Mark,
That's where this gets a little odd.
On 2026-06-23 23:32, Mark Sapiro wrote:
On 6/23/26 9:29 AM, Thomas Ward via Mailman-Developers wrote:
In debugging and diagnosing issues with an archiver I wrote for a
client of mine, I need to see the message as it's seen by Mailman, in
its own Message class. However, I only have the *base* email message
in .eml format. While I know it's derived from the standard
email.message.Message class, is there a function that I can do that
will 'load' that existing email message into a mailman.emaiil.Message
object so I can do some work on it and properly look at the object
I'm working with in the archiver?
Pleasae be more specific as to what you are doing to get the message
to be archived.
The 'message' in question is processed by an archiver class that is
already written. However, to debug an issue we were encountering, I
wanted to see the object as it was a Message object as internally the
archiver process sees it.
What I wanted to do was see the object *exactly* as Mailman3 sees it
internally.
If as I suspect, you are getting messages to be archived by
subscribing your archiver to the list, you could convert that to a
mailman.email.message.Message object with something like
```
import sys
import email
sys.path.insert('/path/to/mailman')
from mailman.email.message import Message
# If the message is in a file
with open('path/to/file', 'rb') as fp:
msg = email.message_from_binary_file(fp, Message)
# Or if the message is piped in
msg = email.message_from_string(sys.stdin, Message)
```
but this won't be the same as Mailman's Message object because of
additional headers added by MTAs and transformations done by Mailman's
out runner.
Yeah, that's what I was wondering and was looking to convert an existing
email file (.eml) to the Message object that reflects the object as the
archiver class / plugin I already rote sees it.
In lieu of this as it won't be the same due to transformations and items
added by the MTA, I did a slightly different approach in the custom
archiver class (which has to preprocess the message into a specific
format for the home-grown environment's expected formats for it to
populate into that archive). During the process, I take the message as
it is from the Message class and wrap it into an
email.message.EmailMessage with
email.message_from_bytes(MessageObject.as_bytes(),
policy=email.policy.default). From there, it was easy to simply
manipulate the data as a traditional Python EmailMessage class (which
has extra functions beyond the base Message class which it seems is
preferred by Python upstream for message manipulation compared to the
traditional Message class) which lets us iterate over all parts of the
message and properly process attachments in the way that this home-grown
archiver that the custom archiver class hands the data off to can parse.
Non-trivial by any nature, but at least I have a workaround and thus a
workable EmailMessage object.
Thomas
_______________________________________________
Mailman-Developers mailing list -- [email protected]
To unsubscribe send an email to [email protected]
https://mail.python.org/mailman3/lists/mailman-developers.python.org/
Mailman FAQ: https://wiki.list.org/x/AgA3
Security Policy: https://wiki.list.org/x/QIA9
