[
https://issues.apache.org/jira/browse/KAFKA-20072?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
David Arthur updated KAFKA-20072:
---------------------------------
Labels: good-first-issue (was: )
> Don't generate IDs with hyphens
> -------------------------------
>
> Key: KAFKA-20072
> URL: https://issues.apache.org/jira/browse/KAFKA-20072
> Project: Kafka
> Issue Type: Improvement
> Reporter: David Arthur
> Priority: Minor
> Labels: good-first-issue
>
> As we continue shifting APIs and tooling away from topic name and towards
> topic IDs, it is likely that topic IDs become more directly used by users.
> Today, we generate topic IDs simply as a random UUID (UUID4) without a
> leading hyphen. This same algorithm is used for generating cluster IDs. We
> exclude the leading hyphen to avoid issues with argument parsing in command
> line tooling.
> However, since we can generate hyphens in the middle of IDs, it leave us with
> a "double click" problem. The default behavior in most OSes and applications
> is that a double click on a string will select a single word. Since a hyphen
> is considered a separator, the selection behavior will not grab the full ID.
> Given the following examples of Uuid#randomUuid, note that ones containing a
> hyphen are not double-clickable. The must be "drag selected"
>
> {code}
> lBZW73x0QUG7p8vjWAxHlQ
> WTIfkBt6QVaxcYMIiSc-jA
> wa5Ni6lkSRS7BIMTmKjL-w
> ouZbg_F3S_q8CDS7kNnEpw
> sWgAlokBRZ-1wqsxN7ivzA
> C_x5y8urTLO_RpEov6f5hg
> wfPu7cf7TsikM_F62Psgfw
> nLBxpI6hT3-LU7zXx6iPyQ
> 3m8iErOSRSOPN3FCYKpJog
> k88Pa2L7S1CjZ61Cg6BzoQ
> {code}
> This can be frustrating as an operator, but also lead to human error. When
> highlighting (click + drag), it is easy to miss some characters.
> As a simple quality of life improvement, we can alter our random ID generator
> algorithm to omit hyphens completely.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
