62640 - Improving the MsgIncomingServer interface for authentication state

Status: NEW

(MailNews Core :: Networking, defect)

Description

[justin]

Both the POP and IMAP code had separate interfaces for setting and querying the
authentication state of the msg server. This patch makes the attribute
"authenticated" accessible from the nsIMsgIncomingServer interface, removes the
old implementations from the POP and IMAP code, and makes everything use the new
attribute on the generic interface.

This allows for future extensions (such as X-Sender) to cleanly (and
generically) access the authentication state, without requiring special code to
deal with each incoming server's implementation.

Comment 1

[justin]

Attachment: Patch to fix interface problems described

Comment 2

[[e-mail] (gone - use [e-mail] instead)]

Sounds good, putterman, sspitzer, mscott?

Comment 3

[scottputterman]

adding patch keyword

Comment 4

[Keyser Sose]

Marking NEW so someone can take a look at this patch.

Comment 5

[Joshua Cranmer [:jcranmer]]

Seems valid enough to me.

Bienvenu?

Comment 6

[David :Bienvenu]

yeah, seems valid. I doubt the patch still applies, but the idea should work.

Comment 7

[Wayne Mery (:wsmwk)]

Patch still relevant?

Comment 8

[Magnus Melin [:mkmelin]]

Yes. Just needs someone to dust off the patch.

Comment 9

[Liscare]

Attachment: Bug 62640 - Improving the MsgIncomingServer interface for authentication state. r=aceman

Comment 10

[Wayne Mery (:wsmwk)]

Thanks lilian. I have changed the review request to Magnus

Comment 11

[Wayne Mery (:wsmwk)]

(In reply to justin from comment #0)

Both the POP and IMAP code had separate interfaces for setting and querying
the
authentication state of the msg server. This patch makes the attribute
"authenticated" accessible from the nsIMsgIncomingServer interface, removes
the
old implementations from the POP and IMAP code, and makes everything use the
new
attribute on the generic interface.

This allows for future extensions (such as X-Sender) to cleanly (and
generically) access the authentication state, without requiring special code
to
deal with each incoming server's implementation.

Magnus wrote the phab, "not sure it [the patch] is wanted if there is no clear use case."

Is comment 0 the use case you are referring to?

Comment 12

[Magnus Melin [:mkmelin]]

I guess it would be a small improvement yes.