns = columns
+

src/db/sqlite.py

@@ -1,7 +1,6 @@
 import os
 import pathlib
 import sqlite3 as sqlite
-from typing import Iterable
 
 dbname = "._cloudbackup.db"
 
@@ -24,8 +23,8 @@ class SQLite():
         return " ".join([s.strip() for s in sql.splitlines()])
 
     # Run SQL query
-    def query(self, sql: str, params: Iterable = ()):
+    def query(self, sql: str):
-        query = self.cursor.execute(sql, params)
+        query = self.cursor.execute(sql)
         self.db.commit()
 
         result = query.fetchall()
@@ -62,8 +61,8 @@ class SQLite():
         if envar:
             return envar
 
-        sql = "SELECT v FROM flags WHERE k = ?"
+        sql = f"SELECT v FROM flags WHERE k = '{key}'"
-        res = self.query(sql, [key])
+        res = self.query(sql)
 
         if not res:
             return False

src/fs/fs.py

@@ -26,8 +26,7 @@ class FileSystem:
 
     @staticmethod
     def zip(item) -> str:
-        name = FileSystem.chksum(item[0])
+        dest = f"{tempfile.gettempdir()}/{str(item[1])}"
-        dest = f"{tempfile.gettempdir()}/{name}"
 
         # Make a temp zip file of single file or folder
         if file_exists(item[0]):