Fix new failures in unit tests #3931
Comments
|
As far as I'm aware, the new failures are EditFieldRender and AdminQueueTemplates, correct? |
|
I can't really tell. My last run was from 2024-09-30. It had some more failing scripts. But this was before the patches done for #3856. So it might be that the current failures are not actually new. |
|
See also #3980 where a test failure for rel-11_1 was fixed. |
|
findings re HumanReadableDataSize.t (I was looking at nightly build pipeline errors earlier) short: looks like fun with Unicode long: possible Problem with Translation data in Kernel/Language/ar_SA.pm, looks like since rel-10_1 (human readable size translations for 10_0 are empty) looking at Kernel/Language/ar_SA.pm line 4993ff I see this No clue regarding proper Unicode representation for Arabic, but for whatever reason the tranlation line for MB looks way different then all the others. Perhaps a copy/paste error thing? Technically, it manages to break the string subst (%s) feature of Translate(). Ah, the power of Unicode!. Sigh. Trying to massage the test data ... |
|
re HumanReadableDataSize.t this would apparently fix it (in the sense of making the test green, probably needs proof-reading): |
Please talk with @stefanhaerter and @KhalilHamada-OTOBO - this has to do with the translations we automatically deploy via weblate, and there was an issue with exactly that. At least for one branch, I believe they fixed it by Khalil correcting the translation on weblate. (Edit: maybe for other branches the test was already changed, I'm not sure. They both worked on this, in any case.) |
|
The failure in EditFieldRender.t seems to be related to the sorting of options, that is whether the empty option is added on top or on bottom. However, I can't tell what the correct behavior is. Here is the unified diff that was generated by hacking |
|
The failure in HumanReadableDataSize.t showed up in rel-10_1 as well. I cherry picked the patch. |
|
After converting EditFieldRender.t to using |
|
Here is the latest run in rel-11_0. Looks like the failures in AdminQueueTemplates.t are sporadic. |
|
For the failures in AgentTicketOwner.t and AgentTicketZoom.t there already is #3563 . |
|
Re EditFieldRender.t possibly related: #3719 |
|
New failure in WebserviceTicketInvoker.t in the latest run: |
|
After some fixes the list of failing tests, this time for rel-11_1, is: @stefanhaerter : I recall that you had some more fixes, but I could not find the patch any more. |
|
After a couple of fixes: AdminQueueTemplates.t might be sporadic |
|
Running the test suite with rel-11_0 under Docker again and we are down to a single failure: |
|
This looks suspiciously related to my work in #3982 - I will try to have a look anytime soon. |
|
@stefanhaerter # Add one additional response header line.
$Selenium->find_element( "#AddValue", 'css' )->click();
but the DOM looks like: <a href="#" id="OutboundHeadersAddValue" class="CallForAction AddValue"><span><i class="fa fa-plus-square"> </i>Add header</span></a>Perhaps change |
|
remaining unittest failure scripts/test/Selenium/Agent/Admin/AgentTicketEmailPGP.t in rel-11_1 (test is green in rel-11_0): otobo@b3d4b0a8eb28:~$ bin/otobo.Console.pl Dev::UnitTest::Run scripts/test/Selenium/Agent/Admin/AgentTicketEmailPGP.t
[11:27:53] scripts/test/Selenium/Agent/Admin/AgentTicketEmailPGP.t .. 46/?
# Failed test 'Signing key is selected'
# at scripts/test/Selenium/Agent/Admin/AgentTicketEmailPGP.t line 276.
[11:27:53] scripts/test/Selenium/Agent/Admin/AgentTicketEmailPGP.t .. 51/?
# Failed test 'Signing key is selected'
# at scripts/test/Selenium/Agent/Admin/AgentTicketEmailPGP.t line 321.
gpg: WARNING: unsafe permissions on homedir '/opt/otobo/var/tmp/pgptest3060000847400000'
[11:27:53] scripts/test/Selenium/Agent/Admin/AgentTicketEmailPGP.t .. 86/? # Seeded srand with seed '20250110' from local date.
# Looks like you failed 2 tests of 92.
[11:27:53] scripts/test/Selenium/Agent/Admin/AgentTicketEmailPGP.t .. Dubious, test returned 2 (wstat 512, 0x200)
Failed 2/92 subtests
[11:28:12]
Test Summary Report
-------------------
scripts/test/Selenium/Agent/Admin/AgentTicketEmailPGP.t (Wstat: 512 (exited 2) Tests: 92 Failed: 2)
Failed tests: 50, 55
Non-zero exit status: 2
Files=1, Tests=92, 19 wallclock secs ( 0.10 usr 0.01 sys + 2.72 cusr 0.59 csys = 3.42 CPU)
Result: FAIL
ran tests for product OTOBO 11.1.x on host b3d4b0a8eb28 .
Steps to reproduce (sigh):
Test Scenario 1 (works):
{"NewUserID":[["","-",false,false,false]],"NewResponsibleID":[["","-",false,false,false]],"SignKeyID":[["","-",false,false,false],["PGP::2AD6BAFF","PGP: [good] 2AD6BAFF [never] John Smith (Test PGP never expired) <[email protected]>",true,true,false],["PGP::667B04B9","PGP: [WARNING: EXPIRED KEY] 667B04B9 [2018-12-19] John Smith (Test PGP expired) <pgptest@exampl[...]",false,false,false]],"NextStateID":[["2","closed successful",false,false,false],["3","closed unsuccessful",false,false,false],["4","open",true,true,false],["7","pending auto close+",false,false,false],["8","pending auto close-",false,false,false],["6","pending reminder",false,false,false]],"EmailSecurityOptions":[["","-",false,false,false],["PGP::-::Encrypt","PGP encrypt",false,false,false],["PGP::Sign::-","PGP sign",true,true,false],["PGP::Sign::Encrypt","PGP sign and encrypt",false,false,false]],"TypeID":[["","-",false,false,false],["1","Unclassified",false,false,false]],"Signature":"<br/>Your Ticket-Team<br/><br/> - -<br/><br/>--<br/> Super Support - Waterford Business Park<br/> 5201 Blue Lagoon Drive - 8th Floor & 9th Floor - Miami, 33126 USA<br/> Email: [email protected] - Web: <a href=\"http://www.example.com/\" title=\"http://www.example.com/\">http://www.example.com/</a><br/>--","Dest":[["","-",false,false,false],["3||Junk","Junk",false,false,false],["4||Misc","Misc",false,false,false],["7||PgpTest","PgpTest",true,true,false],["1||Postmaster","Postmaster",false,false,false],["2||Raw","Raw",false,false,false]],"SLAID":[["","-",false,false,false]],"StandardTemplateID":[["","-",false,false,false]],"CryptKeyID":[["","-",false,false,false]],"ServiceID":[["","-",false,false,false]],"PriorityID":[["1","1 very low",false,false,false],["2","2 low",false,false,false],["3","3 normal",true,true,false],["4","4 high",false,false,false],["5","5 very high",false,false,false]]}Test Scenario 2 - fails as in the unittest:
{}But why?Looking at AgentTicketCompose in rel-11_1, the JSON for AJAXUpdate is built like this: my %Diversity = reverse %Uniformity;
for my $Field ( sort keys %StdFieldValues ) {
push @StdFieldAJAX, {
Name => $Diversity{$Field} || $Field,
Data => $StdFieldValues{$Field},
SelectedID => $GetParam{$Field},
%{ $Attributes{$Field} },
};
}
my $JSON = $LayoutObject->BuildSelectionJSON(
[
@StdFieldAJAX,
@DynamicFieldAJAX,
],
);So, the JSON for standard fields comes from the %StdFieldValues hash. It seems that this hash gets declared in https://github.com/RotherOSS/otobo/blob/rel-11_1/Kernel/Modules/AgentTicketCompose.pm#L1353 - and then never gets assigned to before the end of 'AJAXUpdate'. That would explain the empty JSON array, unless I am overlooking something ... but ... Hu? |
|
Most likely related: #3537 |
|
Current state of the tests. |
A couple of new test failures have crept into rel-11_0:
The text was updated successfully, but these errors were encountered: