Merge pull request #381 from TeamMsgExtractor/next-release

Version 0.43.0

CHANGELOG.md

@@ -1,3 +1,9 @@
+**v0.43.0**
+* [[TeamMsgExtractor #56](https://github.com/TeamMsgExtractor/msg-extractor/issues/56)] [[TeamMsgExtractor #248](https://github.com/TeamMsgExtractor/msg-extractor/issues/248)] Added new function `MessageBase.asEmailMessage` which will convert the `MessageBase` instance, if possible, to an `email.message.EmailMessage` object. If an embedded MSG file on a `MessageBase` object is of a class that does not have this function, it will simply be attached to the instance as bytes.
+* Changed imports in `message_base.py` to help with type checkers.
+* Changed from using `email.parser.EmailParser` to `email.parser.HeaderParser` in `MessageBase.header`.
+* Changed some of the internal code for `MessageBase.header`. This should improve usage of it, and should not have any noticeable negative changes. You man notice some of the values parse slightly differently, but this effect should be mostly suppressed.
+
 **v0.42.2**
 * Fix bug in `AttachmentBase.mimetype` that would cause it to throw an error when accessed. This bug was introduced in `v0.42.0`.
 

README.rst

@@ -242,8 +242,8 @@ your access to the newest major version of extract-msg.
 .. |License: GPL v3| image:: https://img.shields.io/badge/License-GPLv3-blue.svg
    :target: LICENSE.txt
 
-.. |PyPI3| image:: https://img.shields.io/badge/pypi-0.42.2-blue.svg
-   :target: https://pypi.org/project/extract-msg/0.42.2/
+.. |PyPI3| image:: https://img.shields.io/badge/pypi-0.43.0-blue.svg
+   :target: https://pypi.org/project/extract-msg/0.43.0/
 
 .. |PyPI2| image:: https://img.shields.io/badge/python-3.8+-brightgreen.svg
    :target: https://www.python.org/downloads/release/python-3816/

extract_msg/__init__.py

@@ -28,7 +28,7 @@
 
 __author__ = 'Destiny Peterson & Matthew Walker'
 __date__ = '2023-08-02'
-__version__ = '0.42.2'
+__version__ = '0.43.0'
 
 __all__ = [
     # Modules:

extract_msg/msg_classes/message_base.py

@@ -5,6 +5,7 @@
 
 import base64
 import datetime
+import email.message
 import email.utils
 import functools
 import html
@@ -20,7 +21,9 @@
 import compressed_rtf
 import RTFDE
 
-from email.parser import Parser as EmailParser
+from email import policy
+from email.message import EmailMessage
+from email.parser import HeaderParser
 from typing import Callable, List, Optional, Union
 
 from .. import constants
@@ -30,8 +33,8 @@
         BodyTypes, DeencapType, ErrorBehavior, RecipientType, SaveType
     )
 from ..exceptions import (
-        DataNotFoundError, DeencapMalformedData, DeencapNotEncapsulated,
-        IncompatibleOptionsError, WKError
+        ConversionError, DataNotFoundError, DeencapMalformedData,
+        DeencapNotEncapsulated, IncompatibleOptionsError, WKError
     )
 from .msg import MSGFile
 from ..structures.report_tag import ReportTag
@@ -155,6 +158,59 @@ def _genRecipient(self, recipientType, recipientInt : RecipientType) -> Optional
 
         return value
 
+    def asEmailMessage(self) -> EmailMessage:
+        """
+        Returns an instance of EmailMessage used to represent the contents of
+        this message.
+
+        :raises ConversionError: The function failed to convert one of the
+            attachments into a form that it could attach, and the attachment
+            data type was not None.
+        """
+        ret = EmailMessage()
+
+        # Copy the headers.
+        for key, value in self.header.items():
+            ret[key] = value
+
+        # Attach the body to the EmailMessage instance.
+        if self.htmlBody:
+            ret.set_content(self.body, subtype = 'html', cte = 'quoted-printable')
+        elif self.body:
+            ret.set_content(self.body, cte = 'quoted-printable')
+
+        # Process attachments.
+        for att in self.attachments:
+            if att.dataType:
+                if hasattr(att.dataType, 'asEmailMessage'):
+                    # Replace the extension with '.eml'.
+                    filename = att.getFilename()
+                    if filename.lower().endswith('.msg'):
+                        filename = filename[:-4] + '.eml'
+                    ret.add_attachment(
+                                        att.data.asEmailMessage(),
+                                        filename = filename,
+                                        cid = att.contentId)
+                else:
+                    if issubclass(att.dataType, bytes):
+                        data = att.data
+                    elif issubclass(att.dataType, MSGFile):
+                        if hasattr(att.dataType, 'asBytes'):
+                            data = att.asBytes
+                        else:
+                            data = att.data.exportBytes()
+                    else:
+                        raise ConversionError(f'Could not find a suitable method to attach attachment data type "{att.dataType}".')
+                    mime = att.mimetype or 'application/octet-stream'
+                    mainType, subType = mime.split('/')[0], mime.split('/')[-1]
+                    ret.add_attachment(data,
+                                       maintype = mainType,
+                                       subtype = subType,
+                                       filename = att.getFilename(),
+                                       cid = att.contentId)
+
+        return ret
+
     def deencapsulateBody(self, rtfBody : bytes, bodyType : DeencapType) -> Optional[Union[bytes, str]]:
         """
         A function to deencapsulate the specified body from the rtfBody. Returns
@@ -1009,11 +1065,12 @@ def header(self) -> email.message.Message:
         """
         headerText = self.headerText
         if headerText:
-            header = EmailParser().parsestr(headerText)
-            header['date'] = self.date
+            header = HeaderParser(policy = policy.default).parsestr(headerText)
+            del header['Date']
+            header['Date'] = self.date
         else:
             logger.info('Header is empty or was not found. Header will be generated from other streams.')
-            header = EmailParser().parsestr('')
+            header = HeaderParser(policy = policy.default).parsestr('')
             header.add_header('Date', self.date)
             header.add_header('From', self.sender)
             header.add_header('To', self.to)