"""
Message content parsing utilities.
Handles extraction of text from attributedBody blob (typedstream format)
used in macOS Ventura and later for styled messages.
"""
from typing import Optional
def extract_text_from_attributed_body(blob: bytes) -> Optional[str]:
"""
Extract plain text from attributedBody blob.
Since macOS Ventura, message text may be stored in attributedBody
(a typedstream blob) instead of the text column. This is especially
true for messages with styling, mentions, or rich formatting.
Args:
blob: Raw bytes from attributedBody column
Returns:
Extracted plain text or None if parsing fails
"""
if not blob:
return None
# Find NSString marker in typedstream
marker = b"NSString"
idx = blob.find(marker)
if idx == -1:
# Try alternative marker
marker = b"NSMutableString"
idx = blob.find(marker)
if idx == -1:
return None
# Skip past marker and type info bytes
idx += len(marker) + 5
if idx >= len(blob):
return None
# Read string length - can be single byte or multi-byte encoded
if blob[idx] == 0x81:
# Multi-byte length encoding (for strings > 127 chars)
if idx + 3 > len(blob):
return None
length = int.from_bytes(blob[idx+1:idx+3], 'little')
idx += 3
elif blob[idx] == 0x82:
# 3-byte length encoding (for very long strings)
if idx + 4 > len(blob):
return None
length = int.from_bytes(blob[idx+1:idx+4], 'little')
idx += 4
else:
# Single-byte length
length = blob[idx]
idx += 1
# Sanity check length
if length <= 0 or idx + length > len(blob):
return None
# Extract string content
try:
return blob[idx:idx+length].decode('utf-8')
except UnicodeDecodeError:
# Try with error handling
try:
return blob[idx:idx+length].decode('utf-8', errors='replace')
except:
return None
def get_message_text(text: Optional[str], attributed_body: Optional[bytes]) -> Optional[str]:
"""
Get message text, preferring text column but falling back to attributedBody.
Args:
text: Value from message.text column
attributed_body: Value from message.attributedBody column
Returns:
Message text content or None
"""
# Prefer the text column if available
if text:
return text
# Fall back to parsing attributedBody
if attributed_body:
return extract_text_from_attributed_body(attributed_body)
return None
def extract_links(text: str) -> list[str]:
"""
Extract URLs from message text.
Args:
text: Message text content
Returns:
List of URLs found in the text
"""
import re
if not text:
return []
# Simple URL regex - matches http/https URLs
url_pattern = r'https?://[^\s<>"{}|\\^`\[\]]+'
return re.findall(url_pattern, text)
def is_reaction_message(associated_message_type: int) -> bool:
"""
Check if a message is a reaction/tapback.
Args:
associated_message_type: Value from message.associated_message_type
Returns:
True if this is a reaction, False if regular message
"""
if associated_message_type is None:
return False
return associated_message_type >= 2000
def get_reaction_type(associated_message_type: int) -> Optional[str]:
"""
Get the reaction type name from associated_message_type.
Args:
associated_message_type: Value from message.associated_message_type
Returns:
Reaction type name or None
"""
reaction_map = {
2000: 'loved',
2001: 'liked',
2002: 'disliked',
2003: 'laughed',
2004: 'emphasized',
2005: 'questioned',
2006: 'custom_emoji',
2007: 'custom_emoji',
3000: 'removed_love',
3001: 'removed_like',
3002: 'removed_dislike',
3003: 'removed_laugh',
3004: 'removed_emphasis',
3005: 'removed_question',
3006: 'removed_custom_emoji',
3007: 'removed_custom_emoji',
}
return reaction_map.get(associated_message_type)
def reaction_to_emoji(reaction_type: str) -> str:
"""Convert reaction type to emoji representation."""
emoji_map = {
'loved': '\u2764\ufe0f', # ❤️
'liked': '\U0001F44D', # 👍
'disliked': '\U0001F44E', # 👎
'laughed': '\U0001F602', # 😂
'emphasized': '\u203c\ufe0f', # ‼️
'questioned': '\u2753', # ❓
}
return emoji_map.get(reaction_type, '?')
