| Message ID | 001a114592c831ba11052f3ea8b0@google.com |
|---|---|
| State | New |
| Headers | show |
Hi Doug, On 03/30/2016 07:57 AM, Doug Evans wrote: > Hi. > > I was seeing this in gdb: > > (gdb) py print long(gdb.Value(18446744071563607160)) > -2145944456 > > whereas with plain python: > > (gdb) py print long(18446744071563607160) > 18446744071563607160 > Looks like this fell through the cracks? LGTM. > 2016-03-29 Doug Evans <dje@google.com> > > * python/py-value.c (valpy_long): Handle unsigned values. > > testsuite/ > * gdb.python/py-value.exp (test_value_creation): Add test for large > unsigned 64-bit value. Thanks, Pedro Alves
[sorry for the resend ... We've got gdb to use c++ now. Yay. Can we maybe someday get the list to not bounce text/html?] On Tue, Oct 4, 2016 at 7:40 AM, Pedro Alves <palves@redhat.com> wrote: > Hi Doug, > > On 03/30/2016 07:57 AM, Doug Evans wrote: >> Hi. >> >> I was seeing this in gdb: >> >> (gdb) py print long(gdb.Value(18446744071563607160)) >> -2145944456 >> >> whereas with plain python: >> >> (gdb) py print long(18446744071563607160) >> 18446744071563607160 >> > > Looks like this fell through the cracks? > > LGTM. > >> 2016-03-29 Doug Evans <dje@google.com> >> >> * python/py-value.c (valpy_long): Handle unsigned values. >> >> testsuite/ >> * gdb.python/py-value.exp (test_value_creation): Add test for large >> unsigned 64-bit value. > > Thanks, > Pedro Alves > Thanks! I usually do an audit of email sent to gdb-patches every few months. Totally missed this one (and potentially others - a distracting year for many reasons - I'll go through and try to find if there's any more). Committed.
diff --git a/gdb/python/py-value.c b/gdb/python/py-value.c index 7dba0ad..09e8055 100644 --- a/gdb/python/py-value.c +++ b/gdb/python/py-value.c @@ -1518,7 +1518,10 @@ valpy_long (PyObject *self) } END_CATCH - return gdb_py_long_from_longest (l); + if (TYPE_UNSIGNED (type)) + return gdb_py_long_from_ulongest (l); + else + return gdb_py_long_from_longest (l); } /* Implements conversion to float. */ diff --git a/gdb/testsuite/gdb.python/py-value.exp b/gdb/testsuite/gdb.python/py-value.exp index a9dbe97..ca2dc08 100644 --- a/gdb/testsuite/gdb.python/py-value.exp +++ b/gdb/testsuite/gdb.python/py-value.exp @@ -54,10 +54,15 @@ proc test_value_creation {} { if { $gdb_py_is_py3k == 0 } { gdb_py_test_silent_cmd "python i = gdb.Value (5L)" "create long value" 1 } + + gdb_py_test_silent_cmd "python l = gdb.Value(0xffffffff12345678)" "create large unsigned 64-bit value" 1 + gdb_test "python print long(l)" "18446744069720004216" "large unsigned 64-bit int conversion to python" + gdb_py_test_silent_cmd "python f = gdb.Value (1.25)" "create double value" 1 gdb_py_test_silent_cmd "python a = gdb.Value ('string test')" "create 8-bit string value" 1 gdb_test "python print (a)" "\"string test\"" "print 8-bit string"
Hi. I was seeing this in gdb: (gdb) py print long(gdb.Value(18446744071563607160)) -2145944456 whereas with plain python: (gdb) py print long(18446744071563607160) 18446744071563607160 2016-03-29 Doug Evans <dje@google.com> * python/py-value.c (valpy_long): Handle unsigned values. testsuite/ * gdb.python/py-value.exp (test_value_creation): Add test for large unsigned 64-bit value. gdb_test "python print (a.__class__)" "<(type| class) 'gdb.Value'>" "verify type of 8-bit string" + if { $gdb_py_is_py3k == 0 } { gdb_py_test_silent_cmd "python a = gdb.Value (u'unicode test')" "create unicode value" 1 gdb_test "python print (a)" "\"unicode test\"" "print Unicode string"