- Issue created by @Jaypan
- last update
10 months ago Composer error. Unable to continue. - last update
10 months ago Custom Commands Failed - Status changed to RTBC
9 months ago 11:40am 16 February 2024 - π§πͺBelgium wim leers Ghent π§πͺπͺπΊ
Trivial hardening, yes please! π
Thanks!
- π§πͺBelgium wim leers Ghent π§πͺπͺπΊ
This bug was introduced in π Large placeholders are not processed RTBC .
- π§πͺBelgium wim leers Ghent π§πͺπͺπΊ
Can somebody convert this to an MR? ππ
- First commit to issue fork.
- Status changed to Fixed
9 months ago 3:25pm 16 February 2024 - π¬π§United Kingdom catch
Committed/pushed to 11.x and cherry-picked to 10.2.x, thanks!
- π¬π§United Kingdom alexpott πͺπΊπ
I'm pretty sure this is the wrong fix.
We hit this error in π Use selenium/standalone-chrome instead of our chromedriver image Needs work . It occurs from the second call to checkMutation in this code.
function checkMutationAndProcess(node) { if (checkMutation(node)) { processReplacement(node); } // Checks if parent node of target node has not been processed, which can // occur if the script node was first observed with empty content and then // the child text node was added in full later. // @see `@ingroup large_chunk` for more information. else if (checkMutation(node.parentNode)) { processReplacement(node.parentNode); } }
There's no way that node can be a null but there is a way for node.parentNode to be NULL. In that issue we changed the code to:
// occur if the script node was first observed with empty content and then // the child text node was added in full later. // @see `@ingroup large_chunk` for more information. // If an element is added and then immediately (faster than the next // setImmediate is triggered) removed to a watched element of a // MutationObserver, the observer will notice and add a mutation for both // the addedNode and the removedNode - but the referenced element will not // have a parent node. else if (node.parentNode !== null && checkMutation(node.parentNode)) { processReplacement(node.parentNode); } }
Yes the code changed here will result in fixing the problem but I think the comment and code flow are better with the changes in π Use selenium/standalone-chrome instead of our chromedriver image Needs work .
- Status changed to Needs work
9 months ago 5:29pm 16 February 2024 - π¬π§United Kingdom catch
Reverted, let's extract the fix from that issue to here.
- Status changed to Needs review
9 months ago 6:54pm 16 February 2024 - π§πͺBelgium gorkagr
Implemented the suggestion if I understood it correctly
- Status changed to RTBC
9 months ago 10:30am 17 February 2024 - π¬π§United Kingdom alexpott πͺπΊπ
Looks great and from the π Use selenium/standalone-chrome instead of our chromedriver image Needs work issue we know it fixes the problem.
- Status changed to Fixed
9 months ago 3:34pm 17 February 2024 - π¬π§United Kingdom longwave UK
Committed and pushed da7f4e6327 to 11.x and 885932a213 to 10.2.x. Thanks!
-
longwave β
committed 885932a2 on 10.2.x
Issue #3416141 by gorkagr, Jaypan, Wim Leers, alexpott: big_pipe.js'...
-
longwave β
committed 885932a2 on 10.2.x
-
longwave β
committed da7f4e63 on 11.x
Issue #3416141 by gorkagr, Jaypan, Wim Leers, alexpott: big_pipe.js'...
-
longwave β
committed da7f4e63 on 11.x
- π¦πΊAustralia acbramley
Marked π Cannot read properties of null (reading 'nodeType') on node.page.body RTBC as a duplicate of this.
Automatically closed - issue fixed for 2 weeks with no activity.