You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@maven.apache.org by "Robert Scholte (Jira)" <ji...@apache.org> on 2021/04/25 21:30:00 UTC
[jira] [Updated] (MJAVADOC-649) javadoc:aggregate mis-determines
whether a submodule was skipped
[ https://issues.apache.org/jira/browse/MJAVADOC-649?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Robert Scholte updated MJAVADOC-649:
------------------------------------
Description:
The fix for MJAVADOC-613 ends up mis-detecting when a submodule is being skipped in case there is an interplay between command-line and profile activation.
In OpenDaylight we have a setup, where we have a profile doing roughly:
{code:java}
<profile>
<id>q</id>
<properties>
<maven.javadoc.skip>true</maven.javadoc.skip>
<!-- ... and a host of other things -->
</properties>
</profile>
{code}
this profile is defined for each and every artifact we build.
We then have a job dedicated to extraction of javadoc which does:
{noformat}
mvn verify javadoc:aggregate -e -Pq -Dmaven.javadoc.skip=false{noformat}
This works perfectly well with maven-javadoc-plugin-3.1.1, but breaks with 3.2.0: while each submodule generates its javadoc correctly (observing the command-line override of the profile), the aggregate goal ends up skipping all submodules, resulting in no aggregate being generated.
was:
The fix for MJAVADOC-613 ends up mis-detecting when a submodule is being skipped in case there is an interplay between command-line and profile activation.
In OpenDaylight we have a setup, where we have a profile doing roughly:
{code:java}
<profile>
<id>q</id>
<properties>
<maven.javadoc.skip>true</maven.javadoc.skip>
<!-- ... and a host of other things -->
</properties>
</profile>
{code}
this profile is defined for each and every artifact we build.
We then have a job dedicated to extraction of javadoc which does:
{noformat}
mvn clean install javadoc:aggregate -e -Pq -Dmaven.javadoc.skip=false{noformat}
This works perfectly well with maven-javadoc-plugin-3.1.1, but breaks with 3.2.0: while each submodule generates its javadoc correctly (observing the command-line override of the profile), the aggregate goal ends up skipping all submodules, resulting in no aggregate being generated.
> javadoc:aggregate mis-determines whether a submodule was skipped
> ----------------------------------------------------------------
>
> Key: MJAVADOC-649
> URL: https://issues.apache.org/jira/browse/MJAVADOC-649
> Project: Maven Javadoc Plugin
> Issue Type: Bug
> Affects Versions: 3.2.0
> Reporter: Robert Varga
> Priority: Major
> Labels: regression
>
> The fix for MJAVADOC-613 ends up mis-detecting when a submodule is being skipped in case there is an interplay between command-line and profile activation.
> In OpenDaylight we have a setup, where we have a profile doing roughly:
> {code:java}
> <profile>
> <id>q</id>
> <properties>
> <maven.javadoc.skip>true</maven.javadoc.skip>
> <!-- ... and a host of other things -->
> </properties>
> </profile>
> {code}
> this profile is defined for each and every artifact we build.
> We then have a job dedicated to extraction of javadoc which does:
> {noformat}
> mvn verify javadoc:aggregate -e -Pq -Dmaven.javadoc.skip=false{noformat}
> This works perfectly well with maven-javadoc-plugin-3.1.1, but breaks with 3.2.0: while each submodule generates its javadoc correctly (observing the command-line override of the profile), the aggregate goal ends up skipping all submodules, resulting in no aggregate being generated.
>
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)