Oh, and one more question: Is there any way to clean out the bogus
revisions? Like I said, they make the RCS navigation rather useless.
Bye,
Andreas
Andreas Flack schrieb:
Hi!
I just played around a little and it seems that the locking mechanism
really is the culprit. If I just hit F5 in an edit form, two revisions
are produced. When I disable the lock() function in metadata.php, this
stops. When I save, another three revisions are produced (I don't know
why it's sometimes five and sometimes six). When I disable the unlock()
function, this returns to normal.
BTW: Approving an article produces two revisions as well, it would be
good to disable RCS here as well.
Bye,
Andreas
Henri Bergius schrieb:
Hi,
On Tue, Jun 24, 2008 at 1:34 AM, Andreas Flack
<[EMAIL PROTECTED]> wrote:
It happened on another machine as well and on a third one, I at least
see
strange-looking RCS histories like the one above (didn't want to try
on a
production system though). It seems that each save operation produces
something like 6 RCS revisions, with completely bogus line change
information.
Six revisions per save sounds like a lot, yes. I would assume the new
locking system in datamanager2 would cause some duplicate revisioning,
but I have no idea where the rest come from.
This needs to be tracked down.
@Arttu: It would probably be a good idea to set $object->_use_rcs to
false before calling update() to lock or unlock an object, so RCS will
not catch it.
Andreas
/Henri
_______________________________________________
user mailing list
[email protected]
http://lists.midgard-project.org/mailman/listinfo/user
_______________________________________________
user mailing list
[email protected]
http://lists.midgard-project.org/mailman/listinfo/user
