From joel at ardishealth.com  Tue Sep  1 15:09:47 2009
From: joel at ardishealth.com (Joel Turnbull)
Date: Tue, 1 Sep 2009 09:09:47 -0400
Subject: addCondition:labelled:
Message-ID: <26ee8b3e0909010609q242c9935u77828cd834430f25@mail.gmail.com>

addCondition:labelled does not seem to be working the way I expected.

I just want to validate that a given text field is not empty

I have a description definition method and setter:

descriptionTags
^MAStringDescription new
    selectorAccessor: #tagsConverter;
    label: 'Tags';
    priority: 40;
    addCondition: [:value | value notEmpty ]
        labelled: 'Please provide at least one tag';
    yourself

tagsConverter: aString
    tags := aString asUppercase subStrings asSet

On submit, I get an error that nil does not understand asUppercase.
Shouldn't the condition fail before the setter is called? I've tried [
:value | value notNil ], and beRequired as well, got the same error. I also
tried to change the initialize method of the object I'm describing to set
tags to an empty collection but that didn't work either.

Thanks,
Joel
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.iam.unibe.ch/pipermail/smallwiki/attachments/20090901/f2abe4b7/attachment.html>

From johnmci at smalltalkconsulting.com  Tue Sep  1 17:08:19 2009
From: johnmci at smalltalkconsulting.com (John M McIntosh)
Date: Tue, 1 Sep 2009 17:08:19 +0200
Subject: a question about pier syntax
Message-ID: <ED5C1329-2549-485C-AB50-2391B5EB320B@smalltalkconsulting.com>

Let me forward this to the pier list since it's a syntax related  
question.


 From kunicki at gmail.com

> First a quick note of thanks. I am really enjoying using Mobile Wiki  
> Server. I also like that I can edit from my desktop. I have read  
> your site and been digging around piercms.com as well. I was hoping  
> you could help with two questions:
>
> 1) I have tried using Breadcrumb but I cant get it to display in a  
> page or even in the overall structure (Using the design button). How  
> do I use this command?
>
> Basically I would love to have a breadcrumb showing me where I am in  
> the hiearchy of the site. Example:
>
> Home > Grammar > Nouns > Examples
>
> this would allow me to quickly move back up the hiearchy.
>
>
> 2) I like the tree  and navigation components. Is it impossible to  
> embed this into a page? I would like to have a page with some basic  
> content and then have a tree structure listed below. However, it  
> seems i have to add a component page, set it to tree, and thats all  
> i can do with it.
>
> Thanks for your patience with me. Again I thank you for bringing  
> this to the iPhone.
>
> Chris
--
= 
= 
= 
========================================================================
John M. McIntosh <johnmci at smalltalkconsulting.com>   Twitter:   
squeaker68882
Corporate Smalltalk Consulting Ltd.  http://www.smalltalkconsulting.com
= 
= 
= 
========================================================================





From renggli at gmail.com  Wed Sep  2 00:02:30 2009
From: renggli at gmail.com (Lukas Renggli)
Date: Wed, 2 Sep 2009 00:02:30 +0200
Subject: addCondition:labelled:
In-Reply-To: <26ee8b3e0909010609q242c9935u77828cd834430f25@mail.gmail.com>
References: <26ee8b3e0909010609q242c9935u77828cd834430f25@mail.gmail.com>
Message-ID: <67628d690909011502i373695fcn630253adacdd11d5@mail.gmail.com>

> I just want to validate that a given text field is not empty

Normally sending #beRequired to the description should be enough. At
least with my Seaside forms this works well and there is no need to
add an additional check for #isEmpty.

> On submit, I get an error that nil does not understand asUppercase.
> Shouldn't the condition fail before the setter is called?

#beRequired ensures that a nil or empty value is not committed.
Additional conditions will not be evaluated (there are various
examples of this in Pier, for example).

Cheers,
Lukas

-- 
Lukas Renggli
http://www.lukas-renggli.ch

From morph at growler.net  Wed Sep  2 14:29:06 2009
From: morph at growler.net (Grimble Crumble)
Date: Wed, 2 Sep 2009 14:29:06 +0200
Subject: Ultra-newbie question
Message-ID: <0C57C628-6990-4862-AB4C-18612D813965@growler.net>

I have used Squeak, and I have used Seaside, but it has been a while.

I have brought up the Pier environment on a headless server, and it  
looks as if I probably have the ability to configure it, add to it,  
etc.  But there does not appear to be any documentation for Pier that  
is aimed at those who have forgotten what they may ever have learned  
about Seaside.  I'm going back over the Seaside tutorial, but I  
wonder if someone can tell me how to log in to the default Pier  
server, the one that comes up when you make minimal changes out of  
the box (not even running under Apache).  I had a "Log in" link on  
the page at one point, made the attempt with no errors reported, and  
was unable to edit anything.  I think I'm missing a major concept  
somehow.

Can anyone suggest what major item I've probably forgotten about  
Seaside?

Thanks.

GC



From renggli at gmail.com  Wed Sep  2 14:52:02 2009
From: renggli at gmail.com (Lukas Renggli)
Date: Wed, 2 Sep 2009 14:52:02 +0200
Subject: Ultra-newbie question
In-Reply-To: <0C57C628-6990-4862-AB4C-18612D813965@growler.net>
References: <0C57C628-6990-4862-AB4C-18612D813965@growler.net>
Message-ID: <67628d690909020552l33f0f60cgcc47cb85201fe15c@mail.gmail.com>

> minimal changes out of the box (not even running under Apache). ?I had a
> "Log in" link on the page at one point, made the attempt with no errors
> reported, and was unable to edit anything. ?I think I'm missing a major
> concept somehow.

The default username/password of Pier is admin/pier.

The default username/password of Seaside is admin/seaside.

> Can anyone suggest what major item I've probably forgotten about Seaside?

The basics on Pier are documented here: http://www.piercms.com/doc.

Let us know if you have further questions.

Lukas

-- 
Lukas Renggli
http://www.lukas-renggli.ch


From nicolas.roard at gmail.com  Wed Sep  2 16:04:22 2009
From: nicolas.roard at gmail.com (Nicolas Roard)
Date: Wed, 2 Sep 2009 15:04:22 +0100
Subject: Picasa client for Pier
In-Reply-To: <4bd493810908240650o5daec887ka7e0e2f0f460f7e6@mail.gmail.com>
References: <4bd493810908240349k7a5cb7b5l156f9ffba1e79fcf@mail.gmail.com>
	<67628d690908240421p697ff073t13e4439ce0b857f3@mail.gmail.com>
	<4bd493810908240650o5daec887ka7e0e2f0f460f7e6@mail.gmail.com>
Message-ID: <25626500909020704g1633a9e9kecc66c9de1321833@mail.gmail.com>

On Mon, Aug 24, 2009 at 2:50 PM, Ciprian
Teodorov<ciprian.teodorov at gmail.com> wrote:
>
>
> On Mon, Aug 24, 2009 at 1:21 PM, Lukas Renggli <renggli at gmail.com> wrote:
>>
>> > If somebody is interested, I implemented a small picasaweb plugin for
>> > Pier
>> > which permits to embed the albums and the photos from picasa
>> > into a running pier web application. The plugin source is posted on
>> > squeaksource site: http://squeaksource.com/@uxjO7eHSe8tI9dKs/9zFmuhpA.
>>
>> Cool! What is the name of the repository URL? The link you gave is
>> unfortunately not restful.
>
> Sorry for the link -- here you have the good one ;)
> http://www.squeaksource.com/PicasaClient/

I modified it a bit to use lightbox.
See http://www.impro-v-istres.fr/seaside/pier/photos/picasa for an
example of how it looks.
The code is on http://www.squeaksource.com/PicasaClientLightbox/

-- 
Nicolas Roard
"I love deadlines. I like the whooshing sound
they make as they fly by." -- Douglas Adams

From morph at growler.net  Wed Sep  2 20:33:42 2009
From: morph at growler.net (Grimble Crumble)
Date: Wed, 2 Sep 2009 20:33:42 +0200
Subject: Ultra-newbie question
In-Reply-To: <67628d690909020552l33f0f60cgcc47cb85201fe15c@mail.gmail.com>
References: <0C57C628-6990-4862-AB4C-18612D813965@growler.net>
	<67628d690909020552l33f0f60cgcc47cb85201fe15c@mail.gmail.com>
Message-ID: <7953D51F-748B-4A59-BDB4-5BF32102BF2D@growler.net>

On Sep 2, 2009, at 2:52 PM, Lukas Renggli wrote:

>> minimal changes out of the box (not even running under Apache).  I  
>> had a
>> "Log in" link on the page at one point, made the attempt with no  
>> errors
>> reported, and was unable to edit anything.  I think I'm missing a  
>> major
>> concept somehow.
>
> The default username/password of Pier is admin/pier.

I have now achieved login, can't figure out what was happening before,
and may never.  Next, ...

>> Can anyone suggest what major item I've probably forgotten about  
>> Seaside?
>
> The basics on Pier are documented here: http://www.piercms.com/doc.
>
> Let us know if you have further questions.

Thanks very much.  I'll try to make at least marginal sense.

I want to make a random change on the page, for no other reason than
to see that I can have some effect.  So, the way to do that is... Turn
on Halos, I think, and then... OK, in a halo there are icons for Class
Browser, Object Inspector, and CSS Style Editor.  And also a pair of
letters, R-ender and S-ource.  I can edit the behavior of classes, but
if I want to change the color of one of the text objects, how do I do
that?  If I want to change the text itself, how do I do that?

OK, hard things are easy and easy things are hard.  The tools seem to
address issues of much greater complexity than I'm able to think about
yet.  And I don't see the corresponding tools for doing anything
dead simple.  And I have a feeling I'm staring blindly right at it.

Seaside manual maybe?  Any section in particular?  What I'm trying to
do should be the easy stuff, I would have thought.

I thank you in advance for your help.

>
> Lukas
>
> -- 
> Lukas Renggli
> http://www.lukas-renggli.ch
>
> _______________________________________________
> Magritte, Pier and Related Tools ...
> https://www.iam.unibe.ch/mailman/listinfo/smallwiki


From p3anoman at gmail.com  Wed Sep  2 21:40:23 2009
From: p3anoman at gmail.com (John McKeon)
Date: Wed, 2 Sep 2009 15:40:23 -0400
Subject: Ultra-newbie question
In-Reply-To: <7953D51F-748B-4A59-BDB4-5BF32102BF2D@growler.net>
References: <0C57C628-6990-4862-AB4C-18612D813965@growler.net>
	<67628d690909020552l33f0f60cgcc47cb85201fe15c@mail.gmail.com>
	<7953D51F-748B-4A59-BDB4-5BF32102BF2D@growler.net>
Message-ID: <40f637ec0909021240k482b5007m2cf619c0b36cfef1@mail.gmail.com>

On Wed, Sep 2, 2009 at 2:33 PM, Grimble Crumble <morph at growler.net> wrote:

> On Sep 2, 2009, at 2:52 PM, Lukas Renggli wrote:
>
>  minimal changes out of the box (not even running under Apache).  I had a
>>> "Log in" link on the page at one point, made the attempt with no errors
>>> reported, and was unable to edit anything.  I think I'm missing a major
>>> concept somehow.
>>>
>>
>> The default username/password of Pier is admin/pier.
>>
>
> I have now achieved login, can't figure out what was happening before,
> and may never.  Next, ...
>
>  Can anyone suggest what major item I've probably forgotten about Seaside?
>>>
>>
>> The basics on Pier are documented here: http://www.piercms.com/doc.
>>
>> Let us know if you have further questions.
>>
>
> Thanks very much.  I'll try to make at least marginal sense.
>
> I want to make a random change on the page, for no other reason than
> to see that I can have some effect.  So, the way to do that is... Turn
> on Halos, I think, and then... OK, in a halo there are icons for Class
> Browser, Object Inspector, and CSS Style Editor.  And also a pair of
> letters, R-ender and S-ource.  I can edit the behavior of classes, but
> if I want to change the color of one of the text objects, how do I do
> that?  If I want to change the text itself, how do I do that?
>

After you are logged in, commands show up on the bottom of the page. (Where
the Login link was, NOT the halos toolbar)

Click "Edit" to edit the page.
Click "Remove" to remove it.
Click "Add" to add a sub-page or a file or one of the many components
available.
Click "Edit design" to change the CSS or the layout of the page structure.
Play with it, don't be afraid. the thing to understand is that you are not
really writing any seaside code at this point. You are managing content.


> OK, hard things are easy and easy things are hard.  The tools seem to
> address issues of much greater complexity than I'm able to think about
> yet.  And I don't see the corresponding tools for doing anything
> dead simple.  And I have a feeling I'm staring blindly right at it.
>

See above.


>
> Seaside manual maybe?  Any section in particular?  What I'm trying to
> do should be the easy stuff, I would have thought.
>
> I thank you in advance for your help.


Hope this helped
John

>
>
>
>> Lukas
>>
>> --
>> Lukas Renggli
>> http://www.lukas-renggli.ch
>>
>> _______________________________________________
>> Magritte, Pier and Related Tools ...
>> https://www.iam.unibe.ch/mailman/listinfo/smallwiki
>>
>
> _______________________________________________
> Magritte, Pier and Related Tools ...
> https://www.iam.unibe.ch/mailman/listinfo/smallwiki
>



-- 
http://jmck.seasidehosting.st
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.iam.unibe.ch/pipermail/smallwiki/attachments/20090902/0b6552e9/attachment.html>

From morph at growler.net  Wed Sep  2 22:01:06 2009
From: morph at growler.net (Grimble Crumble)
Date: Wed, 2 Sep 2009 22:01:06 +0200
Subject: Ultra-newbie question
In-Reply-To: <40f637ec0909021240k482b5007m2cf619c0b36cfef1@mail.gmail.com>
References: <0C57C628-6990-4862-AB4C-18612D813965@growler.net>
	<67628d690909020552l33f0f60cgcc47cb85201fe15c@mail.gmail.com>
	<7953D51F-748B-4A59-BDB4-5BF32102BF2D@growler.net>
	<40f637ec0909021240k482b5007m2cf619c0b36cfef1@mail.gmail.com>
Message-ID: <6C25B917-C960-480F-9104-D93EC9EE4CC1@growler.net>


On Sep 2, 2009, at 9:40 PM, John McKeon wrote:

>
>
> On Wed, Sep 2, 2009 at 2:33 PM, Grimble Crumble <morph at growler.net>  
> wrote:
> On Sep 2, 2009, at 2:52 PM, Lukas Renggli wrote:
>
> [...]
> The basics on Pier are documented here: http://www.piercms.com/doc.
>
> Let us know if you have further questions.
>
> Thanks very much.  I'll try to make at least marginal sense.
>
> I want to make a random change on the page, for no other reason than
> to see that I can have some effect.  So, the way to do that is... Turn
> on Halos, I think, and then... OK, in a halo there are icons for Class
> Browser, Object Inspector, and CSS Style Editor.  And also a pair of
> letters, R-ender and S-ource.  I can edit the behavior of classes, but
> if I want to change the color of one of the text objects, how do I do
> that?  If I want to change the text itself, how do I do that?
>
> After you are logged in, commands show up on the bottom of the  
> page. (Where the Login link was, NOT the halos toolbar)
>
> Click "Edit" to edit the page.
> Click "Remove" to remove it.
> Click "Add" to add a sub-page or a file or one of the many  
> components available.
> Click "Edit design" to change the CSS or the layout of the page  
> structure.
> Play with it, don't be afraid. the thing to understand is that you  
> are not really writing any seaside code at this point. You are  
> managing content.
>

Sure, OK, that's exactly what I want to do.  I had to stumble a bit  
more, apparently.  I appreciate your patience.

Two questions.
  1. If I click the Class Browser button, then I *am* modifying code,  
right?  I mean, I can make changes to class behavior of, ... the Pier  
classes?

  2. When I clicked Edit, I could totally go to town on the "Welcome  
to Pier!" block of text.  I still haven't worked out how to affect  
the "What's new" block.

>
> [...]
>
> Hope this helped
> John
>

Yes indeed.  Thanks.

> [...]
> -- 
> http://jmck.seasidehosting.st
> _______________________________________________
> Magritte, Pier and Related Tools ...
> https://www.iam.unibe.ch/mailman/listinfo/smallwiki

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.iam.unibe.ch/pipermail/smallwiki/attachments/20090902/655a20a7/attachment-0001.html>

From ciprian.teodorov at gmail.com  Thu Sep  3 20:10:14 2009
From: ciprian.teodorov at gmail.com (Ciprian Teodorov)
Date: Thu, 3 Sep 2009 20:10:14 +0200
Subject: Picasa client for Pier
In-Reply-To: <25626500909020704g1633a9e9kecc66c9de1321833@mail.gmail.com>
References: <4bd493810908240349k7a5cb7b5l156f9ffba1e79fcf@mail.gmail.com> 
	<67628d690908240421p697ff073t13e4439ce0b857f3@mail.gmail.com> 
	<4bd493810908240650o5daec887ka7e0e2f0f460f7e6@mail.gmail.com> 
	<25626500909020704g1633a9e9kecc66c9de1321833@mail.gmail.com>
Message-ID: <4bd493810909031110w36e3d6b0i1e8436b0f3c31fba@mail.gmail.com>

hi Nicolas,
that's really cool but why didn't you publish it as a new version of the
same project?

Cheers,
Cip

On Wed, Sep 2, 2009 at 4:04 PM, Nicolas Roard <nicolas.roard at gmail.com>wrote:

> On Mon, Aug 24, 2009 at 2:50 PM, Ciprian
> Teodorov<ciprian.teodorov at gmail.com> wrote:
> >
> >
> > On Mon, Aug 24, 2009 at 1:21 PM, Lukas Renggli <renggli at gmail.com>
> wrote:
> >>
> >> > If somebody is interested, I implemented a small picasaweb plugin for
> >> > Pier
> >> > which permits to embed the albums and the photos from picasa
> >> > into a running pier web application. The plugin source is posted on
> >> > squeaksource site: http://squeaksource.com/@uxjO7eHSe8tI9dKs/9zFmuhpA
> .
> >>
> >> Cool! What is the name of the repository URL? The link you gave is
> >> unfortunately not restful.
> >
> > Sorry for the link -- here you have the good one ;)
> > http://www.squeaksource.com/PicasaClient/
>
> I modified it a bit to use lightbox.
> See http://www.impro-v-istres.fr/seaside/pier/photos/picasa for an
> example of how it looks.
> The code is on http://www.squeaksource.com/PicasaClientLightbox/
>
> --
> Nicolas Roard
> "I love deadlines. I like the whooshing sound
> they make as they fly by." -- Douglas Adams
> _______________________________________________
> Magritte, Pier and Related Tools ...
> https://www.iam.unibe.ch/mailman/listinfo/smallwiki
>



-- 
Ciprian TEODOROV
Ph.D Student
Lab-STICC/AS CNRS UMR 3192
University of Brest

phone: (+33)(0) 6 08 54 73 48
http://www.teodorov.ro
http://as.univ-brest.fr/ciprian
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.iam.unibe.ch/pipermail/smallwiki/attachments/20090903/fca9218d/attachment.html>

From nicolas.roard at gmail.com  Thu Sep  3 21:20:11 2009
From: nicolas.roard at gmail.com (Nicolas Roard)
Date: Thu, 3 Sep 2009 20:20:11 +0100
Subject: Picasa client for Pier
In-Reply-To: <4bd493810909031110w36e3d6b0i1e8436b0f3c31fba@mail.gmail.com>
References: <4bd493810908240349k7a5cb7b5l156f9ffba1e79fcf@mail.gmail.com>
	<67628d690908240421p697ff073t13e4439ce0b857f3@mail.gmail.com>
	<4bd493810908240650o5daec887ka7e0e2f0f460f7e6@mail.gmail.com>
	<25626500909020704g1633a9e9kecc66c9de1321833@mail.gmail.com>
	<4bd493810909031110w36e3d6b0i1e8436b0f3c31fba@mail.gmail.com>
Message-ID: <25626500909031220u1c9e389bpf721d46fc819f02e@mail.gmail.com>

On Thu, Sep 3, 2009 at 7:10 PM, Ciprian
Teodorov<ciprian.teodorov at gmail.com> wrote:
> hi Nicolas,
> that's really cool but why didn't you publish it as a new version of the
> same project?

I tried, but the project is not writable apparently... and as I didn't
know if you wanted this or not, I thought it was fine to put it
elsewhere for the moment :)

-- 
Nicolas Roard
"I love deadlines. I like the whooshing sound
they make as they fly by." -- Douglas Adams

From asqueaker at gmail.com  Wed Sep  9 21:39:22 2009
From: asqueaker at gmail.com (Chris Muller)
Date: Wed, 9 Sep 2009 14:39:22 -0500
Subject: Starting with an appropriate PRKernel
Message-ID: <b73f92970909091239u56995c6dle949575be581b723@mail.gmail.com>

Hi, I would like to explore Pier from my own development image
(instead of the one-click image).  I used Seaside's Configuration UI
to add a PRPierConfiguration and created an instance of PRKernel with:

  PRKernel named: 'Learning Pier'

But this PRKernel, even the environment page, _looks_ quite different
than the one in the one-click image.  I wish to proceed with a smart,
correct choice of PRKernel instance.  Are the differences between
these Kernels only cosmetic or does each provide more-suitability for
a particular purpose than the others?

Thanks,
  Chris

From obi068 at gmail.com  Wed Sep  9 21:51:00 2009
From: obi068 at gmail.com (Gerhard Obermann)
Date: Wed, 9 Sep 2009 21:51:00 +0200
Subject: Starting with an appropriate PRKernel
In-Reply-To: <b73f92970909091239u56995c6dle949575be581b723@mail.gmail.com>
References: <b73f92970909091239u56995c6dle949575be581b723@mail.gmail.com>
Message-ID: <7b7d752d0909091251i5d8d948cidb340ee6aa6bc932@mail.gmail.com>

PRDistribution new kernel

Gerhard

On Wed, Sep 9, 2009 at 9:39 PM, Chris Muller <asqueaker at gmail.com> wrote:

> Hi, I would like to explore Pier from my own development image
> (instead of the one-click image).  I used Seaside's Configuration UI
> to add a PRPierConfiguration and created an instance of PRKernel with:
>
>  PRKernel named: 'Learning Pier'
>
> But this PRKernel, even the environment page, _looks_ quite different
> than the one in the one-click image.  I wish to proceed with a smart,
> correct choice of PRKernel instance.  Are the differences between
> these Kernels only cosmetic or does each provide more-suitability for
> a particular purpose than the others?
>
> Thanks,
>  Chris
> _______________________________________________
> Magritte, Pier and Related Tools ...
> https://www.iam.unibe.ch/mailman/listinfo/smallwiki
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.iam.unibe.ch/pipermail/smallwiki/attachments/20090909/d04fba66/attachment.html>

From johnmci at smalltalkconsulting.com  Wed Sep  9 22:02:23 2009
From: johnmci at smalltalkconsulting.com (John M McIntosh)
Date: Wed, 9 Sep 2009 21:02:23 +0100
Subject: Starting with an appropriate PRKernel
In-Reply-To: <b73f92970909091239u56995c6dle949575be581b723@mail.gmail.com>
References: <b73f92970909091239u56995c6dle949575be581b723@mail.gmail.com>
Message-ID: <0B6CE1C4-EF10-4F3F-A5CE-98F932D5EEFD@smalltalkconsulting.com>

Chris I've cobbled together a PRKernel setup for WIkiServer. You can  
actually see a copy
by grabbing the one-click or mac/windows/linux installer from

http://mobilewikiserver.com/1.4.html

Sadly I don't have a method to completely recreate it from scratch


On 9-Sep-09, at 8:39 PM, Chris Muller wrote:

> Hi, I would like to explore Pier from my own development image
> (instead of the one-click image).  I used Seaside's Configuration UI
> to add a PRPierConfiguration and created an instance of PRKernel with:
>
>  PRKernel named: 'Learning Pier'
>
> But this PRKernel, even the environment page, _looks_ quite different
> than the one in the one-click image.  I wish to proceed with a smart,
> correct choice of PRKernel instance.  Are the differences between
> these Kernels only cosmetic or does each provide more-suitability for
> a particular purpose than the others?
>
> Thanks,
>  Chris
> _______________________________________________
> Magritte, Pier and Related Tools ...
> https://www.iam.unibe.ch/mailman/listinfo/smallwiki

--
= 
= 
= 
========================================================================
John M. McIntosh <johnmci at smalltalkconsulting.com>   Twitter:   
squeaker68882
Corporate Smalltalk Consulting Ltd.  http://www.smalltalkconsulting.com
= 
= 
= 
========================================================================





From asqueaker at gmail.com  Wed Sep  9 22:50:48 2009
From: asqueaker at gmail.com (Chris Muller)
Date: Wed, 9 Sep 2009 15:50:48 -0500
Subject: Support for standard Squeak (was: Starting with an appropriate 
	PRKernel)
Message-ID: <b73f92970909091350p3f11520br949a110eab80aafa@mail.gmail.com>

> On Wed, Sep 9, 2009 at 2:51 PM, Gerhard Obermann <obi068 at gmail.com> wrote:
> PRDistribution new kernel

To get Pier into my standard Squeak 3.9 image, I originally loaded
Pier from SqueakMap.  However, I can see that this does not obtain
quite a few of the packages available from the Pier squeaksource
repository that are present in the one-click image.

Is there a proper procedure for to loading-up a standard Squeak image
with all of the Pier goodies?

To get PRDistribution I loaded "Pier-Setup" from Lukas' repository.
Doing this exposed PRLoader>>#load loading a bunch of other stuff.
However, I couldn't bring that in because

    ScriptLoader loadLatestPackage: 'Gofer' from: ...

is not part of my ScriptLoader.  The ScriptLoader version in the
one-click image is from the Pharo repository.  Which begs two
important questions for those of us proceeding forward with standard
Squeak images (instead of Pharo images).

1) What is the best way to get a full, proper Pier installed in 3.9 or 3.10?

and.. 2) What are the intentions of the Pier developers, going
forward, for support of the standard Squeak images?

Despite my concern about the Pharo image and its splintering effects
on our community, as the maintainer of Magma, I can say I am
"committed" to maintaining Magma's compatibility with Pharo, even
though I intend to proceed forward with standard Squeak at this time.

I realize you cannot make any guarantee's any more than I can, but may
I at least know what the *intentions* are (as I just have)?  Lukas, it
looks like you are moving forward with Pharo.  But is standard Squeak
compatibility a factor for you at all in any of your decisions
regarding Pier going forward?

Thanks..

From miguel.coba at gmail.com  Wed Sep  9 23:22:27 2009
From: miguel.coba at gmail.com (Miguel Enrique =?ISO-8859-1?Q?Cob=E1?= Martinez)
Date: Wed, 09 Sep 2009 16:22:27 -0500
Subject: Support for standard Squeak (was: Starting with an appropriate
	PRKernel)
In-Reply-To: <b73f92970909091350p3f11520br949a110eab80aafa@mail.gmail.com>
References: <b73f92970909091350p3f11520br949a110eab80aafa@mail.gmail.com>
Message-ID: <1252531348.4299.7.camel@laptop.localdomain>

El mi?, 09-09-2009 a las 15:50 -0500, Chris Muller escribi?:
> > On Wed, Sep 9, 2009 at 2:51 PM, Gerhard Obermann <obi068 at gmail.com> wrote:
> > PRDistribution new kernel
> 
> To get Pier into my standard Squeak 3.9 image, I originally loaded
> Pier from SqueakMap.  However, I can see that this does not obtain
> quite a few of the packages available from the Pier squeaksource
> repository that are present in the one-click image.
> 
> Is there a proper procedure for to loading-up a standard Squeak image
> with all of the Pier goodies?
> 
> To get PRDistribution I loaded "Pier-Setup" from Lukas' repository.
> Doing this exposed PRLoader>>#load loading a bunch of other stuff.
> However, I couldn't bring that in because
> 
>     ScriptLoader loadLatestPackage: 'Gofer' from: ...
> 
> is not part of my ScriptLoader.  The ScriptLoader version in the
> one-click image is from the Pharo repository.  Which begs two
> important questions for those of us proceeding forward with standard
> Squeak images (instead of Pharo images).

Hi Chris,

this is unofficially of course as I am not the mantainer of the Pier
(Lukas or Tudor has better answers) but as I see thing, the Seaside and
Pier packages are going to be developed and distributed on Pharo images,
because the developers are most of them very involved in Pharo. So maybe
the best way is to get a Pharo image (either the PharoCore or the Pharo
images prepared by Damien Cassou from
http://pharo-project.org/pharo-download) and use them for your goals.
Also, to load the packages needed, the best way is to look in the
scripts created and used by Damien to build the dev and web pharo
images, they are distributed with the zips of the dev/web images and
adapt them.

> 
> 1) What is the best way to get a full, proper Pier installed in 3.9 or 3.10?
> 
> and.. 2) What are the intentions of the Pier developers, going
> forward, for support of the standard Squeak images?
> 
> Despite my concern about the Pharo image and its splintering effects
> on our community, as the maintainer of Magma, I can say I am
> "committed" to maintaining Magma's compatibility with Pharo, even
> though I intend to proceed forward with standard Squeak at this time.
> 
> I realize you cannot make any guarantee's any more than I can, but may
> I at least know what the *intentions* are (as I just have)?  Lukas, it
> looks like you are moving forward with Pharo.  But is standard Squeak
> compatibility a factor for you at all in any of your decisions
> regarding Pier going forward?
With time, and as the Pharo reaches 1.0 even more, the squeak
compatibility effort will fade and the packages that now load in squeak
will be made to load in Pharo but not the other way.


> 
> Thanks..
> _______________________________________________
> Magritte, Pier and Related Tools ...
> https://www.iam.unibe.ch/mailman/listinfo/smallwiki
-- 
Miguel Cob?
http://miguel.leugim.com.mx


From renggli at gmail.com  Thu Sep 10 00:11:29 2009
From: renggli at gmail.com (Lukas Renggli)
Date: Thu, 10 Sep 2009 00:11:29 +0200
Subject: Support for standard Squeak (was: Starting with an appropriate 
	PRKernel)
In-Reply-To: <b73f92970909091350p3f11520br949a110eab80aafa@mail.gmail.com>
References: <b73f92970909091350p3f11520br949a110eab80aafa@mail.gmail.com>
Message-ID: <67628d690909091511s3bb61e73vf6d5f095b533f17e@mail.gmail.com>

> 1) What is the best way to get a full, proper Pier installed in 3.9 or 3.10?

It is actually quite simple to load manually. You load Seaside (and
the required dependencies), then Magritte-Core, Magritte-Seaside,
Pier-Core, Pier-Seaside, and the plugins and Pier-All.

> and.. 2) What are the intentions of the Pier developers, going
> forward, for support of the standard Squeak images?

Magritte and Pier probably work in both Pharo and Squeak, in fact the
code runs almost identical on GemStone and VisualWorks. I haven't
tried in Squeak in the past, but I guess it should work there too. I
am certainly willing to include patches if they improve portability,
but I will am not actively maintaining a port to Squeak.

> I realize you cannot make any guarantee's any more than I can, but may
> I at least know what the *intentions* are (as I just have)?  Lukas, it
> looks like you are moving forward with Pharo.  But is standard Squeak
> compatibility a factor for you at all in any of your decisions
> regarding Pier going forward?

The situation is basically the same as for Seaside. Somebody needs to
step up and actively maintain a port, regularly run the tests and
report back problems. For me Squeak is just a platform like VW,
GemStone, GST, VAST, ... for now I chose Pharo as my development
platform of choice.

Cheers,
Lukas

-- 
Lukas Renggli
http://www.lukas-renggli.ch

From ross at biostat.ucsf.edu  Thu Sep 10 00:18:05 2009
From: ross at biostat.ucsf.edu (Ross Boylan)
Date: Wed, 09 Sep 2009 15:18:05 -0700
Subject: Support for standard Squeak (was: Starting with an appropriate
	PRKernel)
In-Reply-To: <67628d690909091511s3bb61e73vf6d5f095b533f17e@mail.gmail.com>
References: <b73f92970909091350p3f11520br949a110eab80aafa@mail.gmail.com>
	<67628d690909091511s3bb61e73vf6d5f095b533f17e@mail.gmail.com>
Message-ID: <1252534685.31590.87.camel@corn.betterworld.us>

On Thu, 2009-09-10 at 00:11 +0200, Lukas Renggli wrote:
> > 1) What is the best way to get a full, proper Pier installed in 3.9
> or 3.10?
> 
> It is actually quite simple to load manually. You load Seaside (and
> the required dependencies), then Magritte-Core, Magritte-Seaside,
> Pier-Core, Pier-Seaside, and the plugins and Pier-All.
Do they need a closure-enabled VM?  Will they eventually?

Ross Boylan



From renggli at gmail.com  Thu Sep 10 00:29:57 2009
From: renggli at gmail.com (Lukas Renggli)
Date: Thu, 10 Sep 2009 00:29:57 +0200
Subject: Support for standard Squeak (was: Starting with an appropriate 
	PRKernel)
In-Reply-To: <1252534685.31590.87.camel@corn.betterworld.us>
References: <b73f92970909091350p3f11520br949a110eab80aafa@mail.gmail.com>
	<67628d690909091511s3bb61e73vf6d5f095b533f17e@mail.gmail.com>
	<1252534685.31590.87.camel@corn.betterworld.us>
Message-ID: <67628d690909091529p4b2123f0s3298389bb973aa14@mail.gmail.com>

>> It is actually quite simple to load manually. You load Seaside (and
>> the required dependencies), then Magritte-Core, Magritte-Seaside,
>> Pier-Core, Pier-Seaside, and the plugins and Pier-All.
>
> Do they need a closure-enabled VM? ?Will they eventually?

No, it worked in images without closures for years. I don't see a
reason why that should be a problem.

Lukas

-- 
Lukas Renggli
http://www.lukas-renggli.ch


From asqueaker at gmail.com  Thu Sep 10 01:08:50 2009
From: asqueaker at gmail.com (Chris Muller)
Date: Wed, 9 Sep 2009 18:08:50 -0500
Subject: Starting with an appropriate PRKernel
In-Reply-To: <0B6CE1C4-EF10-4F3F-A5CE-98F932D5EEFD@smalltalkconsulting.com>
References: <b73f92970909091239u56995c6dle949575be581b723@mail.gmail.com>
	<0B6CE1C4-EF10-4F3F-A5CE-98F932D5EEFD@smalltalkconsulting.com>
Message-ID: <b73f92970909091608y42ed67c6s4d127ac496690ac0@mail.gmail.com>

Wow, seriously cool John.  Too bad it is restricted to wi-fi..

On Wed, Sep 9, 2009 at 3:02 PM, John M McIntosh
<johnmci at smalltalkconsulting.com> wrote:
> Chris I've cobbled together a PRKernel setup for WIkiServer. You can
> actually see a copy
> by grabbing the one-click or mac/windows/linux installer from
>
> http://mobilewikiserver.com/1.4.html
>
> Sadly I don't have a method to completely recreate it from scratch
>

From renggli at gmail.com  Thu Sep 10 14:06:37 2009
From: renggli at gmail.com (Lukas Renggli)
Date: Thu, 10 Sep 2009 14:06:37 +0200
Subject: [ANN] [Book] Dynamic Web Development with Seaside
Message-ID: <67628d690909100506w62d823f3gc01f1216447738fc@mail.gmail.com>

After the first announcement at ESUG 2009, we are proud to present the
release of the free online book
		
        Dynamic Web Development with Seaside
        http://book.seaside.st/

The book is written in the Pier content management system using the
Pier book publishing engine. This allows us to collaboratively edit
the contents and generate different output formats automatically. We
will soon offer a PDF and a printed version, but first we have to iron
out some of the remaining issues.

Over the past years the book got reviewed and revised several times.
We want to thank all the persons who helped us: Torsten Bergmann,
Damien Cassou, Tom Krisch, Philippe Marshall, Ruben Schempp, Roger
Whitney, Julian Fitzell, and Michael Davies carefully reviewed the
book and provided valuable feedback. Martin J. Laubach for his Sudoku
code. Ramon Leon for letting us using his blog ideas and SandStoneDB,
Chris Muller for Magma. Jeff Dorst provided generous financial support
for supporting student text reading. Markus Gaelli for brainstorming
on the book title. Samuel Morello for designing the cover.

Michael Davies is currently helping us getting things ready for print.
And you can help us too by writing down your findings at the bottom of
each page. Also, if you want to contribute a missing chapter or
support the translation of the book please get in touch with us.

We wish to thank the European Smalltalk User Group (ESUG) and
inceptive.be for sponsoring this book. We are looking for additional
sponsors. If you are interested, please contact us. If you are a
publisher and interested in publishing this material, please let us
know.

Please distribute this message widely.

Cheers,
Stef & Lukas & David & Rick


About the authors:
- St?phane Ducasse is the author of a couple of books on Smalltalk.
- Lukas Renggli is core developer of Seaside and Smalltalk consultant.
- David C. Shaffer is Seaside consultant and CS teacher.
- Rick Zaccone is CS teacher.


From girba at iam.unibe.ch  Thu Sep 10 14:12:31 2009
From: girba at iam.unibe.ch (Tudor Girba)
Date: Thu, 10 Sep 2009 14:12:31 +0200
Subject: [Pharo-project] [ANN] [Book] Dynamic Web Development with Seaside
In-Reply-To: <67628d690909100506w62d823f3gc01f1216447738fc@mail.gmail.com>
References: <67628d690909100506w62d823f3gc01f1216447738fc@mail.gmail.com>
Message-ID: <D8AB7AA2-7B31-455F-9387-FD9D17E0031A@iam.unibe.ch>

Congratulations!

It really looks great.

Cheers,
Doru

On 10 Sep 2009, at 14:06, Lukas Renggli wrote:

> After the first announcement at ESUG 2009, we are proud to present the
> release of the free online book
> 		
>        Dynamic Web Development with Seaside
>        http://book.seaside.st/
>
> The book is written in the Pier content management system using the
> Pier book publishing engine. This allows us to collaboratively edit
> the contents and generate different output formats automatically. We
> will soon offer a PDF and a printed version, but first we have to iron
> out some of the remaining issues.
>
> Over the past years the book got reviewed and revised several times.
> We want to thank all the persons who helped us: Torsten Bergmann,
> Damien Cassou, Tom Krisch, Philippe Marshall, Ruben Schempp, Roger
> Whitney, Julian Fitzell, and Michael Davies carefully reviewed the
> book and provided valuable feedback. Martin J. Laubach for his Sudoku
> code. Ramon Leon for letting us using his blog ideas and SandStoneDB,
> Chris Muller for Magma. Jeff Dorst provided generous financial support
> for supporting student text reading. Markus Gaelli for brainstorming
> on the book title. Samuel Morello for designing the cover.
>
> Michael Davies is currently helping us getting things ready for print.
> And you can help us too by writing down your findings at the bottom of
> each page. Also, if you want to contribute a missing chapter or
> support the translation of the book please get in touch with us.
>
> We wish to thank the European Smalltalk User Group (ESUG) and
> inceptive.be for sponsoring this book. We are looking for additional
> sponsors. If you are interested, please contact us. If you are a
> publisher and interested in publishing this material, please let us
> know.
>
> Please distribute this message widely.
>
> Cheers,
> Stef & Lukas & David & Rick
>
>
> About the authors:
> - St?phane Ducasse is the author of a couple of books on Smalltalk.
> - Lukas Renggli is core developer of Seaside and Smalltalk consultant.
> - David C. Shaffer is Seaside consultant and CS teacher.
> - Rick Zaccone is CS teacher.
>
> _______________________________________________
> Pharo-project mailing list
> Pharo-project at lists.gforge.inria.fr
> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
>

--
www.tudorgirba.com

"Beauty is where we see it."





From philippe.marschall at gmail.com  Sat Sep 12 21:57:19 2009
From: philippe.marschall at gmail.com (Philippe Marschall)
Date: Sat, 12 Sep 2009 21:57:19 +0200
Subject: mementos and kind validation on containers
Message-ID: <66666f210909121257t2fcefe1dk81414d248395dec7@mail.gmail.com>

Hi

Recently #kind was made mutable on MADescription. I think this is
great and set it on the containers of my model classes to my model
classes. This allows me to do the following:

aDescription kind new

which is really cool. The problem is that now I always get kind
validation errors. The cause is the following code in MADescription:

validateKind: anObject
	"Validate ==anObject== to be of the right kind."

	(anObject isKindOf: self kind)
		ifFalse: [ MAKindError description: self signal: self kindErrorMessage ]

anObject would be an MACheckedMemento. self kind would previously
always answer Object so this test would always pass. Now however self
kind answers my model class and always fails.

Any ideas?

Cheers
Philippe

From renggli at gmail.com  Sat Sep 12 22:54:39 2009
From: renggli at gmail.com (Lukas Renggli)
Date: Sat, 12 Sep 2009 22:54:39 +0200
Subject: mementos and kind validation on containers
In-Reply-To: <66666f210909121257t2fcefe1dk81414d248395dec7@mail.gmail.com>
References: <66666f210909121257t2fcefe1dk81414d248395dec7@mail.gmail.com>
Message-ID: <67628d690909121354w7b47f701k85371a36a3b5cf8c@mail.gmail.com>

> Any ideas?

I think you should not use #kind to instantiate new objects, but add
your own property to hold the class you want to instantiate.

Cheers,
Lukas

-- 
Lukas Renggli
http://www.lukas-renggli.ch

From philippe.marschall at gmail.com  Sun Sep 13 10:29:01 2009
From: philippe.marschall at gmail.com (Philippe Marschall)
Date: Sun, 13 Sep 2009 10:29:01 +0200
Subject: mementos and kind validation on containers
In-Reply-To: <67628d690909121354w7b47f701k85371a36a3b5cf8c@mail.gmail.com>
References: <66666f210909121257t2fcefe1dk81414d248395dec7@mail.gmail.com>
	<67628d690909121354w7b47f701k85371a36a3b5cf8c@mail.gmail.com>
Message-ID: <66666f210909130129t757c90d2i93777f091879e8bf@mail.gmail.com>

2009/9/12 Lukas Renggli <renggli at gmail.com>:
>> Any ideas?
>
> I think you should not use #kind to instantiate new objects, but add
> your own property to hold the class you want to instantiate.

That doesn't solve the problem that kind validation is borken for containers.

Cheers
Philippe

From renggli at gmail.com  Sun Sep 13 10:35:28 2009
From: renggli at gmail.com (Lukas Renggli)
Date: Sun, 13 Sep 2009 10:35:28 +0200
Subject: mementos and kind validation on containers
In-Reply-To: <66666f210909130129t757c90d2i93777f091879e8bf@mail.gmail.com>
References: <66666f210909121257t2fcefe1dk81414d248395dec7@mail.gmail.com>
	<67628d690909121354w7b47f701k85371a36a3b5cf8c@mail.gmail.com>
	<66666f210909130129t757c90d2i93777f091879e8bf@mail.gmail.com>
Message-ID: <67628d690909130135u40692ad2s133db0feccfdf0c2@mail.gmail.com>

>> I think you should not use #kind to instantiate new objects, but add
>> your own property to hold the class you want to instantiate.
>
> That doesn't solve the problem that kind validation is borken for containers.

#kind is used for validation and that should remain like this. The
default for containers is Object, so it is a no-op that passes the
validation for all objects.

Lukas

-- 
Lukas Renggli
http://www.lukas-renggli.ch

From philippe.marschall at gmail.com  Sun Sep 13 10:58:11 2009
From: philippe.marschall at gmail.com (Philippe Marschall)
Date: Sun, 13 Sep 2009 10:58:11 +0200
Subject: mementos and kind validation on containers
In-Reply-To: <67628d690909130135u40692ad2s133db0feccfdf0c2@mail.gmail.com>
References: <66666f210909121257t2fcefe1dk81414d248395dec7@mail.gmail.com>
	<67628d690909121354w7b47f701k85371a36a3b5cf8c@mail.gmail.com>
	<66666f210909130129t757c90d2i93777f091879e8bf@mail.gmail.com>
	<67628d690909130135u40692ad2s133db0feccfdf0c2@mail.gmail.com>
Message-ID: <66666f210909130158r69ace11ei7ef4f2681a0b0d05@mail.gmail.com>

2009/9/13 Lukas Renggli <renggli at gmail.com>:
>>> I think you should not use #kind to instantiate new objects, but add
>>> your own property to hold the class you want to instantiate.
>>
>> That doesn't solve the problem that kind validation is borken for containers.
>
> #kind is used for validation and that should remain like this. The
> default for containers is Object, so it is a no-op that passes the
> validation for all objects.

Not if you set kind on a container. And the check is not a no-op at
all. It checks the class of the memento against the kind which makes
no sense at all.

Cheers
Philippe

From renggli at gmail.com  Sun Sep 13 11:50:49 2009
From: renggli at gmail.com (Lukas Renggli)
Date: Sun, 13 Sep 2009 11:50:49 +0200
Subject: mementos and kind validation on containers
In-Reply-To: <66666f210909130158r69ace11ei7ef4f2681a0b0d05@mail.gmail.com>
References: <66666f210909121257t2fcefe1dk81414d248395dec7@mail.gmail.com>
	<67628d690909121354w7b47f701k85371a36a3b5cf8c@mail.gmail.com>
	<66666f210909130129t757c90d2i93777f091879e8bf@mail.gmail.com>
	<67628d690909130135u40692ad2s133db0feccfdf0c2@mail.gmail.com>
	<66666f210909130158r69ace11ei7ef4f2681a0b0d05@mail.gmail.com>
Message-ID: <67628d690909130250l5fa83acq472177e2073f5e67@mail.gmail.com>

2009/9/13 Philippe Marschall <philippe.marschall at gmail.com>:
> 2009/9/13 Lukas Renggli <renggli at gmail.com>:
>>>> I think you should not use #kind to instantiate new objects, but add
>>>> your own property to hold the class you want to instantiate.
>>>
>>> That doesn't solve the problem that kind validation is borken for containers.
>>
>> #kind is used for validation and that should remain like this. The
>> default for containers is Object, so it is a no-op that passes the
>> validation for all objects.
>
> Not if you set kind on a container. And the check is not a no-op at
> all. It checks the class of the memento against the kind which makes
> no sense at all.

Of course, for what it is designed for it makes perfectly sense.

As I already said several times, forget about #kind, create your own
property that is not used for validation and do with it whatever you
want.

Lukas

-- 
Lukas Renggli
http://www.lukas-renggli.ch

From philippe.marschall at gmail.com  Sun Sep 13 12:02:21 2009
From: philippe.marschall at gmail.com (Philippe Marschall)
Date: Sun, 13 Sep 2009 12:02:21 +0200
Subject: mementos and kind validation on containers
In-Reply-To: <67628d690909130250l5fa83acq472177e2073f5e67@mail.gmail.com>
References: <66666f210909121257t2fcefe1dk81414d248395dec7@mail.gmail.com>
	<67628d690909121354w7b47f701k85371a36a3b5cf8c@mail.gmail.com>
	<66666f210909130129t757c90d2i93777f091879e8bf@mail.gmail.com>
	<67628d690909130135u40692ad2s133db0feccfdf0c2@mail.gmail.com>
	<66666f210909130158r69ace11ei7ef4f2681a0b0d05@mail.gmail.com>
	<67628d690909130250l5fa83acq472177e2073f5e67@mail.gmail.com>
Message-ID: <66666f210909130302u37857e8fxd1968f2643f8dc56@mail.gmail.com>

2009/9/13 Lukas Renggli <renggli at gmail.com>:
> 2009/9/13 Philippe Marschall <philippe.marschall at gmail.com>:
>> 2009/9/13 Lukas Renggli <renggli at gmail.com>:
>>>>> I think you should not use #kind to instantiate new objects, but add
>>>>> your own property to hold the class you want to instantiate.
>>>>
>>>> That doesn't solve the problem that kind validation is borken for containers.
>>>
>>> #kind is used for validation and that should remain like this. The
>>> default for containers is Object, so it is a no-op that passes the
>>> validation for all objects.
>>
>> Not if you set kind on a container. And the check is not a no-op at
>> all. It checks the class of the memento against the kind which makes
>> no sense at all.
>
> Of course, for what it is designed for it makes perfectly sense.
>
> As I already said several times, forget about #kind, create your own
> property that is not used for validation and do with it whatever you
> want.

But why is #kind mutable on MADescription when setting it results in
failing validation? This is very bad API design. At least it should be
pushed down to MAElement description.

Cheers
Philippe

From renggli at gmail.com  Sun Sep 13 12:14:11 2009
From: renggli at gmail.com (Lukas Renggli)
Date: Sun, 13 Sep 2009 12:14:11 +0200
Subject: mementos and kind validation on containers
In-Reply-To: <66666f210909130302u37857e8fxd1968f2643f8dc56@mail.gmail.com>
References: <66666f210909121257t2fcefe1dk81414d248395dec7@mail.gmail.com>
	<67628d690909121354w7b47f701k85371a36a3b5cf8c@mail.gmail.com>
	<66666f210909130129t757c90d2i93777f091879e8bf@mail.gmail.com>
	<67628d690909130135u40692ad2s133db0feccfdf0c2@mail.gmail.com>
	<66666f210909130158r69ace11ei7ef4f2681a0b0d05@mail.gmail.com>
	<67628d690909130250l5fa83acq472177e2073f5e67@mail.gmail.com>
	<66666f210909130302u37857e8fxd1968f2643f8dc56@mail.gmail.com>
Message-ID: <67628d690909130314k51a76bb4y8f04069dcb36cdb7@mail.gmail.com>

2009/9/13 Philippe Marschall <philippe.marschall at gmail.com>:
> 2009/9/13 Lukas Renggli <renggli at gmail.com>:
>> 2009/9/13 Philippe Marschall <philippe.marschall at gmail.com>:
>>> 2009/9/13 Lukas Renggli <renggli at gmail.com>:
>>>>>> I think you should not use #kind to instantiate new objects, but add
>>>>>> your own property to hold the class you want to instantiate.
>>>>>
>>>>> That doesn't solve the problem that kind validation is borken for containers.
>>>>
>>>> #kind is used for validation and that should remain like this. The
>>>> default for containers is Object, so it is a no-op that passes the
>>>> validation for all objects.
>>>
>>> Not if you set kind on a container. And the check is not a no-op at
>>> all. It checks the class of the memento against the kind which makes
>>> no sense at all.
>>
>> Of course, for what it is designed for it makes perfectly sense.
>>
>> As I already said several times, forget about #kind, create your own
>> property that is not used for validation and do with it whatever you
>> want.
>
> But why is #kind mutable on MADescription when setting it results in
> failing validation? This is very bad API design. At least it should be
> pushed down to MAElement description.

Indeed, you are right, but that has nothing to do with the initial problem.

Maybe the MAMemento should rather implement #isKindOf: to dispatch to
the delegate?

Lukas

-- 
Lukas Renggli
http://www.lukas-renggli.ch

From philippe.marschall at gmail.com  Sun Sep 13 12:24:40 2009
From: philippe.marschall at gmail.com (Philippe Marschall)
Date: Sun, 13 Sep 2009 12:24:40 +0200
Subject: mementos and kind validation on containers
In-Reply-To: <67628d690909130314k51a76bb4y8f04069dcb36cdb7@mail.gmail.com>
References: <66666f210909121257t2fcefe1dk81414d248395dec7@mail.gmail.com>
	<67628d690909121354w7b47f701k85371a36a3b5cf8c@mail.gmail.com>
	<66666f210909130129t757c90d2i93777f091879e8bf@mail.gmail.com>
	<67628d690909130135u40692ad2s133db0feccfdf0c2@mail.gmail.com>
	<66666f210909130158r69ace11ei7ef4f2681a0b0d05@mail.gmail.com>
	<67628d690909130250l5fa83acq472177e2073f5e67@mail.gmail.com>
	<66666f210909130302u37857e8fxd1968f2643f8dc56@mail.gmail.com>
	<67628d690909130314k51a76bb4y8f04069dcb36cdb7@mail.gmail.com>
Message-ID: <66666f210909130324p6cbf46b6n56311eb5f2842980@mail.gmail.com>

2009/9/13 Lukas Renggli <renggli at gmail.com>:
> 2009/9/13 Philippe Marschall <philippe.marschall at gmail.com>:
>> 2009/9/13 Lukas Renggli <renggli at gmail.com>:
>>> 2009/9/13 Philippe Marschall <philippe.marschall at gmail.com>:
>>>> 2009/9/13 Lukas Renggli <renggli at gmail.com>:
>>>>>>> I think you should not use #kind to instantiate new objects, but add
>>>>>>> your own property to hold the class you want to instantiate.
>>>>>>
>>>>>> That doesn't solve the problem that kind validation is borken for containers.
>>>>>
>>>>> #kind is used for validation and that should remain like this. The
>>>>> default for containers is Object, so it is a no-op that passes the
>>>>> validation for all objects.
>>>>
>>>> Not if you set kind on a container. And the check is not a no-op at
>>>> all. It checks the class of the memento against the kind which makes
>>>> no sense at all.
>>>
>>> Of course, for what it is designed for it makes perfectly sense.
>>>
>>> As I already said several times, forget about #kind, create your own
>>> property that is not used for validation and do with it whatever you
>>> want.
>>
>> But why is #kind mutable on MADescription when setting it results in
>> failing validation? This is very bad API design. At least it should be
>> pushed down to MAElement description.
>
> Indeed, you are right, but that has nothing to do with the initial problem.
>
> Maybe the MAMemento should rather implement #isKindOf: to dispatch to
> the delegate?

I though about this first too. However now I'm affraid of the possible
things it would break (debuggers, inspectors, ...). Considering that
something as simple as implementing #asString in Object can crash
VisualWorks this makes me very nervous. Maybe specifically handling
mementos or #magritteIsKindOf: are better.

Cheers
Philippe

From damien.cassou at gmail.com  Tue Sep 15 20:53:08 2009
From: damien.cassou at gmail.com (Damien Cassou)
Date: Tue, 15 Sep 2009 20:53:08 +0200
Subject: Fwd: [Pharo-project] Problem with pharo1.0-10440-BETAweb09.09.1
In-Reply-To: <87755A41-4676-4BC6-AB23-E29CEED3DA32@gmx.de>
References: <87755A41-4676-4BC6-AB23-E29CEED3DA32@gmx.de>
Message-ID: <6ac749c10909151153n58261580qb0b4b37010dbd197@mail.gmail.com>

This seems to be a problem with Pier does not work with the example
browser of Seaside.


---------- Forwarded message ----------
From: Andreas Wacknitz <A.Wacknitz at gmx.de>
Date: 2009/9/15
Subject: [Pharo-project] Problem with pharo1.0-10440-BETAweb09.09.1
To: pharo-project at lists.gforge.inria.fr


I have started to look at Seaside in Pharo. There seems to be a
problem with the examples.
When I enter?http://localhost:8080/seaside/examples/examplebrowser (by
clicking on examples and then examplebrowser)
I will get a Seaside Walkback:

Seaside Walkback

MessageNotUnderstood: receiver of "root" is nil

Debug?Proceed?Full Stack

Possible Causes

the receiver of the message is nil
a class extension hasn't been loaded correctly
you sent the wrong message

Stack Trace

thisContextUndefinedObject(Object)>>doesNotUnderstand:
#rootselfnilaMessagerootexceptionMessageNotUnderstood: receiver of
"root" is nilresumeValuenil
thisContextPRContext class>>kernel:selfPRContextaKernelnil
thisContextPRPierFrame>>initialRequest:selfa PRPierFrameaRequesta
WARequest GET /seaside/examples/examplebrowser
thisContext[] in WARenderLoopMain>>start:selfa
WARenderLoopMainaRequesta WARequest GET
/seaside/examples/examplebrowserroota WAExampleBrowsereacha
PRPierFrame
thisContextPRPierFrame(WAPresenter)>>withNextPresentersDo:selfa
PRPierFrameaBlock[closure] in WARenderLoopMain>>start:

Regards,
Andreas

_______________________________________________
Pharo-project mailing list
Pharo-project at lists.gforge.inria.fr
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project



-- 
Damien Cassou
http://damiencassou.seasidehosting.st

"Lambdas are relegated to relative obscurity until Java makes them
popular by not having them." James Iry


From joel at ardishealth.com  Wed Sep 16 16:06:24 2009
From: joel at ardishealth.com (Joel Turnbull)
Date: Wed, 16 Sep 2009 10:06:24 -0400
Subject: description method cached?
Message-ID: <26ee8b3e0909160706w54b51763x897e8aab4b00c503@mail.gmail.com>

Doing a seaside app using magritte, within pier, within the glass appliance

I created a subclass of MATextInputComponent to do some specific validation,
call it FooInputComponent,

I changed the old description method to send componentClass: like this

descriptionFoo
    ^MAStringDescription new
    componentClass: FooInputComponent;
    selectorAccessor: #foo;
    label: 'Foo';
    priority: 40;
    yourself

However, the change doesn't seem to apply, the container still gets a
MATextInputComponent. I placed a halt in descriptionFoo and it never
triggers.

Is this a caching issue that I can reset?

Thanks
Joel
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.iam.unibe.ch/pipermail/smallwiki/attachments/20090916/75bacd69/attachment.html>

From joel at ardishealth.com  Wed Sep 16 16:16:35 2009
From: joel at ardishealth.com (Joel Turnbull)
Date: Wed, 16 Sep 2009 10:16:35 -0400
Subject: description method cached?
In-Reply-To: <26ee8b3e0909160706w54b51763x897e8aab4b00c503@mail.gmail.com>
References: <26ee8b3e0909160706w54b51763x897e8aab4b00c503@mail.gmail.com>
Message-ID: <26ee8b3e0909160716u687c8a17x55fefcc66cca83d1@mail.gmail.com>

Just found the answer on Dale's blog
http://gemstonesoup.wordpress.com/category/magritte/

MADescriptionBuilder default flush

does the trick


On Wed, Sep 16, 2009 at 10:06 AM, Joel Turnbull <joel at ardishealth.com>wrote:

> Doing a seaside app using magritte, within pier, within the glass appliance
>
> I created a subclass of MATextInputComponent to do some specific
> validation, call it FooInputComponent,
>
> I changed the old description method to send componentClass: like this
>
> descriptionFoo
>     ^MAStringDescription new
>     componentClass: FooInputComponent;
>     selectorAccessor: #foo;
>     label: 'Foo';
>     priority: 40;
>     yourself
>
> However, the change doesn't seem to apply, the container still gets a
> MATextInputComponent. I placed a halt in descriptionFoo and it never
> triggers.
>
> Is this a caching issue that I can reset?
>
> Thanks
> Joel
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.iam.unibe.ch/pipermail/smallwiki/attachments/20090916/fe4751ef/attachment.html>

From joel at ardishealth.com  Wed Sep 16 16:34:53 2009
From: joel at ardishealth.com (Joel Turnbull)
Date: Wed, 16 Sep 2009 10:34:53 -0400
Subject: Pier Users Change Own Passwords
Message-ID: <26ee8b3e0909160734x792ad5exf7bda386a57bb1b8@mail.gmail.com>

Is there a Pier Component that allows non-admin type users to change their
own passwords?

Thanks
Joel
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.iam.unibe.ch/pipermail/smallwiki/attachments/20090916/10156c1c/attachment.html>

From girba at iam.unibe.ch  Wed Sep 16 16:47:50 2009
From: girba at iam.unibe.ch (Tudor Girba)
Date: Wed, 16 Sep 2009 16:47:50 +0200
Subject: Pier Users Change Own Passwords
In-Reply-To: <26ee8b3e0909160734x792ad5exf7bda386a57bb1b8@mail.gmail.com>
References: <26ee8b3e0909160734x792ad5exf7bda386a57bb1b8@mail.gmail.com>
Message-ID: <122E23D1-D6CF-463E-B33E-579951952D23@iam.unibe.ch>

Hi Joel,

In the default configuration, once the user gets logged in, there is a  
link at the bottom with which you can change the password.

Cheers,
Doru


On 16 Sep 2009, at 16:34, Joel Turnbull wrote:

>
> Is there a Pier Component that allows non-admin type users to change  
> their own passwords?
>
> Thanks
> Joel
> _______________________________________________
> Magritte, Pier and Related Tools ...
> https://www.iam.unibe.ch/mailman/listinfo/smallwiki

--
www.tudorgirba.com

"No matter how many recipes we know, we still value a chef."







From razavi at acm.org  Wed Sep 16 17:12:21 2009
From: razavi at acm.org (Reza RAZAVI)
Date: Wed, 16 Sep 2009 17:12:21 +0200
Subject: Pier Users Change Own Passwords
In-Reply-To: <26ee8b3e0909160734x792ad5exf7bda386a57bb1b8@mail.gmail.com
 >
References: <26ee8b3e0909160734x792ad5exf7bda386a57bb1b8@mail.gmail.com>
Message-ID: <200909161512.n8GFCK5R005951@smtpout1.pt.lu>

At 16:34 16/09/2009, Joel Turnbull wrote:
>Is there a Pier Component that allows non-admin type users to change 
>their own passwords?

Joel,

The attached "st" file implements "Edit User": a simple Pier Command 
for editing the attributes of the current user, including its password.

Its mainly a refactoring of code found in PUSecurityWidget. I, 
however, don't know why it has been preferred to implement "Edit 
User" as part of PUSecurityWidget, rather than as a Command. So, 
maybe some cauation would be necessary when using this command.

Hoping this helps,
Reza 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: AAsEditUserCommand.st
Type: application/octet-stream
Size: 1387 bytes
Desc: not available
URL: <http://www.iam.unibe.ch/pipermail/smallwiki/attachments/20090916/ede562ec/attachment.obj>

From renggli at gmail.com  Wed Sep 16 18:02:03 2009
From: renggli at gmail.com (Lukas Renggli)
Date: Wed, 16 Sep 2009 18:02:03 +0200
Subject: description method cached?
In-Reply-To: <26ee8b3e0909160716u687c8a17x55fefcc66cca83d1@mail.gmail.com>
References: <26ee8b3e0909160706w54b51763x897e8aab4b00c503@mail.gmail.com>
	<26ee8b3e0909160716u687c8a17x55fefcc66cca83d1@mail.gmail.com>
Message-ID: <67628d690909160902y3cc9a0d2gb0910163d7a4fd86@mail.gmail.com>

> Just found the answer on Dale's blog
> http://gemstonesoup.wordpress.com/category/magritte/
>
> MADescriptionBuilder default flush
>
> does the trick

Yeah, GemStone does not automatically flush the cache when the code changes.

You might want to implement a description builder that does not cache
at all. Caching makes meta-description heavy apps considerably faster,
but it also introduces all kinds of new problems (e.g. when you change
descriptions on the fly).

Cheers,
Lukas

-- 
Lukas Renggli
http://www.lukas-renggli.ch

From joel at ardishealth.com  Wed Sep 16 18:15:12 2009
From: joel at ardishealth.com (Joel Turnbull)
Date: Wed, 16 Sep 2009 12:15:12 -0400
Subject: Pier Users Change Own Passwords
In-Reply-To: <200909161512.n8GFCK5R005951@smtpout1.pt.lu>
References: <26ee8b3e0909160734x792ad5exf7bda386a57bb1b8@mail.gmail.com> 
	<200909161512.n8GFCK5R005951@smtpout1.pt.lu>
Message-ID: <26ee8b3e0909160915u1be2ca7u938b2536dc1505d@mail.gmail.com>

I don't see that, Doru, but my Gemstone Pier is not up to date

Thanks Reza I will try installing that command.

Joel



On Wed, Sep 16, 2009 at 11:12 AM, Reza RAZAVI <razavi at acm.org> wrote:

> At 16:34 16/09/2009, Joel Turnbull wrote:
>
>> Is there a Pier Component that allows non-admin type users to change their
>> own passwords?
>>
>
> Joel,
>
> The attached "st" file implements "Edit User": a simple Pier Command for
> editing the attributes of the current user, including its password.
>
> Its mainly a refactoring of code found in PUSecurityWidget. I, however,
> don't know why it has been preferred to implement "Edit User" as part of
> PUSecurityWidget, rather than as a Command. So, maybe some cauation would be
> necessary when using this command.
>
> Hoping this helps,
> Reza
> _______________________________________________
> Magritte, Pier and Related Tools ...
> https://www.iam.unibe.ch/mailman/listinfo/smallwiki
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.iam.unibe.ch/pipermail/smallwiki/attachments/20090916/6d8c444b/attachment.html>

From ross at biostat.ucsf.edu  Fri Sep 18 21:56:50 2009
From: ross at biostat.ucsf.edu (Ross Boylan)
Date: Fri, 18 Sep 2009 12:56:50 -0700
Subject: Can't resolve names
Message-ID: <1253303810.22848.96.camel@corn.betterworld.us>

I downloaded and unzipped Pier-1.2.app.zip onto an amd64 Linux system.
./pier.sh initially failed, but when I installed ia32 compatibility libs
it launched.

However, it is unable to resolve names, or at least squeaksource.com;
I've stepped through; the primitive resolving names seems to return
nothing.  I am able to resolve the name on the host system.

I'm suspicious this is a 64 bit compatibity issue.

Any ideas?

Thanks.
Ross Boylan


From dale.henrichs at gemstone.com  Fri Sep 18 22:09:26 2009
From: dale.henrichs at gemstone.com (Dale Henrichs)
Date: Fri, 18 Sep 2009 13:09:26 -0700 (PDT)
Subject: Can't resolve names
In-Reply-To: <1253303810.22848.96.camel@corn.betterworld.us>
Message-ID: <2065614709.3599941253304566271.JavaMail.root@gir>

Ross,

I have seen the same problem with Pharo under 64 bit Ubuntu (64 bit OpenSUSE seems fine)...I work around the problem by overriding NetNameResolver class>>useOldNetwork to unconditionally return true at least it works for me:)

Dale

----- "Ross Boylan" <ross at biostat.ucsf.edu> wrote:

| I downloaded and unzipped Pier-1.2.app.zip onto an amd64 Linux
| system.
| ./pier.sh initially failed, but when I installed ia32 compatibility
| libs
| it launched.
| 
| However, it is unable to resolve names, or at least squeaksource.com;
| I've stepped through; the primitive resolving names seems to return
| nothing.  I am able to resolve the name on the host system.
| 
| I'm suspicious this is a 64 bit compatibity issue.
| 
| Any ideas?
| 
| Thanks.
| Ross Boylan
| 
| _______________________________________________
| Magritte, Pier and Related Tools ...
| https://www.iam.unibe.ch/mailman/listinfo/smallwiki

From marianopeck at gmail.com  Fri Sep 18 22:11:55 2009
From: marianopeck at gmail.com (Mariano Martinez Peck)
Date: Fri, 18 Sep 2009 19:11:55 -0100
Subject: Can't resolve names
In-Reply-To: <1253303810.22848.96.camel@corn.betterworld.us>
References: <1253303810.22848.96.camel@corn.betterworld.us>
Message-ID: <a38fadc50909181311udb369cvb9008b7c9c7b9ce1@mail.gmail.com>

On Fri, Sep 18, 2009 at 6:56 PM, Ross Boylan <ross at biostat.ucsf.edu> wrote:

> I downloaded and unzipped Pier-1.2.app.zip onto an amd64 Linux system.
> ./pier.sh initially failed, but when I installed ia32 compatibility libs
> it launched.
>

Yes. This is explained in the readme file of the VM. I don't know if it is
in Pier one click, but in latest pharo VM it is.


>
> However, it is unable to resolve names, or at least squeaksource.com;
> I've stepped through; the primitive resolving names seems to return
> nothing.  I am able to resolve the name on the host system.
>
>
Yes, this is a Pharo known bug. Look in the bugtracker.


> I'm suspicious this is a 64 bit compatibity issue.
>
> Any ideas?
>
> Thanks.
> Ross Boylan
>
> _______________________________________________
> Magritte, Pier and Related Tools ...
> https://www.iam.unibe.ch/mailman/listinfo/smallwiki
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.iam.unibe.ch/pipermail/smallwiki/attachments/20090918/2d880b8a/attachment.html>

From ross at biostat.ucsf.edu  Sat Sep 19 00:57:37 2009
From: ross at biostat.ucsf.edu (Ross Boylan)
Date: Fri, 18 Sep 2009 15:57:37 -0700
Subject: Universes?
Message-ID: <1253314657.22848.133.camel@corn.betterworld.us>

I just loaded "Pier version current" into a squeak 3.10 image (with
magma).  I noticed later that http://www.piercms.com/dev/code says to
use SqueakMap.  Is the version I got OK, or should I back out and try
again?

Thanks.
Ross


From ross at biostat.ucsf.edu  Sat Sep 19 01:10:19 2009
From: ross at biostat.ucsf.edu (Ross Boylan)
Date: Fri, 18 Sep 2009 16:10:19 -0700
Subject: Universes?
In-Reply-To: <1253314657.22848.133.camel@corn.betterworld.us>
References: <1253314657.22848.133.camel@corn.betterworld.us>
Message-ID: <1253315419.22848.139.camel@corn.betterworld.us>

Sorry; to clarify, I loaded Pier using the "Universe" browser.

This got Pier-Seaside 1.2.424 and Seaside 2.8.4.12.  I also wonder if I
should get any of the Seaside28Jetsam, which sound as if they might be
patches to bring Seaside current.

On Fri, 2009-09-18 at 15:57 -0700, Ross Boylan wrote:
> I just loaded "Pier version current" into a squeak 3.10 image (with
> magma).  I noticed later that http://www.piercms.com/dev/code says to
> use SqueakMap.  Is the version I got OK, or should I back out and try
> again?
> 
> Thanks.
> Ross
> 


From renggli at gmail.com  Sat Sep 19 11:54:41 2009
From: renggli at gmail.com (Lukas Renggli)
Date: Sat, 19 Sep 2009 11:54:41 +0200
Subject: Universes?
In-Reply-To: <1253315419.22848.139.camel@corn.betterworld.us>
References: <1253314657.22848.133.camel@corn.betterworld.us>
	<1253315419.22848.139.camel@corn.betterworld.us>
Message-ID: <67628d690909190254w7a451f70xf6d68a7315db0f4b@mail.gmail.com>

> I just loaded "Pier version current" into a squeak 3.10 image (with
> magma).

You are on flimsy grounds. Neither Pier, Magritte nor Seaside have
been extensively used or tested on Squeak 3.10.

> This got Pier-Seaside 1.2.424 and Seaside 2.8.4.12. ?I also wonder if I
> should get any of the Seaside28Jetsam, which sound as if they might be
> patches to bring Seaside current.

No, these are no updates. These are a collection of hacks (overrides)
for a very specific set of package version.

Lukas

-- 
Lukas Renggli
http://www.lukas-renggli.ch


From ross at biostat.ucsf.edu  Sat Sep 19 17:18:06 2009
From: ross at biostat.ucsf.edu (Ross Boylan)
Date: Sat, 19 Sep 2009 08:18:06 -0700
Subject: Universes?
In-Reply-To: <67628d690909190254w7a451f70xf6d68a7315db0f4b@mail.gmail.com>
References: <1253314657.22848.133.camel@corn.betterworld.us>
	<1253315419.22848.139.camel@corn.betterworld.us>
	<67628d690909190254w7a451f70xf6d68a7315db0f4b@mail.gmail.com>
Message-ID: <1253373486.22848.154.camel@corn.betterworld.us>

On Sat, 2009-09-19 at 11:54 +0200, Lukas Renggli wrote:
> > I just loaded "Pier version current" into a squeak 3.10 image (with
> > magma).
> 
> You are on flimsy grounds. Neither Pier, Magritte nor Seaside have
> been extensively used or tested on Squeak 3.10.
> 
Is Squeak 3.9 a better bet?  Pharo, at least the one that comes with the
Pier download bundled for Linux, isn't working for me because of the
name resolution problems I mentioned in an earlier message.
Ross


From renggli at gmail.com  Sat Sep 19 18:31:00 2009
From: renggli at gmail.com (Lukas Renggli)
Date: Sat, 19 Sep 2009 18:31:00 +0200
Subject: Universes?
In-Reply-To: <1253373486.22848.154.camel@corn.betterworld.us>
References: <1253314657.22848.133.camel@corn.betterworld.us>
	<1253315419.22848.139.camel@corn.betterworld.us>
	<67628d690909190254w7a451f70xf6d68a7315db0f4b@mail.gmail.com>
	<1253373486.22848.154.camel@corn.betterworld.us>
Message-ID: <67628d690909190931s3dd6ab88t1df158ccb6ffd043@mail.gmail.com>

>> You are on flimsy grounds. Neither Pier, Magritte nor Seaside have
>> been extensively used or tested on Squeak 3.10.
>>
> Is Squeak 3.9 a better bet? ?Pharo, at least the one that comes with the
> Pier download bundled for Linux, isn't working for me because of the
> name resolution problems I mentioned in an earlier message.

I am using the download available from piercms.com

Cheers,
Lukas

-- 
Lukas Renggli
http://www.lukas-renggli.ch


From ross at biostat.ucsf.edu  Sat Sep 19 19:19:45 2009
From: ross at biostat.ucsf.edu (Ross Boylan)
Date: Sat, 19 Sep 2009 10:19:45 -0700
Subject: Universes?
In-Reply-To: <67628d690909190931s3dd6ab88t1df158ccb6ffd043@mail.gmail.com>
References: <1253314657.22848.133.camel@corn.betterworld.us>
	<1253315419.22848.139.camel@corn.betterworld.us>
	<67628d690909190254w7a451f70xf6d68a7315db0f4b@mail.gmail.com>
	<1253373486.22848.154.camel@corn.betterworld.us>
	<67628d690909190931s3dd6ab88t1df158ccb6ffd043@mail.gmail.com>
Message-ID: <1253380785.22848.156.camel@corn.betterworld.us>

On Sat, 2009-09-19 at 18:31 +0200, Lukas Renggli wrote:
> >> You are on flimsy grounds. Neither Pier, Magritte nor Seaside have
> >> been extensively used or tested on Squeak 3.10.
> >>
> > Is Squeak 3.9 a better bet?  Pharo, at least the one that comes with the
> > Pier download bundled for Linux, isn't working for me because of the
> > name resolution problems I mentioned in an earlier message.
> 
> I am using the download available from piercms.com
That's the one I  tried.
Ross


From keith_hodges at yahoo.co.uk  Sat Sep 19 23:32:57 2009
From: keith_hodges at yahoo.co.uk (keith)
Date: Sat, 19 Sep 2009 22:32:57 +0100
Subject: Universes?
In-Reply-To: <1253373486.22848.154.camel@corn.betterworld.us>
References: <1253314657.22848.133.camel@corn.betterworld.us>
	<1253315419.22848.139.camel@corn.betterworld.us>
	<67628d690909190254w7a451f70xf6d68a7315db0f4b@mail.gmail.com>
	<1253373486.22848.154.camel@corn.betterworld.us>
Message-ID: <B0631C04-D301-42D8-A754-561596B73070@yahoo.co.uk>


On 19 Sep 2009, at 16:18, Ross Boylan wrote:

> On Sat, 2009-09-19 at 11:54 +0200, Lukas Renggli wrote:
>>> I just loaded "Pier version current" into a squeak 3.10 image (with
>>> magma).
>>
>> You are on flimsy grounds. Neither Pier, Magritte nor Seaside have
>> been extensively used or tested on Squeak 3.10.
>>
> Is Squeak 3.9 a better bet?  Pharo, at least the one that comes with  
> the
> Pier download bundled for Linux, isn't working for me because of the
> name resolution problems I mentioned in an earlier message.
> Ross
>
> __

3.10 is not much different from 3.9

Some of us still have to use squeak, and all of the above currently  
work with squeak as far as I am aware, though I am not up to date with  
everything.

best regards

Keith

From keith_hodges at yahoo.co.uk  Mon Sep 21 05:09:50 2009
From: keith_hodges at yahoo.co.uk (keith)
Date: Mon, 21 Sep 2009 04:09:50 +0100
Subject: Link parameters in tables
Message-ID: <651D4100-AE06-4974-A086-7BC331C3FFA0@yahoo.co.uk>

Hi,

Is it just me, or are link parameters in links that appear within a  
table broken?

e.g.

| +embeddedcomponent|param=value+ | second column

==

I am not up to date with pier, so I dont know if this is still an issue

Keith

From damien.cassou at gmail.com  Mon Sep 21 10:31:14 2009
From: damien.cassou at gmail.com (Damien Cassou)
Date: Mon, 21 Sep 2009 10:31:14 +0200
Subject: Link parameters in tables
In-Reply-To: <651D4100-AE06-4974-A086-7BC331C3FFA0@yahoo.co.uk>
References: <651D4100-AE06-4974-A086-7BC331C3FFA0@yahoo.co.uk>
Message-ID: <6ac749c10909210131j7f72587am5f21b613261d2b8@mail.gmail.com>

On Mon, Sep 21, 2009 at 5:09 AM, keith <keith_hodges at yahoo.co.uk> wrote:
> Is it just me, or are link parameters in links that appear within a table
> broken?
>
> e.g.
>
> | +embeddedcomponent|param=value+ | second column
>
> ==
>
> I am not up to date with pier, so I dont know if this is still an issue

Hi Keith,

I think I fixed that at some point. You may want to check.

Bye

-- 
Damien Cassou
http://damiencassou.seasidehosting.st

"Lambdas are relegated to relative obscurity until Java makes them
popular by not having them." James Iry

From razavi at acm.org  Mon Sep 21 11:48:41 2009
From: razavi at acm.org (Reza RAZAVI)
Date: Mon, 21 Sep 2009 11:48:41 +0200
Subject: Link parameters in tables
In-Reply-To: <651D4100-AE06-4974-A086-7BC331C3FFA0@yahoo.co.uk>
References: <651D4100-AE06-4974-A086-7BC331C3FFA0@yahoo.co.uk>
Message-ID: <200909210948.n8L9me4t014827@smtpout1.pt.lu>

At 05:09 21/09/2009, you wrote:
>Is it just me, or are link parameters in links that appear within a
>table broken?
>
>e.g.
>
>| +embeddedcomponent|param=value+ | second column

Hi Keith,

I had a somehow similar issue with the following markup:

*Launch the Add Command with 'PRPage' as 'type', and 'auto' as 'name' 
(to add a Child to the ROOT)>/|command=PRAddCommand|type=PRPage|name=auto*

Another example for the same markup:
+/tests|command=PBPostTicker|blog=/blog+

I fixed it by simply patching two methods as follows:

MAStringReader >> visitClassDescription: aDescription
         "self shouldNotImplement"
         self visitElementDescription: aDescription

and:

PRChildrenWidget >> unfilteredItems
         | structure children x |
         structure :=  self context structure.
         self level isNil ifFalse: [
                 c := (self childrenOf: (structure parents
                         at: self level
                         ifAbsent: [ ^ #() ])) asOrderedCollection.
                 c remove: structure ifAbsent: [].
                 ^c asArray].
         children := self childrenOf: structure.
         ^ (children isEmpty and: [ structure isRoot not ])
                 ifTrue: [ self childrenOf: structure parent ]
                 ifFalse: [ children ]

Hoping this helps,
Regards,
Reza




From renggli at gmail.com  Mon Sep 21 12:32:52 2009
From: renggli at gmail.com (Lukas Renggli)
Date: Mon, 21 Sep 2009 12:32:52 +0200
Subject: Link parameters in tables
In-Reply-To: <6ac749c10909210131j7f72587am5f21b613261d2b8@mail.gmail.com>
References: <651D4100-AE06-4974-A086-7BC331C3FFA0@yahoo.co.uk>
	<6ac749c10909210131j7f72587am5f21b613261d2b8@mail.gmail.com>
Message-ID: <67628d690909210332ra3474b0p856ad00a2d696ea5@mail.gmail.com>

>> Is it just me, or are link parameters in links that appear within a table
>> broken?
>>
>> e.g.
>>
>> | +embeddedcomponent|param=value+ | second column
>>
>> ==
>>
>> I am not up to date with pier, so I dont know if this is still an issue
>
> I think I fixed that at some point. You may want to check.

I don't think this is fixed, because the test you wrote is still broken.

Lukas

-- 
Lukas Renggli
http://www.lukas-renggli.ch

From keith_hodges at yahoo.co.uk  Mon Sep 21 06:20:10 2009
From: keith_hodges at yahoo.co.uk (keith)
Date: Mon, 21 Sep 2009 05:20:10 +0100
Subject: link alias
Message-ID: <7829B507-CE88-422C-833E-90EFF7F4FBD2@yahoo.co.uk>

Lukas,

I find on a regular basis that I want to be able to provide an alias  
to value "links", e.g. the following displays a link to the current  
structure, with the text "Login", executing the login command, and  
this link is only displayed if the user is not logged in.

+value:structure|link=Login|command=PULogin|loggedout+

This can be implemented very simply in PRValueLink>>#formatStructure:  
aStructure in: aContext

do let me know if you include this addition,

thanks in advance

Keith


=======
PRValueLink>>#formatStructure: aStructure in: aContext

	"Format ==aStructure==. Link it, if the parameter ==link== is given;  
embed it, if the paremeter ==embed== is given."
	
	| string |
	((self selectAndReject: aStructure) isNil
		or: [ (aContext structure: aStructure) isValid not
		or: [ (aStructure propertyAt: #hideFromMenus ifAbsent: [ false ]) ] ])
			ifTrue: [ ^ nil ].

"BEGIN MODIFICATION TO PROVIDE ALIAS TO LINK"			
	string := self parameterAt: 'link'.
	
	string ifNil: [
		string := self
			formatDescribed: aStructure
			default: #title
	].
"END MODIFICATION TO PROVIDE ALIAS TO LINK"	

	^ ((self hasParameter: 'link') or: [ self hasParameter: 'embed' ])
		ifTrue: [
			PRInternalLink new
				embedded: (self hasParameter: 'embed');
				parameters: self parameters;
				target: aStructure;
				addAll: (string isNil
					ifTrue: [ Array new ]
					ifFalse: [ Array with: (PRText content: string) ]);
				yourself ]
		ifFalse: [ string isNil ifFalse: [ PRText content: string ] ]

From ross at biostat.ucsf.edu  Wed Sep 23 01:41:08 2009
From: ross at biostat.ucsf.edu (Ross Boylan)
Date: Tue, 22 Sep 2009 16:41:08 -0700
Subject: security questions
Message-ID: <1253662868.20834.6.camel@markov.biostat.ucsf.edu>

Can anyone give me any pointers about how security works in pier?  In
particular, if I make my own components, how do I apply security to
them?

I've been browsing around and tracing the code and see that the security
descriptors end up as decorations which apply to objectified commands.
However, I don't know what I need to do to set the security of an object
or fiddle with the security policy.

Also, if someone comes in from the outside with a link like
http://my.host.com/seaside/pier/personalpreferences, is there a way I
can ensure that they will be prompted for login and then taken to a
preferences page for them?

Thanks.
Ross


From renggli at gmail.com  Wed Sep 23 08:47:05 2009
From: renggli at gmail.com (Lukas Renggli)
Date: Wed, 23 Sep 2009 08:47:05 +0200
Subject: security questions
In-Reply-To: <1253662868.20834.6.camel@markov.biostat.ucsf.edu>
References: <1253662868.20834.6.camel@markov.biostat.ucsf.edu>
Message-ID: <67628d690909222347y56f0e7eoc97183493db38e86@mail.gmail.com>

> Can anyone give me any pointers about how security works in pier? ?In
> particular, if I make my own components, how do I apply security to
> them?

You can find some documentation in Section 3.5.3 in
http://scg.unibe.ch/archive/masters/Reng06a.pdf.

> I've been browsing around and tracing the code and see that the security
> descriptors end up as decorations which apply to objectified commands.
> However, I don't know what I need to do to set the security of an object
> or fiddle with the security policy.

You can only apply the security decorations to subclasses of
PRStructure (these are the objects that represent a unique URL or
entry point into your application), not to arbitrary objects (out of
the box, at least).

> Also, if someone comes in from the outside with a link like
> http://my.host.com/seaside/pier/personalpreferences, is there a way I
> can ensure that they will be prompted for login and then taken to a
> preferences page for them?

The PRForbiddenView is displayed (with a link to the login page). This
is currently hardcoded into the "Pier-Interpreter", but could probably
be moved to the structures so that they can decide to do something
else, if necessary.

Cheers,
Lukas

-- 
Lukas Renggli
http://www.lukas-renggli.ch


From ross at biostat.ucsf.edu  Thu Sep 24 02:09:03 2009
From: ross at biostat.ucsf.edu (Ross Boylan)
Date: Wed, 23 Sep 2009 17:09:03 -0700
Subject: security questions
In-Reply-To: <67628d690909222347y56f0e7eoc97183493db38e86@mail.gmail.com>
References: <1253662868.20834.6.camel@markov.biostat.ucsf.edu>
	<67628d690909222347y56f0e7eoc97183493db38e86@mail.gmail.com>
Message-ID: <1253750943.803.33.camel@corn.betterworld.us>

I have some questions, but let me put what I've discovered in here too,
for the record.  Also, you can correct me if I'm wrong!

On Wed, 2009-09-23 at 08:47 +0200, Lukas Renggli wrote:
> > Can anyone give me any pointers about how security works in pier?  In
> > particular, if I make my own components, how do I apply security to
> > them?
> 
> You can find some documentation in Section 3.5.3 in
> http://scg.unibe.ch/archive/masters/Reng06a.pdf.
Very useful.
> 
> > I've been browsing around and tracing the code and see that the security
> > descriptors end up as decorations which apply to objectified commands.
> > However, I don't know what I need to do to set the security of an object
> > or fiddle with the security policy.
A little more tracing reveals that when children are added to a
structure all concrete decoration classes are given a chance to do
something to the child.  The security decorator copies from the parent,
which in turn triggers the creation of a security descriptor, using the
current context for user and group, on the page.

I also learned that the children of a PRStructure are kept in a child
decorator; that was not what I would have guessed.
> 
> You can only apply the security decorations to subclasses of
> PRStructure (these are the objects that represent a unique URL or
> entry point into your application), not to arbitrary objects (out of
> the box, at least).

I'm thinking of having a page with a central area that gets swapped
according to what the user is doing; the other material on the page
would stay.  Does that mean there's really only one security decoration
that is relevant?  Or does it mean the children are individually
addressable (e.g., elements of a list)?

Also, I wonder if I need any security other than ensuring someone is
logged in.  A user will only be presented links to stuff that is
relevant (and allowed) for them.  Does adding security to the objects
add any additional security (e.g., perhaps against attempts to
manipulate the URL)?

Finally, in case I do need security, is the ACL based framework still
around?  It's probably more appropriate for the scenarios I have in
mind.

> 
> > Also, if someone comes in from the outside with a link like
> > http://my.host.com/seaside/pier/personalpreferences, is there a way I
> > can ensure that they will be prompted for login and then taken to a
> > preferences page for them?
> 
> The PRForbiddenView is displayed (with a link to the login page). This
> is currently hardcoded into the "Pier-Interpreter", but could probably
> be moved to the structures so that they can decide to do something
> else, if necessary.

Once they login, are they taken to the page they were trying to reach
originally?


Thanks.
Ross



From renggli at gmail.com  Thu Sep 24 09:09:23 2009
From: renggli at gmail.com (Lukas Renggli)
Date: Thu, 24 Sep 2009 09:09:23 +0200
Subject: security questions
In-Reply-To: <1253750943.803.33.camel@corn.betterworld.us>
References: <1253662868.20834.6.camel@markov.biostat.ucsf.edu>
	<67628d690909222347y56f0e7eoc97183493db38e86@mail.gmail.com>
	<1253750943.803.33.camel@corn.betterworld.us>
Message-ID: <67628d690909240009s2e7210a0ia8d3553bd1dec2d0@mail.gmail.com>

> A little more tracing reveals that when children are added to a
> structure all concrete decoration classes are given a chance to do
> something to the child. ?The security decorator copies from the parent,
> which in turn triggers the creation of a security descriptor, using the
> current context for user and group, on the page.

Correct.

> I also learned that the children of a PRStructure are kept in a child
> decorator; that was not what I would have guessed.

Yes, this separates all the "children" behavior into a separate object
and allows all structures to have children or not. This was one of the
major annoyances in SmallWiki the predecessor of Pier, where children
were only allowed in folder.

>> You can only apply the security decorations to subclasses of
>> PRStructure (these are the objects that represent a unique URL or
>> entry point into your application), not to arbitrary objects (out of
>> the box, at least).
>
> I'm thinking of having a page with a central area that gets swapped
> according to what the user is doing; the other material on the page
> would stay. ?Does that mean there's really only one security decoration
> that is relevant? ?Or does it mean the children are individually
> addressable (e.g., elements of a list)?

If you have a page that embeds some of its child-pages using the
+child1+ syntax, chilld1 is only embedded if the user has view
permission on child1 as well. So yeah, sometimes multiple permissions
from different places come into play.

> Also, I wonder if I need any security other than ensuring someone is
> logged in. ?A user will only be presented links to stuff that is
> relevant (and allowed) for them. ?Does adding security to the objects
> add any additional security (e.g., perhaps against attempts to
> manipulate the URL)?

Again, you cannot add security to arbitrary objects, just to
structures that map to particular URLs.

For example for this software engineering course <http://ese.unibe.ch>
there are pages that only the assistants can see and edit (for the
organization of the course), there are pages that only specific groups
of students can edit (for the student teams), there are personal pages
that you can only see when logged in and only a particular student can
edit (for the personal pages), there are pages that all students can
see and edit (for feedback), etc.

> Finally, in case I do need security, is the ACL based framework still
> around? ?It's probably more appropriate for the scenarios I have in
> mind.

The ACL based framework is more powerful, indeed. I don't think it was
used recently and would probably need some updates to work in the
latest version of Pier.

Lukas

-- 
Lukas Renggli
http://www.lukas-renggli.ch


From marianopeck at gmail.com  Thu Sep 24 14:44:31 2009
From: marianopeck at gmail.com (Mariano Martinez Peck)
Date: Thu, 24 Sep 2009 11:44:31 -0100
Subject: Sub bulleted or ordered list
Message-ID: <a38fadc50909240544q315de343hbcce137bb5f9372b@mail.gmail.com>

Hi folks: I need something like the office bullet or ordered list, where you
can have those lists and sublist. Those you use in Word using TAB key.

I tried putting double -- or double ##   but it doesn't work.

Do you know how can I achieve that in Pier ?

Best,

Mariano
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.iam.unibe.ch/pipermail/smallwiki/attachments/20090924/5257fe6f/attachment.html>

From renggli at gmail.com  Thu Sep 24 14:57:26 2009
From: renggli at gmail.com (Lukas Renggli)
Date: Thu, 24 Sep 2009 14:57:26 +0200
Subject: Sub bulleted or ordered list
In-Reply-To: <a38fadc50909240544q315de343hbcce137bb5f9372b@mail.gmail.com>
References: <a38fadc50909240544q315de343hbcce137bb5f9372b@mail.gmail.com>
Message-ID: <67628d690909240557q2427366u26dc4bab25f32a13@mail.gmail.com>

http://www.piercms.com/doc/syntax#99639110

Does that help?

Lukas

2009/9/24 Mariano Martinez Peck <marianopeck at gmail.com>:
> Hi folks: I need something like the office bullet or ordered list, where you
> can have those lists and sublist. Those you use in Word using TAB key.
>
> I tried putting double -- or double ##?? but it doesn't work.
>
> Do you know how can I achieve that in Pier ?
>
> Best,
>
> Mariano
>
> _______________________________________________
> Magritte, Pier and Related Tools ...
> https://www.iam.unibe.ch/mailman/listinfo/smallwiki
>



-- 
Lukas Renggli
http://www.lukas-renggli.ch


From marianopeck at gmail.com  Thu Sep 24 15:07:01 2009
From: marianopeck at gmail.com (Mariano Martinez Peck)
Date: Thu, 24 Sep 2009 12:07:01 -0100
Subject: Sub bulleted or ordered list
In-Reply-To: <67628d690909240557q2427366u26dc4bab25f32a13@mail.gmail.com>
References: <a38fadc50909240544q315de343hbcce137bb5f9372b@mail.gmail.com>
	<67628d690909240557q2427366u26dc4bab25f32a13@mail.gmail.com>
Message-ID: <a38fadc50909240607u965499br88aeec6ac2f40080@mail.gmail.com>

On Thu, Sep 24, 2009 at 11:57 AM, Lukas Renggli <renggli at gmail.com> wrote:

> http://www.piercms.com/doc/syntax#99639110
>
> Does that help?
>
>
Yeah!!! Thanks! It worked. I used to have something like this:

-One bullet

-Two bullet

--a sub bullet of two bullet

That's doesn't work. You must do:

-One bullet

-Two bullet
--a sub bullet of two bullet

I removed a blank line between them!

Thanks.

Mariano




> Lukas
>
> 2009/9/24 Mariano Martinez Peck <marianopeck at gmail.com>:
> > Hi folks: I need something like the office bullet or ordered list, where
> you
> > can have those lists and sublist. Those you use in Word using TAB key.
> >
> > I tried putting double -- or double ##   but it doesn't work.
> >
> > Do you know how can I achieve that in Pier ?
> >
> > Best,
> >
> > Mariano
> >
> > _______________________________________________
> > Magritte, Pier and Related Tools ...
> > https://www.iam.unibe.ch/mailman/listinfo/smallwiki
> >
>
>
>
> --
> Lukas Renggli
> http://www.lukas-renggli.ch
>
> _______________________________________________
> Magritte, Pier and Related Tools ...
> https://www.iam.unibe.ch/mailman/listinfo/smallwiki
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.iam.unibe.ch/pipermail/smallwiki/attachments/20090924/63b837df/attachment.html>

From ross at biostat.ucsf.edu  Fri Sep 25 06:38:03 2009
From: ross at biostat.ucsf.edu (Ross Boylan)
Date: Thu, 24 Sep 2009 21:38:03 -0700
Subject: PRDistribution>>makePrivate:
Message-ID: <1253853483.12875.22.camel@corn.betterworld.us>

This method seems to do the same thing as PRDistribution>>makePublic:.
Am I missing some subtlety, or is this a bug?

Ross


From ross at biostat.ucsf.edu  Fri Sep 25 06:48:30 2009
From: ross at biostat.ucsf.edu (Ross Boylan)
Date: Thu, 24 Sep 2009 21:48:30 -0700
Subject: handling components
Message-ID: <1253854110.12875.33.camel@corn.betterworld.us>

If I have a seaside component and want to make it a page in pier, how do
I do that?  What if I want to make it a part of a page?

I have a feeling the pier way is much different.  Judging by how
PRDistribution sets things up, it looks as if components are wrapped in
PRComponent and have a component class.  But I've been unable to see
where actual rendering happens, or where state goes.

The concrete problem motivating these questions is that I want to
display a list of files whose name match a certain pattern in a certain
directory.  The pattern and the directory could be hardcoded, but it
would be more interesting if they were editable in the web.

This will display a list of the files; if you click on one of them you
will see it.  That sounds like a job for PRFile, though again I'm not
exactly sure how that gets hooked in.

This is all likely to get more complex as I extract relations between
files, parse the contents of the files and show information based on the
parse.  But for now I'm just working on the "simple" stuff.

Any hints?
Thanks.
Ross


From renggli at gmail.com  Fri Sep 25 07:37:26 2009
From: renggli at gmail.com (Lukas Renggli)
Date: Fri, 25 Sep 2009 07:37:26 +0200
Subject: PRDistribution>>makePrivate:
In-Reply-To: <1253853483.12875.22.camel@corn.betterworld.us>
References: <1253853483.12875.22.camel@corn.betterworld.us>
Message-ID: <67628d690909242237xd7641fdud0ea361ab480d77f@mail.gmail.com>

It doesn't look like #makePublic: is used at all. That's the default anyway.

Doru, can the method #makePublic: be removed?

Lukas

2009/9/25 Ross Boylan <ross at biostat.ucsf.edu>:
> This method seems to do the same thing as PRDistribution>>makePublic:.
> Am I missing some subtlety, or is this a bug?
>
> Ross
>
> _______________________________________________
> Magritte, Pier and Related Tools ...
> https://www.iam.unibe.ch/mailman/listinfo/smallwiki
>



-- 
Lukas Renggli
http://www.lukas-renggli.ch

From renggli at gmail.com  Fri Sep 25 07:44:31 2009
From: renggli at gmail.com (Lukas Renggli)
Date: Fri, 25 Sep 2009 07:44:31 +0200
Subject: handling components
In-Reply-To: <1253854110.12875.33.camel@corn.betterworld.us>
References: <1253854110.12875.33.camel@corn.betterworld.us>
Message-ID: <67628d690909242244w2b736358u92f34dc465641961@mail.gmail.com>

> If I have a seaside component and want to make it a page in pier, how do
> I do that?

Add > Select "Component" > Select "Component Class"

> What if I want to make it a part of a page?

You embed it into the page with an embed-link:

    A Seaside component: +/path/to/my/component+

> I have a feeling the pier way is much different. ?Judging by how
> PRDistribution sets things up, it looks as if components are wrapped in
> PRComponent and have a component class. ?But I've been unable to see
> where actual rendering happens, or where state goes.

PRDistribution creates an initial model from a script. Doing it from
the web goes through a series of command objects, but in the end you
have the same model. State is stored in the components and rendering
happens in #renderContentsOn: of the components, just as in default
Seaside.

> The concrete problem motivating these questions is that I want to
> display a list of files whose name match a certain pattern in a certain
> directory. ?The pattern and the directory could be hardcoded, but it
> would be more interesting if they were editable in the web.

You can do this with something like this:

    +value:children|link|target=/path/to/directory|select=File+

> This is all likely to get more complex as I extract relations between
> files, parse the contents of the files and show information based on the
> parse.  But for now I'm just working on the "simple" stuff.

For this requirement you presumably require to add some code of you own.

Cheers,
Lukas

-- 
Lukas Renggli
http://www.lukas-renggli.ch


From ross at biostat.ucsf.edu  Fri Sep 25 18:03:31 2009
From: ross at biostat.ucsf.edu (Ross Boylan)
Date: Fri, 25 Sep 2009 09:03:31 -0700
Subject: handling components
In-Reply-To: <67628d690909242244w2b736358u92f34dc465641961@mail.gmail.com>
References: <1253854110.12875.33.camel@corn.betterworld.us>
	<67628d690909242244w2b736358u92f34dc465641961@mail.gmail.com>
Message-ID: <1253894612.12875.61.camel@corn.betterworld.us>

On Fri, 2009-09-25 at 07:44 +0200, Lukas Renggli wrote:
> > If I have a seaside component and want to make it a page in pier, how do
> > I do that?
> 
> Add > Select "Component" > Select "Component Class"
Is there some registration process to get items to appear on that list?
I tried adding users; that seemed to reuse the existing user list.  At
the same time, I didn't see a direct way to add a user; that is, I had
to add the list, and then hit the add button of the list.
> 
> > What if I want to make it a part of a page?
> 
> You embed it into the page with an embed-link:
> 
>     A Seaside component: +/path/to/my/component+
Does this assume step 1, so that my/component is a page (class
PRComponent) setup with an instance of my component?
> 
> > I have a feeling the pier way is much different.  Judging by how
> > PRDistribution sets things up, it looks as if components are wrapped in
> > PRComponent and have a component class.  But I've been unable to see
> > where actual rendering happens, or where state goes.
> 
> PRDistribution creates an initial model from a script. Doing it from
> the web goes through a series of command objects, but in the end you
> have the same model. State is stored in the components and rendering
> happens in #renderContentsOn: of the components, just as in default
> Seaside.
More precisely, I think the Magritte way is different.  I think I could
subclass PRObject to create a class with instance vars dir and pattern.
On the class side implement descriptionDir and descriptionPattern (which
maybe would create the accessors if I hadn't bothered?).
But then I don't know how to hook that up so one can create an instance
through the web.
And I don't know how to change the default display (the default edit
view is probably fine) so it shows, e.g., the list of files in dir
matching pattern.

When you referred to things being addressable through the web, in
response to my earlier question about security, I had assumed that the
components of a page were not addressable.  I see now that most
components, if included through +path/component+, are addressable (and
presumably can be viewed separately if one is so inclined).
> 
> > The concrete problem motivating these questions is that I want to
> > display a list of files whose name match a certain pattern in a certain
> > directory.  The pattern and the directory could be hardcoded, but it
> > would be more interesting if they were editable in the web.
> 
> You can do this with something like this:
> 
>     +value:children|link|target=/path/to/directory|select=File+
Let's see if I understand.
Do parameters like |foo=bar get translated into method calls foo: "bar"?
And |foo=bar|with=blah becomes sends of foo: "bar" and then with:
"blah", rather than foo: "bar" with: "blah"?
These are directed to my component, not the children?  But the thing
with  +value:children...+ is a PRPage, not my component, so that won't
work.  So maybe my component should subclass PRPage?
So I create a component that implements setters for target: and select:.
When it receives select: (if I can count on the order) it clears it's
existing list of children and repopulates it with appropriate PRFile's.
And then the value:children|link will display them.
?
> 
> > This is all likely to get more complex as I extract relations between
> > files, parse the contents of the files and show information based on the
> > parse.  But for now I'm just working on the "simple" stuff.
> 
> For this requirement you presumably require to add some code of you own.
Well, I already added some code, so maybe I don't understand.
The files in the directory may change (both the names of the files and
the contents), and so I can't use a static list.

Thanks for all your help.
Ross



From renggli at gmail.com  Fri Sep 25 18:22:05 2009
From: renggli at gmail.com (Lukas Renggli)
Date: Fri, 25 Sep 2009 18:22:05 +0200
Subject: handling components
In-Reply-To: <1253894612.12875.61.camel@corn.betterworld.us>
References: <1253854110.12875.33.camel@corn.betterworld.us>
	<67628d690909242244w2b736358u92f34dc465641961@mail.gmail.com>
	<1253894612.12875.61.camel@corn.betterworld.us>
Message-ID: <67628d690909250922s34151762l8a4b6d8492a53a74@mail.gmail.com>

>> Add > Select "Type" > Select "Component Class"
> Is there some registration process to get items to appear on that list?

Type: all PRStructure subclasses that are not abstract
Component Class: all PRWidget classes that are not abstract, all
Seaside root components

> I tried adding users; that seemed to reuse the existing user list. ?At
> the same time, I didn't see a direct way to add a user; that is, I had
> to add the list, and then hit the add button of the list.

I do not understand.

> When you referred to things being addressable through the web, in
> response to my earlier question about security, I had assumed that the
> components of a page were not addressable. ?I see now that most
> components, if included through +path/component+, are addressable (and
> presumably can be viewed separately if one is so inclined).

Yes, if you have the permissions you can individually navigate to the
components, given that there is a link or that you know the URL.
Typically components are not linked directly though, but embedded into
some other page.

>> > The concrete problem motivating these questions is that I want to
>> > display a list of files whose name match a certain pattern in a certain
>> > directory. ?The pattern and the directory could be hardcoded, but it
>> > would be more interesting if they were editable in the web.
>>
>> You can do this with something like this:
>>
>> ? ? +value:children|link|target=/path/to/directory|select=File+
> Let's see if I understand.
> Do parameters like |foo=bar get translated into method calls foo: "bar"?
> And |foo=bar|with=blah becomes sends of foo: "bar" and then with:
> "blah", rather than foo: "bar" with: "blah"?

No method calls. Value links (PRValueLink) are just a special kind of
link that embeds some dynamically created content, similar to a
component but without state. See the documentation at
<http://bit.ly/294VHv>.

> These are directed to my component, not the children? ?But the thing
> with ?+value:children...+ is a PRPage, not my component, so that won't
> work. ?So maybe my component should subclass PRPage?

PRPage is a model object.
WAComponent is a view object.

See <http://www.piercms.com/doc> for details.

Cheers,
Lukas

-- 
Lukas Renggli
http://www.lukas-renggli.ch


From ross at biostat.ucsf.edu  Fri Sep 25 19:16:32 2009
From: ross at biostat.ucsf.edu (Ross Boylan)
Date: Fri, 25 Sep 2009 10:16:32 -0700
Subject: handling components
In-Reply-To: <67628d690909250922s34151762l8a4b6d8492a53a74@mail.gmail.com>
References: <1253854110.12875.33.camel@corn.betterworld.us>
	<67628d690909242244w2b736358u92f34dc465641961@mail.gmail.com>
	<1253894612.12875.61.camel@corn.betterworld.us>
	<67628d690909250922s34151762l8a4b6d8492a53a74@mail.gmail.com>
Message-ID: <1253898992.12875.77.camel@corn.betterworld.us>

On Fri, 2009-09-25 at 18:22 +0200, Lukas Renggli wrote:
> 
> > I tried adding users; that seemed to reuse the existing user list.
>  At
> > the same time, I didn't see a direct way to add a user; that is, I
> had
> > to add the list, and then hit the add button of the list.
> 
> I do not understand.

>From a web interface point of view, I thought there would be a way to
add a new user directly.  As far as I can tell, there isn't.

It seems one must first create a Pier Users component, and then work
within it.

I was a little surprised the Pier Users component seemed to use the
existing users list, rather than create a new one.

>From a development point of view, I was a little puzzled.  I can see
there is a PUUser object, which class side description* methods.  I
think that means it's setup for Magritte.  So I thought that might be
sufficient to make it possible to create one from the web.

Also, I'm confused about the roles of components and models.
I see now that there is a PUUsersWidget; I think this is what I added.
It does not hold a user list, but references it with

members
	^ self context kernel users

I think what this means is that I could produce a quick and dirty widget
that was also a model (holds directory and pattern for files, knows how
to figure out which files are relevant), or I could have an explicit
Magritte object with the model.  But then do I need another widget to
display the magritte object, e.g., for editing the instance variables?


From renggli at gmail.com  Fri Sep 25 19:29:36 2009
From: renggli at gmail.com (Lukas Renggli)
Date: Fri, 25 Sep 2009 19:29:36 +0200
Subject: handling components
In-Reply-To: <1253898992.12875.77.camel@corn.betterworld.us>
References: <1253854110.12875.33.camel@corn.betterworld.us>
	<67628d690909242244w2b736358u92f34dc465641961@mail.gmail.com>
	<1253894612.12875.61.camel@corn.betterworld.us>
	<67628d690909250922s34151762l8a4b6d8492a53a74@mail.gmail.com>
	<1253898992.12875.77.camel@corn.betterworld.us>
Message-ID: <67628d690909251029m1c6cd3dcpe4f6b763b22994f5@mail.gmail.com>

> >From a web interface point of view, I thought there would be a way to
> add a new user directly. ?As far as I can tell, there isn't.
>
> It seems one must first create a Pier Users component, and then work
> within it.

Yes, there are several components to edit users, to edit groups, to
change various settings, to import/export the site, to observe the
server state, to search for broken links, etc.

> I was a little surprised the Pier Users component seemed to use the
> existing users list, rather than create a new one.

Why would they create a new one?

Pier-Security models a unix-like security system, there is a single
global list of users/groups stored in the kernel (root).

The ACL based security package could define users/groups locally, by
adding user/group structures.

> I think what this means is that I could produce a quick and dirty widget
> that was also a model (holds directory and pattern for files, knows how
> to figure out which files are relevant), or I could have an explicit
> Magritte object with the model. ?But then do I need another widget to
> display the magritte object, e.g., for editing the instance variables?

You can embed any Seaside component into Pier. It does not need to use Magritte.

Lukas

-- 
Lukas Renggli
http://www.lukas-renggli.ch


From ross at biostat.ucsf.edu  Fri Sep 25 19:31:13 2009
From: ross at biostat.ucsf.edu (Ross Boylan)
Date: Fri, 25 Sep 2009 10:31:13 -0700
Subject: handling components
In-Reply-To: <67628d690909250922s34151762l8a4b6d8492a53a74@mail.gmail.com>
References: <1253854110.12875.33.camel@corn.betterworld.us>
	<67628d690909242244w2b736358u92f34dc465641961@mail.gmail.com>
	<1253894612.12875.61.camel@corn.betterworld.us>
	<67628d690909250922s34151762l8a4b6d8492a53a74@mail.gmail.com>
Message-ID: <1253899873.12875.83.camel@corn.betterworld.us>

On Fri, 2009-09-25 at 18:22 +0200, Lukas Renggli wrote:
> >> > The concrete problem motivating these questions is that I want to
> >> > display a list of files whose name match a certain pattern in a
> certain
> >> > directory.  The pattern and the directory could be hardcoded, but
> it
> >> > would be more interesting if they were editable in the web.
> >>
> >> You can do this with something like this:
> >>
> >>     +value:children|link|target=/path/to/directory|select=File+
> > Let's see if I understand.
> > Do parameters like |foo=bar get translated into method calls foo:
> "bar"?
> > And |foo=bar|with=blah becomes sends of foo: "bar" and then with:
> > "blah", rather than foo: "bar" with: "blah"?
> 
> No method calls. Value links (PRValueLink) are just a special kind of
> link that embeds some dynamically created content, similar to a
> component but without state. See the documentation at
> <http://bit.ly/294VHv>.
Yes, I saw that.  But I don't understand what happens with the
parameters.  Are you saying the +value...+ above will work without any
new code on my part?  Somehow I have to create the children to iterate
over, don't I?
> 
> > These are directed to my component, not the children?  But the thing
> > with  +value:children...+ is a PRPage, not my component, so that
> won't
> > work.  So maybe my component should subclass PRPage?
> 
> PRPage is a model object.
> WAComponent is a view object.
> 
> See <http://www.piercms.com/doc> for details.


From marianopeck at gmail.com  Fri Sep 25 19:43:33 2009
From: marianopeck at gmail.com (Mariano Martinez Peck)
Date: Fri, 25 Sep 2009 16:43:33 -0100
Subject: [Pharo-project] pharo1.0-10451-BETAweb09.09.3 in FireFox
In-Reply-To: <84f77e270909251039md29beddub77e7f6ccbfefb42@mail.gmail.com>
References: <84f77e270909251039md29beddub77e7f6ccbfefb42@mail.gmail.com>
Message-ID: <a38fadc50909251043g5962d770mbb572dcac013cb85@mail.gmail.com>

You should ask also in Pier mailing list. I cc to it.

2009/9/25 Dennis Schetinin <chaetal at gmail.com>

> I can't log in at the 'main' page (http://127.0.0.1:8080/seaside/pier) in
> Pier. It just ignores all the inputs (both correct and incorrect) simply
> redirecting from input form back to the main page. But from the Blog or
> About page it works correctly. It's all using FireFox 3.5.3. With Chrome
> it's ok everywhere, including 'main' page...
>
> What can be the cause of the problem?
>
> --
> Dennis Schetinin
>
> _______________________________________________
> Pharo-project mailing list
> Pharo-project at lists.gforge.inria.fr
> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.iam.unibe.ch/pipermail/smallwiki/attachments/20090925/55e03250/attachment.html>

From dtrussardi at tiscali.it  Sat Sep 26 13:32:29 2009
From: dtrussardi at tiscali.it (Dario Trussardi)
Date: Sat, 26 Sep 2009 13:32:29 +0200
Subject: Recaptcha as MAgritte component 
Message-ID: <6A8C74C990D14D3D9EB106B4CB293CD3@istcamuno>

Hi,

On Wed, 2009-06-17 at 12:22 -0300, Sebastian Van Lacke wrote:
>> Hi, I need to generate some graphics dinamically for my application
>> and Gemstone doesn't provide draw classes.
>> Specifically I want to implement a Captcha validation in a form, with
>> a distorted text.
>>
>> How can I do it?
>>
> If you don't need to do it on your own you can use the recaptcha
> service (http://www.recaptcha.net). Fortunately I wrote a component
> for this last week. This is available at
>
> http://selfish.org/software/recaptcha
>
> It runs in Glass-dkh.231. If you have problems using it I'm happy
> to help out.
>
> Norbert

Now i use Recaptcha and it work well.

But i,m interesting to use it as Magritte  component ( only for simply use with other MADescription ) and i have do:

1) I copy the RSRecaptchaComponent to MARSRecaptchaComponent  ( as subclass 
of  MAElementComponent ).

2) I create MARSRecaptchaComponentDescription with:
     defaultComponentClass ->  MARSRecaptchaComponent
     kind -> RSRecaptcha

3) Now when rendering a component with a description based on 
MARSRecaptchaComponentDescription :

        task call:  ( collectionOfDescription asComponentOn: model )
                adValidateForm;
                addMessage: 'testo';
                yourself.

    The rendering work well ( i see MARSRecaptchaComponent right )

    But when test the hasValidateResponse with :

        validateSpecific: anObject
            super validateSpecific: anObject.
            anObject hasValidateResponse ifTrue:[  ] .


    the requestBody  ( RSRecaptcha method ) go in error:

     The anRSRecaptcha   parameters dictionary have :
        'challenge' ->nil
        'response' -> nil.

I think registerCallbacksOn: html d'ont work well.

Thank for any consideration.

    Dario

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.iam.unibe.ch/pipermail/smallwiki/attachments/20090926/a7f6dea9/attachment.html>

From anlarionov at gmail.com  Sun Sep 27 13:38:03 2009
From: anlarionov at gmail.com (Andrey Larionov)
Date: Sun, 27 Sep 2009 15:38:03 +0400
Subject: Fwd: [Pharo-project] pharo1.0-10451-BETAweb09.09.3 in FireFox
In-Reply-To: <84f77e270909251039md29beddub77e7f6ccbfefb42@mail.gmail.com>
References: <84f77e270909251039md29beddub77e7f6ccbfefb42@mail.gmail.com>
Message-ID: <b831ed00909270438m26f8b02bic87d024f76ab47bb@mail.gmail.com>

To Pier List

---------- Forwarded message ----------
From: Dennis Schetinin <chaetal at gmail.com>
Date: 2009/9/25
Subject: [Pharo-project] pharo1.0-10451-BETAweb09.09.3 in FireFox
To: pharo-project at lists.gforge.inria.fr


I can't log in at the 'main' page (http://127.0.0.1:8080/seaside/pier)
in Pier. It just ignores all the inputs (both correct and incorrect)
simply redirecting from input form back to the main page. But from the
Blog or About page it works correctly. It's all using FireFox 3.5.3.
With Chrome it's ok everywhere, including 'main' page...

What can be the cause of the problem?

--
Dennis Schetinin

_______________________________________________
Pharo-project mailing list
Pharo-project at lists.gforge.inria.fr
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

From mbany at cincom.com  Mon Sep 28 10:26:45 2009
From: mbany at cincom.com (Bany, Michel)
Date: Mon, 28 Sep 2009 09:26:45 +0100
Subject: Conditional visibility
Message-ID: <BE063FF0656052439E47ADB9BAFB396A017211AB@im04.cincom.com>

I have a requirement for a Magritte-generated Seaside component where
the visibility of some of the model's property is dependent on the
values of other properties. Something like this.

	descriptionMaidenName
	
		^(MAStringDescription new)
			selectorAccessor: 'maidenName';
			label: 'Maiden name';
			visible: [:aModel | aModel
isVisibleInComponent];			
			yourself

I am still using lr.251 and asking the following questions to the list.
1) Is there a response to my requirement in more recent Magritte
versions?
2) Has this topic already been covered in this list?

Thanks,
Michel.   
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.iam.unibe.ch/pipermail/smallwiki/attachments/20090928/aff178bb/attachment.html>

From renggli at gmail.com  Mon Sep 28 10:43:49 2009
From: renggli at gmail.com (Lukas Renggli)
Date: Mon, 28 Sep 2009 10:43:49 +0200
Subject: Conditional visibility
In-Reply-To: <BE063FF0656052439E47ADB9BAFB396A017211AB@im04.cincom.com>
References: <BE063FF0656052439E47ADB9BAFB396A017211AB@im04.cincom.com>
Message-ID: <67628d690909280143x52195bd4i810b62a660a5708e@mail.gmail.com>

> I am still using lr.251 and asking the following questions to the list.
> 1) Is there a response to my requirement in more recent Magritte versions?

No.

> 2) Has this topic already been covered in this list?

Yes, also see the first question on
<http://www.lukas-renggli.ch/smalltalk/magritte/faq>.

Lukas

-- 
Lukas Renggli
http://www.lukas-renggli.ch

From mbany at cincom.com  Mon Sep 28 13:43:40 2009
From: mbany at cincom.com (Bany, Michel)
Date: Mon, 28 Sep 2009 12:43:40 +0100
Subject: Conditional visibility
In-Reply-To: <67628d690909280143x52195bd4i810b62a660a5708e@mail.gmail.com>
References: <BE063FF0656052439E47ADB9BAFB396A017211AB@im04.cincom.com>
	<67628d690909280143x52195bd4i810b62a660a5708e@mail.gmail.com>
Message-ID: <BE063FF0656052439E47ADB9BAFB396A01760E83@im04.cincom.com>

 
Thanks Lukas, I already knew about that trick. But it only works when
the component is initially built, and I would need the component to
change its view as the model changes without creating a new component.

So it looks as if I will have to extend the framework somehow.


> -----Original Message-----
> From: smallwiki-bounces at iam.unibe.ch 
> [mailto:smallwiki-bounces at iam.unibe.ch] On Behalf Of Lukas Renggli
> Sent: lundi, 28. septembre 2009 10:44
> To: Magritte, Pier and Related Tools ...
> Subject: Re: Conditional visibility
> 
> > I am still using lr.251 and asking the following questions 
> to the list.
> > 1) Is there a response to my requirement in more recent 
> Magritte versions?
> 
> No.
> 
> > 2) Has this topic already been covered in this list?
> 
> Yes, also see the first question on
> <http://www.lukas-renggli.ch/smalltalk/magritte/faq>.
> 
> Lukas
> 
> --
> Lukas Renggli
> http://www.lukas-renggli.ch
> _______________________________________________
> Magritte, Pier and Related Tools ...
> https://www.iam.unibe.ch/mailman/listinfo/smallwiki
> 


From chaetal at gmail.com  Mon Sep 28 16:13:57 2009
From: chaetal at gmail.com (Dennis Schetinin)
Date: Mon, 28 Sep 2009 18:13:57 +0400
Subject: Pier for Seaside 3.0 plans
Message-ID: <84f77e270909280713l4e4371a4lf976de9155c629d0@mail.gmail.com>

... are there any?

As there are plans of porting (recent) Pier to VW, maybe it's better to port
a newer version (if it exist in any form ofcourse)?

-- 
Dennis Schetinin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.iam.unibe.ch/pipermail/smallwiki/attachments/20090928/f43518b4/attachment.html>

From renggli at gmail.com  Mon Sep 28 16:35:16 2009
From: renggli at gmail.com (Lukas Renggli)
Date: Mon, 28 Sep 2009 16:35:16 +0200
Subject: Pier for Seaside 3.0 plans
In-Reply-To: <84f77e270909280713l4e4371a4lf976de9155c629d0@mail.gmail.com>
References: <84f77e270909280713l4e4371a4lf976de9155c629d0@mail.gmail.com>
Message-ID: <67628d690909280735k6ca619d5q4e345132582746ce@mail.gmail.com>

> ... are there any?

Julian is currently working on integrating Pier with Seaside 3.0.

Lukas

-- 
Lukas Renggli
http://www.lukas-renggli.ch

From ducasse at iam.unibe.ch  Mon Sep 28 22:36:25 2009
From: ducasse at iam.unibe.ch (=?ISO-8859-1?Q?st=E9phane_ducasse?=)
Date: Mon, 28 Sep 2009 22:36:25 +0200
Subject: Pier for Seaside 3.0 plans
In-Reply-To: <84f77e270909280713l4e4371a4lf976de9155c629d0@mail.gmail.com>
References: <84f77e270909280713l4e4371a4lf976de9155c629d0@mail.gmail.com>
Message-ID: <A34F070B-53C4-44BA-B058-52F2EC15340A@iam.unibe.ch>


On Sep 28, 2009, at 4:13 PM, Dennis Schetinin wrote:

> ... are there any?
>
> As there are plans of porting (recent) Pier to VW,

who wants to do that?



> maybe it's better to port a newer version (if it exist in any form  
> ofcourse)?
>
> -- 
> Dennis Schetinin
> _______________________________________________
> Magritte, Pier and Related Tools ...
> https://www.iam.unibe.ch/mailman/listinfo/smallwiki


From keith_hodges at yahoo.co.uk  Tue Sep 29 02:45:40 2009
From: keith_hodges at yahoo.co.uk (keith)
Date: Tue, 29 Sep 2009 01:45:40 +0100
Subject: Conditional visibility
In-Reply-To: <67628d690909280143x52195bd4i810b62a660a5708e@mail.gmail.com>
References: <BE063FF0656052439E47ADB9BAFB396A017211AB@im04.cincom.com>
	<67628d690909280143x52195bd4i810b62a660a5708e@mail.gmail.com>
Message-ID: <95157C61-679E-4737-88F6-7614AFAFAB87@yahoo.co.uk>


On 28 Sep 2009, at 09:43, Lukas Renggli wrote:

>> I am still using lr.251 and asking the following questions to the  
>> list.
>> 1) Is there a response to my requirement in more recent Magritte  
>> versions?
>
> No.
>
>> 2) Has this topic already been covered in this list?
>
> Yes, also see the first question on
> <http://www.lukas-renggli.ch/smalltalk/magritte/faq>.
>
> Lukas
>

The magritte-scriptaculous package does what you want dynamically.

This defines additional properties on the accessor as in the following  
example.

descriptionLicenceForCA
	^MAStringDescription new
		accessor: (#licenceForCA asAccessor visible: #isInCalifornia);
		label: 'Licence # (California)';
		addCondition: [ :v | v size between: 5 and: 10 ] labelled: 'Please  
enter licence number';
		priority: 80;
		bePersisted;
		beRequired;
		beEditable;
		yourself

The sub-accessors are stored as properties of the main accessor.
Additional sub-accessors available are:

#tooltip:
#readonly:
#visible:
#options: (dynamically compiled lists for select lists)
#cssClasses:

Mementos
========

Since the additional accessors refer to methods on the real model  
object you cant use the conventional mementos because they hold the  
state of the form in a dictionary and this has none of the behaviour  
of a real object. To solve this there is an alternative Magritte- 
RealMemento package which supplies almost equivalent mementos which  
use copies of the actual model object to store the cached state.

e.g. MARealCheckedMemento etc.

The main difference being that only one cache of the whole object  
structure is kept in the MARealCheckedMemento, related objects (sub- 
forms aka internal/external edited objects) must use an MAStraitMemento.

A little care has to be taken to make the mementos work with sub-form  
objects. The #copyForMemento method is available for any fine tuning  
of the memento creation process. For example related objects need to  
be copied, and any internal structure within the object needs to be  
consistent. In the example bellow the related objects are explicitly  
copied, and the internal relationship between the personal, and  
billing object is re-established between the copies.

E.g.

copyForRealMemento
		
	super copyForRealMemento.
	shipping := shipping  copyForRealMemento.
	billing := billing  copyForRealMemento.
	personal syncPersonalWithBilling: billing

Components
==========

descriptionState

	^ super descriptionState
		beRequired;
		componentClass: MASelectListComponentSU;
		yourself

MASelectListComponentSU - differs from the standard SelectList in that  
it triggers an scriptaculous update of the entire form, when the  
selection is changed.

In this case, when the user selects the state of california in the  
billing address sub-form, the "Licence for CA" field appears in the  
personal data sub-form.

best regards

Keith

From keith_hodges at yahoo.co.uk  Tue Sep 29 02:55:31 2009
From: keith_hodges at yahoo.co.uk (keith)
Date: Tue, 29 Sep 2009 01:55:31 +0100
Subject: Conditional visibility
In-Reply-To: <95157C61-679E-4737-88F6-7614AFAFAB87@yahoo.co.uk>
References: <BE063FF0656052439E47ADB9BAFB396A017211AB@im04.cincom.com>
	<67628d690909280143x52195bd4i810b62a660a5708e@mail.gmail.com>
	<95157C61-679E-4737-88F6-7614AFAFAB87@yahoo.co.uk>
Message-ID: <EC534441-DAA3-4410-BE38-9A541B5853D2@yahoo.co.uk>

>

erratum

> copyForRealMemento
> 		

postCopyForRealMemento

	super postCopyForRealMemento.

> 	shipping := shipping  copyForRealMemento.
> 	billing := billing  copyForRealMemento.
> 	personal syncPersonalWithBilling: billing

Keith
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.iam.unibe.ch/pipermail/smallwiki/attachments/20090929/f2381ebb/attachment.html>

From chaetal at gmail.com  Tue Sep 29 08:52:33 2009
From: chaetal at gmail.com (Dennis Schetinin)
Date: Tue, 29 Sep 2009 10:52:33 +0400
Subject: Pier for Seaside 3.0 plans
In-Reply-To: <A34F070B-53C4-44BA-B058-52F2EC15340A@iam.unibe.ch>
References: <84f77e270909280713l4e4371a4lf976de9155c629d0@mail.gmail.com> 
	<A34F070B-53C4-44BA-B058-52F2EC15340A@iam.unibe.ch>
Message-ID: <84f77e270909282352x689589f9h9616595a5f1ac29d@mail.gmail.com>

>> ... are there any?
>>
>> As there are plans of porting (recent) Pier to VW,
>
> who wants to do that?

It's in Cincom, AFAIK. I'm not yet sure it's ok to discuss details,
and anyway I don't know much...

-- 
Dennis Schetinin

From keith_hodges at yahoo.co.uk  Tue Sep 29 19:41:35 2009
From: keith_hodges at yahoo.co.uk (keith)
Date: Tue, 29 Sep 2009 18:41:35 +0100
Subject: Conditional visibility
In-Reply-To: <EC534441-DAA3-4410-BE38-9A541B5853D2@yahoo.co.uk>
References: <BE063FF0656052439E47ADB9BAFB396A017211AB@im04.cincom.com>
	<67628d690909280143x52195bd4i810b62a660a5708e@mail.gmail.com>
	<95157C61-679E-4737-88F6-7614AFAFAB87@yahoo.co.uk>
	<EC534441-DAA3-4410-BE38-9A541B5853D2@yahoo.co.uk>
Message-ID: <B0B9ACD0-B897-46DB-B9AD-A40CCBD42943@yahoo.co.uk>

addendum:

The Magritte-Scriptaculous package also defines a third accessor to  
all objects.

in addition to readUsing: and write:Using: there is added update:Using.

This allows the changing component (e.g. MASelectListComponentSU to  
apply any changes to a different selector, so that the model can  
respond dynamically to the change.

E.g.

description010CustomerName

	^ MAStringDescription new
		accessor: (#customerName asAccessor update: #customerNameChangedTo;  
options: #customerNamesAvailable);
		label: 'Customer Name';
		fieldSize: 30;
  		propertyAt: #filter put: #matchLeft;
		componentClass: MAAutoCompletingTextInputComponent;
  		beSearchable;
		bePersisted;
		yourself

best regards

Keith