The current download_attachment method always has to write to file:
|
def download_attachment( |
|
self, |
|
attachment: EmailBodyPart, |
|
file_name: Union[str, Path], |
|
) -> None: |
|
if not file_name: |
|
raise Exception("Destination file name is required") |
|
file_name = Path(file_name) |
|
blob_url = self.jmap_session.download_url.format( |
|
accountId=self.account_id, |
|
blobId=attachment.blob_id, |
|
name=attachment.name, |
|
type=attachment.type, |
|
) |
|
r = self.requests_session.get( |
|
blob_url, stream=True, timeout=REQUEST_TIMEOUT |
|
) |
|
r.raise_for_status() |
|
with open(file_name, "wb") as f: |
|
f.write(r.raw.data) |
It would be handy to optionally write to memory, ie. being able to use io.BytesIO objects. This would make the library more flexible to be used as a library. A use case being for example an ETL tool that processes the contents of email attachments.
I'm happy to contribute as well.
The current
download_attachmentmethod always has to write to file:jmapc/jmapc/client.py
Lines 151 to 170 in f8200fa
It would be handy to optionally write to memory, ie. being able to use
io.BytesIOobjects. This would make the library more flexible to be used as a library. A use case being for example an ETL tool that processes the contents of email attachments.I'm happy to contribute as well.