Wait, what?

When you break a test unintentionally and someone else kindly points that out to you, you'd say "oh, thank you very much", then you go on to try to fix a problem.

Instead you decide to attack the test, claiming that it's brittle, and demand that someone else fixes it.

That is not nice.



It looks like the change we are talking about is [1]. This involves pulling out a hard-coded functionality (LDAP group filter) and demoting it to an implementation of an extension point.

This is a big change that spans both UI and functionality. Of course it breaks a test, and that's a test's way of asking you if you really meant to change that part of the behaviour.

So you change the test, to re-confirm that you intended it. Just like you had to update src/test/java/.../LDAPSecurityRealm_Test.java in this change. This is a test working as it is supposed to be.


These tests cover a lot of things, including UI. I don't understand the notion that this is somehow bad and that we actively should avoid testing the UI, when it is one of the problems we often fail to catch in unit tests.

They are used to qualify LTS releases, and they are currently the only tests that actually run LDAP plugin with a real LDAP server. I'm sure there are ways to make it better, but these tests have unquestionably positive values as they stand today already.

But you and I disagree, and I respect your choice of not writing these tests for your code. Likewise, please show some respect to what other people are doing in the community.



[1] https://github.com/jenkinsci/ldap-plugin/commit/a375a65b639447d22807f3d03d9d5a3e73637fc1

On 05/24/2014 06:49 AM, Stephen Connolly wrote:
I am not planning to update UI driven tests of non UI fubctionality as I
fundamentally disagree with that approach. I will probably replace these tests
with a non-UI driven version when I have integrated my scalability test
framework into the acceptance test harness which will enable the writing of
sensible tests and leave UI driven tests for what they are best for... Ie
testing the ui

On Saturday, 24 May 2014, Ullrich Hafner <[email protected]
<mailto:[email protected]>> wrote:

     Are you planning to update the test?

     Am Freitag, 23. Mai 2014 schrieb Stephen Connolly :

         Well if the tests are non UI tests driven through the UI then they can
         be overly brittle.

         @Kohsuke this is a case in point

         On Thursday, 22 May 2014, Ulli Hafner <[email protected]> wrote:

             Seems that the new plugin breaks the acceptance tests for the LDAP
             plugin:
             
https://github.com/jenkinsci/acceptance-test-harness/blob/master/src/test/java/plugins/LdapPluginTest.java

             Am 22.05.2014 um 17:12 schrieb Stephen Connolly
             <[email protected]>:

            OK, so there is now rumoured to be a faster and better way to look
            up the groups that a user belongs to in the LDAP 1.10 plugin.

            I say rumoured because due to the complexities of Active Directory
            server configurations, one can never be quite sure until one has
            had a fair amount of testing.

            To that end, please could you set up a simple test Jenkins
            instance and upgrade to ldap:1.10 and configure the `Parse user
            attribute for list of groups` group membership strategy (again
            rumour has it that on Active Directory the attribute `memberOf` is
            the magic attribute.

            See if that ends up giving you the same JENKINS_URL/whoAmI list of
            groups as when you have the `Search for groups containing user`
            set with the filter being
            `(member:1.2.840.113556.1.4.1941:={0})`... though the `Parse user
            attribute for list of groups` should be very very fast for login
            while the `Search for groups containing user` could take *ages*.

            Respond back here with your findings so that I can remove the Red
            text on the version history about this being a rumour

            --
            You received this message because you are subscribed to the Google
            Groups "Jenkins Users" group.
            To unsubscribe from this group and stop receiving emails from it,
            send an email to [email protected].
            For more options, visit https://groups.google.com/d/optout.



         --
         Sent from my phone

         --
         You received this message because you are subscribed to the Google
         Groups "Jenkins Users" group.
         To unsubscribe from this group and stop receiving emails from it, send
         an email to [email protected].
         For more options, visit https://groups.google.com/d/optout.

     --
     You received this message because you are subscribed to the Google Groups
     "Jenkins Users" group.
     To unsubscribe from this group and stop receiving emails from it, send an
     email to [email protected]
     
<javascript:_e(%7B%7D,'cvml','jenkinsci-users%[email protected]');>.
     For more options, visit https://groups.google.com/d/optout.



--
Sent from my phone

--
You received this message because you are subscribed to the Google Groups
"Jenkins Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected]
<mailto:[email protected]>.
For more options, visit https://groups.google.com/d/optout.



--
Kohsuke Kawaguchi | CloudBees, Inc. | http://cloudbees.com/
Try Jenkins Enterprise, our professional version of Jenkins

--
You received this message because you are subscribed to the Google Groups "Jenkins 
Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to