commit

.gitignore

@@ -0,0 +1,2 @@
+attachments/
+__pycache__/

Dockerfile

@@ -0,0 +1,23 @@
+# Use an official Ubuntu image as a parent image
+FROM ubuntu:20.04
+
+# Set the working directory in the container
+WORKDIR /app
+
+# Update the package repository and install Python 3
+RUN apt-get update && apt-get install -y python3
+
+# Copy the current directory contents into the container at /app
+
+COPY main.py mbox_parser.py requirements.txt /app/
+
+RUN mkdir /app/target_files
+
+# Install any needed Python packages specified in requirements.txt for your Python application
+RUN apt-get install -y python3-pip && pip3 install -r requirements.txt
+
+# Install "readpst" in the container
+RUN apt-get install -y pst-utils
+
+# Run the Python script (replace "main.py" with the actual entry point of your Python application)
+CMD ["python3", "main.py"]

docker-compose.yml

@@ -0,0 +1,6 @@
+version: '3'
+services:
+  myapp:
+    image: my-python-app:latest
+    volumes:
+      - target_files:/app/target_files

main.py

@@ -0,0 +1,27 @@
+import os
+from mbox_parser import parse_mbox_file
+
+
+def list_mbox_files(target_dir):
+    mbox_files = []
+    for root, _, files in os.walk(target_dir):
+        for file in files:
+            if file.endswith(".mbox"):
+                mbox_files.append(file)
+    return mbox_files
+
+
+if __name__ == '__main__':
+    # Specify the directory containing .mbox files
+    target_dir = "target_files"  # This should match the volume mount point in Docker
+
+    mbox_file_list = list_mbox_files(target_dir)
+    print("List of .mbox files:", mbox_file_list)
+    for mbox_file in mbox_file_list:
+        print("\n#################################################################")
+        print(f'working on {mbox_file}')
+        data = parse_mbox_file(f'target_files/{mbox_file}')
+        print("#################################################################\n")
+        print(data)
+
+

mbox_parser.py

@@ -0,0 +1,53 @@
+import mailbox
+import email.utils
+import os
+
+
+def parse_mbox_file(mbox_file):
+    data = []
+    mbox = mailbox.mbox(mbox_file)
+    for message in mbox:
+        # Extract email details
+        subject = message['subject']
+        sender_info = message['from']
+        date = message['date']
+        receiver_info = message['to']
+        sender_name, sender_email = email.utils.parseaddr(sender_info)
+        receiver_name, receiver_email = email.utils.parseaddr(receiver_info)
+
+        for part in message.walk():
+            if part.get_content_maintype() == 'multipart':
+                continue
+            if part.get('Content-Disposition') is None:
+                continue
+
+            attachment_name = part.get_filename()
+            content_type = part.get_content_type()
+            attachment_data = part.get_payload(decode=True)
+
+            if attachment_data:
+                save_dir = 'attachments'
+                if not os.path.exists(save_dir):
+                    os.makedirs(save_dir)
+                attachment_path = os.path.join(save_dir, attachment_name)
+
+                with open(attachment_path, 'wb') as file:
+                    file.write(attachment_data)
+
+            data.append({
+                'subject': subject,
+                'sender_name': sender_name,
+                'seder_email': sender_email,
+                'receiver_name': receiver_name,
+                'receiver_email,': receiver_email,
+                'date': date,
+                'attachment name': attachment_name,
+                'content type': content_type
+            })
+    return data
+
+
+# print(parse_mbox_file('D:\Languages\Python\Email Attachments\cindyloh3333_gmail.com.mbox')) 
+
+
+

requirements.txt

@@ -0,0 +1 @@
+��

target_files/testttttt.txt

@@ -0,0 +1 @@
+helooooooooooooooooo