Hi David, thanks for reporting. You're right, ScanResult.extractBeanArchiveId() does not handle windows separator correctly. And in the attached document, you're also right the context path is not guaranteed to be part of the bean archive path. The problem is we cannot determine the separator reliably for bean archives within WEB-INF/lib. The only solution I can see would be to completely switch to ServletContext.getResourcePaths() based solution (right now the filesystem is used if possible). I think we should try to go this way and see if it's feasible.
|