| commit | c0f8ce0fe56e8d885d0712120293f37bb740e759 | |
| author | pier11 <pier11@operamail.com> | |
| Mon, 29 Mar 2010 13:01:08 +0000 (29 14:01 +0100) | ||
| committer | pier11 <pier11@operamail.com> | |
| Mon, 29 Mar 2010 13:01:08 +0000 (29 14:01 +0100) | ||
| tree | 51c0352e67d1390f6cb8517c1bab31ccb4aa3fac | treesnapshot (tar.gz zip) |
| parent | 1976eae0b9bbd2cf75cb963e17c860bd69e3a761 | commitdiff |
access: optimization
to not explicitly add user to container
when user implicitly would belong to it
(For example if "sameEnterprise" is already
member of the container and user IS in that
enterprise).
For that we modify our container cache to
would-be state and calculate projected access
level.
If the AL is already that what we want, then
we don't add user explicitly to the container.
Complex? By design of the protocol...
to not explicitly add user to container
when user implicitly would belong to it
(For example if "sameEnterprise" is already
member of the container and user IS in that
enterprise).
For that we modify our container cache to
would-be state and calculate projected access
level.
If the AL is already that what we want, then
we don't add user explicitly to the container.
Complex? By design of the protocol...
| src/core/sipe.c | diffblobblamehistory |