Mailing-List: contact harmony-dev-help@incubator.apache.org; run by ezmlm
Precedence: bulk
Reply-To: harmony-dev@incubator.apache.org
Received-SPF: pass (asf.osuosl.org: domain of mloenko@gmail.com designates
 66.249.92.175 as permitted sender)
DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws;
        s=beta; d=gmail.com;
        h=received:message-id:date:from:to:subject:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references;
        b=KXMoLH6Dq29b+4OiZGXJ4OxPZobe4vzjJ8cO5ObIWXNkjIe152DXUZ/0jpenb8K80BTlwlmLmfIGBabKPCeVRTSuokQmMLTJ62I7MDvS0sojNn88oUrsYocGKiXsEF34RqlzL3/cm4W7C4N08tkevr9ACRrCFOOmJDy5kyw93Ys=
Message-ID: <906dd82e0605180320g13443b58l88669dc503d76941@mail.gmail.com>
Date: Thu, 18 May 2006 17:20:46 +0700
From: "Mikhail Loenko" <mloenko@gmail.com>
To: harmony-dev@incubator.apache.org
Subject: Re: [classlib] Layout of tests in crypto module
In-Reply-To: <446C4328.80702@googlemail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline
References: <906dd82e0605142112k3bce88a2i9842b6375c03f734@mail.gmail.com>
	 <446900B7.3040206@googlemail.com>
	 <906dd82e0605160032p7a1f373do81d71f52222c2291@mail.gmail.com>
	 <4469A679.80805@googlemail.com>
	 <906dd82e0605160519q4fbedfceu4ffedd11d9d059d8@mail.gmail.com>
	 <446AEBF0.1060708@googlemail.com>
	 <906dd82e0605170528y666155daq9b77dfaa303571ad@mail.gmail.com>
	 <446C4328.80702@googlemail.com>

That sounds very reasonable, but I have a problem:

I tried to implement it and found that as far as we put all test results
into a single directory and generate a single report, we can't have
different tests with the same name.

For example we can't have impl and api tests of
o.a.h.module.tests.java.lang.SomeClassTest
Looks like we should put some differentiating mark to a class or package na=
me.

Thanks,
Mikhail

2006/5/18, George Harley <george.c.harley@googlemail.com>:
> Mikhail Loenko wrote:
> > Hi George,
> >
> > I use ant to build and run the tests, so I'm likely unaware of some
> > Eclipse
> > problems.
> >
> > If we put classpath test classes to
> >   impl/java and api/java
> > and bootclasspath ones to something like
> >   impl/java.injected and api/java.injected
> > will it solve the problem you describe?
> >
> > Thanks,
> > Mikhail
> >
> >
>
> Hi Mikhail,
>
> Yes, I think that compiling to separate bin folders would do the trick ;
> a bin for stuff that will go on the classpath and a bin for stuff that
> will be loaded on the bootclasspath. In order to reach that goal the
> test source itself will AFAIK need to be laid out in a similar manner.
> That is, the directory tree will look something like the following (may
> need to be viewed in wide-screen):
>
>
> src/test/api/java             <--- implementation-independent tests
> intended to be loaded by system class loader, compiled to bin
>
> src/test/api/java.injected    <--- implementation-independent tests
> intended to be loaded by boot class loader, compiled to bin.injected
>
> src/test/impl/java             <--- Harmony-specific tests intended to
> be loaded by system class loader, compiled to bin
>
> src/test/impl/java.injected    <--- Harmony-specific tests intended to
> be loaded by boot class loader, compiled to bin.injected
>
>
>
> Does that sound reasonable ?
>
>
> Best regards,
> George
>
>
> > 2006/5/17, George Harley <george.c.harley@googlemail.com>:
> >> Mikhail Loenko wrote:
> >> > 2006/5/16, George Harley <george.c.harley@googlemail.com>:
> >> >> Mikhail Loenko wrote:
> >> >> > Hi George, see below
> >> >> >
> >> >> > 2006/5/16, George Harley <george.c.harley@googlemail.com>:
> >> >> >> Hi Mikhail,
> >> >> >>
> >> >> >> I have a couple of minor comments about your proposal for a test
> >> >> >> layouts. I should have responded sooner, I know, but I have
> >> suffered
> >> >> >> from a number of hardware problems in the past few weeks that
> >> slowed
> >> >> >> things down somewhat for me. Anyway, it's all great but it would
> >> >> be nice
> >> >> >> to get answers to the following ...
> >> >> >>
> >> >> >> 1) The section on "Location of the tests in the directory tree"
> >> >> >> proposes <modulename>/src/tests/impl for Harmony specific tests
> >> and
> >> >> >> <modulename>/src/tests/api for implementation-independent tests.
> >> >> Where
> >> >> >> would tests go for Harmony API's that are not part of the J2SE
> >> >> spec but
> >> >> >> are still accessible ? Strictly speaking they are API as well as
> >> >> being
> >> >> >> specific to Harmony.
> >> >> >
> >> >> > The main idea is to separate tests that must pass on any conforma=
nt
> >> >> > implementation from the tests passing on Harmony only.
> >> >> >
> >> >> > When these are separated we can e.g. easily validate
> >> "implementation-
> >> >> > independent" tests by running them against RI. Actually I would n=
ot
> >> >> > like to
> >> >> > start an endless "philosophical" discussion here about what are
> >> >> > unit vs. api vs. whatever tests.
> >> >>
> >> >> Me ? Start a philosophical discussion ? :-)
> >> >>
> >> >>
> >> >> >
> >> >> > "api" is just a name (suggested by Tim) and might be changed to a=
ny
> >> >> > unconfusing one.
> >> >> >
> >> >> > So, back to your question, those must go to 'impl' as far as
> >> they fail
> >> >> > on RI:
> >> >> > "<modulename>/src/tests/impl - Harmony specific tests"
> >> >> >
> >> >> >> What about the location of tests designed to run on
> >> >> >> the classpath and tests designed to run on the bootclasspath ?
> >> >> That does
> >> >> >> not seem to be addressed in this section. When the tests are run
> >> >> how are
> >> >> >
> >> >> > Directory defines the root of the test suite, then the package
> >> goes,
> >> >> > see "Package names for different types of the tests" section:
> >> >> > "If the test is designed to be run from bootclasspath then its
> >> package
> >> >> > is the same as the package of the class under test"
> >> >> >
> >> >>
> >> >> Is it the case that all of the classes under a particular source
> >> folder
> >> >> (say src/tests/api/java) get compiled to one output folder ?
> >> >
> >> > We have not discuss it yet. So, suggestions are welcome!  :)
> >> >
> >> > Thanks,
> >> > Mikhail
> >> >
> >> >
> >>
> >> Hi Mikhail,
> >>
> >>  OK, great. I'm wondering about how the test classes intended to be
> >> loaded by the bootclasspath loader get distinguished from the classes
> >> that are intended to be loaded by the system loader. For the sake of
> >> discussion, imagine that all of the test classes under a given source
> >> folder (e.g. src/tests/api/java) get compiled to just one output folde=
r
> >> (e.g. bin). At test runtime we only want the classes that are "in the
> >> same package as the class under test" to be on the bootclasspath - but
> >> how is this enforced ? If the tests are being run from an Ant script
> >> then I think that it is possible to do this using Ant's path-like
> >> structures where wildcards can be specified among the classpath and
> >> bootclasspath elements. But how can an equivalent degree of separation
> >> be enforced if the tests are being run from an IDE such as Eclipse ?
> >>
> >>  I am not exactly an Eclipse "power-user" so it is possible that I am
> >> missing some piece of magic here but it appears to me that if I want t=
o
> >> configure the classpath and bootclasspath for running a particular set
> >> of unit tests I can only work at the output folder level. That is, it
> >> can be configured that classes under a root folder of "bin" are on the
> >> classpath and that classes under the root folder "bin-other" are
> >> similarly on the bootclasspath. But there is nothing available that on=
ly
> >> puts classes in package java.util.* under "bin" on the bootclasspath a=
nd
> >> everything else on the classpath.
> >>
> >>  All of which makes me wonder if we should aim at compiling classes
> >> intended for the bootclasspath to one bin and classes intended for the
> >> classpath to another bin. It wouldn't really make much difference to A=
nt
> >> users but as far as I can tell it would offer more flexibility to IDE
> >> users who would then have the potential to run the tests in a way that
> >> was more familiar to them (runtime configuration, green bar etc).
> >>
> >>
> >> Best regards,
> >> George
> >>
> >> >
> >> >>
> >> >>
> >> >>
> >> >> >> the "bootclasspath" and "classpath" tests distinguished ?
> >> Purely on
> >> >> >> package name ? Did you ever see the append I wrote to the list a
> >> >> couple
> >> >> >> of weeks ago on this topic ? [1]
> >> >> >
> >> >> > Yes, I've seen it. As you could notice we had more test categorie=
s.
> >> >> > They are described in the same thread. We might have
> >> >> > "independent" tests running from bootclasspath and "specific" one=
s
> >> >> > running from classpath.
> >> >> >
> >> >> >>
> >> >> >> 2) Still in the "Location of the tests in the directory tree"
> >> >> section,
> >> >> >> shouldn't the suggested source folder names include "java" in
> >> there ?
> >> >> >> e.g. <modulename>/src/tests/java/api ? What is wrong with the
> >> >> >> src/test/java (below here is Java code), src/test/resources (bel=
ow
> >> >> here
> >> >> >> is non-code test artefacts) convention ? I notice that at
> >> present the
> >> >> >> page does not include any mention of where test resources would
> >> go.
> >> >> >
> >> >> > Yes, you are right. It's missing.
> >> >> >
> >> >> > It was supposed to be under test "type", like:
> >> >> > module/src/test/api/java
> >> >> > module/src/test/api/resources
> >> >> >
> >> >> >
> >> >> >>
> >> >> >> 3) What does the sentence "Tests are not separated by
> >> functionality
> >> >> >> under test, e.g. tests against clone() methods are not
> >> separated from
> >> >> >> tests against equals() methods" actually mean ?
> >> >> >
> >> >> > That was to address Tim's concern:
> >> >> > "  > 1. Tests are separated on directory level by 'intention'
> >> >> >
> >> >> > I agree where the intention is broad (i.e. functional tests vs.
> >> >> > performance tests vs. stress tests) but I'm not convinced that
> >> we need
> >> >> > to separate to a precise 'intent', at least I've never been in a
> >> >> > position where I've wished that my my serialization tests are
> >> separate
> >> >> > from my cloning tests or whatever."
> >> >> >
> >> >> > Feel free to reword if something is not clear in the proposal
> >> >> >
> >> >> > Thanks,
> >> >> > Mikhail
> >> >> >
> >> >> >
> >> >> >>
> >> >> >>
> >> >> >> Best regards,
> >> >> >> George
> >> >> >>
> >> >> >> [1]
> >> >> >>
> >> >>
> >> http://mail-archives.apache.org/mod_mbox/incubator-harmony-dev/200604.=
mbox/%3c4450A1CC.1060803@googlemail.com%3e
> >>
> >> >>
> >> >> >>
> >> >> >>
> >> >> >>
> >> >> >>
> >> >> >>
> >> >> >> Mikhail Loenko wrote:
> >> >> >> > Hello
> >> >> >> >
> >> >> >> > I would like to make some changes in the crypto module:
> >> >> >> >
> >> >> >> > - Separate implemetation-independent tests from implementation
> >> >> >> specific
> >> >> >> > ones.
> >> >> >> >
> >> >> >> > - Fix layout according to the proposed scheme [1]
> >> >> >> >
> >> >> >> > Please let me know if you do any changes in that module then
> >> >> >> > I'll delay restruct.
> >> >> >> >
> >> >> >> > Thanks,
> >> >> >> > Mikhail
> >> >> >> >
> >> >> >> > [1]
> >> >> >> >
> >> >> >>
> >> >>
> >> http://incubator.apache.org/harmony/subcomponents/classlibrary/testing=
.html
> >>
> >> >>
>
>
> ---------------------------------------------------------------------
> Terms of use : http://incubator.apache.org/harmony/mailing.html
> To unsubscribe, e-mail: harmony-dev-unsubscribe@incubator.apache.org
> For additional commands, e-mail: harmony-dev-help@incubator.apache.org
>
>

---------------------------------------------------------------------
Terms of use : http://incubator.apache.org/harmony/mailing.html
To unsubscribe, e-mail: harmony-dev-unsubscribe@incubator.apache.org
For additional commands, e-mail: harmony-dev-help@incubator.apache.org