? ? Pending

User tests: Successful: Unsuccessful:

avatar alikon
alikon
22 May 2020

Pull Request for Issue #29196 .

Summary of Changes

hide db credential fields

Testing Instructions

curl -H "Authorization: Bearer YOUR_TOKEN" {{URL}}/api/index.php/v1/config/application

Expected result

db credentials not exposed

Actual result

exposed

Documentation Changes Required

?

avatar alikon alikon - open - 22 May 2020
avatar alikon alikon - change - 22 May 2020
Status New Pending
avatar alikon alikon - change - 22 May 2020
Labels Added: ? ?
avatar brianteeman
brianteeman - comment - 22 May 2020

No need to reinvent the wheel. The "safe" data has already been defined for use by the admin sysinfo

* Remove sections of data marked as private in the privateSettings

avatar alikon
alikon - comment - 22 May 2020

good catch

avatar toivo toivo - test_item - 23 May 2020 - Tested successfully
avatar toivo
toivo - comment - 23 May 2020

I have tested this item successfully on d51fb22

Tested in Nightly Build of May 23 successfully.


This comment was created with the J!Tracker Application at issues.joomla.org/tracker/joomla-cms/29198.

avatar wilsonge
wilsonge - comment - 24 May 2020

Made this a draft so it's clear to others not to merge this until discussions in original issue are complete

avatar zero-24
zero-24 - comment - 25 May 2020

Well I personally would say that this PR here that is just hiding the sensible fields for now is the right thing todo. Removing the endpoint at all is discussed in the other issue. Or do you think we should expose the sensible fields via the API too? We even don't do that in the UI for all fields.

avatar wilsonge
wilsonge - comment - 25 May 2020

Yes. I'd expose the fields if you have the permissions to view the fields. You have to be a super admin. so why not? Database in the API allows you to do proper DevOps deploys of your website from say a staging to a production site (restore a backup and update the global configuration to the Prod values)

avatar particthistle
particthistle - comment - 18 Jul 2020

Reading #29196 which was closed June 3 should this now be closed @alikon to clear it out of the patch tester and pending PR list?

Still getting the hang of the protocol on things here so feel free to correct my perspective if that's not how these types of things are managed.

Discovered during PBF July 18 while testing other API PRs.

avatar richard67
richard67 - comment - 18 Jul 2020

I am a bit confused now. Issue #29196 has been closed as intended behavior. But maybe this PR still makes sense, now not as a bug fix but as a new feature? The discussion above seems to be open to me, but also stale, nothing new since a while. @alikon @wilsonge @zero-24 how to continue here?

avatar alikon
alikon - comment - 18 Jul 2020

i really don't know what i can do more
i've opened this pr as a RFC cause of #29196
after, it has been demoted to draft
after #29196 has been closed
...after 2 months...
so i guess the best think i can do
it is simply to close this one

avatar alikon alikon - change - 18 Jul 2020
Status Pending Closed
Closed_Date 0000-00-00 00:00:00 2020-07-18 14:21:12
Closed_By alikon
avatar alikon alikon - close - 18 Jul 2020

Add a Comment

Login with GitHub to post a comment