9893[BUGFIX] Fix for complex content object re-use
Commit message
Review
-
Code review CR 0
-
Verified V 0
Issues 1
Files 2
Review log 29
Show 4 earlier messages
-
Tolleiv Nietsch commented on patch set 1 05/27/13, 17:19
Looks good to me, approved Would you mind to share the typoscript which shows the bug? -
Stefan Neufeind commented on patch set 1 05/27/13, 17:19
Looks good to me, but someone else must approve reading commit-msg and code it looks obvious. It was incremented at the beginning and it seems to be correct to then also decrement it again (nest-count).and 6.0 of course -
Stefan Neufeind commented on patch set 1 05/27/13, 17:19
(1 inline comment) -
Sebastian Fischer commented on patch set 1 05/27/13, 17:19
Looks good to me, but someone else must approve
-
Philipp Gampe commented on patch set 1 05/27/13, 17:19
Looks good to me, but someone else must approve -
Mr. Jenkins commented on patch set 1 05/27/13, 17:19
Build Successful https://ci.typo3.org/job/typo3-v4-core-gerrit/3659/ : SUCCESS -
Alexander Achenbach commented on patch set 1 05/27/13, 17:19
To Tolleiv Nietsch: Thanks for your review. As mentioned in the bug report, it should be visible without any custom typoscript if you just put two CSS styled content section menus (Menu/Sitemap: Section index) as content elements onto the same page. The first section menu should render ok, the second would be empty (without the patch). There are more complex scenarios, such as re-use of a whole content set in a conditional (eg to test for presence of content before rendering wrappers or other content) if the content contains nested cases (such as a section index). The common case I hit here was a tt_content object with a renderObject, applied subsequently two or more times on the same page. -
Wouter Wolters uploaded patch set 2 05/28/13, 18:02
-
Mr. Jenkins commented on patch set 2 05/28/13, 18:03
Build Successful https://ci.typo3.org/job/typo3-v4-core-gerrit/8906/ : SUCCESS -
Sebastian Fischer commented on patch set 2 05/29/13, 06:27
Looks good to me, but someone else must approve -
Dmitry Dulepov commented on patch set 2 05/29/13, 13:34
Looks good to me, approved -
Stefan Neufeind commented on patch set 2 CR +1 10/14/13, 12:03
Needs to be tested, but looks quite obvious from the code (incremented on top, to be decremented at the bottom). -
Helmut Hummel commented on patch set 2 CR -1 10/14/13, 12:24
Needs tests -
Stefan Neufeind commented on patch set 2 11/14/13, 18:10
Could somebody try to create a small test? -
Benni Mack updated the commit message in patch set 3 11/24/14, 18:27
-
Benni Mack voted on patch set 3 CR +1 11/24/14, 18:27
-
Benni Mack commented on patch set 3 11/24/14, 18:30
actually, the example you mentioned works for me. But the fix should go in anyway to be clean and consistent. -
Benni Mack commented on patch set 3 11/24/14, 18:31
... works for me without the patch. -
Philipp Gampe commented on patch set 3 11/25/14, 15:05
This patch feels really lonely without a functional test. Please add one to ensure that we cover the various edge cases. This might be a pair coding task for the next code sprint. -
Georg Ringer updated the commit message in patch set 4 11/27/14, 12:09
-
Stefan Neufeind voted on patch set 4 CR +1 11/27/14, 12:35
-
Mathias Schreiber wrote 12/17/14, 20:36
Topic set to cObj -
Markus Klein voted on patch set 4 CR +1 12/23/14, 00:05
-
Helmut Hummel commented on patch set 4 CR -1 V -1 12/23/14, 00:43
There is no way to verify such a change. Please provide at least an example configuration of forge that shows actual and expected behavior so that anybody can go ahead and write a functional test for that change. -
Benni Mack commented on patch set 4 12/30/14, 15:56
So, if I read this correctly here, I consider abandoning the patch within the next 7 days for the following reasons: 1. I could not reproduce this behaviour (the explanation works for me without the patch), seems like nobody else could within the last year. 2. The original author did not respond to any comments in the last year. 3. No one (who can reproduce this issue) comes up with a test. -
Stefan Neufeind commented on patch set 4 12/30/14, 20:25
While I agree with you in terms of "we need to be able to reproduce it" just looking at the code I wonder how we could have such a counter and it wouldn't make a difference if it is properly decremented or not. So reading the code it looks "obvious". But I also didn't yet figure out how to reproduce the problem. Weird! -
Alexander Achenbach commented on patch set 1 12/30/14, 21:59
Hi everybody, sorry for the late reply, I had to locate the commits and original (less abstract) motivation. At that time we had implemented an automaketemplate-based web site that required conditional rendering of certain page parts (eg right column) based on whether those page parts actually produced content or not. For technical reasons the conditional rendering had to be made CSS-based, and I originally tried something like page.bodyTagCObject = COA page.bodyTagCObject { stdWrap.wrap = <body class="|"> stdWrap.required = 1 10 = TEXT 10.value = show-right-column 10.override { cObject = TEXT cObject { value = hide-right-column if.value = if.equals.cObject < temp.bodyInstance if.equals.cObject.workOnSubpart = right } } } and (partially) failed: While the above (IIRC) worked for static content handling, it did not work with section menus and other CONTENT objects that re-use CONTENT objects during their render() phase. In other words: Using an approach like the above and embedding eg a section menu into the right column should allow to reproduce the problem in an unpatched TYPO3, at least for the versions of TYPO3 I originally submitted the patch for. -
Benni Mack commented on patch set 4 01/01/15, 20:08
seems like this was fixed (once you look at the code in master) by Markus Klein on 9.2.2014: https://review.typo3.org/25512 - can we abandon this one (incl the forge issue) then? -
Stefan Neufeind abandoned the change 01/01/15, 20:48
I can confirm this was already fixed by https://review.typo3.org/25512 for 6.0 and above. And I don't think we'll backport this change to 4.5 (not critical enough imho).