2009/9/16 Stephen Connolly <[email protected]>:
> FYI,
>
> as far as I know, this was the first time I saw that trick/hack
>
> http://mail-archives.apache.org/mod_mbox/maven-dev/200801.mbox/browser
>

dang it to heck

http://mail-archives.apache.org/mod_mbox/maven-dev/200801.mbox/%[email protected]%3e



> so if we can get Michael McCallum to tell us where he picked it up and
> if he has any pom.xml's in the wild that depend on the feature we may
> get further.
>
> 2009/9/15 Benjamin Bentmann <[email protected]>:
>> Hi,
>>
>> the ITs of the versions-maven-plugin employ version ranges like "[,3.0-!)".
>> The interesting question is how something like "3.0-!" should compare to
>> usual version numbers. For instance, we currently have
>>
>> Maven 2.x:
>>
>>  3.0-! < 3.0-SNAPSHOT < 3.0
>>
>> Maven 3.x, cf. [0]:
>>
>>  3.0-SNAPSHOT < 3.0 < 3.0-!
>>
>
> personally, I have no strong attachment to the range ordering...
>
> but does this imply that 3.0 is now less that 3.0-classifier
>
> that would seem to be a big regression / breaking change from 2.x
>
> -Stephen
>> As you see, there's some disagreement about the ordering which can affect
>> the result of version ranges that use "3.0-!" as a boundary.
>>
>> I originally thought those strange version numbers ending in "-!" would be
>> limited to the plugin ITs but while talking with Stephen he mentioned he
>> picked this syntax from some other projects.
>>
>> So, I'm not sure how to handle this discrepancy in version ordering. Has
>> anybody any insight how common those -! versions really are in wild
>> projects? Do we need to extend the new version comparison to handle
>> qualifiers like "!" and some unknown set of similar ones like 2.x does or is
>> this an acceptable change in behavior?
>>
>>
>> Benjamin
>>
>>
>> [0] http://docs.codehaus.org/display/MAVEN/Versioning
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: [email protected]
>> For additional commands, e-mail: [email protected]
>>
>>
>

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to