Ok so that patch isn't safe when there's whitespace in the expanded
version of "~", not sure if that is even allowed though.
Anyway, it illustrates the point. It seems that in version 3.5 it did
this internally, at least the argument wasn't added in the wrapper
script for 3.5, which explains why I see this as a regression in 3.7
This is the maintainer address of Debian's Java team
debian-j...@lists.debian.org for discussions and questions.