The Trunk: Monticello-bf.525.mcz

> Bert Freudenberg uploaded a new version of Monticello to project The Trunk:
> http://source.squeak.org/trunk/Monticello-bf.525.mcz
>
> ==================== Summary ====================
>
> Name: Monticello-bf.525
> Author: bf
> Time: 20 September 2012, 2:07:49.849 pm
> UUID: 35b8d257-a1b4-4dda-baa1-f497de4d4328
> Ancestors: Monticello-eem.524
>
> SnapshotBrowser: label class extensions as *inst or *class
>
> =============== Diff against Monticello-eem.524 ===============

> Hi,
>
> Am 20.09.2012 15:37, schrieb Frank Shearar:
>
>>
>> I haven't posted a progress report in a while. As of this commit we
>> have only 9 failing tests!
>>
>> http://squeakci.org/job/SqueakTrunk/lastCompletedBuild/testReport/
>>
>> One of those is entirely my fault: the HudsonBuildTools needs either
>> fleshing out to pull in RB and friends or, better, stripping down to
>> remove references to same.
>
> too slow network to download the zip, so I updated the last one to 12229 and
> got:
> 3222 run, 3176 passes, 18 expected failures, 7 failures, 21 errors, 0
> unexpected passes
> http://i.minus.com/iLqjljOy965AF.png

> On Thu, 20 Sep 2012, Frank Shearar wrote:
>
>> On 20 September 2012 16:52, Herbert König <[hidden email]> wrote:
>>>
>>> Hi,
>>>
>>> Am 20.09.2012 15:37, schrieb Frank Shearar:
>>>
>>>>
>>>> I haven't posted a progress report in a while. As of this commit we
>>>> have only 9 failing tests!
>>>>
>>>> http://squeakci.org/job/SqueakTrunk/lastCompletedBuild/testReport/
>>>>
>>>> One of those is entirely my fault: the HudsonBuildTools needs either
>>>> fleshing out to pull in RB and friends or, better, stripping down to
>>>> remove references to same.
>>>
>>>
>>> too slow network to download the zip, so I updated the last one to 12229
>>> and
>>> got:
>>> 3222 run, 3176 passes, 18 expected failures, 7 failures, 21 errors, 0
>>> unexpected passes
>>> http://i.minus.com/iLqjljOy965AF.png
>>
>>
>> Cool, we're making progress!
>>
>> #testNoObsoleteClasses fails because of the HudsonBuildTools.st we use.
>> #testMinimumNegativeInte... is, if I recall correctly, a marker for a
>> bugfix (Nicolas?) as is #testClassRemoval...
>
>
> #testMinimumNegativeIntegerArithmetic is failing because of a bug in the VM.
> You'll need a VM with Nicolas's patch applied to make it pass.

> #testMethodsWithUnboundGlobals,
> #testClassRemovalAndRecompilcationWontCreateDuplicateVariableBindings and
> #testNoObsoleteClasses fail (the last one might pass depending on the order
> of the tests) because if you remove a class from the system which is being
> referred to and then create a class with the same name, then the existing
> bindings will point to the old class (which is obsolete) instead of the new.
>
> #testUndeclared is failing because of HudsonBuildTools referencing RB and
> Gofer. I made a version where I only kept HDReport and HDTestReport and it
> just works.

>
> Levente
>
>>
>> frank
>>
>>> on Win 7, 64 Bit
>>>
>>> Cheers
>>>
>>> Herbert
>>>
>>
>
>
>

> we KNOW it works with VMx X, Y and Z on platforms A, B and C" as a
> minimum.
>
>> #testMethodsWithUnboundGlobals,
>> #testClassRemovalAndRecompilcationWontCreateDuplicateVariableBindings and
>> #testNoObsoleteClasses fail (the last one might pass depending on the order
>> of the tests) because if you remove a class from the system which is being
>> referred to and then create a class with the same name, then the existing
>> bindings will point to the old class (which is obsolete) instead of the new.
>>
>> #testUndeclared is failing because of HudsonBuildTools referencing RB and
>> Gofer. I made a version where I only kept HDReport and HDTestReport and it
>> just works.
>
> Great - if you send that stripped down version my way I'll update the
> CI scripts accordingly.

> On Thu, 20 Sep 2012, Frank Shearar wrote:
>
>> On 20 September 2012 18:35, Levente Uzonyi <[hidden email]> wrote:
>>>
>>>
>>> #testMinimumNegativeIntegerArithmetic is failing because of a bug in the
>>> VM.
>>> You'll need a VM with Nicolas's patch applied to make it pass.
>>
>>
>> OK, but then for the Official Squeak 4.4 VMs we'll want to use a
>> patched VM, right? I'd like to be able to say "with the released 4.4
>
>
> Hopefully yes. :)
>
>
>> we KNOW it works with VMx X, Y and Z on platforms A, B and C" as a
>> minimum.
>>
>>> #testMethodsWithUnboundGlobals,
>>> #testClassRemovalAndRecompilcationWontCreateDuplicateVariableBindings and
>>> #testNoObsoleteClasses fail (the last one might pass depending on the
>>> order
>>> of the tests) because if you remove a class from the system which is
>>> being
>>> referred to and then create a class with the same name, then the existing
>>> bindings will point to the old class (which is obsolete) instead of the
>>> new.
>>>
>>> #testUndeclared is failing because of HudsonBuildTools referencing RB and
>>> Gofer. I made a version where I only kept HDReport and HDTestReport and
>>> it
>>> just works.
>>
>>
>> Great - if you send that stripped down version my way I'll update the
>> CI scripts accordingly.
>
>
> I uploaded my modified HudsonBuildTools.st and tests.st here:
> http://leves.web.elte.hu/squeak/jenkins
>
> HudsonBuildTools.st is stripped down and sets/restores #authorInitials
> automatically. tests.st uses MC to find the packages to be tested and
> doesn't set #authorInitials anymore, since it's done by HudsonBuildTools.