Okay. I think the best approach is to recognise the userinfo but just
remove it when constructing
URLPermissions thereby effectively ignoring it.
This is what the http protocol handler (and other support classes) have
been doing all the time
since the field is not directly of interest to http itself. That doesn't
prevent higher level software from
using it, as in the case here that provoked the bug report (the Java GIT
client used in netbeans and eclipse)
Michael
On 02/12/13 11:00, Michael McMahon wrote:
It looks like userinfo is not permitted in http URLs anyway (in rfc
2616). And even if clients
are permissive about allowing it, any userinfo would most likely not
be seen by a server
since the request URI only contains the path component of the original
URI.
I need to look at the bug report, to see how this situation arose in
the first place.
Michael
On 02/12/13 10:41, Weijun Wang wrote:
Is it possible to just ignore the userinfo part? I wonder if people
will complain why "user:pass" is not the same as "user".
--Max
On 12/2/13, 18:00, Michael McMahon wrote:
This means http://example.com does not imply
http://some...@example.com. Is this intended?
http://cr.openjdk.java.net/~michaelm/8029354/webrev.1/
