On Fri, May 7, 2010 at 11:34 AM, ant elder <[email protected]> wrote: > On Fri, May 7, 2010 at 10:39 AM, Simon Laws <[email protected]> wrote: > >> >> Right, and I think we need to decide once and for all what we're going >> to do about launchers. It seems complicated at the moment. >> > > Ok, lets for a moment look at the various ways we currently have in the > samples: > > 1- programatically with NodeFactory > 2- programatically with NodeLauncherEquinox > 3- Maven Tuscany plugin with mvn truscany:run > 4- Maven Tuscany OSGi JUnit plugin > 5- command line with binary distribution bin/tuscany.bat (or .sh) > 6- Ant scripts invoking binary distribution bin/ launcher.jar > 7- Ant scripts invoking NodeLauncher manual classpath > 8- Ant scripts invoking NodeLauncher using generated manifest classpath jar > 9- webapp embedded runtime with .war as contribution > 10- webapp embedded runtime with contributions separately embeded in webapp > 11- Tomcat deep integration with tuscany.war distribution > > and slightly different from those the Tuscany SCAClient can be used > programatically eg samples/helloworld-scaclient with mvn exec:java or > a manually created classpath with java -cp ... > > Is anything missing from that list? > > ...ant >
Touche, a good list and you've got a couple of things that I missed. I'd like to approach this from what we expect the users to be asking for then we can map our technical solutions to it. For example, from the sample launching thread... A. running a contribution from the command line B. running a contribution from Maven C. running a contribution from JUnit D. running a contribution from eclipse (or any other IDE) E. running a contribution from Ant F. running a contribution from OSGi G. running a contribution from a webapp H. running a contribution from Tomcat I. running a contribution from JSE program I just added "I" Simon -- Apache Tuscany committer: tuscany.apache.org Co-author of a book about Tuscany and SCA: tuscanyinaction.com
