You are viewing a plain text version of this content. The canonical link for it is here.
Posted to fop-dev@xmlgraphics.apache.org by Eric Douglas <ed...@blockhouse.com> on 2010/08/10 16:48:18 UTC
classpath
I see there's still a classpath statement in the 1.0 Manifest.mf file.
Is this necessary? Could this be removed in the next version, and
document a classpath parameter to pass for manually executing? I run
this with embedded code from an app which already has a classpath, so I
have to recompile as the manifest classpath interferes, and having this
classpath hardcoded prevents testing with updated versions of the
referenced jars.
Re: classpath
Posted by Jeremias Maerki <de...@jeremias-maerki.ch>.
No objects heard, so I went ahead: http://svn.apache.org/viewvc?rev=985921&view=rev
On 12.08.2010 14:31:21 Jeremias Maerki wrote:
> I agree that we should remove this. The manifest's Classpath is quite
> restricted in its usefulness. Furthermore, I've long ago added some code
> that builds up the classpath dynamically when necessary. Hardly anyone
> will call FOP by "java -jar fop.jar". Most people will use one of the
> scripts (I do, too). If noone objects, I'll remove it.
>
> On 10.08.2010 16:48:18 Eric Douglas wrote:
> > I see there's still a classpath statement in the 1.0 Manifest.mf file.
> > Is this necessary? Could this be removed in the next version, and
> > document a classpath parameter to pass for manually executing? I run
> > this with embedded code from an app which already has a classpath, so I
> > have to recompile as the manifest classpath interferes, and having this
> > classpath hardcoded prevents testing with updated versions of the
> > referenced jars.
Jeremias Maerki
Re: classpath
Posted by Jeremias Maerki <de...@jeremias-maerki.ch>.
I agree that we should remove this. The manifest's Classpath is quite
restricted in its usefulness. Furthermore, I've long ago added some code
that builds up the classpath dynamically when necessary. Hardly anyone
will call FOP by "java -jar fop.jar". Most people will use one of the
scripts (I do, too). If noone objects, I'll remove it.
On 10.08.2010 16:48:18 Eric Douglas wrote:
> I see there's still a classpath statement in the 1.0 Manifest.mf file.
> Is this necessary? Could this be removed in the next version, and
> document a classpath parameter to pass for manually executing? I run
> this with embedded code from an app which already has a classpath, so I
> have to recompile as the manifest classpath interferes, and having this
> classpath hardcoded prevents testing with updated versions of the
> referenced jars.
Jeremias Maerki