sharedkernel 2.2.7__tar.gz → 2.4.0__tar.gz

{sharedkernel-2.2.7 → sharedkernel-2.4.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sharedkernel
-Version: 2.2.7
+Version: 2.4.0
 Summary: sharekernel is a shared package between all python projects
 Author: Smilinno
 Description-Content-Type: text/markdown
@@ -29,6 +29,10 @@ Dynamic: summary
 this is a shared kernel package
 
 # Change Log
+### Version 2.4.0
+- Implement file validation
+### Version 2.3.0
+- Implement multipart uploader for S3
 ### Version 2.2.7
 - add presigned url in s3 uploader
 ### Version 2.2.6

{sharedkernel-2.2.7 → sharedkernel-2.4.0}/README.md

@@ -2,6 +2,10 @@
 this is a shared kernel package
 
 # Change Log
+### Version 2.4.0
+- Implement file validation
+### Version 2.3.0
+- Implement multipart uploader for S3
 ### Version 2.2.7
 - add presigned url in s3 uploader
 ### Version 2.2.6

{sharedkernel-2.2.7 → sharedkernel-2.4.0}/setup.py

@@ -37,7 +37,7 @@ setup(
         "deepdiff",
     ],
     # *strongly* suggested for sharing
-    version="2.2.7",
+    version="2.4.0",
     description="sharekernel is a shared package between all python projects",
     long_description=long_description,
     long_description_content_type="text/markdown",

sharedkernel-2.4.0/sharedkernel/file_validation.py

@@ -0,0 +1,83 @@
+from functools import wraps
+from fastapi import HTTPException
+import magic
+
+DEFAULT_ALLOWED_MIME_TYPES = (
+    "image/jpeg",
+    "image/png",
+    "image/gif",
+    "image/webp",
+    "application/pdf",
+    "application/msword",
+    "application/x-ole-storage",
+    "application/vnd.openxmlformats-officedocument.wordprocessingml.document",
+    "application/vnd.ms-excel",
+    "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet",
+    "application/vnd.ms-powerpoint",
+    "application/vnd.openxmlformats-officedocument.presentationml.presentation",
+    "text/plain",
+    "text/csv",
+    "audio/mpeg",
+    "audio/wav",
+    "audio/mp3",
+)
+
+
+def get_mime_type(fileobj) -> str:
+    """
+    Detects the real MIME type of a file based on its content (magic bytes),
+    not the client-provided extension or Content-Type header.
+    """
+    try:
+        fileobj.seek(0)
+        sample = fileobj.read(2048)
+        fileobj.seek(0)
+        m = magic.Magic(mime=True)
+        mime = m.from_buffer(sample)
+        return mime or "application/octet-stream"
+    except Exception:
+        return "application/octet-stream"
+
+
+def validate_file_type(fileobj, allowed_mimes=None):
+    """
+    Validates the uploaded file’s real MIME type against the allowed list.
+    """
+    allowed_mimes = allowed_mimes or DEFAULT_ALLOWED_MIME_TYPES
+    mime = get_mime_type(fileobj)
+    if mime not in allowed_mimes:
+        raise ValueError(f"نوع فایل مجاز نیست: {mime}")
+    return mime
+
+
+def is_upload_file(obj):
+    return (
+        hasattr(obj, "file") and
+        hasattr(obj, "filename") and
+        callable(getattr(obj.file, "read", None)) and
+        callable(getattr(obj.file, "seek", None))
+    )
+
+
+def validate_upload_file(allowed_mimes=None):
+    allowed_mimes = allowed_mimes or DEFAULT_ALLOWED_MIME_TYPES
+
+    def decorator(func):
+            @wraps(func)
+            async def wrapper(*args, **kwargs):
+                for value in list(args) + list(kwargs.values()):
+                    if is_upload_file(value):
+                        try:
+                            validate_file_type(value.file, allowed_mimes)
+                        except ValueError as e:
+                            raise HTTPException(status_code=400, detail=str(e))
+                    elif isinstance(value, list):
+                        for f in value:
+                            if is_upload_file(f):
+                                try:
+                                    validate_file_type(f.file, allowed_mimes)
+                                except ValueError as e:
+                                    raise HTTPException(status_code=400, detail=str(e))
+                return await func(*args, **kwargs)
+            return wrapper
+    return decorator

sharedkernel-2.4.0/sharedkernel/multipart_upload.py

@@ -0,0 +1,69 @@
+class MultipartUploadSession:
+    MIN_PART_SIZE = 5 * 1024 * 1024  # 5MB
+
+    def __init__(self, s3_client, bucket, object_key, acl="private"):
+        self.s3 = s3_client
+        self.bucket = bucket
+        self.key = object_key
+        self.acl = acl
+
+        self.upload_id = None
+        self.parts = []
+        self.part_number = 1
+        self.buffer = bytearray()
+
+        self._start()
+
+    def _start(self):
+        resp = self.s3.create_multipart_upload(
+            Bucket=self.bucket,
+            Key=self.key,
+            ACL=self.acl,
+        )
+        self.upload_id = resp["UploadId"]
+
+    def upload_chunk(self, chunk: bytes):
+        """
+        Receives small chunks (KB) and buffers them
+        """
+        self.buffer.extend(chunk)
+
+        if len(self.buffer) >= self.MIN_PART_SIZE:
+            self._upload_part(bytes(self.buffer))
+            self.buffer.clear()
+
+    def _upload_part(self, data: bytes):
+        resp = self.s3.upload_part(
+            Bucket=self.bucket,
+            Key=self.key,
+            UploadId=self.upload_id,
+            PartNumber=self.part_number,
+            Body=data,
+        )
+
+        self.parts.append({
+            "ETag": resp["ETag"],
+            "PartNumber": self.part_number
+        })
+        self.part_number += 1
+
+    def complete(self):
+        # upload remaining buffer (can be < 5MB)
+        if self.buffer:
+            self._upload_part(bytes(self.buffer))
+            self.buffer.clear()
+
+        self.s3.complete_multipart_upload(
+            Bucket=self.bucket,
+            Key=self.key,
+            UploadId=self.upload_id,
+            MultipartUpload={"Parts": self.parts},
+        )
+
+    def abort(self):
+        if self.upload_id:
+            self.s3.abort_multipart_upload(
+                Bucket=self.bucket,
+                Key=self.key,
+                UploadId=self.upload_id,
+            )

{sharedkernel-2.2.7 → sharedkernel-2.4.0}/sharedkernel/s3_uploader.py

@@ -4,6 +4,7 @@ import boto3
 from io import BytesIO
 import uuid
 import os
+from sharedkernel.multipart_upload import MultipartUploadSession
 
 class S3Uploader:
     def __init__(self, endpoint_url, bucket, access_key, secret_key):
@@ -82,5 +83,24 @@ class S3Uploader:
         file_obj = BytesIO(response.content)
         return self.upload_file_object(file_obj=file_obj, object_name=object_name, file_extension=file_extension, folder_name=folder_name)
 
+    def create_multipart_session(
+        self,
+        object_name=None,
+        file_extension=None,
+        folder_name=None,
+        public_read=True,
+    ):
+        object_key = self.__generate_object_name(
+            object_name, file_extension, folder_name
+        )
+
+        acl = "public-read" if public_read else "private"
+
+        return MultipartUploadSession(
+            s3_client=self.s3,
+            bucket=self.bucket,
+            object_key=object_key,
+            acl=acl,
+        )
 
 

{sharedkernel-2.2.7 → sharedkernel-2.4.0}/sharedkernel.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sharedkernel
-Version: 2.2.7
+Version: 2.4.0
 Summary: sharekernel is a shared package between all python projects
 Author: Smilinno
 Description-Content-Type: text/markdown
@@ -29,6 +29,10 @@ Dynamic: summary
 this is a shared kernel package
 
 # Change Log
+### Version 2.4.0
+- Implement file validation
+### Version 2.3.0
+- Implement multipart uploader for S3
 ### Version 2.2.7
 - add presigned url in s3 uploader
 ### Version 2.2.6

{sharedkernel-2.2.7 → sharedkernel-2.4.0}/sharedkernel.egg-info/SOURCES.txt

@@ -4,7 +4,9 @@ sharedkernel/common.py
 sharedkernel/data_format_converter.py
 sharedkernel/date_converter.py
 sharedkernel/diff_utils.py
+sharedkernel/file_validation.py
 sharedkernel/jwt_service.py
+sharedkernel/multipart_upload.py
 sharedkernel/regex_masking.py
 sharedkernel/s3_uploader.py
 sharedkernel/string_extentions.py