Bugzilla

This diff attempts to get a window for url "folderstatus" but one was never associated with it, AFAICT. So those code changes should probably be removed from the patch.

This also has a change to obtain a stored password for a "windowless" url like folderstatus. This avoids aborting the login attempt as now occurs and you see in the log "login failed entirely", e.g., if folderstatus is the first URL. 

So if the url doesn't have a window we no longer abort the login. If a password is stored, we go ahead and use it instead of going into a function that potentially prompts for the password if one is not stored. With the new proposed method, if no password is stored, there is still a login failure for windowless URLs.

In some cases the "login failed entirely" log line seems innocuous (as the reporter is seeing). In other cases it is not, like when copying a folder or messages to/from a currently disconnected imap account.

This diff attempts to get a window for url "folderstatus" but one was never associated with it, AFAICT. So those code changes should probably be removed from the patch.

This also has a change to obtain a stored password for a "windowless" url like folderstatus. This avoids aborting the login attempt as now occurs and you see in the log "login failed entirely", e.g., if folderstatus is the first URL. 

So if the url doesn't have a window we no longer always abort the login. If a password is stored, we go ahead and use it instead of going into a function that potentially prompts for the password if one is not stored. With the new proposed method, if no password is stored, there is still a login failure for windowless URLs.

In some cases the "login failed entirely" log line seems innocuous (as the reporter is seeing). In other cases it is not, like when copying a folder or messages to/from a currently disconnected imap account.

This diff attempts to get a window for url "folderstatus" but one was never associated with it, AFAICT. So those code changes should probably be removed from the patch.

This also has a change to obtain a stored password for a "windowless" url like folderstatus. This avoids aborting the login attempt as now occurs and you see in the log "login failed entirely", e.g., if folderstatus is the first URL. 

So if the url doesn't have a window we no longer always abort the login. If a password is stored, we go ahead and use it instead of going into a function that potentially prompts for the password if one is not stored. With the new proposed method, if no password is stored there is still a login failure for windowless URLs.

In some cases the "login failed entirely" log line seems innocuous (as the reporter is seeing). In other cases it is not, like when copying a folder or messages to/from a currently disconnected imap account.