From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([140.186.70.92]:47185) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Qx0Zx-0002c1-LY for qemu-devel@nongnu.org; Fri, 26 Aug 2011 13:52:30 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1Qx0Zw-0006we-KJ for qemu-devel@nongnu.org; Fri, 26 Aug 2011 13:52:29 -0400 Received: from goliath.siemens.de ([192.35.17.28]:20626) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Qx0Zw-0006wC-8l for qemu-devel@nongnu.org; Fri, 26 Aug 2011 13:52:28 -0400 Message-ID: <4E57DD4B.4090302@siemens.com> Date: Fri, 26 Aug 2011 19:52:11 +0200 From: Jan Kiszka MIME-Version: 1.0 References: <4E57CE74.5070708@siemens.com> <20110826144325.0a41947a@doriath> In-Reply-To: <20110826144325.0a41947a@doriath> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Subject: [Qemu-devel] [PATCH] Fix qjson test of solidus encoding List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Luiz Capitulino Cc: Anthony Liguori , qemu-devel , "mdroth@linux.vnet.ibm.com" On 2011-08-26 19:43, Luiz Capitulino wrote: > On Fri, 26 Aug 2011 18:48:52 +0200 > Jan Kiszka wrote: > >> Hi all, >> >> current master has some qregressions: >> >> # ./check-qdict >> Running suite(s): QDict test-suite >> 93%: Checks: 16, Failures: 1, Errors: 0 >> check-qdict.c:289:F:Stress Test:qdict_stress_test:0: Assertion 'test_file != NULL' failed > > I guess you build qemu from a different directory, right? This test only > works from the source dir. I'll fix it (actually I thought I had already fixed it!). Yes, found that reason already. Works fine when fed with the test data as expected. > >> # ./check-qjson >> Running suite(s): QJSON test-suite >> 95%: Checks: 21, Failures: 1, Errors: 0 >> check-qjson.c:59:F:String Literals:escaped_string:0: / != \/ > > I found this problem while reading the parser code and added the test > so that we don't forget fixing it. However, at the time I looked at the > issue, I didn't know how to fix it. > > Michael, would you mind to take a look at it? After reading the spec I tend to think it's a test bug: -------8<------- "\/" is supposed to be decoded as "/", but there is no need to encode "/" via escape. Fix the existing test and add a second one expressing this. Signed-off-by: Jan Kiszka --- check-qjson.c | 3 ++- 1 files changed, 2 insertions(+), 1 deletions(-) diff --git a/check-qjson.c b/check-qjson.c index 64fcdcb..36d4ac2 100644 --- a/check-qjson.c +++ b/check-qjson.c @@ -33,7 +33,8 @@ START_TEST(escaped_string) { "\"\\n\"", "\n" }, { "\"\\r\"", "\r" }, { "\"\\t\"", "\t" }, - { "\"\\/\"", "\\/" }, + { "\"/\"", "/" }, + { "\"\\/\"", "/", .skip = 1 }, { "\"\\\\\"", "\\" }, { "\"\\\"\"", "\"" }, { "\"hello world \\\"embedded string\\\"\"", -- 1.7.3.4