olm2eml/olm_xml_to_eml.py

#!/usr/bin/env python3
"""
Convert emails extracted from an OLM file (as XML) to standard .eml format.

Usage:
    python olm_xml_to_eml.py /path/to/emails/root /path/to/output_dir
"""

import argparse
import re
import sys
import xml.etree.ElementTree as ET
from email import encoders
from email.mime.base import MIMEBase
from email.mime.image import MIMEImage
from email.mime.multipart import MIMEMultipart
from email.mime.text import MIMEText
from email.utils import formataddr, formatdate
from pathlib import Path

EMAILS_ROOT_ATTR = '{http://www.w3.org/XML/1998/namespace}space'
ADDR_ADDRESS = 'OPFContactEmailAddressAddress'
ADDR_NAME = 'OPFContactEmailAddressName'
ADDR_TYPE = 'OPFContactEmailAddressType'

ATTACH_EXT = 'OPFAttachmentContentExtension'
ATTACH_SIZE = 'OPFAttachmentContentFileSize'
ATTACH_CID = 'OPFAttachmentContentID'
ATTACH_TYPE = 'OPFAttachmentContentType'
ATTACH_NAME = 'OPFAttachmentName'
ATTACH_URL = 'OPFAttachmentURL'


def parse_addresses(parent_elem):
    addresses = []
    if parent_elem is None:
        return addresses
    for child in parent_elem:
        if child.tag == 'emailAddress':
            addr = child.get(ADDR_ADDRESS, '')
            name = child.get(ADDR_NAME, '')
            addresses.append((name, addr))
    return addresses


def format_address_list(addresses):
    parts = []
    for name, addr in addresses:
        if name and addr:
            parts.append(formataddr((name, addr)))
        elif addr:
            parts.append(addr)
    return ', '.join(parts) if parts else ''


def parse_attachment_list(parent_elem):
    attachments = []
    if parent_elem is None:
        return attachments
    for child in parent_elem:
        if child.tag == 'messageAttachment':
            att = {
                'name': child.get(ATTACH_NAME, 'attachment'),
                'content_type': child.get(ATTACH_TYPE, 'application/octet-stream'),
                'extension': child.get(ATTACH_EXT, ''),
                'content_id': child.get(ATTACH_CID, ''),
                'url': child.get(ATTACH_URL, ''),
                'size': child.get(ATTACH_SIZE, '0'),
            }
            attachments.append(att)
    return attachments


def find_email_xmls(root_dir):
    xml_files = []
    root_path = Path(root_dir)
    for path in root_path.rglob('message_*.xml'):
        xml_files.append(path)
    return sorted(xml_files)


def get_email_element(xml_path):
    tree = ET.parse(str(xml_path))
    root = tree.getroot()
    return root[0]


def text_of(elem):
    if elem is None:
        return ''
    return (elem.text or '').strip()


def convert_email_xml_to_eml(xml_path, emails_root):
    email = get_email_element(xml_path)

    subject = text_of(email.find('OPFMessageCopySubject'))

    from_addrs = parse_addresses(email.find('OPFMessageCopyFromAddresses'))
    to_addrs = parse_addresses(email.find('OPFMessageCopyToAddresses'))
    cc_addrs = parse_addresses(email.find('OPFMessageCopyCCAddresses'))
    reply_to_addrs = parse_addresses(email.find('OPFMessageCopyReplyToAddresses'))

    sender_addr = text_of(email.find('OPFMessageCopySenderAddress'))
    display_to = text_of(email.find('OPFMessageCopyDisplayTo'))

    sent_time = text_of(email.find('OPFMessageCopySentTime'))
    received_time = text_of(email.find('OPFMessageCopyReceivedTime'))
    mod_time = text_of(email.find('OPFMessageCopyModDate'))
    server_time = text_of(email.find('ExchangeServerLastModifiedTime'))

    message_id = text_of(email.find('OPFMessageCopyMessageID'))
    in_reply_to = text_of(email.find('OPFMessageCopyInReplyTo'))
    references = text_of(email.find('OPFMessageCopyReferences'))
    thread_topic = text_of(email.find('OPFMessageCopyThreadTopic'))
    conversation_id = text_of(email.find('OPFMessageCopyExchangeConversationId'))

    html_body = text_of(email.find('OPFMessageCopyHTMLBody'))
    body = text_of(email.find('OPFMessageCopyBody'))

    attachments = parse_attachment_list(email.find('OPFMessageCopyAttachmentList'))

    mime_msg = MIMEMultipart('mixed')

    mime_msg['Subject'] = subject

    from_str = format_address_list(from_addrs)
    if from_str:
        mime_msg['From'] = from_str
    elif sender_addr:
        mime_msg['From'] = sender_addr
    else:
        mime_msg['From'] = 'unknown@unknown.com'

    if from_addrs and display_to:
        mime_msg['To'] = display_to
    else:
        to_str = format_address_list(to_addrs)
        if to_str:
            mime_msg['To'] = to_str
        else:
            mime_msg['To'] = 'undisclosed-recipients:;'

    cc_str = format_address_list(cc_addrs)
    if cc_str:
        mime_msg['Cc'] = cc_str

    reply_to_str = format_address_list(reply_to_addrs)
    if reply_to_str:
        mime_msg['Reply-To'] = reply_to_str

    date_str = sent_time or received_time or mod_time or server_time
    if date_str:
        mime_msg['Date'] = date_str
    else:
        mime_msg['Date'] = formatdate(localtime=True)

    if message_id:
        mime_msg['Message-ID'] = message_id

    if in_reply_to:
        mime_msg['In-Reply-To'] = in_reply_to

    if references:
        mime_msg['References'] = references

    if thread_topic:
        mime_msg['Thread-Topic'] = thread_topic

    if conversation_id:
        mime_msg['Thread-Index'] = conversation_id

    mime_msg['MIME-Version'] = '1.0'

    has_html = bool(html_body)
    has_text = bool(body)

    if has_html or has_text:
        plain_text = strip_html(body) if body else ''
        if has_html:
            body_part = MIMEMultipart('alternative')
            if plain_text:
                body_part.attach(MIMEText(plain_text, 'plain', 'utf-8'))
            body_part.attach(MIMEText(html_body, 'html', 'utf-8'))
            mime_msg.attach(body_part)
        elif plain_text:
            mime_msg.attach(MIMEText(plain_text, 'plain', 'utf-8'))
        else:
            mime_msg.attach(MIMEText('', 'plain', 'utf-8'))
    else:
        mime_msg.attach(MIMEText('', 'plain', 'utf-8'))

    for att in attachments:
        att_url = att['url']
        if not att_url:
            continue
        att_path = Path(emails_root) / att_url
        if not att_path.exists():
            continue
        try:
            data = att_path.read_bytes()
            content_type = att['content_type'] or 'application/octet-stream'
            if '/' in content_type:
                main_type, sub_type = content_type.split('/', 1)
            else:
                main_type, sub_type = 'application', 'octet-stream'

            if main_type == 'image':
                part = MIMEImage(data, _subtype=sub_type)
            else:
                part = MIMEBase(main_type, sub_type)
                part.set_payload(data)
                encoders.encode_base64(part)

            filename = att['name']
            cid = att['content_id']
            if cid:
                part.add_header('Content-ID', f'<{cid}>')
                part.add_header('Content-Disposition', 'inline', filename=filename)
            else:
                part.add_header('Content-Disposition', 'attachment', filename=filename)

            mime_msg.attach(part)
        except Exception as e:
            print(f"  Warning: failed to attach {att_url}: {e}", file=sys.stderr)

    return mime_msg.as_bytes()


def strip_html(html_text):
    text = re.sub(r'<style[^>]*>.*?</style>', '', html_text, flags=re.DOTALL)
    text = re.sub(r'<script[^>]*>.*?</script>', '', text, flags=re.DOTALL)
    text = re.sub(r'<[^>]+>', ' ', text)
    text = re.sub(r'\s+', ' ', text)
    return text.strip()


def build_output_path(xml_path, emails_root, output_dir, preserve_structure):
    if preserve_structure:
        rel = xml_path.parent.relative_to(Path(emails_root).resolve())
        out = Path(output_dir) / rel / f"{xml_path.stem}.eml"
    else:
        out = Path(output_dir) / f"{xml_path.stem}.eml"
    out.parent.mkdir(parents=True, exist_ok=True)
    return out


def main():
    parser = argparse.ArgumentParser(description='Convert OLM-extracted XML emails to EML format')
    parser.add_argument('input_dir', help='Root directory of the extracted OLM emails')
    parser.add_argument('output_dir', help='Output directory for EML files')
    parser.add_argument('--preserve-structure', action='store_true',
                        help='Preserve folder structure in output (mirrors account/folder hierarchy)')
    parser.add_argument('--overwrite', action='store_true', help='Overwrite existing EML files')

    args = parser.parse_args()

    input_dir = Path(args.input_dir).resolve()
    output_dir = Path(args.output_dir).resolve()

    if not input_dir.is_dir():
        print(f"Error: input directory not found: {input_dir}", file=sys.stderr)
        sys.exit(1)

    output_dir.mkdir(parents=True, exist_ok=True)

    xml_files = find_email_xmls(input_dir)
    if not xml_files:
        print(f"No message_*.xml files found in {input_dir}", file=sys.stderr)
        sys.exit(1)

    print(f"Found {len(xml_files)} email XML files")

    success = 0
    skipped = 0
    errors = 0

    for xml_path in xml_files:
        out_path = build_output_path(xml_path, input_dir, output_dir, args.preserve_structure)

        if out_path.exists() and not args.overwrite:
            skipped += 1
            continue

        try:
            eml_bytes = convert_email_xml_to_eml(xml_path, input_dir)
            out_path.write_bytes(eml_bytes)
            success += 1
        except Exception as e:
            print(f"Error converting {xml_path}: {e}", file=sys.stderr)
            errors += 1

    print(f"\nDone. {success} converted, {skipped} skipped, {errors} errors.")


if __name__ == '__main__':
    main()