[
https://issues.apache.org/jira/browse/JCR-3833?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15223779#comment-15223779
]
Unico Hommes commented on JCR-3833:
-----------------------------------
[~takahito.kikuchi] Because the iterator is indeed ordered, all the keys of the
items that could not be deleted will be at the beginning of the iterator and
you will skip them because you counted them all. However, I agree with Clay
Ferguson on the dev list that it would be more robust to first collect all the
keys to delete in a separate list and then use the keys in the separate list to
remove the items from the cache. That way there is no concurrent modification
of the list you are iterating over. My only concern was the added overhead in
terms of memory and processing that comes with creating a separate list but
that is probably quite acceptable.
> Compatibility with Java 8
> -------------------------
>
> Key: JCR-3833
> URL: https://issues.apache.org/jira/browse/JCR-3833
> Project: Jackrabbit Content Repository
> Issue Type: Task
> Reporter: Unico Hommes
> Assignee: Unico Hommes
> Fix For: 2.9.1, 2.8.1
>
>
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
