This is the mail archive of the java-patches@gcc.gnu.org mailing list for the Java project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: Patch: FYI: important memory-trashing fix


>>>>> "Scott" == Scott Gilbertson <scottg@mantatest.com> writes:

Scott> Double.parseDouble no longer parses strings with leading
Scott> spaces.  My app previously worked but now gets a
Scott> NumberFormatException.  A small test case (attached) sees the
Scott> same thing.

Scott> It seems to me that "length + start" is guaranteed to be no greater than
Scott> "str->length()", so the old version was correct.

You are right.  Sorry about that, and thanks for catching this.

Fixed as appended.

There's already a Mauve test for this.  I'm not sure why my nightly
testing didn't pick up this failure, since running Mauve by hand
showed it :-(

Tom

Index: ChangeLog
from  Tom Tromey  <tromey@redhat.com>

	* java/lang/natDouble.cc (parseDouble): Reverted patch of
	2003-11-13.

Index: java/lang/natDouble.cc
===================================================================
RCS file: /cvs/gcc/gcc/libjava/java/lang/natDouble.cc,v
retrieving revision 1.17
diff -u -r1.17 natDouble.cc
--- java/lang/natDouble.cc 14 Nov 2003 01:48:30 -0000 1.17
+++ java/lang/natDouble.cc 26 Nov 2003 17:29:48 -0000
@@ -183,7 +183,7 @@
     {
       // Note that UTF can expand 3x.
       char *data = (char *) __builtin_alloca (3 * length + 1);
-      jsize blength = _Jv_GetStringUTFRegion (str, start, str->length(), data);
+      jsize blength = _Jv_GetStringUTFRegion (str, start, length, data);
       data[blength] = 0; 
 
       struct _Jv_reent reent;  


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]