I’m going to add a test where the “newer” pom has an incompatible schema
with only modelVersion retained to ensure the parser errors get dismissed
and we bomb early with the modelVersion complaint

On Sat 1 Jun 2019 at 22:05, Stephen Connolly <
[email protected]> wrote:

> Ok I've switched to a dedicated comparator. The comparator you suggested
> is not available on the classpath in that module and the other alternatives
> you suggested could leak unintentionally side-effects.
>
> On Sat, 1 Jun 2019 at 16:03, Robert Scholte <[email protected]> wrote:
>
>> This is a bit awkward, a modelVersion is not an artifactVersion even
>> though they have the same characteristics.
>> Comparison will work, as would JavaVersion.
>> I'd prefer a dedicated comparator or GenericVersionScheme
>>
>> Robert
>>
>> [1]
>>
>> https://maven.apache.org/resolver/apidocs/org/eclipse/aether/util/version/GenericVersionScheme.html
>>
>>
>>
>>
>> On Sat, 01 Jun 2019 14:48:00 +0200, Stephen Connolly
>> <[email protected]> wrote:
>>
>> > CI passed:
>> > https://builds.apache.org/job/maven-box/job/maven/job/mng-6667/1/
>> >
>> > On Sat 1 Jun 2019 at 13:01, Stephen Connolly <
>> > [email protected]> wrote:
>> >
>> >> Finally got some time to do some work on Maven again... who knows how
>> >> long
>> >> it will last!!!
>> >>
>> >> https://issues.apache.org/jira/browse/MNG-6667
>> >>
>> >>
>> >>
>> https://gitbox.apache.org/repos/asf?p=maven.git;a=commit;h=7376a99093984c459f6a70cd1f508bbcf5ef26f7
>> >>
>> >> WDYT?
>> >>
>> >> If CI passes are we good to merge?
>> >>
>> >> -Stephen
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: [email protected]
>> For additional commands, e-mail: [email protected]
>>
>> --
Sent from my phone

Reply via email to