Review main Merged

13032[FEATURE] New hook for managing page cache

Francois Suter CR +2 V +2 Easy +15 −1 updated 06/03/13, 20:49

Commit message

When pages are cached using some extra system (e.g. nc_staticfilecache)
it is useless to store these as well in whatever cache backend is used.
Thanks to a new hook, storage of a page in cache can be disabled
selectively.

Review

  • Code review CR +2
    • Francois Suter +2
    • Julien Henchoz +1
    • Christian Kuhn +1
  • Verified V +2
    • Francois Suter +2
    • Julien Henchoz +1
    • Christian Kuhn +1

Files 1

Review log 9

  1. Ingo Renner commented on patch set 1 07/30/12, 07:36

    Please see my comment in forge. I don't think this is the correct/clean approach here.
  2. Francois Suter commented on patch set 1 07/30/12, 07:36

    (1 inline comment) Thanks for the review. I must say I have no idea how to design a test in such a circumstance... I also answered your comment.
  3. Francois Suter commented on patch set 1 07/30/12, 07:36

    I suppose you're right. My problem is that I have no resources to implement such a sophisticated solution. So either this goes in as is or I'll keep patching my Core... This hook could still be useful until a better solution is developed.
  4. Christian Kuhn commented on patch set 1 07/30/12, 07:36

    Ah, btw: it would be cool to see a basic test for this hook (a callback test maybe with a test if the page cache call is calles if usePageCache is manipulated), but I have no time for this currently and do not demand it therefore.

    class.tslib_fe.php:2800

    Giving $usePageCache to the hook means, that extension authors must handle the outcome of possible other hooks that are called before properly. There are several variants to circumvent this, but it is also pretty unlikely that more than one extension uses this hook in one instance. For now, I'm fine with the given construct, especially together with the comment in front of the hook.
  5. Christian Kuhn commented on patch set 1 1 comment 07/30/12, 07:36

    Verified; Looks good to me, but someone else must approve
  6. Mr. Jenkins commented on patch set 1 07/30/12, 07:36

  7. Julien Henchoz commented on patch set 1 07/30/12, 07:36

    Verified; Looks good to me, but someone else must approve Tried it on a Introduction package v6, works fine.
  8. Francois Suter commented on patch set 1 V 0 07/30/12, 07:36

    by team member; Looks good to me, approved Thanks for the reviews.
  9. Francois Suter merged patch set 1 07/30/12, 07:36