Re: [RFA] Java DWARF fixes

["Ulrich Weigand" <uweigand@de.ibm.com>]

Keith Seitz wrote:
> On 03/25/2010 01:05 PM, Tom Tromey wrote:
> > This all seems pretty reasonable to me.  It is ok.
> 
> I have committed the patches. Thank you for reviewing them.

Thanks!  Unfortunately, the test case still fails for me:
FAIL: gdb.java/jmisc.exp: ptype jmisc

This time the problem seems to be that the constructor
is shown with a "void" argument list:

type = class jmisc  extends java.lang.Object {

    void main(java.lang.String[])void;
    jmisc(void)void;
}

while the test expects this to be just "jmisc()":

    send_gdb "ptype jmisc\n"
    gdb_expect {
        -re "type = class jmisc  extends java.lang.Object \{\[\r\n\ \t]+void main\\(java\.lang\.String\\\[]\\);\[\r\n\ \t]+jmisc\\(\\);\[\r\n\ \t]+\}\[\r\n\ \t]+$gdb_prompt $"
            { pass "ptype jmisc" }
        -re "type = class jmisc  extends java.lang.Object \{\[\r\n\ \t]+void main\\(java\.lang\.String\\\[]\\)void;\[\r\n\ \t]+jmisc\\(\\);\[\r\n\ \t]+\}\[\r\n\ \t]+$gdb_prompt $" {
            # Just because GCC includes the signature doesn't mean we
            # should print it here.  We already show the return type.
            kfail "ptype jmisc" gdb/2215
        }
        -re ".*$gdb_prompt $"             { fail "ptype jmisc" }
        timeout { fail "ptype jmisc (timeout)" ; return }
    }

Is this a testcase problem or still something wrong in
the symbol reader?

Bye,
Ulrich

-- 
  Dr. Ulrich Weigand
  GNU Toolchain for Linux on System z and Cell BE
  Ulrich.Weigand@de.ibm.com