Github user jvwing commented on a diff in the pull request:
https://github.com/apache/nifi/pull/2751#discussion_r194140856
--- Diff:
nifi-nar-bundles/nifi-aws-bundle/nifi-aws-processors/src/main/java/org/apache/nifi/processors/aws/s3/ListS3.java
---
@@ -307,6 +328,20 @@ private boolean commit(final ProcessContext context,
final ProcessSession sessio
return willCommit;
}
+ private Map<String, String> writeObjectTags(AmazonS3 client,
S3VersionSummary versionSummary) {
+ final GetObjectTaggingResult taggingResult =
client.getObjectTagging(new
GetObjectTaggingRequest(versionSummary.getBucketName(),
versionSummary.getKey()));
--- End diff --
I agree with @pvillard31 that it should be off by default. From comments
on the users/developer email lists, I understand ListS3 is used to process very
large lists of objects, easily 10,000+ on a regular basis. Even if the
additional API calls are quick, it will add up to be a lot of API calls.
Unfortunately, it does not look like the
[S3ObjectSummary](https://docs.aws.amazon.com/AWSJavaSDK/latest/javadoc/com/amazonaws/services/s3/model/S3ObjectSummary.html)
returned by the listing contains any hints on the number of tags present, if
any.
---
