Postponed (maintainer needs more info)
Project:
Drupal core
Version:
main
Component:
config.module
Priority:
Major
Category:
Task
Assigned:
Unassigned
Reporter:
Created:
22 Sep 2015 at 09:07 UTC
Updated:
30 Dec 2025 at 18:17 UTC
Jump to comment: Most recent, Most recent file
Comments
Comment #2
eli-tComment #5
eli-tComment #6
eli-tAdding interdiff to configuration-2247291-98.patch, the most recent patch on the parent issue from which this was split and has previously been RTBC.
Comment #7
eli-tComment #8
jhodgdonHm... I have a few questions/concerns about this patch:
a) Why is the confirm form being generated from a form state redirect? That is not the normal way to get to a confirm form. Normally, when you have a destructive operation, the trigger for getting to the confirm form is an ordinary link, while here for some reason it is a form submit with redirect set, which doesn't make sense to me and might not always work (sometimes the form system doesn't redirect the way you think it should, like for instance if some URL query parameters are set).
I guess I don't understand why the original page is still a form at all, since the submit just redirects. Couldn't it be an ordinary page with a link that would be used to import the shown config?
b) The new confirm form class has no documentation header. It definitely must have one.
c) For future reviews, a screen shot would be helpful. It's not so easy to set up a site to test this.
Comment #9
eli-tJust to manage expectations, I'm away for the next two weeks so will address points above on my return.
Comment #10
tkoleary commentedComment #11
eli-tIs that a remaining task, or a related issue?
Comment #12
tkoleary commentedAdded to config management usability meta #2642404: [meta] Usability improvements to configuration management post 8.0
Comment #13
xjmComment #14
cilefen commentedFrom the issue summary:
The argument can be made that this is a critical bug.
Comment #15
swentel commentedReroll + diff + screenshot
Comment #17
claudiu.cristeaNot in use. Remove.
Well, I guess no contrib tried to extend this because this is a BC break. In a conservative approach I would suggest to keep the signature and create a protected getter for the key-values store. Also we add a @todo to inject the service in 9.0.x. But normally there should be no use-case ti extend this form. Opinions?
Look as not-related?
$this->t()
s/array()/[]
$this->t()
Why line break? The 2nd phrase should continue on the 1st line.
s/array()/[]
Comment #18
eli-tReroll against latest 8.1.x before addressing points in #2572503-17: Add confirm dialog before full configuration import
Comment #19
eli-thttps://www.drupal.org/files/issues/add_confirm_dialog-2572503-18.patch is bad as it applies cleanly but doesn't take the changed code that made it failed to apply into account when it moves it elsewhere. I will fix later.
Comment #33
smustgrave commentedThank you for creating this issue to improve Drupal.
We are working to decide if this task is still relevant to a currently supported version of Drupal. There hasn't been any discussion here for over 8 years which suggests that this has either been implemented or is no longer relevant. Your thoughts on this will allow a decision to be made.
Since we need more information to move forward with this issue, the status is now Postponed (maintainer needs more info). If we don't receive additional information to help with the issue, it may be closed after three months.
Thanks!
Comment #34
smustgrave commentedwanted to bump 1 more time if still needed.