41798[BUGFIX] Respect rootline order for getPagesTSconfig()
Commit message
Review
-
Code review CR +2
- Stefan Neufeind +2
- Ralf Merz +1
- Markus Klein +1
-
Verified V +2
- Stefan Neufeind +2
- Markus Klein +1
Issues 1
Files 1
Review log 21
-
Daniel Goerz uploaded patch set 1 07/21/15, 22:23
-
Daniel Goerz voted on patch set 1 V +1 07/22/15, 05:51
-
Hans Mayer voted on patch set 1 CR +1 07/22/15, 07:35
-
Hans Mayer voted on patch set 1 V +1 07/22/15, 16:24
-
Markus Klein updated the commit message in patch set 2 07/22/15, 20:13
-
Markus Klein commented on patch set 2 07/22/15, 20:13
Is there any chance to get a unit test maybe for this? -
Daniel Goerz commented on patch set 2 07/22/15, 20:28
Hmmm I wonder how to deal with the static calls in a unit tests. Would be curios to see a working test but I'm afraid I cant provide one myself :( -
Daniel Goerz voted on patch set 2 V +1 07/22/15, 20:29
-
Markus Klein commented on patch set 2 07/22/15, 20:31
TypoScriptFrontendController.php:4167can you maybe add a comment here why we have to reverse that stuff? what order is the rootline in? -
Daniel Goerz commented on patch set 2 07/22/15, 20:39
TypoScriptFrontendController.php:4167ReplySure. I will add a comment. We have a related situation BackenUtility::getPagesTSconfig() btw where the $rootline is simply reversed by ksort() (see line 1234) as the keys are the rootlevel. The root page (level 0) is the last entry in the array and therefore always "wins" if the array is not reversed. -
Daniel Goerz uploaded patch set 3 07/22/15, 20:44
-
Daniel Goerz commented on patch set 2 07/22/15, 20:45
Removed the unneeded array_unshift() as well -
Daniel Goerz voted on patch set 3 V +1 07/22/15, 20:45
-
Ralf Merz commented on patch set 3 CR +1 07/23/15, 08:28
good by reading -
Markus Klein commented on patch set 3 CR -1 07/27/15, 07:19
TypoScriptFrontendController.php:4164I took a closer look now at the implementation for BE. I must say... throw this thing away and use the BE implementation. This is wrong/broken beyond repair. a) uses different parser than BE b) does not emit the signal, which allows custom processing for extensions If you could make that one working with the BE function, it would be really nice. We also need some way to test that in FE (extension?), since this is only used for a few things by Core. -
Daniel Goerz commented on patch set 3 07/27/15, 07:48
> (1 comment) Hi Markus. I agree that the implementation is suboptimal. Anyway the code is in there and returns currently plain wrong data. The patch fixes that without changing too much. I'd say we should merge it to get at least correct return value and think about the refactoring after that, becauses that implies a lot more work. I could provide a simple reproduce extension or post code in the Ticket if you are not totally blocking this one =) -
Daniel Goerz commented on patch set 3 V 0 07/28/15, 18:18
Added an extension to reproduce the issue and test the patch in the forge ticket. -
Markus Klein commented on patch set 3 CR +1 V +1 07/28/15, 18:28
Thanks for the test extension. works nicely. will do the refactoring later (tm) -
Stefan Neufeind commented on patch set 3 CR +1 V +1 07/28/15, 19:02
Well done! Unit-test for it would be super, but I'd be fine without one for the moment (to get this through). -
Stefan Neufeind commented on patch set 3 CR +2 V +2 07/28/15, 19:08
as Markus said: > this is basically a copy of BEutility which has tests So I'd be fine without tests here for the moment. Should be refactored anyhow. -
Stefan Neufeind merged patch set 3 07/28/15, 19:08