[jira] [Commented] (MRESOLVER-352) Duplicate METADATA_DOWNLOADING event is being sent

["Tamas Cservenak (Jira)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/MRESOLVER-352?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17715003#comment-17715003 ] 

Tamas Cservenak commented on MRESOLVER-352:
-------------------------------------------

Thanks for this detailed report! Yes, the "attempt" will redo almost everything (when lock change happens), and event emission is an oversight...

> Duplicate METADATA_DOWNLOADING event is being sent
> --------------------------------------------------
>
>                 Key: MRESOLVER-352
>                 URL: https://issues.apache.org/jira/browse/MRESOLVER-352
>             Project: Maven Resolver
>          Issue Type: Bug
>          Components: Resolver
>    Affects Versions: 1.9.8
>         Environment: Maven 3.9.2-SANPSHOT, Maven 4.0.0-SNAPSHOT, linux, windows, macos
>            Reporter: Alexey Venderov
>            Priority: Major
>
> When using Maven 3.9.2-SNAPSHOT and Maven 4.0.0-SNAPSHOT we've noticed that the METADATA_DOWNLOADING event is being sent twice for the same metadata file + repository pair. And there is only one METADATA_DOWNLOADED event that corresponds to the second METADATA_DOWNLOADING event. The duplicated event might break event consumers that try to track downloads using METADATA_DOWNLOADING/METADATA_DOWNLOADED event pairs.
> The regression was introduced in this PR https://github.com/apache/maven-resolver/commit/4c5e9ea98f8815c6df8bf26baa9032c8d9cd5f2d. When the lock "upgrade" is happening there is a second pass through the resolution logic that leads to the duplicate event. 



--
This message was sent by Atlassian Jira
(v8.20.10#820010)