maven-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Curtis Rueden <>
Subject Re: JNI jars dependencies
Date Mon, 07 Jan 2013 17:51:06 GMT
Hi all,


Some notes about this code:

1) The above URL is to an SVN repository that has since been migrated to
Git. The current URL for the project is:

2) The SciJava native-lib-loader project is a fork [1] of Richard van der
Hoff's mx-native-loader project, which is also available from its own Maven
repository [2], and has several other forks floating around on GitHub, too.
Since we forked (at version 1.7), the author has released a newer version
(1.8), which we have not reconciled with the SciJava code.



On Fri, Jan 4, 2013 at 2:45 PM, Stuart Maclean <>wrote:

> Sorry Martin, I missed some of your points:
>  Hi from a newbie.MG>a newbie with extensive experience with maven and
>>> make files!
> Ah yes, I meant new to mailing list ;)
> com.wapmx.native loader can unpack it at runtime. MG>is this NativeLoader
> what you're alluding toMG>**
> trac/software/browser/**branches/maven/projects/**
> native-library-util/src/main/**java/com/wapmx/nativeutils/**
> jniloader/**rev=7574<>
> Essentially yes, that is it.  I did try contacting the original author, to
> no avail.  I can't recall now if I was able to get that artifact from a
> repo or had to install it manually.
> of a poor man's AOL (from the nar plugin, which I ended up NOT using).
> MG>did'nt catch the reasons why you do'nt want to use nar..explain please
> I didn't use NAR as I was worried it was unsupported,fragmented and overly
> optimistic on what it was trying to achieve.  I could live with the exec
> plugin firing make, I am happy in make myself.tion across many
>  platform Makefiles. MG>smart ..OS and compiler specific variables should
>>> be aggregated in OS and compiler specific profiles
> Yes.  I cannot fathom why there is no easier way to run javah from Maven.
>  It certainly isn't 'native' the way gcc is 'native'.  It's a standard Java
> tool running on .class files.  OK, it produces C headers but that doesn't
> make it 'native'.  I may switch to the antrun-plugin for the javah step,
> since the native-maven-plugin looks a bit alpha itself, and I don't use it
> for any 'heavy lifting', I use make.
>  MG>app code?
>> MG>JNI ?MG>How is the app PATHed to native binaries?
> See last post, you don't need any LD_LIBRARY_PATH nor -Djava.library.path.
> I should qualify this by saying that in PRINCIPLE this should all work on
> Windows, I have NOT actually tested it, I have just tested Linux 32 and 64
> bit.
> Stuart
> ------------------------------**------------------------------**---------
> To unsubscribe, e-mail: users-unsubscribe@maven.**<>
> For additional commands, e-mail:

  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message