No LDAP Plugin in 4.0 new installs. If needed it can be installed from a joomla-extensions repo.
There is a unmaintained LDAP Plugin in the core that is disabled by default.
Please leave your opinions on this topic below ;)
Labels |
Added:
?
|
Sure we should drop the client lib too from the core install.
It is maintained and it is used.
Category | ⇒ | Authentication com_installer |
Status | New | ⇒ | Discussion |
It is maintained
Without any way to test it as it require a complex setup and to real test it you need a original Active Directory.
and it is used
Would be great to know who we can ask to test changes. Weblinks was also used and i guess much more than ldap similar apply to geshi they got dropped too. ;)
Without any way to test it as it require a complex setup and to real test it you need a original Active Directory.
Actually, I just put in a lot of work writing functional tests for the LDAP client so we actually do for the first time ever have real testing of our LDAP code against a real LDAP source. And half a clue how awkward our API actually is as a result.
We definitely need LDAP. It's almost essential for any sort of scaling company
I'm going to make an executive decision and reject this one
Status | Discussion | ⇒ | Closed |
Closed_Date | 0000-00-00 00:00:00 | ⇒ | 2017-10-13 16:13:51 |
Closed_By | ⇒ | wilsonge |
Hmm this does not solve the issue that we can't test that thing on real sites. Nor test changes pre release. (like the last one)
We definitely need LDAP. It's almost essential for any sort of scaling company
Sure that is the reason i suggested to move it to joomla-extensions.
:(
Yes, we can test the thing. There are instructions to set up LDAP servers for many platforms, and we have real (non-mocked) functional testing of the LDAP client code against a real LDAP server. So the only thing that does not have proper testing is how the plugin itself behaves. Moving it out of core and into an abandonware repo doesn't fix that particular concern.
joomla-framework/ldap@e0c19e6 here are the tests btw. I merged them from @mbabker last week
Explain how the plugin's unmaintained? The fact it isn't updated as frequently as other parts of the system doesn't mean it's unmaintained. So, what are we really gaining by dropping this (and IMO if we drop the plugin we should drop the library client class too because it is even worse than the plugin as far as structure and maintainability goes) other than removing a low-maintenance burden?