RE: [cp-patches] RFC: prevent URL degeneration

[Jeroen Frijters]

Robert Schuster wrote:
> 2005-10-07  Robert Schuster  <address@hidden>
> 
>     * java/net/URLStreamHandler.java:
>     (toExternalForm): Remove superfluous leading slashes from URL
>     paths.

This patch introduces several Mauve regressions for me (on Windows):

FAIL: gnu.testlet.java.net.URL.URLTest: new URL(string) (number 1)
FAIL: gnu.testlet.java.net.URL.URLTest: new URL(string) (number 19)
FAIL: gnu.testlet.java.net.URL.URLTest: new URL(string) (number 31)
FAIL: gnu.testlet.java.net.URL.URLTest: new URL(string) (number 40)
FAIL: gnu.testlet.java.net.URL.URLTest: new URL(protocol, host, file)
(number 29)
FAIL: gnu.testlet.java.net.URL.URLTest: new URL(protocol, host, file)
(number 33)
FAIL: gnu.testlet.java.net.URL.URLTest: new URL(protocol, host, file)
(number 37)
FAIL: gnu.testlet.java.net.URL.URLTest: new URL(protocol, host, file)
(number 41)
FAIL: gnu.testlet.java.net.URL.URLTest: new URL(protocol, host, file)
(number 45)
FAIL: gnu.testlet.java.net.URL.URLTest: new URL(protocol, host, file)
(number 49)
FAIL: gnu.testlet.java.net.URL.URLTest: new URL(protocol, host, file)
(number 53)

I also disagree with it on principle grounds. We should match the JDK in
cases like this, not try to do the "right thing". It's very easy to
break applications that depend on the JDK URL implementation and it is
not reasonable to expect them to be fixed if the JDK is never going to
get fixed (which I can assure you, it isn't).

Regards,
Jeroen