Assuming it produces the same binaries (and that's sort of the point of this, right?) IMHO it should not be optional. Every option is an additional thing to document and test, and in this case, for no particular user benefit. If you can make the code noticeably faster while producing the same output, do so. Don't give the user an option to choose the slow road or the fast road.
On Tue, Apr 21, 2020 at 6:20 PM Robert Oxspring <[email protected]> wrote: > > Hi all, > > I’m working on a multi module project where should-be no-op “mvn install > -DskipTests” takes multiple minutes and "assembly:single" appears to be among > the offenders - needlessly recreating archives when none of the inputs have > been modified. I’m considering modifying the plugin to avoid this recreation, > leading to the following questions… > > Would such a change be welcome? Is it important that the modified behaviour > is optional and not default? I figure the next step would be to start hacking > out a poc, but if you need me to raise a Jira issue or something then please > let me know. > > Should I be targeting my change at the assembly plugin or would it make more > sense to try and get this deeper into plexus-archiver? Do you have any > further advice where best to start?? > > Are there reasons why this hasn’t been done before? - e.g. folks have died on > this hill and I should give up now. > > What other questions should I be asking!? > > Thanks, > > Rob > --------------------------------------------------------------------- > To unsubscribe, e-mail: [email protected] > For additional commands, e-mail: [email protected] > -- Elliotte Rusty Harold [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
