[
https://issues.apache.org/jira/browse/SOLR-12638?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16892757#comment-16892757
]
David Smiley commented on SOLR-12638:
-------------------------------------
CC [~moshebla]
I found the cause of this bug or maybe non-bug depending on how you want to
look at it. For partial updates to nested documents to work, it is _not only_
necessary to have the \_nest_path_ field in the schema, but the \_root_ field
must be stored=true. If either are false then any such partial updates will
wipe out child documents. So the cause here is actually very similar to
SOLR-13523.
org.apache.solr.update.processor.NestedAtomicUpdateTest#testBlockAtomicStack
exercises partial updates and it passes... but it passes because this test
suite uses the test schema-nest.xml with root stored=true.
When I was working on SOLR-13523 I started to work on wiping out all traces of
stored=true in all schemas but it got to be a chunk of work that was a bit
distracting from a simple bug fix so I had tabled it. But that would have
surfaced this problem then had I continued. Perhaps _for now_, root with
stored=true this is simply a requirement for partial updates to work in the
presence of nested docs. We don't document that and it's a problem. I also
think this requirement stinks... I'd much rather this feature work without
having to toggle the stored attribute because I think it's a source of errors
(e.g. this issue), something to document, something to want to test in
different ways, and ultimately not truly necessary if we code this better.
> Support atomic updates of nested/child documents for nested-enabled schema
> --------------------------------------------------------------------------
>
> Key: SOLR-12638
> URL: https://issues.apache.org/jira/browse/SOLR-12638
> Project: Solr
> Issue Type: Sub-task
> Reporter: mosh
> Assignee: David Smiley
> Priority: Major
> Fix For: 8.1
>
> Attachments: SOLR-12638-delete-old-block-no-commit.patch,
> SOLR-12638-nocommit.patch, SOLR-12638.patch, SOLR-12638.patch
>
> Time Spent: 17h 10m
> Remaining Estimate: 0h
>
> I have been toying with the thought of using this transformer in conjunction
> with NestedUpdateProcessor and AtomicUpdate to allow SOLR to completely
> re-index the entire nested structure. This is just a thought, I am still
> thinking about implementation details. Hopefully I will be able to post a
> more concrete proposal soon.
--
This message was sent by Atlassian JIRA
(v7.6.14#76016)
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org
