Joe Witt created NIFI-16314:
-------------------------------
Summary: Trim trailing spaces in Split*, UnpackContent, and
ExtractEmailAttachments @WritesAttribute names
Key: NIFI-16314
URL: https://issues.apache.org/jira/browse/NIFI-16314
Project: Apache NiFi
Issue Type: Task
Reporter: Joe Witt
Assignee: Joe Witt
Several processors document FlowFile attributes with a trailing space in
@WritesAttribute. Generated NAR docs therefore advertise names that do not
exist.
Runtime already writes the unspaced keys
(FragmentAttributes.SEGMENT_ORIGINAL_FILENAME, CoreAttributes.FILENAME,
ExtractEmailAttachments.ATTACHMENT_ORIGINAL_FILENAME). This is documentation
only.
NIFI-16313 fixed the same paste on SegmentContent, including a duplicate
annotation that also claimed a filename rewrite. That rewrite claim is unique
to SegmentContent and is already handled there. This ticket is the remaining
mechanical trim.
Processors / attributes:
- SplitText, SplitJson, SplitXml, SplitContent, SplitRecord, PartitionRecord,
SplitAvro, UnpackContent: "segment.original.filename "
- ExtractEmailAttachments: "filename " and "email.attachment.parent.filename "
Out of scope:
- Do not change what onTrigger writes.
- Do not re-open UnpackContent CRC behavior (NIFI-15718).
- Do not include ParseSyslog/ParseSyslog5424 syslog.sender, UpdateDatabaseTable
output.path, or RetryFlowFile property-as-key docs.
Proposed change:
- Trim the attribute names in the annotations to match the constants actually
put on FlowFiles.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
