[
https://issues.apache.org/jira/browse/SOLR-5488?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13938167#comment-13938167
]
Tomás Fernández Löbbe commented on SOLR-5488:
---------------------------------------------
{quote}
Second problem:
Everything else is failing, things like:
Expected :[25.0, 26.0, 28.5, 27.0, 22.5, 23.5, 24.5, 25.5, 26.5, 27.5, 24.0]
Actual :[25.0, 26.0, 27.0, 22.5, 23.5, 24.5, 25.5, 26.5, 27.5, 28.5, 24.0]
Same numbers, just not in the same order. This one from:
at
org.apache.solr.analytics.facet.FieldFacetTest.medianFacetAscTest(FieldFacetTest.java:561)
So is ordering important here or should the values be sorted before passing to
assertEquals? As you an tell I have no real deep-level understanding of what
this code is supposed to do, which makes it difficult to have confidence if I
sorted the expected and actual before passing to assertEquals or used .
{quote}
See my comments in SOLR-5716. The test assumed that the results would be always
in the same order, but since SOLR-5685 was fixed, the test started creating
multiple segments, which made the component return different ordering (random,
depending on when commit/merges occur). I was not aware of this issue, so my
attempt to solve this is in SOLR-5716. I changed the test to allow unordered
results... but, if this is an undesired behavior of the AnalyticsComponent,
then it's a code issue and not a test one.
> Fix up test failures for Analytics Component
> --------------------------------------------
>
> Key: SOLR-5488
> URL: https://issues.apache.org/jira/browse/SOLR-5488
> Project: Solr
> Issue Type: Bug
> Affects Versions: 4.7, 5.0
> Reporter: Erick Erickson
> Assignee: Erick Erickson
> Attachments: SOLR-5488.patch, SOLR-5488.patch, SOLR-5488.patch,
> SOLR-5488.patch, SOLR-5488.patch, SOLR-5488.patch, SOLR-5488.patch, eoe.errors
>
>
> The analytics component has a few test failures, perhaps
> environment-dependent. This is just to collect the test fixes in one place
> for convenience when we merge back into 4.x
--
This message was sent by Atlassian JIRA
(v6.2#6252)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
