[GitHub] nifi pull request #2666: NIFI-5130 ExecuteInfluxDBQuery processor chunking s...

[MikeThomsen <gi...@git.apache.org>]

Github user MikeThomsen commented on a diff in the pull request:

    https://github.com/apache/nifi/pull/2666#discussion_r186124698
  
    --- Diff: nifi-nar-bundles/nifi-influxdb-bundle/nifi-influxdb-processors/src/main/java/org/apache/nifi/processors/influxdb/ExecuteInfluxDBQuery.java ---
    @@ -86,6 +93,18 @@
                 .expressionLanguageSupported(ExpressionLanguageScope.FLOWFILE_ATTRIBUTES)
                 .build();
     
    +    public static final PropertyDescriptor INFLUX_DB_QUERY_CHUNK_SIZE = new PropertyDescriptor.Builder()
    +            .name("influxdb-query-chunk-size")
    +            .displayName("Results chunk size")
    +            .description("Chunking can be used to return results in a stream of smaller batches "
    +                + "(each has a partial results up to a chunk size) rather than as a single response. "
    +                + "Chunking queries can return an unlimited number of rows. Note: Chunking is enable when result chunk size is greater than 0")
    +            .defaultValue(String.valueOf(DEFAULT_INFLUX_RESPONSE_CHUNK_SIZE))
    +            .expressionLanguageSupported(ExpressionLanguageScope.FLOWFILE_ATTRIBUTES)
    +            .addValidator(StandardValidators.createLongValidator(0, Integer.MAX_VALUE, true))
    --- End diff --
    
    If the default query result limit is 10k, what is the likelihood that there is a good reason for this to be above 10k? Even if we go beyond that for chunks, Integer.MAX_VALUE is way past a sane limit.


---