seaside February 2011

seaside@lists.squeakfoundation.org

60 participants
76 discussions

Reef Error
by recursive68＠gmail.com 22 Feb '11

22 Feb '11

Hi, I was successfully using Reef to build the examples at: http://smallworks.com.ar/en/community/Reef However, I'm now getting an error when using Pharo-1.1.1 with COG (OSX): SystemDictionary(Object)>>error: SystemDictionary(Dictionary)>>errorKeyNotFound [] in SystemDictionary(Dictionary)>>at: SystemDictionary(Dictionary)>>at:ifAbsent: SystemDictionary(Dictionary)>>at: REApplication>>createRoot REApplication>>initialize REApplication class(GRObject class)>>new WARenderLoopMain>>createRoot WARenderLoopMain>>start WASession>>start [] in WASession>>handleFiltered: WAMergedRequestFields>>at:ifAbsent: WASession>>handleFiltered: WAMutualExclusionFilter(WARequestFilter)>>handleFiltered: [] in WAMutualExclusionFilter>>handleFiltered: BlockClosure>>ensure: [] in WAMutex>>critical: [] in Semaphore>>critical: BlockClosure>>ensure: Any ideas ? Thanks

2 1

TFLogin refactored
by Tony Fleig 22 Feb '11

22 Feb '11

Published refactored TFLogin components: http://www.tonyfleig.com/smallthoughts/tflogin/update18feb2011. (TFLogin offers basic user login, registration and account management for Seaside.) Regards, TF

2 1

Seaside 3.0 in Visual Works 7.7.1
by Ivan, Liliana 22 Feb '11

22 Feb '11

Hi The package Grease-Namespace that is loaded by Seaside-All 3.0-7 turns on IEEE floating point arithmetics: Parcel: #('Grease-Namespace') PrintStringCache: (3.0 - 3,mlucas-smith) Version: 3.0 - 3 Post-Load Block: [:package | ObjectMemory registerObject: true withEngineFor: 'IEEEFloatingPointPrimitives'] We are in the process of upgrading our app from VW 7.6 to 7.7.1 and we require division by 0 to raise an exception rather than returning INF. We turned this back off by executing: ObjectMemory registerObject: false withEngineFor: 'IEEEFloatingPointPrimitives' The web site passes all our unit and selenium tests; so we'll leave it turned off - I just wanted to ask : 1. why was it enabled - which parts of seaside require it 2. how/when can it affect us if it's turned off Thanks [cid:image001.gif@01CBD273.C1F12730] LILIANA IVAN Business Technology Enterprise Applications Rand Merchant Bank Telephone +27 112821960 Mobile +27 828476682 http://www.philotechnia.co.za<http://www.philotechnia.co.za/> (Rand Merchant Bank, a division of FirstRand Bank Limited, is an Authorised Financial Services Provider) This e-mail is subject to a disclaimer, available at http://www.rmb.co.za/web/elements.nsf/online/disclaimer-communications.html

1 0

Why - XHTML perfectly good names are now given en Seaside new names
by Fritz Schenk 21 Feb '11

21 Feb '11

I don't think that the decision of using new names (for example paragraph for <p>) is such a good idea. Why was it done? Thanks P.A. Wow the captcha for this post is 'crusade'

15 31

external browser in SeasideTesting ?
by Andrei Vasile Chis 20 Feb '11

20 Feb '11

Hi, Does anyone know if, at the moment, the the external browser functionality is working in SeasideTesting? Cheers, Andrei

2 2

chained WATasks need to be in #children?
by michal-list＠auf.net 19 Feb '11

19 Feb '11

hi - I stumbled upon this surprising property of WATask in Seaside3 today (because it broke my old code): chaining two tasks works only if the second is returned by #children (though non-tasks don't need to be returned by #children). Could someone confirm this, and maybe explain it? Here is a much simplified scenario to reproduce the issue: Task1>>go self call: Task2 new Task2>>go self inform: Time now printString The above doesn't work, failing with WATaskNotDelegated. This on the other hand works: Task>>go self call: Component new Component>>renderContentOn: html html heading: Time now printString Furthermore, as soon as Task1 is modified to answer the Task2 instance in its #children, the first example starts working (regardless of whether it is the same Task2 instance across calls, or a new one on each cycle). So it looks like "chained tasks count as children components" and must be declared as such in #children, while "chained non-task components do not count as children components" (and hence do not need to be answered in #children). I couldn't find any info on that in tutorials or googles, so: is that right? And if so, why is there this asymmetry? thanks, Michal

3 2

[OT] JQuery Terminal Emulator
by Torsten Bergmann 19 Feb '11

19 Feb '11

Nice plugin that may be helpfull for debugging/development: http://terminal.jcubic.pl/#demo When embedded on a (Seaside) page you can evaluate Javascript code. -- Schon gehört? GMX hat einen genialen Phishing-Filter in die Toolbar eingebaut! http://www.gmx.net/de/go/toolbar

2 1

Loading seaside into Pharo core (stable)
by Pablo 19 Feb '11

19 Feb '11

I almost get it but when the script finish, and I go to localhost:8080 I get an error: Error: you are forbidden to access "/" Can someone explain what I have to do to, thanks PD: I use this script and Pharo 1.1.1-stable: "Clean the image" ScriptLoader new cleanUpForProduction. "Load the latest stable seaside" Gofer new squeaksource: 'MetacelloRepository'; package: 'ConfigurationOfSeaside30'; load. ((Smalltalk at: #ConfigurationOfSeaside30 ) project stableVersion) load: #( 'Base' 'Seaside-Adaptors-Comanche' 'JQuery-UI' 'Seaside-Examples'). "Start the server" WAComancheAdaptor startOn: 8085.

4 4

Re: Why - XHTML perfectly good names are now given en Seaside new names
by Stephan Eggermont 18 Feb '11

18 Feb '11

Fritz Schenk wrote: >Robert, depends on the beholder of the language. For people having used html >since 1992, introducing something new is questionable and rather than beautiful >it is a trap for new users. I fail to see where something new is introduced. RFC 1866 might not be from 1992, but it still is 15 years old. Stephan

4 7

Strange script behavior
by Robert Sirois 18 Feb '11

18 Feb '11

Has anyone had a problem where you load a script like this... renderMeOn: html self renderPlatformOn: html. html div script: ((self renderObjectsOn: html) timeout: 1 second). renderObjectsOn: html | js | js := OrderedCollection new. self class objects do: [:o | js add: ((html jQuery: o cssSelector) html: o). ]. ^js javascriptItemsConcatenated. ... and the first query never selects an id? Oddly enough, if I add a meaningless item first: renderObjectsRevisedOn: html | js | js := OrderedCollection with: (html jQuery: 'blah'). self class objects do: [:o | js add: ((html jQuery: o cssSelector) html: o). ]. ^js javascriptItemsConcatenated. It works just fine? RS

2 2

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

seaside February 2011