[XML-SIG] "Borrowed" tests

[Martin v. Loewis]

> Your logic here escapes me.  Why would this distinction 
> be necessary?  It just seems like extra work with no
> additional benifit.  As I find bugs in my software I 
> add it to my regression test, and usually name the
> test after the bug.

Outright doing so with bugs found by others might violate the
copyright of the others; there should be an indication of authorship,
and there must be, of course, a permission of the author of the test
case that copying it is allowed. It is often hard to determine whether
the author would agree to redistribution of a test case, so it is
clearly better to keep them separate from the cases with clear
licensing conditions.

The GCC team could not ship its entire regression test suite with gcc
2.95 because authorship of many of the tests cannot be established.

Regards,
Martin