Hi Stephen,
I confirmed that issuing parseStrict() was irrelevant. The incorrect
text won't throw the exception in "lenient" mode as well. In addition,
"builder" is always instantiated in AbstractTestPrinterParser on each
TestNG invocation and "strict" is the default mode. Thus I did not
explicitly issue parseStrict() in the test.
BTW, I found typos wrt the default parse mode. Included the corrections
in this webrev as well:
http://cr.openjdk.java.net/~naoto/8230136/webrev.01/
Naoto
On 9/11/19 2:01 AM, Stephen Colebourne wrote:
The bug references parseStrict() but the test does not. Is the builder
already set to parseStrict() ? Anyway, if the bug is to be clearly
squished, parseStrict() should appear somewhere.
Stephen
On Tue, 10 Sep 2019 at 23:42, Joe Wang <huizhe.w...@oracle.com> wrote:
+1, looks good to me.
Best regards,
Joe
On 9/10/19 2:20 PM, naoto.s...@oracle.com wrote:
Hello,
Please review the fix to the following issue:
https://bugs.openjdk.java.net/browse/JDK-8230136
The proposed changeset is located at:
http://cr.openjdk.java.net/~naoto/8230136/webrev.00/
The fix is to correct the condition of the invalid case, as suggested
in the bug report.
Naoto