Hi! I again and again find errors in scannotation. Some of them are caused by using an URL also for files which causes 'funny' things like new File(url.getPath())
If the path contains a space then the URL contains %20. Also, URLs are pretty error prone and sometimes causes bad problems with opening connections on some method invocations (e.g. URL#equals is a really bad thing!) Thus we should imo really get rid of them in the ScannerService. I'd rather like to drop all that sh*t and invoke AnnotationDB#scanClass(InputStream) directly. wdyt? LieGrue, strub https://issues.apache.org/jira/browse/OWB-524
