seaside July 2005

seaside@lists.squeakfoundation.org

33 participants
41 discussions

Re: [Seaside] Template support in Seaside 2.5
by Adnan Zaman 07 Feb '06

07 Feb '06

Hi Alain, Your fix works so thanks for that. However, I see that my html text was not "rendered" as you would expect to see in a browser. All I saw was the same html text that I wanted to be rendered. Not sure if this has anything to do with your another email in which you asked about rendering problem in 2.5. - Adnan >>Date: Tue, 21 Jun 2005 22:42:30 +0200 >>From: Alain Fischer <alain.fischer(a)bluewin.ch> >>Subject: Re: [Seaside] Template support in Seaside 2.5 >>To: "The Squeak Enterprise Aubergines Server - general discussion." >> <seaside(a)lists.squeakfoundation.org> >>Message-ID: <daafe26dd5ae1e8b1e8f16ded82f3fdf(a)bluewin.ch> >>Content-Type: text/plain; charset=ISO-8859-1; format=flowed >> >>Hi Adnan, >> >>For the first issue, I doesn't got a dependency warning. >>Here is how I setup my image: >>- downloaded prebuild image from Avi >>(http://beta4.com/seaside2/Seaside-2.5.zip) >>- used SqueakMap Package Loader to install SmaCC Smalltalk >>Compiler-Compiler-Runtime version: 4 >>- opened a repository on the disk space containing >>SeasideTemplate-AFi.1.mcz >>- loaded this package in the image. >> >>For the second issue here is a fix: >> >>STTTemplate>>compileIn: aClass >> >> | methodNode | >> methodNode _ Compiler new >> compile: self asSmalltalkMethod >> in: aClass >> notifying: nil >> ifFail: [^ nil]. >> aClass >> addSelector: methodNode selector >> withMethod: (methodNode generate: #(0 0 0 0)). >> aClass organization >> classify: methodNode selector >> under: 'generated'. >> "aClass compileProgrammatically: self asSmalltalkMethod classified: >>'generated'" >> "| node method | >> node := Compiler new >> compile: self asSmalltalkMethod >> in: aClass >> notifying: nil >> ifFail: [self error: 'Syntax Error']. >> method := node generate: #(0 0 0 0). >> aClass addSelector: node selector withMethod: method." >> >> >>and it should work. >> >>If you try to run STTTest in Test Runner, you will notice: >>'13 run, 12 passes, 0 expected failures, 0 failures, 1 errors, 0 >>unexpected passes' >> >> >>Alain __________________________________ Yahoo! Mail Mobile Take Yahoo! Mail with you! Check email on your mobile phone. http://mobile.yahoo.com/learn/mail

4 4

Version 2.5b6 for VW is ready
by Bany, Michel 06 Nov '05

06 Nov '05

I just published Seaside 2.5b6 for VW on the public repository. It is based upon Seaside2.5b6-avi.8.mcz 15-Jan-2005 16:15 i.e. the very latest version of 2.5b6. This port includes all the fixes that were discussed here: - file upload - lost halo icons and missing styles Note: I also fixed Seaside 2.5b5 on the public repository. For the fun, I added the Seaside logo on some of the pages. Bundle to load : either SeasideForWebToolkit or SeasideForSwazoo. Enjoy, Michel.

3 4

Nested form tags
by Bany, Michel 04 Nov '05

04 Nov '05

I came accross some weird situation when using a component that is a composition of other components. A subcomponent may in turn be a composition of some more components. To make things even more complex, some deeply nested subcomponents may call some other component. Some of the subcomponents may contain some input html tag and as such will require to be contained in a form html tag. In such a complex construct it may be difficult to decide exactly what component should have the responsibility to generate the form tag. Also, it happened frequently to me that form tags become nested especially when a subcomponent calls another component that happens to be a dialog. Therefore I am making the following suggestion : Seaside should prevent form tags from being nested. It should discard any form tag that occurs within a previously opened form tag. With this in place, I can design a component that needs a form with a form construct knowing that if my component happens to be used inside a form, then my form construct will be discarded. Otherwise my form construct will be preserved. If my component calls a dialog component, the form generated by the dialog component will be discarded if my component happens to be used from inside a form. Michel.

6 6

Re: [ShoreComponents] WebTree, SVG demo and dynamic files
by Jan B. Krejčí 26 Sep '05

26 Sep '05

> From: radoslav hodnicak <rh(a)4096.sk> > > 3) a decorator allowing to preserve page scrolling coordinates between > > submits, so if you have scrolled 5 feet down to find the button you > > want to click, you will see the same page location after the response > > is loaded from server. I saw this in .NET and I like it very much. > > Do you know how is it implemented in .not? Page anchors won't scroll > vertically (at least not in IE). I'd like to have this one too so I'm > willing to implement it. sure. please have a look into enclosed HTML file. It's done this way: 1) reister onScroll handler to <body> 2) in the handler store the coordinates into hidden form fields 3) make sure they are submitted with each clickable component you need 4) put a script at the end of the page causing the scroll position is restored after the page is loaded (or use <body onLoad="..."> that's all > I will try to pick some interesting pieces from my app and add it to the > collection but really, can we pick a better name than ShoreComponents? It > hurts my fingers. Something shorter, like Seadgets or whatever... uh, well, I do not really insist on "ShoreComponents" name, but please consider what impact would that have. How about creating some voting component, expose it in the demo and count other's opinions? :) -- .:jbk:. Jan B. Krejci

11 12

Issue with #checkForCookies
by German Viscuso 31 Aug '05

31 Aug '05

Hello list. When I try to use #checkForCookies in a seaside component I obtain the following message in the Transcript: "Warning: You have tried to set a cookie and a redirect in the same page, responding to the request /seaside/go/myapp. Some browsers will not handle this correctly." Has this issue been already reported or solved? (I'm using VW 7.3 nc and the web toolkit) My test code is: --------------------------------------------------------------------------- renderContentOn: html self session checkForCookies ifTrue: [ self inform: 'Cookies enabled!'. self answer: 'ok'] ifFalse: [ self inform: 'You must have cookies enabled to proceed.'. self answer: nil ] --------------------------------------------------------------------------- Regards and thanks in advance, GV

3 4

RE: [Seaside] trouble with callbacks
by Ramon Leon 01 Aug '05

01 Aug '05

Show code. > -----Original Message----- > From: seaside-bounces(a)lists.squeakfoundation.org > [mailto:seaside-bounces@lists.squeakfoundation.org] On Behalf > Of Jason Rogers > Sent: Friday, July 29, 2005 12:45 PM > To: Seaside List > Subject: [Seaside] trouble with callbacks > > I am getting an error while trying to go to the next page of > a BatchedList (subclass of ShoreBatchedList): 'Components not > found while processing callbacks: #(an > ATHEntryAttributesTableBatchedList)'. > I have implemented #children which returns an instance > variable array of instance variable subcomponents (so I feel > certain the subcomponents are the same instance at callback time). > > Does anyone know what else could be wrong? > > I could provide the code if necessary. > > -- > Jason Rogers > > "I am crucified with Christ: nevertheless I live; yet not I, > but Christ liveth in me: and the life which I now live in the > flesh I live by the faith of the Son of God, who loved me, > and gave himself for me." > Galatians 2:20 > _______________________________________________ > Seaside mailing list > Seaside(a)lists.squeakfoundation.org > http://lists.squeakfoundation.org/listinfo/seaside >

2 2

trouble with callbacks
by Jason Rogers 29 Jul '05

29 Jul '05

I am getting an error while trying to go to the next page of a BatchedList (subclass of ShoreBatchedList): 'Components not found while processing callbacks: #(an ATHEntryAttributesTableBatchedList)'. I have implemented #children which returns an instance variable array of instance variable subcomponents (so I feel certain the subcomponents are the same instance at callback time). Does anyone know what else could be wrong? I could provide the code if necessary. -- Jason Rogers "I am crucified with Christ: nevertheless I live; yet not I, but Christ liveth in me: and the life which I now live in the flesh I live by the faith of the Son of God, who loved me, and gave himself for me." Galatians 2:20

1 0

test
by Jason Rogers 27 Jul '05

27 Jul '05

The traffic here has been so lax. This is a test to make sure the lists are working. -- Jason Rogers "I am crucified with Christ: nevertheless I live; yet not I, but Christ liveth in me: and the life which I now live in the flesh I live by the faith of the Son of God, who loved me, and gave himself for me." Galatians 2:20

2 1

ShoreComponents test cases -- worth doing?
by David Shaffer 23 Jul '05

23 Jul '05

I've started looking at writing tests for the ShoreComponents (the name is fine for me, by the way). One problem I've been having is the "canBeRoot" check. In SeasideTesting components that don't answer "true" to this method can still be the root of a "test" application. ShoreComponents breaks this with: ShoreComponent>>new "Returns a new instance of the receiver" self canBeRoot ifTrue: [ ^ super new ] ifFalse: [ self error: 'use parent:'. ]. This makes writing a test case for any ShoreComponent very inconvenient. It seems to me to be more of a "self check" rather than a necessity. I understand that some components _require_ a parent but this is independent of the #canBeRoot mechanism (which, despite its name, is really just a flag to make the config tool's list shorter). So, rather than me hacking a workaround into SeasideTesting would you consider adding #mustBeChild to the class side of ShoreComponent (default to false) and check it instead of canBeRoot. That would, of course, mean that you'd have to go through and pick the components that require a parent and set their #mustBeChild to true. Of course this is only reasonable if most components don't actually require a parent...which I assume they don't. If I'm wrong here then I'd be happy to hear other suggestions. David

4 7

Accessing configuration values from the configuration
by William E Harford 23 Jul '05

23 Jul '05

If there any way to access a site's user inputted configuration values from a subclass of WASystemConfiguration ? What I am trying to do is present the site administrator with a series of questions in the config biased on answers that were given in the config. For example In this application there might be 3 components. The administrator would put the number 3 for a config value than save the config. The config might than ask for the 3 component classes and there names. I have only been able to retrieve the default values for a SystemConfiguration using. self localConfiguration valueAt: #numberOfContainers. I have to admit even though I have read a couple notes on seasides config component I still don't fully understand how it works. Thanks Will -- William E Harford <seaside(a)harford.org>

4 10

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

seaside July 2005