[ 
https://issues.apache.org/jira/browse/CAMEL-27179?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Claus Ibsen resolved CAMEL-27179.
---------------------------------
    Resolution: Fixed

Fixed by https://github.com/apache/camel/pull/30439

> camel-aws2-timestream - query has no next token header, results cannot be 
> paged
> -------------------------------------------------------------------------------
>
>                 Key: CAMEL-27179
>                 URL: https://issues.apache.org/jira/browse/CAMEL-27179
>             Project: Camel
>          Issue Type: Improvement
>          Components: camel-aws2
>            Reporter: Claus Ibsen
>            Assignee: Claus Ibsen
>            Priority: Minor
>             Fix For: 4.24.0
>
>
> h3. Problem
> The {{query}} operation of {{aws2-timestream}} (without {{pojoRequest}}) 
> builds the {{QueryRequest}} from two headers only: 
> {{CamelAwsTimestreamQueryString}} and {{CamelAwsTimestreamQueryMaxRows}}. 
> Timestream returns a result in pages, with a {{nextToken}} for the next page, 
> but there is no header to send a next token, and none is set from the 
> response. So only the first page of a result can be read. Setting 
> {{CamelAwsTimestreamQueryMaxRows}} limits the page size, it does not give 
> paging. The endpoint documentation already says that "the next page needs a 
> QueryRequest with its nextToken and pojoRequest=true", that is Java code.
> The Kamelet {{aws-timestream-query-sink}} cannot offer paging because of this.
> h3. Where
> {{components/camel-aws/camel-aws2-timestream/src/main/java/org/apache/camel/component/aws2/timestream/query/Timestream2QueryProducer.java}},
>  {{query(...)}}: the {{else}} branch reads only {{QUERY_STRING}} and 
> {{QUERY_MAX_ROWS}}, and the body is set to the {{QueryResponse}} without any 
> header from it.
> h3. Suggested change
> One or both of:
> * A header {{CamelAwsTimestreamQueryNextToken}}: read it into 
> {{QueryRequest.Builder.nextToken(...)}}, and set it on the result from 
> {{QueryResponse.nextToken()}} (absent on the last page), so that a route can 
> loop over the pages, e.g. with {{loop doWhile}}.
> * An endpoint option (e.g. {{queryAllPages}}, default {{false}}) that follows 
> the pages with {{timestreamQueryClient.queryPaginator(...)}} and gives all 
> the rows in one result.
> Document the new header and option in {{Timestream2Constants}} and the 
> component docs.



--
This message was sent by Atlassian Jira
(v10.3.26#10030026)

Reply via email to