Well put, all. I think I would be -1 on such a change without
better/specific reasons as to what is broken/bad.
Multiple versions of Maven can be installed side-by-side (and we don't
have esoteric requirements). As such, I don't see the need for such a
change.
On 12/31/17 2:27 AM, Julian Hyde wrote:
I’m -0 also. There is a cost - including a .jar in our source repo and distro
is a code smell to some people - and I don’t see a clear benefit, when the
industry standard is to use maven.
Julian
On Dec 30, 2017, at 8:50 PM, Michael Mior <[email protected]> wrote:
I agree with Christian. We don't require a particularly recent version of
Maven, so I don't think there's much benefit in having the wrapper.
However, I also don't think there's a major downside to including it, so
I'll come in at -0.
--
Michael Mior
[email protected]
2017-12-29 18:11 GMT-05:00 Christian Beikov <[email protected]>:
I didn't have the need for something like that up until now in any of the
projects I worked on. I guess if there is some feature from a very recent
Maven release that we would like to make use of, it would make sense to use
that wrapper script. Since the current minimum version that is required for
a build seems to be satisfied by the installations most users have, I don't
see a reason for doing this yet. I'd suggest we do this when a new Maven
model is released that might not be too widespread yet.
Mit freundlichen Grüßen,
------------------------------------------------------------------------
*Christian Beikov*
Am 29.12.2017 um 21:42 schrieb Julian Hyde:
We have a pull request for a maven wrapper[1]. People would not need to
install maven to build Calcite, but we would include shell/cmd scripts and
a bootstrap .jar. It would allow us to use a specific version of maven (not
that this has been a problem to date). It would make our release process a
bit more complicated (we’d be shipping a binary in the .jar file) and we’d
have to change instructions in several locations.
What do you all think of the idea?
I don’t think we should change something as fundamental as our build
process unless there is a substantial majority in favor.
Julian
[1] https://issues.apache.org/jira/browse/CALCITE-2112 <
https://issues.apache.org/jira/browse/CALCITE-2112>
