[buddypress-trac] [BuddyPress Trac] #7298: xprofile fied user visibility settings regression

[buddypress-trac]

#7298: xprofile fied user visibility settings regression
-----------------------------------+--------------------
 Reporter:  hnla                   |       Owner:
     Type:  defect (bug)           |      Status:  new
 Priority:  high                   |   Milestone:  2.7.1
Component:  Extended Profile       |     Version:
 Severity:  major                  |  Resolution:
 Keywords:  has-patch 2nd-opinion  |
-----------------------------------+--------------------
Changes (by boonebgorges):

 * keywords:  needs-patch => has-patch 2nd-opinion


Comment:

 Thanks for the research on this so far.

 I agree that something like 2 or 3 is probably the best route for the time
 being. [attachment:7298.diff] is a somewhat less horrendous technique than
 2 (a special case hardcoded into `bp_user_can()`) while avoiding a huge
 amount of infrastructure to handle logged-out user caps. At the moment,
 the cases where we need to add caps for logged-out users are quite
 limited, and using a separate callback function for each makes it very
 explicit.

 It seems to me that WP will not easily be able to eliminate the
 inconsistency noted by DJPaul, for back compat reasons. BP demonstrates
 why cap checks for user 0 need to go through the same capability mapping
 filters as other cap checks.

 A side note: if we go with something like [attachment:7298.diff], we might
 want to consider adding a dynamic filter `bp_user_can_{$capability}`. This
 will make the callbacks cleaner, and will reduce greatly reduce overhead
 (`'bp_user_can'` can be called hundreds of times on a pageload).

--
Ticket URL: <https://buddypress.trac.wordpress.org/ticket/7298#comment:9>
BuddyPress Trac <http://buddypress.org/>
BuddyPress Trac