Seaside 3.0 Final for Dolphin 6 Available

Folks:

Follow http://twit.io/2Fa to the full announce on
comp.lang.smalltalk.dolphin newsgroup.

Best Regards
Jose Sebastian Calvo

2010/11/6 GallegO <[hidden email]>:
> Folks:
>
> Follow http://twit.io/2Fa to the full announce on
> comp.lang.smalltalk.dolphin newsgroup.

That's great news, can you tell us whether the install instructions
[1] still apply? Can you tell us which tests are failing and why? If
the functional tests are green then likely everything is OK but it
could also be a bug in the tests not taking different semantics on
different platforms into account.

 [1] http://www.seaside.st/download/dolphin

Cheers
Philippe
_______________________________________________
seaside mailing list
[hidden email]
http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside

Philippe Marschall <philippe.marschall <at> gmail.com> writes:
> That's great news, can you tell us whether the install instructions
> [1] still apply?

The install instruction are the contained in the Seaside 3 download at
http://www.infoil.com.ar/seaside
Instructions at [1] are outdated.

For Dolphin you only need the download and "File-in" of Install Seaside.st and
you will end with your default internet browser open on the Welcome
application page.

Can you tell us which tests are failing and why?

I will write the details explaining the failure reasons. Most are related to
utf-encoding, bom character, I think none  is failing due to problems related
to platform diferences.

One thing to take into account is that this version actually does not supports
continuations due to the lack of Partial Continuations support in Dolphin
(only full continuations). I think that is possible but requires some
technical knowledge or at least lots of free time to get them working.

>  [1] http://www.seaside.st/download/dolphin
>
> Cheers
> Philippe
>

Cheers
  Sebastian Calvo



_______________________________________________
seaside mailing list
[hidden email]
http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside

Philip:

As promised here are the tests failing in Dolphin 6:

1) GRDolphinColorTest>>testFromStringThreeDigit
Not yet implemented. I forget implement this correctly so for the next update
will be OK

2) GRDolphinPlatformTest>>testMessageSendValueWithPossibleArguments
I will post to the Dolphin list about this. The problem is with
MessageSend>>valueWithArguments:. The Dolphin implementation does not allow to
pass additional arguments if the MessageSend was initially created with empty
arguments so is not valid send #valueWithArguments: (Array with: 2) to
(MessageSend receiver: 1 selector: #+). Strange, but is no so easy to change.
#valueWithPossibleArguments: is an alternative.

3) GRPlatformTest>>testCharacterAsUnicode
#asUnicode is a message present in the UTF support for Dolphin (from Udo
Schneider) and there are no senders of #asUnicode others than the test so
currently there is no reason to patch to support the behavior that Grease
expects. Maybe I can ask to Udo if it is safe change to answer the codePoint as
Grease expects.


4) GRPlatformTest>>testGreaseIntegerOnCharacter
    GRUtf8CodecTest>>testCodecUtf8Bom
    GRUtf8CodecTest>>testCodecUtf8ShortestForm

This are related to some UTF strings not handled same as Pharo. I will ask Udo,
I'm not an expert here.

5) WADebugErrorHandlerTest>>testMechanismProceed
The mechanism proceed works exactly as the test expects but you have hit (IMO) a
Dolphin bug here just in the test case, when using 1/0 to simulate an error.
For this I will post to the Dolphin list too.

Again, you should remember that currently we don't support Seaside Flow

Will the installation instructions on the Seaside.st site be updated?

Best Regards
Sebastian Calvo




_______________________________________________
seaside mailing list
[hidden email]
http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside

2010/11/9 Sebastian Calvo <[hidden email]>:
> Philip:
>
> As promised here are the tests failing in Dolphin 6:
>
> 1) GRDolphinColorTest>>testFromStringThreeDigit
> Not yet implemented. I forget implement this correctly so for the next update
> will be OK

Seaside does not send this, so this should have no effect.

> 2) GRDolphinPlatformTest>>testMessageSendValueWithPossibleArguments
> I will post to the Dolphin list about this. The problem is with
> MessageSend>>valueWithArguments:. The Dolphin implementation does not allow to
> pass additional arguments if the MessageSend was initially created with empty
> arguments so is not valid send #valueWithArguments: (Array with: 2) to
> (MessageSend receiver: 1 selector: #+). Strange, but is no so easy to change.
> #valueWithPossibleArguments: is an alternative.

It looks like Seaside is only sending #valueWithPossibleArguments: so
this does not seem critical.

> 3) GRPlatformTest>>testCharacterAsUnicode
> #asUnicode is a message present in the UTF support for Dolphin (from Udo
> Schneider) and there are no senders of #asUnicode others than the test so
> currently there is no reason to patch to support the behavior that Grease
> expects. Maybe I can ask to Udo if it is safe change to answer the codePoint as
> Grease expects.

Seaside does not send this, so this should have no effect.

> 4) GRPlatformTest>>testGreaseIntegerOnCharacter
>    GRUtf8CodecTest>>testCodecUtf8Bom
>    GRUtf8CodecTest>>testCodecUtf8ShortestForm

The second one is not so important, at least Pharo and VW don't pass
it as well. What's the problem with the first one, does Dolphin return
the BOM (U+FEFF)?

> This are related to some UTF strings not handled same as Pharo. I will ask Udo,
> I'm not an expert here.
>
> 5) WADebugErrorHandlerTest>>testMechanismProceed
> The mechanism proceed works exactly as the test expects but you have hit (IMO) a
> Dolphin bug here just in the test case, when using 1/0 to simulate an error.
> For this I will post to the Dolphin list too.

Does 1 / 0 not signal an error on Dolphin?

> Again, you should remember that currently we don't support Seaside Flow
>
> Will the installation instructions on the Seaside.st site be updated?

Yes, right now I have problems logging in but they will be updated.

Cheers
Philippe
_______________________________________________
seaside mailing list
[hidden email]
http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside

Philippe:

> > 4) GRPlatformTest>>testGreaseIntegerOnCharacter
> >    GRUtf8CodecTest>>testCodecUtf8Bom
> >    GRUtf8CodecTest>>testCodecUtf8ShortestForm
>
> The second one is not so important, at least Pharo and VW don't pass
> it as well. What's the problem with the first one, does Dolphin return
> the BOM (U+FEFF)?

Dolphin actually does the encoding using the Windows API wideCharTo...
multiByteCharTo... and I think that BOM is not handled by the API. I have not
found info about that.
For GRPlatformTest>>testGreaseIntegerOnCharacter
The problem is in the statement previous to the last assert, in
#decodeUtf8Character:

"SMP (4 byte)"
        manna := self decodeUtf8Character: #(240 144 140 188 ).
        self assert: manna greaseInteger = 66364

#decodeUtf8Character: internally executes one additional assert "self assert:
decoded size = 1" but the decoded size is actually 2 composed by #(U+D800
U+DF3C). What do you think?

 
> > 5) WADebugErrorHandlerTest>>testMechanismProceed
> > The mechanism proceed works exactly as the test expects but you have hit
(IMO) a
> > Dolphin bug here just in the test case, when using 1/0 to simulate an error.
> > For this I will post to the Dolphin list too.
>
> Does 1 / 0 not signal an error on Dolphin?

:) Yes but two times, you can read about this problem here http://twit.io/2Fk 

Cheers
Sebastian Calvo

_______________________________________________
seaside mailing list
[hidden email]
http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside

2010/11/9 Sebastian Calvo <[hidden email]>:
Done

Cheers
Philippe
_______________________________________________
seaside mailing list
[hidden email]
http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside

2010/11/10 Sebastian Calvo <[hidden email]>:
You know, this is why I had so much fun writing these tests ;-). The
problem here is not the BOM but that the win32 API is limited to 16bit
characters. For characters outside the BMP it has to use surrogate
pairs [1] that's why the string has two characters.

We could split the test in two, one for BMP and the other for SMP and
then you would pass the BMP one and fail the SMP one which I believe
would be a fair assessment of the situation.

>> > 5) WADebugErrorHandlerTest>>testMechanismProceed
>> > The mechanism proceed works exactly as the test expects but you have hit
> (IMO) a
>> > Dolphin bug here just in the test case, when using 1/0 to simulate an error.
>> > For this I will post to the Dolphin list too.
>>
>> Does 1 / 0 not signal an error on Dolphin?
>
> :) Yes but two times, you can read about this problem here http://twit.io/2Fk

Ooops, possible VM bug. Could we use something like like:

1 isDispatcher

And for MessageSend there's GRDelayedSend which has the semantics we
need so this shouldn't be a problem.

 [1] http://msdn.microsoft.com/en-us/library/dd374069(VS.85).aspx

Cheers
Philippe
_______________________________________________
seaside mailing list
[hidden email]
http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside