[
https://issues.apache.org/jira/browse/SLING-11799?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17802891#comment-17802891
]
Thomas Mueller commented on SLING-11799:
----------------------------------------
> What would be needed is the (first) alias that has the most occurrences.
>From the data I have analyzed, some of the values have more than 10'000
>entries. So it's better to use a dynamic limit:
* select without limit, order by first alias
* read from the result set _until_ the first alias is different, but at least
1'000 (both conditions need to be met)
* the next page can then continue with the new value
> use keyset pagination for sling:alias query
> -------------------------------------------
>
> Key: SLING-11799
> URL: https://issues.apache.org/jira/browse/SLING-11799
> Project: Sling
> Issue Type: Bug
> Components: ResourceResolver
> Reporter: Julian Reschke
> Assignee: Julian Reschke
> Priority: Major
>
> Similar to what was done for vanity paths (SLING-11581), we should use keyset
> pagination for queries for sling:alias.
> This would work the same in principle, however we should keep in mind that -
> due to the different type of data - we may have way more duplicates, so in
> edge cases the pages might be bigger than expected. However, every size
> reduction of "results of a single query" should help.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
