RE: Experiments building security2

2006-01-16 Thread Mishura, Stepan M
Tim Ellison wrote: >I think the immediate goal should be to make the 'security2' code a >replacement for 'security'. > >Since security is part of the componentized-build bootstrapping code >(i.e. required to run the compiler in self-hosting mode) then we need to >integrate the security2 build scrip

RE: Test suite layout

2006-01-13 Thread Mishura, Stepan M
t;>>> Agreed. But can we please *resist* the temptation to do this by >>>> incorporating JIRA issue numbers into test case names (e.g. calling >>>> unit test methods test_26() or test_JIRA_26()). I've seen this kind >>>> of approach adopted in a couple of proje

RE: Test suite layout

2006-01-12 Thread Mishura, Stepan M
more sense for >> the particular issue. Then marking certain code as being for regression >> test purposes could be done in comments that include the URL of the JIRA >> issue. Perhaps an agreed tag like "JIRA" or "BUG" etc could be used as an >> ey

RE: regression test suite

2006-01-11 Thread Mishura, Stepan M
Hello, Tim Ellison wrote: [snip] > >What is the useful distinction for regression tests being kept separate? > I can see that you may distinguish unit and 'system-level' tests just >because of the difference in frameworks etc. required, but why do I care >if the test was written due to a JIRA i

Kerberos service provider

2005-12-07 Thread Mishura, Stepan M
A service provider mechanism is aimed to provide an access to some service implementation. A service can be implemented by different vendors in quite different ways (different features, favorite bugs :-), backward compatibility and so on). We can consider Kerberos as some type of security service a

RE: componentization : security.jar

2005-12-05 Thread Mishura, Stepan M
>However, I just noticed that the security.jar file which gets built from >the Intel security code contribution has a MANIFEST.MF that does not >contain anything like the componentization information present in the >security.jar we can build from today's repository code. You are right. Thank you f

RE: Contribution of security, crypto, and x-net libraries

2005-11-29 Thread Mishura, Stepan M
>The 'full' implementation areas still need work too! Besides bringing >the APIs up to 5.0 we have some known bugs which I'll be happy to help >fix if the contribution is accepted into svn. > >It seems that you are falling over the stub implementation of >java.math.BigInteger. I'll also volunteer

RE: failed to use xml formatter on contributions combination

2005-11-29 Thread Mishura, Stepan M
> ah... same stack trace as Ant Bugzilla Report 36733: > http://issues.apache.org/bugzilla/show_bug.cgi?id=36733 Thank you for pointing out to the bug report. I added all jars mentioned in the bug report comments to class path and this resolved the issue. It made possible to use JUnitReport task t

failed to use xml formatter on contributions combination

2005-11-28 Thread Mishura, Stepan M
While running security unit test on contributions combination (lang/util/io/net/nio plus security/crypto/x-net) I tried to generate tests report with 'xml' formatter. So I added xml jars to class path and changed formatter type in build.xml to 'xml'. But all tests failed with [junit] Exce

RE: Contribution of security, crypto, and x-net libraries

2005-11-24 Thread Mishura, Stepan M
On Mon, Nov 21, 2005 at 09:08:28PM +0300, Loenko, Mikhail Y wrote: >> >> Some folks at Intel spent some time over the last week trying to >> figure >> >> out whether or not it is going to work with the recent contribution >> of >> >> lang/util/io/net/nio packages by IBM, and the general consensus