attachment listing,download and reading from gmail inbox functionality added

[Mustafa-tech-20]

Added Gmail Attachment Handling Functionality

Implemented listing of email attachments from the Gmail inbox

Added functionality to download attachments

Enabled reading and processing of attachment content

[Copilot]

Pull Request Overview

This PR adds Gmail attachment handling functionality to enable users to interact with email attachments programmatically. The changes introduce comprehensive attachment management capabilities including listing, downloading, and reading attachment content from various file formats.

• Added attachment listing and metadata extraction from Gmail messages
• Implemented attachment download functionality with security controls and file size limits
• Added content reading support for multiple file formats (PDF, DOCX, XLSX, CSV, HTML, JSON, text files)

Reviewed Changes

Copilot reviewed 2 out of 3 changed files in this pull request and generated 8 comments.

File	Description
pyproject.toml	Added dependencies for document processing libraries (beautifulsoup4, mammoth, pandas, pdfplumber, pyxlsb)
gmail/gmail_tools.py	Implemented comprehensive attachment handling with 6 new functions and support for multiple file formats

Comments suppressed due to low confidence (1)

gmail/gmail_tools.py:833

• The variable attachment_id may be undefined in the error case if the attachment lookup failed before assignment. Initialize attachment_id at the beginning of the function or use a default value.

            'attachment_id': attachment_id,

[Copilot]

Function _extract_message_body is defined twice (lines 76 and 111). The second definition will override the first, potentially causing unexpected behavior. Remove the duplicate function definition.

Suggested change

	# (Lines 111-144 removed; no replacement needed)

[Copilot]

[nitpick] Inconsistent spacing around commas in the import statement. Should be consistent spacing: Literal, Any, Tuple instead of Literal , Any , Tuple.

Suggested change

	from typing import Optional, List, Dict, Literal , Any , Tuple
	from typing import Optional, List, Dict, Literal, Any, Tuple

[Copilot]

The log message is incorrect. attachment_info is a single attachment object, not a count. Should be len(attachments) to log the actual count of attachments.

Suggested change

	logger.info(f"Found {attachment_info} attachments in message {message_id}")
	logger.info(f"Found {len(attachments)} attachments in message {message_id}")

[Copilot]

This line will raise a TypeError if attachment_info is None (when attachment is not found), but the None check happens after this assignment. Move this assignment inside the if block after the None check.

Suggested change

	if not attachment_info:
	logger.error(f"Attachment {attachment_name} not found. Available attachments: {[att['attachment_id'] for att in attachments]}")
	# Try to find by filename if attachment_id doesn't match
	if attachments:
	logger.info(f"Using first available attachment: {attachments[0]['filename']}")
	attachment_info = attachments[0]
	else:
	raise Exception(f"Attachment {attachment_name} not found in message {message_id}")
	attachment_id = attachment_info['attachment_id']

[Copilot]

The variable attachment_id may be undefined if attachment_info is None and the previous line failed. Use attachment_name instead since that's the parameter that was passed in.

[Copilot]

The function read_gmail_attachment_content expects attachment_name parameter, but attachment_id is being passed. This will cause the function call to fail.

Suggested change

	attachment_id=attachment['attachment_id'],
	attachment_name=attachment['filename'],

[Copilot]

The csv module is part of Python's standard library and should not be wrapped in a try-except ImportError block. This will never raise an ImportError.

Suggested change

	import csv
	CSV_AVAILABLE = True

[Copilot]

Path traversal protection is incomplete. The check only ensures the resolved path is under the base directory, but malicious filenames could still overwrite important files in subdirectories. Consider sanitizing the filename itself and using a dedicated downloads directory.

Suggested change

	# Use a dedicated downloads directory
	downloads_dir = Path.cwd() / "downloads"
	downloads_dir.mkdir(parents=True, exist_ok=True)
	# Sanitize the filename
	def sanitize_filename(filename):
	# Remove any path components
	filename = os.path.basename(filename)
	# Allow only safe characters: letters, numbers, dot, dash, underscore
	filename = re.sub(r'[^A-Za-z0-9._-]', '_', filename)
	return filename
	safe_filename = sanitize_filename(str(save_path))
	final_path = downloads_dir / safe_filename
	with open(final_path, 'wb') as f:
	f.write(file_data)
	result['saved_path'] = str(final_path)
	logger.info(f"Attachment saved to {final_path}")

[taylorwilsdon]

Love the idea, thanks for the PR! I'm a bit concerned about how much overhead this adds though - it's introducing a huge number of new and in some cases (ie pandas) very heavy dependencies. Also looks like some redundant code. Maybe a refac is in order?