Richard S. Hall wrote: > > The two can be used to entirely the same thing (i.e., convert a JAR file > to a bundle): > > * For wrapping, you simply add OSGi metadata to your normally > generated JAR file manifest (i.e., you just have one JAR file). > * For embedding, you create a container JAR file with OSGi metadata > in its manifest and add the existing JAR file you want to embed > into the container JAR file and its its location to the > Bundle-ClassPath (i.e., you have two JAR files, one containing the > other). > > The assumption here is you might not be able to modify an existing JAR > file so you embed it rather than wrap it to convert it to a bundle. > > However, embedding provides a use case that wrapping cannot, which is > having your own private copy. In some cases, you don't want to convert a > JAR file into a bundle, you just want to use it inside your bundle. In > that case you embed it, add it to your Bundle-ClassPath and don't expose > it via Export-Package or whatnot, then it is your bundle's own private > library. >
Thanks Richard. What you've written makes sense. With that new knowledge, let me throw out a couple statements that seem correct, and please let me know if I'm wrong: * an embedded jar doesn't "wrap", it is "wrapped" by the "wrapping" (enclosing and possibly exposing) bundle * a wrapper bundle may or may not contain embedded jars -- View this message in context: http://old.nabble.com/wrapping-vs.-embedding-jars-tp27920331p27922108.html Sent from the Apache Felix - Users mailing list archive at Nabble.com. --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
