You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
IMAP search terms and servers have limitations on the operations that can be performed server side. By default, AngusMail will fallback to client side searching which will get all message ids and perform client side searching on the email envelopes or body which is slow if the mailbox contains a lot of messages. I think there are some cases we could still use the server side search to narrow the number message ids that would then be used in the client side search. Currently the code is either all server side or all client side.
An example would be searching for UTF-8 subject in outlook365 which doesn't allow UTF-8 in IMAP search but can store emails with UTF-8 encoded subjects. For hybrid searching, we could examine the SubjectTerm to see if the UTF-8 data contained runs of ascii data that we could then pick the largest run or largest 2 runs of ascii data to create an AND term for the server side search. Then apply the client side search on the smaller set of message ids to match the full subject which contains the UTF-8 data to filter false positive matches.
The text was updated successfully, but these errors were encountered:
IMAP search terms and servers have limitations on the operations that can be performed server side. By default, AngusMail will fallback to client side searching which will get all message ids and perform client side searching on the email envelopes or body which is slow if the mailbox contains a lot of messages. I think there are some cases we could still use the server side search to narrow the number message ids that would then be used in the client side search. Currently the code is either all server side or all client side.
An example would be searching for UTF-8 subject in outlook365 which doesn't allow UTF-8 in IMAP search but can store emails with UTF-8 encoded subjects. For hybrid searching, we could examine the SubjectTerm to see if the UTF-8 data contained runs of ascii data that we could then pick the largest run or largest 2 runs of ascii data to create an AND term for the server side search. Then apply the client side search on the smaller set of message ids to match the full subject which contains the UTF-8 data to filter false positive matches.
The text was updated successfully, but these errors were encountered: