On 16/01/2020 11:53, Rémy Maucherat wrote:
> On Thu, Jan 16, 2020 at 11:58 AM Mark Thomas <ma...@apache.org
> <mailto:ma...@apache.org>> wrote:
> 
>     On 16/01/2020 10:27, Rémy Maucherat wrote:
>     > On Thu, Jan 16, 2020 at 11:06 AM <r...@apache.org
>     <mailto:r...@apache.org>
>     > <mailto:r...@apache.org <mailto:r...@apache.org>>> wrote:
>     >
>     >     This is an automated email from the ASF dual-hosted git
>     repository.
>     >
>     >     remm pushed a commit to branch master
>     >     in repository
>     >     https://gitbox.apache.org/repos/asf/tomcat-jakartaee-migration.git
>     >
>     >
>     >     The following commit(s) were added to refs/heads/master by
>     this push:
>     >          new 86bc4ee  Shade and add a main class
>     >     86bc4ee is described below
>     >
>     >     commit 86bc4eea2804fa5df6a5089dd4a232847b709c91
>     >     Author: remm <r...@apache.org <mailto:r...@apache.org>
>     <mailto:r...@apache.org <mailto:r...@apache.org>>>
>     >     AuthorDate: Thu Jan 16 11:06:25 2020 +0100
>     >
>     >         Shade and add a main class
>     >
>     >         It seems easier to use to me: "java -jar
>     overlylongjarname.jar src
>     >         dest". The overlylongjarname could be trimmed down a bit.
>     >
>     >
>     > So there's an option between this way and the assembly that was added
>     > (which I'm not sure how to use).
> 
>     If I understand your changes correctly, they create a single "fat" jar.
>     Is that right?
> 
> 
> Yes, so it includes BCEL.
>  
> 
> 
>     My thinking with the assembly approach (and I don't think the two
>     approaches are mutually exclusive) was:
> 
>     - looking towards a formal ASF release where we would want bin and
>       source releases
> 
>     - to make it easier to "plug-in" additional convertors for specific file
>       types.
> 
>     I'm just guessing at requirements at this point. Given where things are
>     right now, I think a single JAR makes most sense. It is much easier to
>     work with during development: (build, use) rather than (build,
>     unpack, use)
> 
>     At the moment the script just sets up the class path. I have no
>     objection to removing the assembly plumbing for now. We can always add
>     it back if we need it.
> 
> Well, that's not a bad idea since that's how the releases are done and I
> didn't think about the plugin idea, I get it better now.
> I guess I can keep my hack pom :)

Sorry, I should have been clearer. I'm in favour of keeping the single
JAR via the shade plug-in- it makes it easier to work with during
development / testing.

We can still use the assembly to produce a final release (if we get to
the point of formally releasing this).

Mark

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org

Reply via email to