Actually I think using lineDelimitor can't work because of UtilIO.parseObject which is used in checkBasicReadWriteObject by
UtilIO.readObject
The line is
if (buffer[length - 1] == '\n')
Not sure it's related to the issue though...
Jacques
From: "Jacques Le Roux" <[email protected]>
Ha... there is also one '\n' in checkBasicReadWriteObject...
Actually, tomorrow, I will try to replace all EOLs in this file by platforms
independent ones!
Jacques
From: "Jacques Le Roux" <[email protected]>
Yes same line, ie now 156 indeed
junit.framework.ComparisonFailure: write builder java.lang.Boolean expected:<...
...> but was:<......>
at
org.ofbiz.base.test.GenericTestCaseBase.assertEquals(GenericTestCaseBase.java:313)
at
org.ofbiz.base.util.test.UtilIOTests.checkBasicReadWriteObject(UtilIOTests.java:152)
at
org.ofbiz.base.util.test.UtilIOTests.testReadWriteObject(UtilIOTests.java:156)
at org.ofbiz.testtools.TestRunContainer.start(TestRunContainer.java:146)
at org.ofbiz.base.container.ContainerLoader.start(ContainerLoader.java:230)
at org.ofbiz.base.start.Start.startStartLoaders(Start.java:310)
at org.ofbiz.base.start.Start.start(Start.java:289)
at org.ofbiz.base.start.Start.main(Start.java:119)
Jacques
From: "Jacopo Cappellato" <[email protected]>
Thanks Jacques... are you sure it is happening at the same line (156) and not in the next (157) that it is still using the
hardcoded '\n'?
Jacopo
On Dec 17, 2011, at 8:22 AM, Jacques Le Roux wrote:
Hi Jacopo,
Unfortunately not: same error, pretty vicious thing, sorry no time to help
today :/
Jacques
From: "Jacopo Cappellato" <[email protected]>
Hi Jacques,
I suspect this is a problem with the end of line... are you running on Windows? Could you please try if the patch below fixes
the test?
Thanks,
Jacopo
Index: framework/base/src/org/ofbiz/base/util/test/UtilIOTests.java
===================================================================
--- framework/base/src/org/ofbiz/base/util/test/UtilIOTests.java (revision
1215052)
+++ framework/base/src/org/ofbiz/base/util/test/UtilIOTests.java (working copy)
@@ -39,6 +39,7 @@
@SourceMonitored
public class UtilIOTests extends GenericTestCaseBase {
+ private String lineSeparator = System.getProperty("line.separator");
private static final byte[] trademarkBytes = new byte[] {
(byte) 0xE2, (byte) 0x84, (byte) 0xA2
};
@@ -152,7 +153,7 @@
}
public void testReadWriteObject() throws Exception {
- checkBasicReadWriteObject(Boolean.TRUE, "java.lang.Boolean:true\n");
+ checkBasicReadWriteObject(Boolean.TRUE, "java.lang.Boolean:true" +
lineSeparator);
checkBasicReadWriteObject(Byte.valueOf("1"), "java.lang.Byte:1\n");
checkBasicReadWriteObject(Double.valueOf("1.0"),
"java.lang.Double:1.0\n");
checkBasicReadWriteObject(Float.valueOf("1.0"), "java.lang.Float:1.0\n");
On Dec 16, 2011, at 5:51 PM, Jacques Le Roux wrote:
Same Boolean issue as expected
Jacques
From: "Jacques Le Roux" <[email protected]>
OK Guys,
I will check that tonight...
Thanks for the effort!
Jacques
From: "Adrian Crum" <[email protected]>
I believe the base test failures are a timing issue, but I can't be sure
because I haven't looked into it thoroughly. I am
confident they have nothing to do with locale, however.
-Adrian
On 12/16/2011 2:25 PM, Jacopo Cappellato wrote:
Yes, right, a few minutes ago I have committed the fix you suggested.
I doubt it will fix the issue Jacques reported... but that may be unrelated to locale.... Jacques, is there any chance
that you
have some local modifications?
Jacopo
On Dec 16, 2011, at 3:17 PM, Adrian Crum wrote:
There is no need to investigate - I wrote that code and I can assure you the locale and time zone is always set when
OFBiz
starts. The problem is an incorrect (or missing) setting in one of the properties files - see my previous reply to
Jacques.
-Adrian
On 12/16/2011 1:43 PM, Jacopo Cappellato wrote:
No, I meant to say that I am pretty sure that the code that sets the locale is not executed when the tests are run... I
am
investigating into this now
Jacopo
On Dec 16, 2011, at 12:39 PM, Adrian Crum wrote:
Huh? Start.java is not executed when run-tests is called??!!
-Adrian
On 12/16/2011 11:15 AM, Jacopo Cappellato wrote:
Adrian,
I doubt that that code is executed when run-tests is called.
Jacopo
On Dec 16, 2011, at 11:14 AM, Adrian Crum wrote:
Jacopo,
Setting the language on the command line should not have any effect because OFBiz sets the language when it starts
up. See
the ofbiz.locale.default setting in start.properties.
I can duplicate the test failures by changing that setting to some other locale. That is why I thought it is a local
issue -
the tests are based on OOTB settings, and if you change the settings the tests
will fail.
-Adrian
On 12/16/2011 9:12 AM, Jacopo Cappellato wrote:
Jacques, Hans,
could you please try if with the below modifications the tests are successful?
Thanks,
Jacopo
