From squeakman at gmail.com Sat Aug 1 00:35:48 2009 From: squeakman at gmail.com (Squeaker) Date: Sat Aug 1 00:36:14 2009 Subject: [Seaside] Greetings, My questions concerns the generation of the in an html file. The html that is generated by Seaside is: I would like to generate: the "; initialize()" is the new bit. How do I do this? I have looked into WAHtmlRoot but I think it is the wrong path. Thanks, Frank From boris at deepcovelabs.com Sat Aug 1 00:51:13 2009 From: boris at deepcovelabs.com (Boris Popov) Date: Sat Aug 1 00:51:35 2009 Subject: [Seaside] self session addLoadScript: 'initialize();' should have the same effect. -Boris (via BlackBerry) ----- Original Message ----- From: seaside-bounces@lists.squeakfoundation.org To: seaside@lists.squeakfoundation.org Sent: Fri Jul 31 17:35:48 2009 Subject: [Seaside] in an html file. The html that is generated by Seaside is: I would like to generate: the "; initialize()" is the new bit. How do I do this? I have looked into WAHtmlRoot but I think it is the wrong path. Thanks, Frank _______________________________________________ seaside mailing list seaside@lists.squeakfoundation.org http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.squeakfoundation.org/pipermail/seaside/attachments/20090731/4032a8e3/attachment.htm From pierce at netmemetic.com Sat Aug 1 10:53:47 2009 From: pierce at netmemetic.com (Pierce Ng) Date: Sat Aug 1 11:26:46 2009 Subject: [Seaside] Mime type static files Kom In-Reply-To: References: Message-ID: <20090801105347.GK24289@beastie.netmemetic.com> On Sat, Jul 25, 2009 at 01:24:03PM +0200, Francois Stephany wrote: > When serving static files with Kom, the Content-type is always set on > 'text/html'. I've committed a new version of KomHttpServer that should > correct this bug. As i'm not familiar with Kom, can someone have a quick > look at it? On the 3rd-from-last line of HttpResponse class>>fromStream: changing "aStream contentType" to "aStream mimeType" does it for me. From pierce at netmemetic.com Sat Aug 1 10:53:47 2009 From: pierce at netmemetic.com (Pierce Ng) Date: Sat Aug 1 11:26:49 2009 Subject: [Seaside] Mime type static files Kom In-Reply-To: References: Message-ID: <20090801105347.GK24289@beastie.netmemetic.com> On Sat, Jul 25, 2009 at 01:24:03PM +0200, Francois Stephany wrote: > When serving static files with Kom, the Content-type is always set on > 'text/html'. I've committed a new version of KomHttpServer that should > correct this bug. As i'm not familiar with Kom, can someone have a quick > look at it? On the 3rd-from-last line of HttpResponse class>>fromStream: changing "aStream contentType" to "aStream mimeType" does it for me. From john.chludzinski at gmail.com Sun Aug 2 07:43:36 2009 From: john.chludzinski at gmail.com (John Chludzinski) Date: Sun Aug 2 07:43:40 2009 Subject: [Seaside] Problems with Seaside 2.9 libraries Message-ID: <27a48dd60908020043v1975137el241c5413b402945f@mail.gmail.com> I'm porting code from 2.8 to 2.9. I have a chat app (currently model on CTChat) using Comet. I also have an app that uses SUTreeReport (using Scriptaculous). If I add the libraries in the following order, the app using SUTreeReport works: addLibrary: CTLibrary; addLibrary: MMAFileLibrary; addLibrary: MMALibrary; addLibrary: JQDevelopmentLibrary; addLibrary: PTDevelopmentLibrary; addLibrary: SUComponentLibrary; addLibrary: SUDevelopmentLibrary; If I add libraries in the following order my chat app works: addLibrary: PTDeploymentLibrary; addLibrary: CTLibrary; addLibrary: JQDeploymentLibrary; addLibrary: MMAFileLibrary; addLibrary: SUDeploymentLibrary; addLibrary: SUComponentLibrary; addLibrary: MMALibrary; It looks like PTDeploymentLibrary clashes with something else, likely JQDeploymentLibrary. ---John -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.squeakfoundation.org/pipermail/seaside/attachments/20090802/06c0c152/attachment.htm From renggli at gmail.com Sun Aug 2 09:32:59 2009 From: renggli at gmail.com (Lukas Renggli) Date: Sun Aug 2 09:33:01 2009 Subject: [Seaside] Problems with Seaside 2.9 libraries In-Reply-To: <27a48dd60908020043v1975137el241c5413b402945f@mail.gmail.com> References: <27a48dd60908020043v1975137el241c5413b402945f@mail.gmail.com> Message-ID: <67628d690908020232g616b66a1v82c44c3503e73918@mail.gmail.com> > addLibrary: MMAFileLibrary; > addLibrary: MMALibrary; I don't know the above libraries. > addLibrary: JQDevelopmentLibrary; "JQuery" > addLibrary: PTDevelopmentLibrary; "Protype and ..." > addLibrary: SUComponentLibrary; "Scriptaculous" > addLibrary: SUDevelopmentLibrary; JQuery and Prototype/Scriptaculous do not work together, unless you take extra precautions . If you are porting an existing application you presumably only use Prototype/Scriptaculous, so you should not include JQDevelopmentLibrary; similar if you use JQuery you don't want to include PTDevelopmentLibrary, SUComponentLibrary and SUDevelopmentLibrary. Lukas -- Lukas Renggli http://www.lukas-renggli.ch From damien.cassou at gmail.com Sun Aug 2 11:20:43 2009 From: damien.cassou at gmail.com (Damien Cassou) Date: Sun Aug 2 11:21:06 2009 Subject: [Seaside] The rendering context XHTML document is nil on page refresh Message-ID: <6ac749c10908020420l697679b1mc5107ca7df0a74e7@mail.gmail.com> Hi, sometimes, when I hit the refresh button of my web browser, I get the following stack trace from Seaside. It seems the WARenderingContext has a document which is nil. I'm using pharo1.0beta-10401web09.07.5.image which contains Seaside 2.8a1-pmm.593. Can somebody help me please? Thank you 2 August 2009 1:09:39 pm VM: Win32 - IX86 - NT - Squeak3.10.2 of '5 June 2008' [latest update: #7179] Image: Pharo1.0beta [Latest update: #10401] SecurityManager state: Restricted: false FileAccess: true SocketAccess: true Working Dir E:\Squeak VM 3.11.3 pharo 9.7.5 Trusted Dir E:\Squeak VM 3.11.3 pharo 9.7.5\Parents2 Untrusted Dir C:\Documents and Settings\Parents2\Mes documents\My Squeak UndefinedObject(Object)>>doesNotUnderstand: #openTag:attributes:closed: Receiver: nil Arguments and temporary variables: aMessage: openTag: 'div' attributes: a WAHtmlAttributes closed: false exception: MessageNotUnderstood: receiver of "openTag:attributes:closed:" is ni...etc... resumeValue: nil Receiver's instance variables: nil WADivTag(WATagBrush)>>openTag Receiver: a WADivTag Arguments and temporary variables: Receiver's instance variables: canvas: a WARenderCanvas parent: nil closed: false attributes: a WAHtmlAttributes WADivTag(WATagBrush)>>with: Receiver: a WADivTag Arguments and temporary variables: anObject: [closure] in GdoHeader>>renderContentOn: Receiver's instance variables: canvas: a WARenderCanvas parent: nil closed: false attributes: a WAHtmlAttributes GdoHeader>>renderContentOn: Receiver: a GdoHeader Arguments and temporary variables: html: a WARenderCanvas Receiver's instance variables: decoration: a WAValueHolder contents: a GdoHeader titre: 'Gestion des op?rations' search: a GdoHeaderRecherches chainesARechercher: nil identif: nil gdoMain: a GdoMain GdoHeader(WAPresenter)>>renderWithContext: Receiver: a GdoHeader Arguments and temporary variables: aRenderingContext: a WARenderingContext html: a WARenderCanvas callbacks: a WACallbackRegistry Receiver's instance variables: decoration: a WAValueHolder contents: a GdoHeader titre: 'Gestion des op?rations' search: a GdoHeaderRecherches chainesARechercher: nil identif: nil gdoMain: a GdoMain [] in GdoHeader(WAComponent)>>renderOn: Receiver: a GdoHeader Arguments and temporary variables: aRenderer: a GdoHeader each: a WARenderCanvas Receiver's instance variables: decoration: a WAValueHolder contents: a GdoHeader titre: 'Gestion des op?rations' search: a GdoHeaderRecherches chainesARechercher: nil identif: nil gdoMain: a GdoMain GdoHeader(WAComponent)>>decorationChainDo: Receiver: a GdoHeader Arguments and temporary variables: aBlock: [closure] in GdoHeader(WAComponent)>>renderOn: Receiver's instance variables: decoration: a WAValueHolder contents: a GdoHeader titre: 'Gestion des op?rations' search: a GdoHeaderRecherches chainesARechercher: nil identif: nil gdoMain: a GdoMain GdoHeader(WAComponent)>>renderOn: Receiver: a GdoHeader Arguments and temporary variables: aRenderer: a WARenderCanvas Receiver's instance variables: decoration: a WAValueHolder contents: a GdoHeader titre: 'Gestion des op?rations' search: a GdoHeaderRecherches chainesARechercher: nil identif: nil gdoMain: a GdoMain WARenderCanvas(WACanvas)>>render: Receiver: a WARenderCanvas Arguments and temporary variables: anObject: a GdoHeader Receiver's instance variables: currentBrush: nil parentBrush: nil context: a WARenderingContext callbacks: a WACallbackRegistry [] in GdoMain>>renderContentOn: Receiver: a GdoMain Arguments and temporary variables: html: a WARenderCanvas Receiver's instance variables: decoration: a WAValueHolder contents: a GdoMain gdo: a Gdo chainesEnCours: nil collectOperations: nil mesPreferences: a Dictionary('gdoActionDefaut'->'Initialisation' 'gdoClassement...etc... equipe: an Equipe collectOpRecherchees: nil msgStatus: nil dicoForms: a Dictionary(#bandeau->a GdoHeader #barreOutils->a GdoHeaderToolsBar...etc... dicoBools: a Dictionary(#boolIdent->false #chainesARechercher->nil #selAffectat...etc... dicoIdent: a Dictionary(#chaineIdent->nil #identifiant->'Cassou' #initiales->'J...etc... BlockClosure>>renderOn: Receiver: [closure] in GdoMain>>renderContentOn: Arguments and temporary variables: aRenderer: a WARenderCanvas Receiver's instance variables: outerContext: GdoMain>>renderContentOn: startpc: 111 numArgs: 0 [] in WAFormTag(WATagBrush)>>with: Receiver: a WAFormTag Arguments and temporary variables: anObject: [closure] in GdoMain>>renderContentOn: Receiver's instance variables: canvas: a WARenderCanvas parent: nil closed: true attributes: a WAHtmlAttributes defaultAction: nil BlockClosure>>renderOn: Receiver: [closure] in WAFormTag(WATagBrush)>>with: Arguments and temporary variables: aRenderer: a WARenderCanvas Receiver's instance variables: outerContext: WAFormTag(WATagBrush)>>with: startpc: 46 numArgs: 0 WARenderCanvas(WACanvas)>>nest: Receiver: a WARenderCanvas Arguments and temporary variables: aBlock: [closure] in WAFormTag(WATagBrush)>>with: Receiver's instance variables: currentBrush: nil parentBrush: nil context: a WARenderingContext callbacks: a WACallbackRegistry WAFormTag(WABrush)>>with: Receiver: a WAFormTag Arguments and temporary variables: aBlock: [closure] in WAFormTag(WATagBrush)>>with: Receiver's instance variables: canvas: a WARenderCanvas parent: nil closed: true attributes: a WAHtmlAttributes defaultAction: nil WAFormTag(WATagBrush)>>with: Receiver: a WAFormTag Arguments and temporary variables: anObject: [closure] in GdoMain>>renderContentOn: Receiver's instance variables: canvas: a WARenderCanvas parent: nil closed: true attributes: a WAHtmlAttributes defaultAction: nil WARenderCanvas(WAHtmlCanvas)>>form: Receiver: a WARenderCanvas Arguments and temporary variables: aBlock: [closure] in GdoMain>>renderContentOn: Receiver's instance variables: currentBrush: nil parentBrush: nil context: a WARenderingContext callbacks: a WACallbackRegistry GdoMain>>renderContentOn: Receiver: a GdoMain Arguments and temporary variables: html: a WARenderCanvas Receiver's instance variables: decoration: a WAValueHolder contents: a GdoMain gdo: a Gdo chainesEnCours: nil collectOperations: nil mesPreferences: a Dictionary('gdoActionDefaut'->'Initialisation' 'gdoClassement...etc... equipe: an Equipe collectOpRecherchees: nil msgStatus: nil dicoForms: a Dictionary(#bandeau->a GdoHeader #barreOutils->a GdoHeaderToolsBar...etc... dicoBools: a Dictionary(#boolIdent->false #chainesARechercher->nil #selAffectat...etc... dicoIdent: a Dictionary(#chaineIdent->nil #identifiant->'Cassou' #initiales->'J...etc... GdoMain(WAPresenter)>>renderWithContext: Receiver: a GdoMain Arguments and temporary variables: aRenderingContext: a WARenderingContext html: a WARenderCanvas callbacks: a WACallbackRegistry Receiver's instance variables: decoration: a WAValueHolder contents: a GdoMain gdo: a Gdo chainesEnCours: nil collectOperations: nil mesPreferences: a Dictionary('gdoActionDefaut'->'Initialisation' 'gdoClassement...etc... equipe: an Equipe collectOpRecherchees: nil msgStatus: nil dicoForms: a Dictionary(#bandeau->a GdoHeader #barreOutils->a GdoHeaderToolsBar...etc... dicoBools: a Dictionary(#boolIdent->false #chainesARechercher->nil #selAffectat...etc... dicoIdent: a Dictionary(#chaineIdent->nil #identifiant->'Cassou' #initiales->'J...etc... [] in GdoMain(WAComponent)>>renderOn: Receiver: a GdoMain Arguments and temporary variables: aRenderer: a GdoMain each: a WARenderCanvas Receiver's instance variables: decoration: a WAValueHolder contents: a GdoMain gdo: a Gdo chainesEnCours: nil collectOperations: nil mesPreferences: a Dictionary('gdoActionDefaut'->'Initialisation' 'gdoClassement...etc... equipe: an Equipe collectOpRecherchees: nil msgStatus: nil dicoForms: a Dictionary(#bandeau->a GdoHeader #barreOutils->a GdoHeaderToolsBar...etc... dicoBools: a Dictionary(#boolIdent->false #chainesARechercher->nil #selAffectat...etc... dicoIdent: a Dictionary(#chaineIdent->nil #identifiant->'Cassou' #initiales->'J...etc... GdoMain(WAComponent)>>decorationChainDo: Receiver: a GdoMain Arguments and temporary variables: aBlock: [closure] in GdoMain(WAComponent)>>renderOn: Receiver's instance variables: decoration: a WAValueHolder contents: a GdoMain gdo: a Gdo chainesEnCours: nil collectOperations: nil mesPreferences: a Dictionary('gdoActionDefaut'->'Initialisation' 'gdoClassement...etc... equipe: an Equipe collectOpRecherchees: nil msgStatus: nil dicoForms: a Dictionary(#bandeau->a GdoHeader #barreOutils->a GdoHeaderToolsBar...etc... dicoBools: a Dictionary(#boolIdent->false #chainesARechercher->nil #selAffectat...etc... dicoIdent: a Dictionary(#chaineIdent->nil #identifiant->'Cassou' #initiales->'J...etc... GdoMain(WAComponent)>>renderOn: Receiver: a GdoMain Arguments and temporary variables: aRenderer: a WARenderCanvas Receiver's instance variables: decoration: a WAValueHolder contents: a GdoMain gdo: a Gdo chainesEnCours: nil collectOperations: nil mesPreferences: a Dictionary('gdoActionDefaut'->'Initialisation' 'gdoClassement...etc... equipe: an Equipe collectOpRecherchees: nil msgStatus: nil dicoForms: a Dictionary(#bandeau->a GdoHeader #barreOutils->a GdoHeaderToolsBar...etc... dicoBools: a Dictionary(#boolIdent->false #chainesARechercher->nil #selAffectat...etc... dicoIdent: a Dictionary(#chaineIdent->nil #identifiant->'Cassou' #initiales->'J...etc... WARenderCanvas(WACanvas)>>render: Receiver: a WARenderCanvas Arguments and temporary variables: anObject: a GdoMain Receiver's instance variables: currentBrush: nil parentBrush: nil context: a WARenderingContext callbacks: a WACallbackRegistry [] in [] in WAToolFrame>>renderChildOn: Receiver: a WAToolFrame Arguments and temporary variables: html: a WARenderCanvas Receiver's instance variables: decoration: a WAValueHolder contents: a WAToolFrame root: a GdoMain plugins: a SortedCollection(a WANewSessionToolPlugin a WAConfigureToolPlugin a ...etc... showHalos: a WAValueHolder contents: false processTime: nil renderTime: 344 deprecated: a Set() BlockClosure>>on:do: Receiver: [closure] in [] in WAToolFrame>>renderChildOn: Arguments and temporary variables: exception: WADeprecatedApi handlerAction: [closure] in WAToolFrame>>withDeprecatedHandlerDo: handlerActive: true Receiver's instance variables: outerContext: [] in WAToolFrame>>renderChildOn: startpc: 86 numArgs: 0 WAToolFrame>>withDeprecatedHandlerDo: Receiver: a WAToolFrame Arguments and temporary variables: aBlock: [closure] in [] in WAToolFrame>>renderChildOn: Receiver's instance variables: decoration: a WAValueHolder contents: a WAToolFrame root: a GdoMain plugins: a SortedCollection(a WANewSessionToolPlugin a WAConfigureToolPlugin a ...etc... showHalos: a WAValueHolder contents: false processTime: nil renderTime: 344 deprecated: a Set() [] in WAToolFrame>>renderChildOn: Receiver: a WAToolFrame Arguments and temporary variables: html: a WARenderCanvas Receiver's instance variables: decoration: a WAValueHolder contents: a WAToolFrame root: a GdoMain plugins: a SortedCollection(a WANewSessionToolPlugin a WAConfigureToolPlugin a ...etc... showHalos: a WAValueHolder contents: false processTime: nil renderTime: 344 deprecated: a Set() Time class>>millisecondsToRun: Receiver: Time Arguments and temporary variables: timedBlock: [closure] in WAToolFrame>>renderChildOn: initialMilliseconds: 17986803 Receiver's instance variables: superclass: Magnitude methodDict: a MethodDictionary(#<->a CompiledMethod (2992) #=->a CompiledMethod...etc... format: 134 instanceVariables: #('seconds' 'nanos') organization: ('ansi protocol' < = duration hash hour hour12 hour24 meridianAbb...etc... subclasses: nil name: #Time classPool: nil sharedPools: an OrderedCollection(ChronologyConstants) environment: Smalltalk category: #'Kernel-Chronology' traitComposition: nil localSelectors: nil WAToolFrame>>renderChildOn: Receiver: a WAToolFrame Arguments and temporary variables: html: a WARenderCanvas Receiver's instance variables: decoration: a WAValueHolder contents: a WAToolFrame root: a GdoMain plugins: a SortedCollection(a WANewSessionToolPlugin a WAConfigureToolPlugin a ...etc... showHalos: a WAValueHolder contents: false processTime: nil renderTime: 344 deprecated: a Set() WAToolFrame>>renderContentOn: Receiver: a WAToolFrame Arguments and temporary variables: html: a WARenderCanvas Receiver's instance variables: decoration: a WAValueHolder contents: a WAToolFrame root: a GdoMain plugins: a SortedCollection(a WANewSessionToolPlugin a WAConfigureToolPlugin a ...etc... showHalos: a WAValueHolder contents: false processTime: nil renderTime: 344 deprecated: a Set() WAToolFrame(WAPresenter)>>renderWithContext: Receiver: a WAToolFrame Arguments and temporary variables: aRenderingContext: a WARenderingContext html: a WARenderCanvas callbacks: a WACallbackRegistry Receiver's instance variables: decoration: a WAValueHolder contents: a WAToolFrame root: a GdoMain plugins: a SortedCollection(a WANewSessionToolPlugin a WAConfigureToolPlugin a ...etc... showHalos: a WAValueHolder contents: false processTime: nil renderTime: 344 deprecated: a Set() [] in WARenderContinuation>>processRendering: Receiver: a WARenderContinuation Arguments and temporary variables: < Receiver's instance variables: root: a WAToolFrame states: a WASnapshot context: a WARenderingContext WAToolFrame(WAComponent)>>decorationChainDo: Receiver: a WAToolFrame Arguments and temporary variables: aBlock: [closure] in WARenderContinuation>>processRendering: Receiver's instance variables: decoration: a WAValueHolder contents: a WAToolFrame root: a GdoMain plugins: a SortedCollection(a WANewSessionToolPlugin a WAConfigureToolPlugin a ...etc... showHalos: a WAValueHolder contents: false processTime: nil renderTime: 344 deprecated: a Set() WARenderContinuation>>processRendering: Receiver: a WARenderContinuation Arguments and temporary variables: aResponse: a WAResponse 200 document: a WAHtmlStreamDocument htmlRoot: a WAHtmlRoot Receiver's instance variables: root: a WAToolFrame states: a WASnapshot context: a WARenderingContext [] in WARenderContinuation>>render Receiver: a WARenderContinuation Arguments and temporary variables: response: a WAResponse 200 Receiver's instance variables: root: a WAToolFrame states: a WASnapshot context: a WARenderingContext WARenderContinuation(WASessionContinuation)>>respond: Receiver: a WARenderContinuation Arguments and temporary variables: aBlock: [closure] in WARenderContinuation>>render response: a WAResponse 200 Receiver's instance variables: root: a WAToolFrame states: a WASnapshot context: a WARenderingContext WARenderContinuation>>render Receiver: a WARenderContinuation Arguments and temporary variables: Receiver's instance variables: root: a WAToolFrame states: a WASnapshot context: a WARenderingContext [] in WARenderContinuation>>run Receiver: a WARenderContinuation Arguments and temporary variables: Receiver's instance variables: root: a WAToolFrame states: a WASnapshot context: a WARenderingContext BlockClosure>>on:do: Receiver: [closure] in WARenderContinuation>>run Arguments and temporary variables: exception: an ExceptionSet handlerAction: [closure] in WARenderContinuation>>withNotificationHandler: handlerActive: true Receiver's instance variables: outerContext: WARenderContinuation>>run startpc: 26 numArgs: 0 --- The full stack --- UndefinedObject(Object)>>doesNotUnderstand: #openTag:attributes:closed: WADivTag(WATagBrush)>>openTag WADivTag(WATagBrush)>>with: GdoHeader>>renderContentOn: GdoHeader(WAPresenter)>>renderWithContext: [] in GdoHeader(WAComponent)>>renderOn: GdoHeader(WAComponent)>>decorationChainDo: GdoHeader(WAComponent)>>renderOn: WARenderCanvas(WACanvas)>>render: [] in GdoMain>>renderContentOn: BlockClosure>>renderOn: [] in WAFormTag(WATagBrush)>>with: BlockClosure>>renderOn: WARenderCanvas(WACanvas)>>nest: WAFormTag(WABrush)>>with: WAFormTag(WATagBrush)>>with: WARenderCanvas(WAHtmlCanvas)>>form: GdoMain>>renderContentOn: GdoMain(WAPresenter)>>renderWithContext: [] in GdoMain(WAComponent)>>renderOn: GdoMain(WAComponent)>>decorationChainDo: GdoMain(WAComponent)>>renderOn: WARenderCanvas(WACanvas)>>render: [] in [] in WAToolFrame>>renderChildOn: BlockClosure>>on:do: WAToolFrame>>withDeprecatedHandlerDo: [] in WAToolFrame>>renderChildOn: Time class>>millisecondsToRun: WAToolFrame>>renderChildOn: WAToolFrame>>renderContentOn: WAToolFrame(WAPresenter)>>renderWithContext: [] in WARenderContinuation>>processRendering: WAToolFrame(WAComponent)>>decorationChainDo: WARenderContinuation>>processRendering: [] in WARenderContinuation>>render WARenderContinuation(WASessionContinuation)>>respond: WARenderContinuation>>render [] in WARenderContinuation>>run BlockClosure>>on:do: - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - WARenderContinuation>>withNotificationHandler: WARenderContinuation>>run WARenderLoop>>run WARenderLoopMain>>start: WASession>>start: [] in WASession>>performRequest: Dictionary>>at:ifAbsent: WASession>>performRequest: [] in [] in [] in WASession>>responseForRequest: BlockClosure>>on:do: [] in WASession>>withErrorHandler: BlockClosure>>on:do: WASession>>withErrorHandler: [] in [] in WASession>>responseForRequest: BlockClosure>>on:do: WACurrentSession class(WADynamicVariable class)>>use:during: [] in WASession>>responseForRequest: [] in WASession>>withEscapeContinuation: EscapeContinuation class(Continuation class)>>currentDo: WASession>>withEscapeContinuation: WASession>>responseForRequest: [] in WASession>>incomingRequest: BlockClosure>>on:do: [] in [] in [] in WAProcessMonitor>>critical:ifError: BlockClosure>>ensure: [] in [] in WAProcessMonitor>>critical:ifError: [] in BlockClosure>>newProcess -- Damien Cassou http://damiencassou.seasidehosting.st "Lambdas are relegated to relative obscurity until Java makes them popular by not having them." James Iry From aaron at cocoanutstech.com Sun Aug 2 12:52:58 2009 From: aaron at cocoanutstech.com (Aaron Rosenzweig) Date: Sun Aug 2 12:53:01 2009 Subject: [Seaside] launchd headless on Mac OS X Message-ID: <640AD31B-DB73-4242-B8B5-3D8C9C9848CE@cocoanutstech.com> Hello Everyone, I want to run either a Squeak or Pharo VM as a "service" in Mac OS X. That means a few things: * headless operation * automatically loads when the OS boots * safely quits when the OS shuts down * preferably controlled by launchd (launchctl) I'm sorry if some of us are seeing "double" as I started posting first on the Pier list. There was some discussion there that has led me to start experimenting with VMMaker to try and solve my problem but while I'm doing that, I thought I'd post here. This really shouldn't be so hard. I see so many screencasts for Seaside done on a Mac and the one-click installers work so well on a Mac. Surely someone has tackled this issue before. The one-click installers work pretty good. They even launch headless if you start them from the command line with the "-headless" option but only from a terminal window. You can't start them headless from launchd because it seems they are Carbon apps which require some minimal communication with the window server during startup. Sounds like a contradicted myself but it's true. If you have a login session, fire up terminal, then run headless that works. If you try to use launchd you get an error like this: -headless: kCGErrorRangeCheck : Window Server communications from outside of session allowed for root and console user only So, I'm about to try the Pier one-click installer and use VMMaker to produce a "unix-only" source tree that I compile that does not rely on Carbon to see what will happen. Any help appreciated :-) Thanks in advance, -- Aaron From cputney at wiresong.ca Sun Aug 2 14:14:20 2009 From: cputney at wiresong.ca (Colin Putney) Date: Sun Aug 2 14:14:26 2009 Subject: [Seaside] launchd headless on Mac OS X In-Reply-To: <640AD31B-DB73-4242-B8B5-3D8C9C9848CE@cocoanutstech.com> References: <640AD31B-DB73-4242-B8B5-3D8C9C9848CE@cocoanutstech.com> Message-ID: <73E51FD5-562D-490B-9FED-5E9D382B38F9@wiresong.ca> On 2-Aug-09, at 5:52 AM, Aaron Rosenzweig wrote: > So, I'm about to try the Pier one-click installer and use VMMaker to > produce a "unix-only" source tree that I compile that does not rely > on Carbon to see what will happen. No need to compile yourself; try the unix vm here: http://www.squeakvm.org/unix/ . Launch with -vm-display-null instead of headless. Colin From aaron at cocoanutstech.com Sun Aug 2 14:23:29 2009 From: aaron at cocoanutstech.com (Aaron Rosenzweig) Date: Sun Aug 2 14:23:33 2009 Subject: [Seaside] Re: Headless on Mac OS X In-Reply-To: References: <86fxcc6h5u.fsf@blue.stonehenge.com> <4D5D67FB-2D84-4A86-97BE-CA11D70406A4@cocoanutstech.com> <38F82E1E-222E-46C0-92E1-9310E6224DC4@cocoanutstech.com> Message-ID: Stop the presses! I got it working :-) Sorry for cross posting. I'll write it up in more detail later, on my own Pier installation even! But here's the rough notes at the moment: 1) I'm on Mac OS 10.4 Tiger on Intel 2) I downloaded the Pier 1.2 one-click install from www.piercms.com that is running on Pharo VM I believe 3) I enabled LSBackgroundOnly = true in the "info.plist" file inside the .app bundle resources - The purpose of this is to not let the app startup in the dock... I had somehow missed this before because my dock is normally hidden. 4) Needed to use "-headless" option as that is the only one that works with this vm. The -vm-display-null option does not work. Here's the command I use to load into launchd: sudo launchctl load /Library/LaunchDaemons/com.squeak.seaside.plist (That also shows the path to where I put the custom plist file in for launchd). Below is the contents of the launchd plist file Disabled Label com.squeak.seaside OnDemand ProgramArguments /Applications/Smalltalk/Pier-1.2.app/Contents/MacOS/Squeak VM Opt -headless /Applications/Smalltalk/Pier-1.2.app/Contents/Resources/ pier.image ServiceIPC StandardErrorPath /Library/Logs/com.squeak.seaside.log StandardOutPath /Library/Logs/com.squeak.seaside.log UserName root Again, I'll write more later, and clarify once I get Pier set up on the desired server Mac. Thanks everyone, -- Aaron From merlyn at stonehenge.com Sun Aug 2 16:04:55 2009 From: merlyn at stonehenge.com (Randal L. Schwartz) Date: Sun Aug 2 16:04:57 2009 Subject: [Seaside] launchd headless on Mac OS X In-Reply-To: <73E51FD5-562D-490B-9FED-5E9D382B38F9@wiresong.ca> (Colin Putney's message of "Sun, 2 Aug 2009 07:14:20 -0700") References: <640AD31B-DB73-4242-B8B5-3D8C9C9848CE@cocoanutstech.com> <73E51FD5-562D-490B-9FED-5E9D382B38F9@wiresong.ca> Message-ID: <86bpmy2od4.fsf@blue.stonehenge.com> >>>>> "Colin" == Colin Putney writes: Colin> No need to compile yourself; try the unix vm here: Colin> http://www.squeakvm.org/unix/ . Colin> Launch with -vm-display-null instead of headless. Is there a version of this with the closure upgrades yet? -- Randal L. Schwartz - Stonehenge Consulting Services, Inc. - +1 503 777 0095 Smalltalk/Perl/Unix consulting, Technical writing, Comedy, etc. etc. See http://methodsandmessages.vox.com/ for Smalltalk and Seaside discussion From ramiro at diaztrepat.name Sun Aug 2 17:23:23 2009 From: ramiro at diaztrepat.name (Ramiro Diaz Trepat) Date: Sun Aug 2 17:23:28 2009 Subject: [Seaside] 2.9 alpha 4 Message-ID: <313dbe090908021023m4f73c3d4n40334348cc655c6c@mail.gmail.com> Hi, I wanted to install Seaside 2.9 to give it a try. I got a Pharo dev image, and I downloaded the Monticello Configurations file from builder.seaside.st This mcm failed during the installation process. Then I tried the .cs and that one did install without problems, but when I access http://localhost:8080/seaside I get a response from Kom, as if it were unrelated to Seaside. Cheers r -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.squeakfoundation.org/pipermail/seaside/attachments/20090802/4a039b77/attachment.htm From lewis at mail.msen.com Sun Aug 2 18:20:51 2009 From: lewis at mail.msen.com (David T. Lewis) Date: Sun Aug 2 18:34:47 2009 Subject: [Seaside] launchd headless on Mac OS X In-Reply-To: <86bpmy2od4.fsf@blue.stonehenge.com> References: <640AD31B-DB73-4242-B8B5-3D8C9C9848CE@cocoanutstech.com> <73E51FD5-562D-490B-9FED-5E9D382B38F9@wiresong.ca> <86bpmy2od4.fsf@blue.stonehenge.com> Message-ID: <20090802182051.GA3626@shell.msen.com> On Sun, Aug 02, 2009 at 09:04:55AM -0700, Randal L. Schwartz wrote: > >>>>> "Colin" == Colin Putney writes: > > Colin> No need to compile yourself; try the unix vm here: > Colin> http://www.squeakvm.org/unix/ . > > Colin> Launch with -vm-display-null instead of headless. > > Is there a version of this with the closure upgrades yet? Yes, Ian's latest Unix VM build (3.10-6) on http://www.squeakvm.org/unix/ includes closure support. Dave From merlyn at stonehenge.com Sun Aug 2 18:39:36 2009 From: merlyn at stonehenge.com (Randal L. Schwartz) Date: Sun Aug 2 18:39:40 2009 Subject: [Seaside] launchd headless on Mac OS X In-Reply-To: <20090802182051.GA3626@shell.msen.com> (David T. Lewis's message of "Sun, 2 Aug 2009 14:20:51 -0400") References: <640AD31B-DB73-4242-B8B5-3D8C9C9848CE@cocoanutstech.com> <73E51FD5-562D-490B-9FED-5E9D382B38F9@wiresong.ca> <86bpmy2od4.fsf@blue.stonehenge.com> <20090802182051.GA3626@shell.msen.com> Message-ID: <86ws5m12mv.fsf@blue.stonehenge.com> >>>>> "David" == David T Lewis writes: David> Yes, Ian's latest Unix VM build (3.10-6) on David> http://www.squeakvm.org/unix/ includes closure support. Wonderful! Sorry for needing to ask... I've been distracted with $dayjobnotsmalltalk for a bit. -- Randal L. Schwartz - Stonehenge Consulting Services, Inc. - +1 503 777 0095 Smalltalk/Perl/Unix consulting, Technical writing, Comedy, etc. etc. See http://methodsandmessages.vox.com/ for Smalltalk and Seaside discussion From philippe.marschall at gmail.com Sun Aug 2 19:17:34 2009 From: philippe.marschall at gmail.com (Philippe Marschall) Date: Sun Aug 2 19:17:36 2009 Subject: [Seaside] 2.9 alpha 4 In-Reply-To: <313dbe090908021023m4f73c3d4n40334348cc655c6c@mail.gmail.com> References: <313dbe090908021023m4f73c3d4n40334348cc655c6c@mail.gmail.com> Message-ID: <66666f210908021217p66a1a551kc8e47b21a1aeecfd@mail.gmail.com> 2009/8/2 Ramiro Diaz Trepat : > Hi, > I wanted to install Seaside 2.9 to give it a try. > I got a Pharo dev image, and I downloaded the Monticello Configurations file > from builder.seaside.st > This mcm failed during the installation process. > Then I tried the .cs and that one did install without problems, but when I > access > http://localhost:8080/seaside > I get a response from Kom, as if it were unrelated to Seaside. > Cheers Can you be a bit more specific about the errors you encounter? Cheers Philippe From ramiro at diaztrepat.name Sun Aug 2 20:51:26 2009 From: ramiro at diaztrepat.name (Ramiro Diaz Trepat) Date: Sun Aug 2 20:51:29 2009 Subject: [Seaside] 2.9 alpha 4 In-Reply-To: <66666f210908021217p66a1a551kc8e47b21a1aeecfd@mail.gmail.com> References: <313dbe090908021023m4f73c3d4n40334348cc655c6c@mail.gmail.com> <66666f210908021217p66a1a551kc8e47b21a1aeecfd@mail.gmail.com> Message-ID: <313dbe090908021351h7ed3c6ecj36bb0a51f6eecb16@mail.gmail.com> Thanks for replying Philippe, I got it working in the end.I had to start the Seaside Control Panel and add the WAComancheAdaptor. It seems that that last step is not performed but the installatoion .cs. I can be more specific of course, but if you try it you will see what is going on. In any case, here are the steps to reproduce: Case 1: 1. Download the latest versoin of Pharo DEV 2. Download the Monticello Configurations file from builder.seaside.st 3. Open a Monticello browser and load it. You will get an error during installation. I didn't debug, I went and tried the script. Case 2: 1. Download the latest version of Pharo DEV 2. Download the .cs scritp to install Seaside 2.9 from builder.seaside.st 3. Install it. 4. Execute: WAKom startOn: 8080 5. Point your browser to: http://localhost:8080/seaside You will get an error from Kom. On the same image, I successfuly loaded Seaside 2.9 using ScriptLoader loadSeaside29 Then, the Seaside Control Panel opens and you can add the adaptor that is not added by default. Since you can choose which server to use in the nifty tool provided in builder.seaside.st, I believe it would be good that the adapter for it is created after finishing the installation. Cheers r. On Sun, Aug 2, 2009 at 8:17 PM, Philippe Marschall < philippe.marschall@gmail.com> wrote: > 2009/8/2 Ramiro Diaz Trepat : > > Hi, > > I wanted to install Seaside 2.9 to give it a try. > > I got a Pharo dev image, and I downloaded the Monticello Configurations > file > > from builder.seaside.st > > This mcm failed during the installation process. > > Then I tried the .cs and that one did install without problems, but when > I > > access > > http://localhost:8080/seaside > > I get a response from Kom, as if it were unrelated to Seaside. > > Cheers > > Can you be a bit more specific about the errors you encounter? > > Cheers > Philippe > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.squeakfoundation.org/pipermail/seaside/attachments/20090802/7af422ce/attachment-0001.htm From jfitzell at gmail.com Mon Aug 3 01:04:24 2009 From: jfitzell at gmail.com (Julian Fitzell) Date: Mon Aug 3 01:04:27 2009 Subject: [Seaside] The rendering context XHTML document is nil on page refresh In-Reply-To: <6ac749c10908020420l697679b1mc5107ca7df0a74e7@mail.gmail.com> References: <6ac749c10908020420l697679b1mc5107ca7df0a74e7@mail.gmail.com> Message-ID: <4a5dbd430908021804r59bfe00bj77c520381eff912f@mail.gmail.com> Can you show GdoHeader>>renderContentOn: ? Can you define "sometimes"? Do you mean certain pages do it and certain pages don't or, on any given page, sometimes it happens and sometimes not? Julian On Sun, Aug 2, 2009 at 4:20 AM, Damien Cassou wrote: > Hi, > > sometimes, when I hit the refresh button of my web browser, I get the > following stack trace from Seaside. It seems the WARenderingContext > has a document which is nil. I'm using > pharo1.0beta-10401web09.07.5.image which contains Seaside > 2.8a1-pmm.593. > > Can somebody help me please? > > Thank you From damien.cassou at gmail.com Mon Aug 3 09:00:29 2009 From: damien.cassou at gmail.com (Damien Cassou) Date: Mon Aug 3 09:00:52 2009 Subject: [Seaside] The rendering context XHTML document is nil on page refresh In-Reply-To: <4a5dbd430908021804r59bfe00bj77c520381eff912f@mail.gmail.com> References: <6ac749c10908020420l697679b1mc5107ca7df0a74e7@mail.gmail.com> <4a5dbd430908021804r59bfe00bj77c520381eff912f@mail.gmail.com> Message-ID: <6ac749c10908030200g252a2228y7a53403333b044c7@mail.gmail.com> Hi Julian, On Mon, Aug 3, 2009 at 3:04 AM, Julian Fitzell wrote: > Can you show GdoHeader>>renderContentOn: ? renderContentOn: html html div id: 'bdGlobal'; with: [ html div id: 'bdTitre'; with: [ html heading with: titre ]. self renderRecherche: html. self renderLogo: html ] renderLogo: html html div id: 'bdIdentLogoGlobal'; with: [ html div id: 'bdLogo'; with: [ html image url: (GdoFileLibrary urlOf: #logoLP390x80Png); altText: 'LaPoste' ]] renderRecherche: html html div id: 'bdRechGlobal'; with: [ html div id: 'bdNomRechGlobal'; with: [ html render: self recherches ] ] recherches chainesARechercher = '#none#' ifTrue: [^ nil]. ^ search ifNil: [ search := GdoHeaderRecherches new chainesARechercher: chainesARechercher; gdoMain: gdoMain; yourself ] children ^ {self recherches} > Can you define "sometimes"? Do you mean certain pages do it and > certain pages don't or, on any given page, sometimes it happens and > sometimes not? 1) The first page displays correctly the first time 2) If I hit refresh then I get the error message 3) If instead I go to another page, I never get the message again when clicking on refresh, even if I go back to the first page and hit refresh Does that help? -- Damien Cassou http://damiencassou.seasidehosting.st "Lambdas are relegated to relative obscurity until Java makes them popular by not having them." James Iry From karsten at heeg.de Mon Aug 3 09:30:52 2009 From: karsten at heeg.de (Karsten) Date: Mon Aug 3 09:30:55 2009 Subject: [Seaside] The rendering context XHTML document is nil on page refresh In-Reply-To: <6ac749c10908030200g252a2228y7a53403333b044c7@mail.gmail.com> References: <6ac749c10908020420l697679b1mc5107ca7df0a74e7@mail.gmail.com> <4a5dbd430908021804r59bfe00bj77c520381eff912f@mail.gmail.com> <6ac749c10908030200g252a2228y7a53403333b044c7@mail.gmail.com> Message-ID: <4A76AE4C.4080001@heeg.de> Hi Damien, if i remember correctly this might have something to do with ajax callbacks but i just can't remember the details. The RenderContext's document is nil, probably because it was released. Maybe that info helps. Karsten Damien Cassou wrote: > Hi Julian, > > On Mon, Aug 3, 2009 at 3:04 AM, Julian Fitzell wrote: > >> Can you show GdoHeader>>renderContentOn: ? >> > > renderContentOn: html > html div > id: 'bdGlobal'; > with: > [ html div > id: 'bdTitre'; > with: [ html heading with: titre ]. > self renderRecherche: html. > self renderLogo: html ] > > renderLogo: html > html div > id: 'bdIdentLogoGlobal'; > with: > [ html div > id: 'bdLogo'; > with: > [ html image > url: (GdoFileLibrary > urlOf: #logoLP390x80Png); > altText: 'LaPoste' ]] > > renderRecherche: html > html div > id: 'bdRechGlobal'; > with: > [ html div > id: 'bdNomRechGlobal'; > with: [ html render: self recherches ] ] > > > recherches > chainesARechercher = '#none#' ifTrue: [^ nil]. > ^ search ifNil: > [ search := GdoHeaderRecherches new > chainesARechercher: chainesARechercher; > gdoMain: gdoMain; > yourself ] > > children > ^ {self recherches} > > >> Can you define "sometimes"? Do you mean certain pages do it and >> certain pages don't or, on any given page, sometimes it happens and >> sometimes not? >> > > 1) The first page displays correctly the first time > 2) If I hit refresh then I get the error message > 3) If instead I go to another page, I never get the message again when > clicking on refresh, even if I go back to the first page and hit > refresh > > Does that help? > > -- Karsten Kusche - Dipl.Inf. - karsten@heeg.de Tel: +49 3496 21 43 29 Georg Heeg eK - K?then Handelsregister: Amtsgericht Dortmund A 12812 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.squeakfoundation.org/pipermail/seaside/attachments/20090803/02913eba/attachment.htm From karsten at heeg.de Mon Aug 3 12:47:59 2009 From: karsten at heeg.de (Karsten) Date: Mon Aug 3 12:48:23 2009 Subject: [Seaside] [Seaside 2.9 VW] WARenderLoopContinuation>>updateRoot: Message-ID: <4A76DC7F.1010400@heeg.de> Hi, I've noticed that WARenderLoopContinuation>>updateRoot: only visits the visible presenters. This doesn't take into account #lightbox:, where the caller is still visible, even though the Lightbox decoration is a delegation. The result is that the caller's updateRoot: is not called and the caller's stylesheets are not loaded. Maybe WAVisiblePresenterGuide>>shouldVisitNextPresenterOf: should ask the decoration if its next presenter is visible or not. Kind Regards Karsten -- Karsten Kusche - Dipl.Inf. - karsten@heeg.de Tel: +49 3496 21 43 29 Georg Heeg eK - K?then Handelsregister: Amtsgericht Dortmund A 12812 From jfitzell at gmail.com Mon Aug 3 16:34:22 2009 From: jfitzell at gmail.com (Julian Fitzell) Date: Mon Aug 3 16:34:25 2009 Subject: [Seaside] The rendering context XHTML document is nil on page refresh In-Reply-To: <6ac749c10908030200g252a2228y7a53403333b044c7@mail.gmail.com> References: <6ac749c10908020420l697679b1mc5107ca7df0a74e7@mail.gmail.com> <4a5dbd430908021804r59bfe00bj77c520381eff912f@mail.gmail.com> <6ac749c10908030200g252a2228y7a53403333b044c7@mail.gmail.com> Message-ID: <4a5dbd430908030934k9c40c8ew9bf04dccf73fd60f@mail.gmail.com> Hi Damien, I don't see an obvious problem, I'm afraid. As Karsten mentions, a RequestContext has its variables nil'ed out when it is finished and its response is returned, and that's the most likely cause of the document being nil, but I certainly don't see anything in the code you posted that would explain that. Usually that is caused by somebody storing the canvas and using it again in another request (like, an ajax rendering callback). So make sure you're not storing the canvas or request context anywhere and that any ajax callbacks are using the rendering that is passed to the block, and not the one in the rendering method (Slime checks for the latter). If neither of those is your problem, try stripping your code down to the simplest example that demonstrates the problem and someone will take a look at it. Julian On Mon, Aug 3, 2009 at 2:00 AM, Damien Cassou wrote: > Hi Julian, > > On Mon, Aug 3, 2009 at 3:04 AM, Julian Fitzell wrote: >> Can you show GdoHeader>>renderContentOn: ? > > renderContentOn: html > ? ? ? html div > ? ? ? ? ? ? ? ? ? ? ? id: 'bdGlobal'; > ? ? ? ? ? ? ? ? ? ? ? with: > ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? [ html div > ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? id: 'bdTitre'; > ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? with: [ html heading with: titre ]. > ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? self renderRecherche: html. > ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? self renderLogo: html ] > > renderLogo: html > ? ? ? html div > ? ? ? ? ? ? ? id: 'bdIdentLogoGlobal'; > ? ? ? ? ? ? ? with: > ? ? ? ? ? ? ? ? ? ? ? [ html div > ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? id: 'bdLogo'; > ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? with: > ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? [ html image > ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? url: (GdoFileLibrary > urlOf: #logoLP390x80Png); > ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? altText: 'LaPoste' ]] > > renderRecherche: html > ? ? ? html div > ? ? ? ? ? ? ? id: 'bdRechGlobal'; > ? ? ? ? ? ? ? with: > ? ? ? ? ? ? ? ? ? ? ? [ html div > ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? id: 'bdNomRechGlobal'; > ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? with: [ html render: self recherches ] ] > > > recherches > ? ? ? chainesARechercher = '#none#' ifTrue: [^ nil]. > ? ? ? ^ search ifNil: > ? ? ? ? ? ? ? [ search := GdoHeaderRecherches new > ? ? ? ? ? ? ? ? ? ? ? chainesARechercher: chainesARechercher; > ? ? ? ? ? ? ? ? ? ? ? gdoMain: gdoMain; > ? ? ? ? ? ? ? ? ? ? ? yourself ] > > children > ? ? ? ^ {self recherches} > >> Can you define "sometimes"? Do you mean certain pages do it and >> certain pages don't or, on any given page, sometimes it happens and >> sometimes not? > > 1) The first page displays correctly the first time > 2) If I hit refresh then I get the error message > 3) If instead I go to another page, I never get the message again when > clicking on refresh, even if I go back to the first page and hit > refresh > > Does that help? > > -- > Damien Cassou > http://damiencassou.seasidehosting.st > > "Lambdas are relegated to relative obscurity until Java makes them > popular by not having them." James Iry > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside > From jfitzell at gmail.com Mon Aug 3 16:42:10 2009 From: jfitzell at gmail.com (Julian Fitzell) Date: Mon Aug 3 16:42:12 2009 Subject: [Seaside] [Seaside 2.9 VW] WARenderLoopContinuation>>updateRoot: In-Reply-To: <4A76DC7F.1010400@heeg.de> References: <4A76DC7F.1010400@heeg.de> Message-ID: <4a5dbd430908030942j34b8a578x71959996e45a1638@mail.gmail.com> I think #isDelegation is probably being overloaded with too much meaning, alright. I think we need to clarify what that message means and the go from there. Can you file an issue for this? Julian On Mon, Aug 3, 2009 at 5:47 AM, Karsten wrote: > Hi, > > I've noticed that WARenderLoopContinuation>>updateRoot: only visits the > visible presenters. This doesn't take into account #lightbox:, where the > caller is still visible, even though the Lightbox decoration is a > delegation. The result is that the caller's updateRoot: is not called and > the caller's stylesheets are not loaded. > > Maybe WAVisiblePresenterGuide>>shouldVisitNextPresenterOf: should ask the > decoration if its next presenter is visible or not. > > Kind Regards > Karsten > > > -- > Karsten Kusche - Dipl.Inf. - karsten@heeg.de > Tel: +49 3496 21 43 29 > Georg Heeg eK - K?then > Handelsregister: Amtsgericht Dortmund A 12812 > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside > From stephane.ducasse at free.fr Sun Aug 2 15:32:00 2009 From: stephane.ducasse at free.fr (stephane ducasse) Date: Mon Aug 3 17:17:18 2009 Subject: [Seaside] choiceDialog specifying the default selected item? Message-ID: <309CA71B-128E-43CB-AC46-EB4D502B4D9C@free.fr> Hi all is there a way to specify the default selected item in a ChoiceDialog? Should I really build my own Dialog :( ? Stef I did the following but I get a Request handling aborted :( -------------- next part -------------- A non-text attachment was scrubbed... Name: EnhChoiceDialog.1.cs Type: application/octet-stream Size: 1210 bytes Desc: not available Url : http://lists.squeakfoundation.org/pipermail/seaside/attachments/20090802/d4a66a1e/EnhChoiceDialog.1.obj -------------- next part -------------- From philippe.marschall at gmail.com Mon Aug 3 18:27:06 2009 From: philippe.marschall at gmail.com (Philippe Marschall) Date: Mon Aug 3 18:27:07 2009 Subject: [Seaside] 2.9 alpha 4 In-Reply-To: <313dbe090908021351h7ed3c6ecj36bb0a51f6eecb16@mail.gmail.com> References: <313dbe090908021023m4f73c3d4n40334348cc655c6c@mail.gmail.com> <66666f210908021217p66a1a551kc8e47b21a1aeecfd@mail.gmail.com> <313dbe090908021351h7ed3c6ecj36bb0a51f6eecb16@mail.gmail.com> Message-ID: <66666f210908031127o5af73378wfd2edc5571083ab5@mail.gmail.com> 2009/8/2 Ramiro Diaz Trepat : > Thanks for replying Philippe, I got it working in the end. > I had to start the Seaside Control Panel and add the WAComancheAdaptor. Ah this bug, sorry about that, others have bumped into this before you. It happens when you start Seaside with WAKom. > It > seems that that last step is not performed but the installatoion .cs. > I can be more specific of course, but if you try it you will see what is > going on. ?In any case, here are the steps to reproduce: > Case 1: > 1. Download the latest versoin of Pharo DEV > 2. Download the Monticello Configurations file from builder.seaside.st > 3. Open a Monticello browser and load it. > You will get an error during installation. ?I didn't debug, I went and tried > the script. > Case 2: > 1. Download the latest version of Pharo DEV > 2. Download the .cs scritp to install Seaside 2.9 from builder.seaside.st > 3. Install it. > 4. Execute: ?WAKom startOn: 8080 > 5. Point your browser to: ? http://localhost:8080/seaside > You will get an error from Kom. > > On the same image, I successfuly loaded Seaside 2.9 using > ScriptLoader loadSeaside29 > Then, the Seaside Control Panel opens and you can add the adaptor that is > not added by default. > Since you can choose which server to use in the nifty tool provided in > builder.seaside.st, I believe it would be good that the adapter for it is > created after finishing the installation. I opened a bug: http://code.google.com/p/seaside/issues/detail?id=421 Cheers Philippe From renggli at gmail.com Mon Aug 3 18:29:30 2009 From: renggli at gmail.com (Lukas Renggli) Date: Mon Aug 3 18:29:33 2009 Subject: [Seaside] choiceDialog specifying the default selected item? In-Reply-To: <309CA71B-128E-43CB-AC46-EB4D502B4D9C@free.fr> References: <309CA71B-128E-43CB-AC46-EB4D502B4D9C@free.fr> Message-ID: <67628d690908031129p2662f0c9j1c131346d87d6173@mail.gmail.com> Hi Stef > is there a way to specify the default selected item in a ChoiceDialog? > Should I really build my own Dialog :( ? Yes, that seems to be an omission. I guess nobody ever used that thing, except for some tutorials. > I did the following but I get a Request handling aborted :( It looks like in your code you send #options:seletion: to the instance, while it is only defined in the class. I integrated your (fixed) changes in the latest Seaside 2.9 code. http://code.google.com/p/seaside/issues/detail?id=422 Lukas -- Lukas Renggli http://www.lukas-renggli.ch From stan.shepherd414 at gmail.com Mon Aug 3 21:55:59 2009 From: stan.shepherd414 at gmail.com (stan shepherd) Date: Mon Aug 3 21:56:02 2009 Subject: [Seaside] ignore Code Critics rules per class. (Was Re: Slime in one click image) In-Reply-To: <67628d690907310412t5a8b19f0sb2901125c28ee0d1@mail.gmail.com> References: <24736487.post@talk.nabble.com> <4a5dbd430907300912r449f9c73w5521c223195124f9@mail.gmail.com> <102261990907310000x1562970fic70e635c632e9b29@mail.gmail.com> <67628d690907310412t5a8b19f0sb2901125c28ee0d1@mail.gmail.com> Message-ID: <24798870.post@talk.nabble.com> Hi, When I subclass WASession, I get 'Class not referenced' from code critics; there's one instance of the problem for each method in the class. I've tried to set a pragma to ignore the rule as per: http://www.lukas-renggli.ch/blog/ignoring-lint-rules , but can't get it to ignore: currentNeatSheet ^ currentNeatSheet or currentNeatSheet ^ currentNeatSheet or currentNeatSheet ^ currentNeatSheet But none of these succeed in bypassing the rule. Can anyone point me at what I'm doing wrong please? ...Stan Lukas Renggli wrote: > > Sorry for the omission. I've quickly back-ported Slime to Seaside 2.8 > and loaded it into the one-click image. You can either download a new > image from the website, or simply load the latest Slime code from > . > > Cheers, > Lukas > > > -- View this message in context: http://www.nabble.com/Slime-in-one-click-image-tp24736487p24798870.html Sent from the Squeak - Seaside mailing list archive at Nabble.com. From damien.cassou at gmail.com Tue Aug 4 09:03:17 2009 From: damien.cassou at gmail.com (Damien Cassou) Date: Tue Aug 4 09:03:41 2009 Subject: [Seaside] The rendering context XHTML document is nil on page refresh In-Reply-To: <4a5dbd430908030934k9c40c8ew9bf04dccf73fd60f@mail.gmail.com> References: <6ac749c10908020420l697679b1mc5107ca7df0a74e7@mail.gmail.com> <4a5dbd430908021804r59bfe00bj77c520381eff912f@mail.gmail.com> <6ac749c10908030200g252a2228y7a53403333b044c7@mail.gmail.com> <4a5dbd430908030934k9c40c8ew9bf04dccf73fd60f@mail.gmail.com> Message-ID: <6ac749c10908040203x10f42948g3655917e6b010a0e@mail.gmail.com> Hi, On Mon, Aug 3, 2009 at 6:34 PM, Julian Fitzell wrote: > I don't see an obvious problem, I'm afraid. As Karsten mentions, a > RequestContext has its variables nil'ed out when it is finished and > its response is returned, and that's the most likely cause of the > document being nil, but I certainly don't see anything in the code you > posted that would explain that. Usually that is caused by somebody > storing the canvas and using it again in another request (like, an > ajax rendering callback). Thank you and Karsten for you explanations. I will come back to you later with the solution or a small piece of code. Thank you -- Damien Cassou http://damiencassou.seasidehosting.st "Lambdas are relegated to relative obscurity until Java makes them popular by not having them." James Iry From mark at markross.de Tue Aug 4 16:42:05 2009 From: mark at markross.de (Mark Ross) Date: Tue Aug 4 16:42:12 2009 Subject: [Seaside] Generated links (anchor -> callback) In-Reply-To: References:

Message-ID: <1249404125.6751.2.camel@scaleo> Hi James, Sorry for replying a month later - I had a bad accident with my push-bike. > As Dale mentioned, GemStone doesn't quite handle the block the way you > expect. The workaround is to help the compiler see that this is really > a new context. I generally do this with a new method: > > renderContentOn: canvas > > 1 to: 10 do: [:index | self renderIndex: index on: canvas]. > % > renderIndex: anInteger on: aCanvas > > aCanvas anchor > callback: [self value: anInteger]; > with: anInteger. > aCanvas space. > % > > Because the refactoring tends to give smaller, better named methods, I > generally don't find this GemStone "bug" to be much of a burden. Thanks for the hint - this seems to be a very convenient and in a propgramming manner clear way to implement the links :) Best regards Mark From stephan at stack.nl Wed Aug 5 06:54:22 2009 From: stephan at stack.nl (Stephan Eggermont) Date: Wed Aug 5 06:54:26 2009 Subject: [Seaside] Seaside one-click 2.8.4 experience on windows Message-ID: - I downloaded the 2.8.4 one-click yesterday on windows, and noticed the right mouse button doesn't work by default. - I managed to get the OBBrowser to not handle ctrl-s and not ask for author initials. Haven't been able to reproduce, though. - I installed Atom-avi-7 from squeaksource. Its dependency on XMLDomParser is not shown. Shouldn't the one-click have XML support? - I tried to periodically load a feed with Scheduler,but that seems to be incompatible with closure images. Stephan From marianopeck at gmail.com Wed Aug 5 18:40:50 2009 From: marianopeck at gmail.com (Mariano Martinez Peck) Date: Wed Aug 5 18:40:53 2009 Subject: [Seaside] Questions about double (or more) request Message-ID: Hi! I am trying to understand how seaside handle double or more, request. I started with the simpler example: WACounter, which has: renderContentOn: html Transcript show: 'renderContentOn'; cr. html heading: count. html anchor callback: [ self increase ]; with: '++'. html space. html anchor callback: [ self decrease ]; with: '--' increase Transcript show: 'increase'; cr. count := count + 1. (Delay forSeconds: 3) wait. Now, suppose the counter is in 0. I click one time on '++' and wait and I see in the transcipt: increase renderContentOn Ok, perfect. New session, 0 again. Now I click on '++' but before it finish, I click 3 times more (very quickly). I can see this in the transcipt: increase increase increase increase renderContentOn So, as I can see: 1) the callback (self increase in this case) is called for every click. 2) the renderContentOn isn't call until all of the self increase of that session finish. 3) when renderContentOn is called, the webpage shows the number 1 (which is correct instead of 4). Now the questions are: a) how does seaside can do that? every request has its own variables? I think this has to be with continuations but I want to be sure. b) which of all of the self increase is the one that is finally rendered? the first one? c) how can I handle the famous double commit? suppose now self increase does something in a relational database (like persisting an order or something like that) and I don't want it to get persisted twice or I don't want an error because of duplicated PK. Thanks for the help. I just want to understand how this is handle! best, Mariano -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.squeakfoundation.org/pipermail/seaside/attachments/20090805/1d97a761/attachment.htm From Smalltalk at JGFoster.net Wed Aug 5 18:53:09 2009 From: Smalltalk at JGFoster.net (James Foster) Date: Wed Aug 5 18:53:19 2009 Subject: [Seaside] Questions about double (or more) request In-Reply-To: References: Message-ID: <1060FC6C-17B8-4422-B89A-99307E116C5B@JGFoster.net> Hi Mariano, This is covered in chapter 7 of my tutorial (see http://seaside.gemstone.com/tutorial/) . James On Aug 5, 2009, at 11:40 AM, Mariano Martinez Peck wrote: > Hi! I am trying to understand how seaside handle double or more, > request. I started with the simpler example: WACounter, which has: > > renderContentOn: html > Transcript show: 'renderContentOn'; cr. > html heading: count. > html anchor > callback: [ self increase ]; > with: '++'. > html space. > html anchor > callback: [ self decrease ]; > with: '--' > > > increase > Transcript show: 'increase'; cr. > count := count + 1. > (Delay forSeconds: 3) wait. > > > Now, suppose the counter is in 0. I click one time on '++' and wait > and I see in the transcipt: > > increase > renderContentOn > > Ok, perfect. New session, 0 again. Now I click on '++' but before it > finish, I click 3 times more (very quickly). I can see this in the > transcipt: > > increase > increase > increase > increase > renderContentOn > > So, as I can see: > > 1) the callback (self increase in this case) is called for every > click. > 2) the renderContentOn isn't call until all of the self increase of > that session finish. > 3) when renderContentOn is called, the webpage shows the number 1 > (which is correct instead of 4). > > Now the questions are: > > a) how does seaside can do that? every request has its own > variables? I think this has to be with continuations but I want to > be sure. > b) which of all of the self increase is the one that is finally > rendered? the first one? > c) how can I handle the famous double commit? suppose now self > increase does something in a relational database (like persisting an > order or something like that) and I don't want it to get persisted > twice or I don't want an error because of duplicated PK. > > Thanks for the help. I just want to understand how this is handle! > > best, > > Mariano > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside From diogenes.moreira at gmail.com Wed Aug 5 18:54:56 2009 From: diogenes.moreira at gmail.com (Diogenes Alberto Moreira) Date: Wed Aug 5 18:55:10 2009 Subject: [Seaside] Questions about double (or more) request In-Reply-To: References: Message-ID: <0CFDFBD8-CA90-40C0-93FA-F36CB79099A0@gmail.com> Bueno te lo contesto en castellano y despues trato de traducirlo. en otro mail (sorry for english speakers, my English is horrible) El punto es este, por cada click que haces se ejecuta un continuation que esta identificado por el _s para la session _k para el componete un numero de &. en punto que cuando vos le asignas un bloque al call back el bloque se lleva el contexto, donde count tiene un valor x, por ejemplo 20 por mas que ejecutes 70 veces, en el contexto del bloque count vale 20 e incease ejecuta count := 20 + 1. por eso en el transcript, te aperece increase, incease, rendercontentOn y el valor que toma count es simpre 21. (para nuestro ejemplo). y por eso lo importante del fixtemp en squeak y la mejora que significo Pharo. cual se ejecuta,,, todos.... pero todos hacen lo mismo... Espero ser claro.... PD: mariano ponele onda y traducilo. Saludos El 05/08/2009, a las 15:40, Mariano Martinez Peck escribi?: > Hi! I am trying to understand how seaside handle double or more, > request. I started with the simpler example: WACounter, which has: > > renderContentOn: html > Transcript show: 'renderContentOn'; cr. > html heading: count. > html anchor > callback: [ self increase ]; > with: '++'. > html space. > html anchor > callback: [ self decrease ]; > with: '--' > > > increase > Transcript show: 'increase'; cr. > count := count + 1. > (Delay forSeconds: 3) wait. > > > Now, suppose the counter is in 0. I click one time on '++' and wait > and I see in the transcipt: > > increase > renderContentOn > > Ok, perfect. New session, 0 again. Now I click on '++' but before it > finish, I click 3 times more (very quickly). I can see this in the > transcipt: > > increase > increase > increase > increase > renderContentOn > > So, as I can see: > > 1) the callback (self increase in this case) is called for every > click. > 2) the renderContentOn isn't call until all of the self increase of > that session finish. > 3) when renderContentOn is called, the webpage shows the number 1 > (which is correct instead of 4). > > Now the questions are: > > a) how does seaside can do that? every request has its own > variables? I think this has to be with continuations but I want to > be sure. > b) which of all of the self increase is the one that is finally > rendered? the first one? > c) how can I handle the famous double commit? suppose now self > increase does something in a relational database (like persisting an > order or something like that) and I don't want it to get persisted > twice or I don't want an error because of duplicated PK. > > Thanks for the help. I just want to understand how this is handle! > > best, > > Mariano > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside From diogenes.moreira at gmail.com Wed Aug 5 19:11:24 2009 From: diogenes.moreira at gmail.com (Diogenes Alberto Moreira) Date: Wed Aug 5 19:11:36 2009 Subject: [Seaside] Questions about double (or more) request In-Reply-To: <0CFDFBD8-CA90-40C0-93FA-F36CB79099A0@gmail.com> References: <0CFDFBD8-CA90-40C0-93FA-F36CB79099A0@gmail.com> Message-ID: The point is that, for every click you do, seaside run a continuation it is identified by the _s _f and a number. At this point when you assign a block to call back the block is leads to the context, where count is a value x, eg 20 if you run 70 times the same continuation, count value will be the same 21 (20 +1). because that, Transcript show increase, increase, increase, renderContentOn and the count value is the same 21. because that, fixtemp is very important in squeak, and Pharo have better behavior. wich click is running all... but al run the same block regard (sorry for my google english). El 05/08/2009, a las 15:54, Diogenes Alberto Moreira escribi?: > Bueno te lo contesto en castellano y despues trato de traducirlo. en > otro mail (sorry for english speakers, my English is horrible) > > El punto es este, por cada click que haces se ejecuta un > continuation que esta identificado por el _s para la session _k para > el componete un numero de &. > en punto que cuando vos le asignas un bloque al call back el bloque > se lleva el contexto, donde count tiene un valor x, por ejemplo 20 > > por mas que ejecutes 70 veces, en el contexto del bloque count vale > 20 e incease ejecuta count := 20 + 1. > > por eso en el transcript, te aperece increase, incease, > rendercontentOn y el valor que toma count es simpre 21. (para > nuestro ejemplo). y por eso lo importante del fixtemp en squeak y la > mejora que significo Pharo. > > cual se ejecuta,,, todos.... pero todos hacen lo mismo... > > Espero ser claro.... > > PD: mariano ponele onda y traducilo. > > Saludos > > El 05/08/2009, a las 15:40, Mariano Martinez Peck escribi?: > >> Hi! I am trying to understand how seaside handle double or more, >> request. I started with the simpler example: WACounter, which has: >> >> renderContentOn: html >> Transcript show: 'renderContentOn'; cr. >> html heading: count. >> html anchor >> callback: [ self increase ]; >> with: '++'. >> html space. >> html anchor >> callback: [ self decrease ]; >> with: '--' >> >> >> increase >> Transcript show: 'increase'; cr. >> count := count + 1. >> (Delay forSeconds: 3) wait. >> >> >> Now, suppose the counter is in 0. I click one time on '++' and wait >> and I see in the transcipt: >> >> increase >> renderContentOn >> >> Ok, perfect. New session, 0 again. Now I click on '++' but before >> it finish, I click 3 times more (very quickly). I can see this in >> the transcipt: >> >> increase >> increase >> increase >> increase >> renderContentOn >> >> So, as I can see: >> >> 1) the callback (self increase in this case) is called for every >> click. >> 2) the renderContentOn isn't call until all of the self increase of >> that session finish. >> 3) when renderContentOn is called, the webpage shows the number 1 >> (which is correct instead of 4). >> >> Now the questions are: >> >> a) how does seaside can do that? every request has its own >> variables? I think this has to be with continuations but I want to >> be sure. >> b) which of all of the self increase is the one that is finally >> rendered? the first one? >> c) how can I handle the famous double commit? suppose now self >> increase does something in a relational database (like persisting >> an order or something like that) and I don't want it to get >> persisted twice or I don't want an error because of duplicated PK. >> >> Thanks for the help. I just want to understand how this is handle! >> >> best, >> >> Mariano >> _______________________________________________ >> seaside mailing list >> seaside@lists.squeakfoundation.org >> http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside > From marianopeck at gmail.com Wed Aug 5 19:45:59 2009 From: marianopeck at gmail.com (Mariano Martinez Peck) Date: Wed Aug 5 19:46:23 2009 Subject: [Seaside] Questions about double (or more) request In-Reply-To: References: <0CFDFBD8-CA90-40C0-93FA-F36CB79099A0@gmail.com> Message-ID: Thanks to all. Now I get it. But...I still wonder how can I manage double request. c) how can I handle the famous double commit? suppose now self increase does something in a relational database (like persisting an order or something like that) and I don't want it to get persisted twice or I don't want an error because of duplicated PK thanks for the help! mariano On Wed, Aug 5, 2009 at 6:11 PM, Diogenes Alberto Moreira < diogenes.moreira@gmail.com> wrote: > The point is that, for every click you do, seaside run a continuation > it is identified by the _s _f and a number. > > At this point when you assign a block to call back the block is > leads to the context, where count is a value x, eg 20 > > if you run 70 times the same continuation, count value will be the same 21 > (20 +1). > > because that, Transcript show increase, increase, increase, > renderContentOn and the count value is the same 21. > > because that, fixtemp is very important in squeak, and Pharo have better > behavior. > > wich click is running all... but al run the same block > > regard (sorry for my google english). > > El 05/08/2009, a las 15:54, Diogenes Alberto Moreira escribi?: > > > Bueno te lo contesto en castellano y despues trato de traducirlo. en otro >> mail (sorry for english speakers, my English is horrible) >> >> El punto es este, por cada click que haces se ejecuta un continuation que >> esta identificado por el _s para la session _k para el componete un numero >> de &. >> en punto que cuando vos le asignas un bloque al call back el bloque se >> lleva el contexto, donde count tiene un valor x, por ejemplo 20 >> >> por mas que ejecutes 70 veces, en el contexto del bloque count vale 20 e >> incease ejecuta count := 20 + 1. >> >> por eso en el transcript, te aperece increase, incease, rendercontentOn y >> el valor que toma count es simpre 21. (para nuestro ejemplo). y por eso lo >> importante del fixtemp en squeak y la mejora que significo Pharo. >> >> cual se ejecuta,,, todos.... pero todos hacen lo mismo... >> >> Espero ser claro.... >> >> PD: mariano ponele onda y traducilo. >> >> Saludos >> >> El 05/08/2009, a las 15:40, Mariano Martinez Peck escribi?: >> >> Hi! I am trying to understand how seaside handle double or more, request. >>> I started with the simpler example: WACounter, which has: >>> >>> renderContentOn: html >>> Transcript show: 'renderContentOn'; cr. >>> html heading: count. >>> html anchor >>> callback: [ self increase ]; >>> with: '++'. >>> html space. >>> html anchor >>> callback: [ self decrease ]; >>> with: '--' >>> >>> >>> increase >>> Transcript show: 'increase'; cr. >>> count := count + 1. >>> (Delay forSeconds: 3) wait. >>> >>> >>> Now, suppose the counter is in 0. I click one time on '++' and wait and I >>> see in the transcipt: >>> >>> increase >>> renderContentOn >>> >>> Ok, perfect. New session, 0 again. Now I click on '++' but before it >>> finish, I click 3 times more (very quickly). I can see this in the >>> transcipt: >>> >>> increase >>> increase >>> increase >>> increase >>> renderContentOn >>> >>> So, as I can see: >>> >>> 1) the callback (self increase in this case) is called for every click. >>> 2) the renderContentOn isn't call until all of the self increase of that >>> session finish. >>> 3) when renderContentOn is called, the webpage shows the number 1 (which >>> is correct instead of 4). >>> >>> Now the questions are: >>> >>> a) how does seaside can do that? every request has its own variables? I >>> think this has to be with continuations but I want to be sure. >>> b) which of all of the self increase is the one that is finally rendered? >>> the first one? >>> c) how can I handle the famous double commit? suppose now self increase >>> does something in a relational database (like persisting an order or >>> something like that) and I don't want it to get persisted twice or I don't >>> want an error because of duplicated PK. >>> >>> Thanks for the help. I just want to understand how this is handle! >>> >>> best, >>> >>> Mariano >>> _______________________________________________ >>> seaside mailing list >>> seaside@lists.squeakfoundation.org >>> http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside >>> >> >> > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.squeakfoundation.org/pipermail/seaside/attachments/20090805/d77d797e/attachment-0001.htm From miguel.coba at gmail.com Wed Aug 5 19:51:17 2009 From: miguel.coba at gmail.com (Miguel Enrique =?ISO-8859-1?Q?Cob=E1?= Martinez) Date: Wed Aug 5 19:51:29 2009 Subject: [Seaside] Questions about double (or more) request In-Reply-To: References: <0CFDFBD8-CA90-40C0-93FA-F36CB79099A0@gmail.com> Message-ID: <1249501877.4231.3.camel@laptop.localdomain> El mi?, 05-08-2009 a las 18:45 -0100, Mariano Martinez Peck escribi?: > Thanks to all. Now I get it. But...I still wonder how can I manage > double request. > > c) how can I handle the famous double commit? suppose now self > increase does something in a relational database (like persisting an > order or something like that) and I don't want it to get persisted > twice or I don't want an error because of duplicated PK > check the isolate: message of seaside (I think that it is from WAComponent) for processing a request just one time. I think it can help you Miguel Cob? > thanks for the help! > > mariano > > On Wed, Aug 5, 2009 at 6:11 PM, Diogenes Alberto Moreira > wrote: > The point is that, for every click you do, seaside run a > continuation > it is identified by the _s _f and a number. > > At this point when you assign a block to call back the block > is > leads to the context, where count is a value x, eg 20 > > if you run 70 times the same continuation, count value will > be the same 21 (20 +1). > > because that, Transcript show increase, increase, increase, > renderContentOn and the count value is the same 21. > > because that, fixtemp is very important in squeak, and Pharo > have better behavior. > > wich click is running all... but al run the same block > > regard (sorry for my google english). > > El 05/08/2009, a las 15:54, Diogenes Alberto Moreira escribi?: > > > > Bueno te lo contesto en castellano y despues trato de > traducirlo. en otro mail (sorry for english speakers, > my English is horrible) > > El punto es este, por cada click que haces se ejecuta > un continuation que esta identificado por el _s para > la session _k para el componete un numero de &. > en punto que cuando vos le asignas un bloque al call > back el bloque se lleva el contexto, donde count tiene > un valor x, por ejemplo 20 > > por mas que ejecutes 70 veces, en el contexto del > bloque count vale 20 e incease ejecuta count := 20 + > 1. > > por eso en el transcript, te aperece increase, > incease, rendercontentOn y el valor que toma count es > simpre 21. (para nuestro ejemplo). y por eso lo > importante del fixtemp en squeak y la mejora que > significo Pharo. > > cual se ejecuta,,, todos.... pero todos hacen lo > mismo... > > Espero ser claro.... > > PD: mariano ponele onda y traducilo. > > Saludos > > El 05/08/2009, a las 15:40, Mariano Martinez Peck > escribi?: > > Hi! I am trying to understand how seaside > handle double or more, request. I started with > the simpler example: WACounter, which has: > > renderContentOn: html > Transcript show: 'renderContentOn'; cr. > html heading: count. > html anchor > callback: [ self increase ]; > with: '++'. > html space. > html anchor > callback: [ self decrease ]; > with: '--' > > > increase > Transcript show: 'increase'; cr. > count := count + 1. > (Delay forSeconds: 3) wait. > > > Now, suppose the counter is in 0. I click one > time on '++' and wait and I see in the > transcipt: > > increase > renderContentOn > > Ok, perfect. New session, 0 again. Now I click > on '++' but before it finish, I click 3 times > more (very quickly). I can see this in the > transcipt: > > increase > increase > increase > increase > renderContentOn > > So, as I can see: > > 1) the callback (self increase in this case) > is called for every click. > 2) the renderContentOn isn't call until all of > the self increase of that session finish. > 3) when renderContentOn is called, the > webpage shows the number 1 (which is correct > instead of 4). > > Now the questions are: > > a) how does seaside can do that? every request > has its own variables? I think this has to be > with continuations but I want to be sure. > b) which of all of the self increase is the > one that is finally rendered? the first one? > c) how can I handle the famous double commit? > suppose now self increase does something in a > relational database (like persisting an order > or something like that) and I don't want it to > get persisted twice or I don't want an error > because of duplicated PK. > > Thanks for the help. I just want to understand > how this is handle! > > best, > > Mariano > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside > > > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside > > > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside From diogenes.moreira at gmail.com Wed Aug 5 20:38:24 2009 From: diogenes.moreira at gmail.com (Diogenes Alberto Moreira) Date: Wed Aug 5 20:38:55 2009 Subject: [Seaside] Questions about double (or more) request In-Reply-To: References: <0CFDFBD8-CA90-40C0-93FA-F36CB79099A0@gmail.com> Message-ID: <3C8E2BC6-3762-47C3-A78C-FCC45C75A649@gmail.com> Puaggggggggggg, relational... Well, you can replace the block for a empty block in the continuation...but this is advance seaside :P, You can extend WAApplication and extend handle request... regards. El 05/08/2009, a las 16:45, Mariano Martinez Peck escribi?: > Thanks to all. Now I get it. But...I still wonder how can I manage > double request. > > c) how can I handle the famous double commit? suppose now self > increase does something in a relational database (like persisting an > order or something like that) and I don't want it to get persisted > twice or I don't want an error because of duplicated PK > > thanks for the help! > > mariano > > On Wed, Aug 5, 2009 at 6:11 PM, Diogenes Alberto Moreira > wrote: > The point is that, for every click you do, seaside run a continuation > it is identified by the _s _f and a number. > > At this point when you assign a block to call back the blk is > leads to the context, where count is a value x, eg 20 > > if you run 70 times the same continuation, count value will be the > same 21 (20 +1). > > because that, Transcript show increase, increase, increase, > renderContentOn and the count value is the same 21. > > because that, fixtemp is very important in squeak, and Pharo have > better behavior. > > wich click is running all... but al run the same block > > regard (sorry for my google english). > > El 05/08/2009, a las 15:54, Diogenes Alberto Moreira escribi?: > > > Bueno te lo contesto en castellano y despues trato de traducirlo. en > otro mail (sorry for english speakers, my English is horrible) > > El punto es este, por cada click que haces se ejecuta un > continuation que esta identificado por el _s para la session _k para > el componete un numero de &. > en punto que cuando vos le asignas un bloque al call back el bloque > se lleva el contexto, donde count tiene un valor x, por ejemplo 20 > > por mas que ejecutes 70 veces, en el contexto del bloque count vale > 20 e incease ejecuta count := 20 + 1. > > por eso en el transcript, te aperece increase, incease, > rendercontentOn y el valor que toma count es simpre 21. (para > nuestro ejemplo). y por eso lo importante del fixtemp en squeak y la > mejora que significo Pharo. > > cual se ejecuta,,, todos.... pero todos hacen lo mismo... > > Espero ser claro.... > > PD: mariano ponele onda y traducilo. > > Saludos > > El 05/08/2009, a las 15:40, Mariano Martinez Peck escribi?: > > Hi! I am trying to understand how seaside handle double or more, > request. I started with the simpler example: WACounter, which has: > > renderContentOn: html > Transcript show: 'renderContentOn'; cr. > html heading: count. > html anchor > callback: [ self increase ]; > with: '++'. > html space. > html anchor > callback: [ self decrease ]; > with: '--' > > > increase > Transcript show: 'increase'; cr. > count := count + 1. > (Delay forSeconds: 3) wait. > > > Now, suppose the counter is in 0. I click one time on '++' and wait > and I see in the transcipt: > > increase > renderContentOn > > Ok, perfect. New session, 0 again. Now I click on '++' but before it > finish, I click 3 times more (very quickly). I can see this in the > transcipt: > > increase > increase > increase > increase > renderContentOn > > So, as I can see: > > 1) the callback (self increase in this case) is called for every > click. > 2) the renderContentOn isn't call until all of the self increase of > that session finish. > 3) when renderContentOn is called, the webpage shows the number 1 > (which is correct instead of 4). > > Now the questions are: > > a) how does seaside can do that? every request has its own > variables? I think this has to be with continuations but I want to > be sure. > b) which of all of the self increase is the one that is finally > rendered? the first one? > c) how can I handle the famous double commit? suppose now self > increase does something in a relational database (like persisting an > order or something like that) and I don't want it to get persisted > twice or I don't want an error because of duplicated PK. > > Thanks for the help. I just want to understand how this is handle! > > best, > > Mariano > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside > > > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside > > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.squeakfoundation.org/pipermail/seaside/attachments/20090805/0b650333/attachment.htm From jfitzell at gmail.com Wed Aug 5 21:56:09 2009 From: jfitzell at gmail.com (Julian Fitzell) Date: Wed Aug 5 21:56:12 2009 Subject: [Seaside] Questions about double (or more) request In-Reply-To: <1249501877.4231.3.camel@laptop.localdomain> References: <0CFDFBD8-CA90-40C0-93FA-F36CB79099A0@gmail.com> <1249501877.4231.3.camel@laptop.localdomain> Message-ID: <4a5dbd430908051456r6c4b481bg3e5227e307862de3@mail.gmail.com> 2009/8/5 Miguel Enrique Cob? Martinez : > El mi?, 05-08-2009 a las 18:45 -0100, Mariano Martinez Peck escribi?: >> Thanks to all. Now I get it. But...I still wonder how can I manage >> double request. >> >> c) how can I handle the famous double commit? suppose now self >> increase does something in a relational database (like persisting an >> order or something like that) and I don't want it to get persisted >> twice or I don't want an error because of duplicated PK >> > > check the isolate: message of seaside (I think that it is from > WAComponent) for processing a request just one time. I think it can help > you As Miguel says, #isolate: is the usual way of handling this. Julian From philippe.marschall at gmail.com Thu Aug 6 05:16:54 2009 From: philippe.marschall at gmail.com (Philippe Marschall) Date: Thu Aug 6 05:16:56 2009 Subject: [Seaside] Questions about double (or more) request In-Reply-To: References: Message-ID: <66666f210908052216p63546c83mf1362d2250a2db91@mail.gmail.com> 2009/8/5 Mariano Martinez Peck : > Hi! I am trying to understand how seaside handle double or more, request. I > started with the simpler example: WACounter, which has: > > renderContentOn: html > ??? Transcript show: 'renderContentOn'; cr. > ??? html heading: count. > ??? html anchor > ??? ??? callback: [ self increase ]; > ??? ??? with: '++'. > ??? html space. > ??? html anchor > ??? ??? callback: [ self decrease ]; > ??? ??? with: '--' > > > increase > ??? Transcript show: 'increase'; cr. > ??? count := count + 1. > ??? (Delay forSeconds: 3) wait. > > > Now, suppose the counter is in 0. I click one time on '++' and wait and I > see in the transcipt: > > increase > renderContentOn > > Ok, perfect. New session, 0 again. Now I click on '++' but before it finish, > I click 3 times more (very quickly). I can see this in the transcipt: > > increase > increase > increase > increase > renderContentOn > > So, as I can see: > > 1) the callback (self increase in this case) is called for every click. > 2) the renderContentOn isn't call until all of the self increase of that > session finish. > 3) when renderContentOn? is called, the webpage shows the number 1 (which is > correct instead of 4). > > Now the questions are: > > a) how does seaside can do that? every request has its own variables? I > think this has to be with continuations but I want to be sure. I order of you observations: 1) You (your browser) makes three callback requests and discards the redirect to render for the first two 2) There is a lock that allows only one request per session at a time, otherwise your components would have to be thread safe 3) state backtracking, before every request the state of the counter is reverted to 0 > b) which of all of the self increase is the one that is finally rendered? > the first one? That should be the last one. The browser should discard processing the previous requests. Disable backtracking to be sure. Cheers Philippe From philippe.marschall at gmail.com Thu Aug 6 05:22:51 2009 From: philippe.marschall at gmail.com (Philippe Marschall) Date: Thu Aug 6 05:22:53 2009 Subject: [Seaside] Seaside one-click 2.8.4 experience on windows In-Reply-To: References: Message-ID: <66666f210908052222m505cebbdr5912a9ff7ccf321a@mail.gmail.com> 2009/8/5 Stephan Eggermont : > - I downloaded the 2.8.4 one-click yesterday on windows, and noticed > the right mouse button doesn't work by default. Probably it's swapped with the middle mouse button. Can you check and file an issue if that is the case? > - I managed to get the OBBrowser to not handle ctrl-s and not ask for > author initials. Haven't been able to reproduce, though. Probably Ctrl+Alt are swapped or duplicated. Can you check and file an issue if that is the case? > - I installed Atom-avi-7 from squeaksource. Its dependency on XMLDomParser > is not shown. Shouldn't the one-click have XML support? I'm not sure. Nothing in the One-Click needs XML parsing support so why should it be there? > - I tried to periodically load a feed with Scheduler,but ?that seems to be > incompatible with closure images. That would be a bug in Atom then. Cheers Philippe From p3anoman at gmail.com Thu Aug 6 11:23:08 2009 From: p3anoman at gmail.com (John McKeon) Date: Thu Aug 6 11:23:12 2009 Subject: [Seaside] Seaside one-click 2.8.4 experience on windows In-Reply-To: <66666f210908052222m505cebbdr5912a9ff7ccf321a@mail.gmail.com> References: <66666f210908052222m505cebbdr5912a9ff7ccf321a@mail.gmail.com> Message-ID: <40f637ec0908060423p72b87aeckf7e2c870dbd57f53@mail.gmail.com> On Thu, Aug 6, 2009 at 1:22 AM, Philippe Marschall < philippe.marschall@gmail.com> wrote: > 2009/8/5 Stephan Eggermont : > > - I downloaded the 2.8.4 one-click yesterday on windows, and noticed > > the right mouse button doesn't work by default. > > Probably it's swapped with the middle mouse button. Can you check and > file an issue if that is the case? If its a Pharo image, Alt-left click should do the trick. > > > > - I managed to get the OBBrowser to not handle ctrl-s and not ask for > > author initials. Haven't been able to reproduce, though. > > Probably Ctrl+Alt are swapped or duplicated. Can you check and file an > issue if that is the case? Don't all images ask for the initials on the first change? I'm no expert, but I assume this is to mark the change sets? In any event it only occurs once in my experience. > > > - I installed Atom-avi-7 from squeaksource. Its dependency on > XMLDomParser > > is not shown. Shouldn't the one-click have XML support? > > I'm not sure. Nothing in the One-Click needs XML parsing support so > why should it be there? > > > - I tried to periodically load a feed with Scheduler,but that seems to > be > > incompatible with closure images. > > That would be a bug in Atom then. > > Cheers > Philippe > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside > -- http://jmck.seasidehosting.st -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.squeakfoundation.org/pipermail/seaside/attachments/20090806/5d1711e5/attachment.htm From marianopeck at gmail.com Thu Aug 6 13:14:36 2009 From: marianopeck at gmail.com (Mariano Martinez Peck) Date: Thu Aug 6 13:14:39 2009 Subject: [Seaside] Questions about double (or more) request In-Reply-To: <66666f210908052216p63546c83mf1362d2250a2db91@mail.gmail.com> References: <66666f210908052216p63546c83mf1362d2250a2db91@mail.gmail.com> Message-ID: On Thu, Aug 6, 2009 at 4:16 AM, Philippe Marschall < philippe.marschall@gmail.com> wrote: > 2009/8/5 Mariano Martinez Peck : > > Hi! I am trying to understand how seaside handle double or more, request. > I > > started with the simpler example: WACounter, which has: > > > > renderContentOn: html > > Transcript show: 'renderContentOn'; cr. > > html heading: count. > > html anchor > > callback: [ self increase ]; > > with: '++'. > > html space. > > html anchor > > callback: [ self decrease ]; > > with: '--' > > > > > > increase > > Transcript show: 'increase'; cr. > > count := count + 1. > > (Delay forSeconds: 3) wait. > > > > > > Now, suppose the counter is in 0. I click one time on '++' and wait and I > > see in the transcipt: > > > > increase > > renderContentOn > > > > Ok, perfect. New session, 0 again. Now I click on '++' but before it > finish, > > I click 3 times more (very quickly). I can see this in the transcipt: > > > > increase > > increase > > increase > > increase > > renderContentOn > > > > So, as I can see: > > > > 1) the callback (self increase in this case) is called for every click. > > 2) the renderContentOn isn't call until all of the self increase of that > > session finish. > > 3) when renderContentOn is called, the webpage shows the number 1 (which > is > > correct instead of 4). > > > > Now the questions are: > > > > a) how does seaside can do that? every request has its own variables? I > > think this has to be with continuations but I want to be sure. > > I order of you observations: > 1) You (your browser) makes three callback requests and discards the > redirect to render for the first two Ok, but which is the "condition" to be discarded? if the callback request is finished but there are others callbacks request then it is discarded? > > 2) There is a lock that allows only one request per session at a time, > otherwise your components would have to be thread safe > 3) state backtracking, before every request the state of the counter > is reverted to 0 that's because there was the method states ? > > > > b) which of all of the self increase is the one that is finally rendered? > > the first one? > > That should be the last one. The browser should discard processing the > previous requests. Disable backtracking to be sure. what do you mean with Disable backtracking ? remove the method states from the component? Thanks! mariano > > > Cheers > Philippe > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.squeakfoundation.org/pipermail/seaside/attachments/20090806/88f18e7e/attachment.htm From stephan at stack.nl Thu Aug 6 14:28:48 2009 From: stephan at stack.nl (stephan@stack.nl) Date: Thu Aug 6 14:28:52 2009 Subject: [Seaside] Re: Seaside one-click 2.8.4 experience on windows Message-ID: <20090806162848.11502ozpl8saj1k4@webmail.stack.nl> >> - I downloaded the 2.8.4 one-click yesterday on windows, and noticed >> the right mouse button doesn't work by default. Hmm, no 3 button mouse available. There is also no preference to do something with mouse buttons. Alt-left click works, but is not acceptable for a multi-button mouse. >> - I managed to get the OBBrowser to not handle ctrl-s and not ask for >> author initials. Haven't been able to reproduce, though. The OBBrowser doesn't handle having empty selections very well. Adding a category doesn't ask for Author initials > I'm not sure. Nothing in the One-Click needs XML parsing support so > why should it be there? Because it used to be there? Or am I mistaken? Was that just in developer images? Groetjes, Stephan From marianopeck at gmail.com Thu Aug 6 14:42:51 2009 From: marianopeck at gmail.com (Mariano Martinez Peck) Date: Thu Aug 6 14:42:53 2009 Subject: [Seaside] Re: Seaside one-click 2.8.4 experience on windows In-Reply-To: <20090806162848.11502ozpl8saj1k4@webmail.stack.nl> References: <20090806162848.11502ozpl8saj1k4@webmail.stack.nl> Message-ID: On Thu, Aug 6, 2009 at 1:28 PM, wrote: > - I downloaded the 2.8.4 one-click yesterday on windows, and noticed >>> the right mouse button doesn't work by default. >>> >> > Hmm, no 3 button mouse available. There is also no preference to do > something with mouse buttons. As you may know, Seaside 2.8.4 one click is based in Pharo. There were some bugs and changes in Pharo some time ago related with this. Could you please download the beta image from Pharo ( http://www.pharo-project.org/pharo-download) and see if you still have the problem? If so, please open a ticket in Pharo because this is not seaside related: http://code.google.com/p/pharo/issues/list > > > Alt-left click works, but is not acceptable for a multi-button mouse. > > - I managed to get the OBBrowser to not handle ctrl-s and not ask for >>> author initials. Haven't been able to reproduce, though. >>> >> > The OBBrowser doesn't handle having empty selections very well. Adding > a category doesn't ask for Author initials Agian, I think this is pharo related and should be asked there. > > > I'm not sure. Nothing in the One-Click needs XML parsing support so >> why should it be there? >> > > Because it used to be there? Or am I mistaken? Was that just in developer > images? > > Groetjes, > Stephan > > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.squeakfoundation.org/pipermail/seaside/attachments/20090806/0b9bba39/attachment.htm From kmr1642 at yahoo.com Thu Aug 6 19:12:32 2009 From: kmr1642 at yahoo.com (Michael Ross-1) Date: Thu Aug 6 19:12:34 2009 Subject: [Seaside] Porting Lively Kernel into Seaside Message-ID: <24852667.post@talk.nabble.com> Hi, I'm trying to get an idea of how much work would be involved in setting up Lively Kernel (LK) to work in Seaside. (I've done a few webpages in Seaside but have no experience enhancing Seaside code). LK is written in JavaScript, so I imagine that the Scriptaculous package might serve as a starting model for porting LK into Seaside, like storing the LK code as is done in the SULibrary class. It's the Seaside linkage that puzzles me - I don't have a clue what's required here. But I want to keep it simple, at least until the LK demo page is working. So right now, when the demo page starts in the browser, I just want Seaside to serve the LK javascript code to the browser and see what happens. Any tips on how to proceed? what to modify? level of difficulty? Comments? Thanks, -Michael Lively Kernel HomePage at Sun Labs http://research.sun.com/projects/lively/ (best viewed with Safari) -- View this message in context: http://www.nabble.com/Porting-Lively-Kernel-into-Seaside-tp24852667p24852667.html Sent from the Squeak - Seaside mailing list archive at Nabble.com. From merlyn at stonehenge.com Thu Aug 6 19:18:12 2009 From: merlyn at stonehenge.com (Randal L. Schwartz) Date: Thu Aug 6 19:18:15 2009 Subject: [Seaside] Porting Lively Kernel into Seaside In-Reply-To: <24852667.post@talk.nabble.com> (Michael Ross's message of "Thu, 6 Aug 2009 12:12:32 -0700 (PDT)") References: <24852667.post@talk.nabble.com> Message-ID: <86prb8sqdn.fsf@blue.stonehenge.com> >>>>> "Michael" == Michael Ross-1 writes: Michael> LK is written in JavaScript, so I imagine that the Michael> Scriptaculous package might serve as a starting model for porting LK Michael> into Seaside, like storing the LK code as is done in the SULibrary Michael> class. It might be closer to MOOTools, which I understand when mixed with Seaside provides a transparent bidirectional proxy somehow. Michael Lucas-Smith has more information on that. -- Randal L. Schwartz - Stonehenge Consulting Services, Inc. - +1 503 777 0095 Smalltalk/Perl/Unix consulting, Technical writing, Comedy, etc. etc. See http://methodsandmessages.vox.com/ for Smalltalk and Seaside discussion From mlucas-smith at cincom.com Thu Aug 6 19:24:51 2009 From: mlucas-smith at cincom.com (Michael Lucas-Smith) Date: Thu Aug 6 19:24:59 2009 Subject: [Seaside] Porting Lively Kernel into Seaside In-Reply-To: <86prb8sqdn.fsf@blue.stonehenge.com> References: <24852667.post@talk.nabble.com> <86prb8sqdn.fsf@blue.stonehenge.com> Message-ID: <4A7B2E03.9020305@cincom.com> Randal L. Schwartz wrote: >>>>>> "Michael" == Michael Ross-1 writes: >>>>>> > > Michael> LK is written in JavaScript, so I imagine that the > Michael> Scriptaculous package might serve as a starting model for porting LK > Michael> into Seaside, like storing the LK code as is done in the SULibrary > Michael> class. > > It might be closer to MOOTools, which I understand when mixed with > Seaside provides a transparent bidirectional proxy somehow. > > Michael Lucas-Smith has more information on that. > > Well, Lively is a bit different to the regular communications that are done with Seaside. So you'd have to decide what it is you want to transmit back and forth and how. But, you can register any kind of callback you want in Seaside and transmit any kind of data in those payloads. You can also make your own WARequestHandler subclass to handle incoming requests if you want too. Seaside is very pluggable that way. Cheers, Michael From thierry.thelliez.tech at gmail.com Thu Aug 6 19:37:21 2009 From: thierry.thelliez.tech at gmail.com (Thierry Thelliez) Date: Thu Aug 6 19:37:24 2009 Subject: [Seaside] Seaside + JQuery example page on port 9090? Message-ID: <42f5c4430908061237h3ab5cbf6vb75bddc63756e671@mail.gmail.com> Having these examples on port 9090 is making access difficult when you are behind a firewall. Could they be moved to a regular port? Thierry From philippe.marschall at gmail.com Thu Aug 6 20:22:58 2009 From: philippe.marschall at gmail.com (Philippe Marschall) Date: Thu Aug 6 20:23:00 2009 Subject: [Seaside] Questions about double (or more) request In-Reply-To: References: <66666f210908052216p63546c83mf1362d2250a2db91@mail.gmail.com> Message-ID: <66666f210908061322y51c924e0x4b53d49613e3333d@mail.gmail.com> 2009/8/6 Mariano Martinez Peck : > > > On Thu, Aug 6, 2009 at 4:16 AM, Philippe Marschall > wrote: >> >> 2009/8/5 Mariano Martinez Peck : >> > Hi! I am trying to understand how seaside handle double or more, >> > request. I >> > started with the simpler example: WACounter, which has: >> > >> > renderContentOn: html >> > ??? Transcript show: 'renderContentOn'; cr. >> > ??? html heading: count. >> > ??? html anchor >> > ??? ??? callback: [ self increase ]; >> > ??? ??? with: '++'. >> > ??? html space. >> > ??? html anchor >> > ??? ??? callback: [ self decrease ]; >> > ??? ??? with: '--' >> > >> > >> > increase >> > ??? Transcript show: 'increase'; cr. >> > ??? count := count + 1. >> > ??? (Delay forSeconds: 3) wait. >> > >> > >> > Now, suppose the counter is in 0. I click one time on '++' and wait and >> > I >> > see in the transcipt: >> > >> > increase >> > renderContentOn >> > >> > Ok, perfect. New session, 0 again. Now I click on '++' but before it >> > finish, >> > I click 3 times more (very quickly). I can see this in the transcipt: >> > >> > increase >> > increase >> > increase >> > increase >> > renderContentOn >> > >> > So, as I can see: >> > >> > 1) the callback (self increase in this case) is called for every click. >> > 2) the renderContentOn isn't call until all of the self increase of that >> > session finish. >> > 3) when renderContentOn? is called, the webpage shows the number 1 >> > (which is >> > correct instead of 4). >> > >> > Now the questions are: >> > >> > a) how does seaside can do that? every request has its own variables? I >> > think this has to be with continuations but I want to be sure. >> >> I order of you observations: >> 1) You (your browser) makes three callback requests and discards the >> redirect to render for the first two > > Ok, but which is the "condition" to be discarded? if the callback request is > finished but there are others callbacks request then it is discarded? Seaside doesn't discard anything, the browser has to. Here's a rough sketch of what happens: - You initiate a content exchange for the "callbacks page". - The browser sends a request. - The Seaside callback machinery kicks in. - Before Seaside can send a response you initiate three more requests resulting in three more requests. Seaside doesn't really have a problem with this, it will process these requests one after another, meaning "executing" the same "callback page" several times and send the redirect response to the render page for each one. However the browser has, he has just one pane where he can render the resonse. So he'll have to discard all but one of the responses. The least bad thing is to discard all but the last one. >> >> 2) There is a lock that allows only one request per session at a time, >> otherwise your components would have to be thread safe >> 3) state backtracking, before every request the state of the counter >> is reverted to 0 > > that's because there was the method states ? Yes >> >> > b) which of all of the self increase is the one that is finally >> > rendered? >> > the first one? >> >> That should be the last one. The browser should discard processing the >> previous requests. Disable backtracking to be sure. > > what do you mean with Disable backtracking ? remove the method states from > the component? Yes Cheers Philippe From pdebruic at gmail.com Thu Aug 6 20:28:11 2009 From: pdebruic at gmail.com (Paul DeBruicker) Date: Thu Aug 6 20:28:12 2009 Subject: [Seaside] Seaside + JQuery example page on port 9090? In-Reply-To: <42f5c4430908061237h3ab5cbf6vb75bddc63756e671@mail.gmail.com> References: <42f5c4430908061237h3ab5cbf6vb75bddc63756e671@mail.gmail.com> Message-ID: In Pharo you can open the Seaside Control Panel under Tools->More...-> Seaside Control Panel. Once its open right click in the upper area and choose Add Adaptor... then choose the type of adaptor and enter the Port you want the Adaptor to run on. The examples will then run on that port On Thu, Aug 6, 2009 at 3:37 PM, Thierry Thelliez wrote: > Having these examples on port 9090 is making access difficult when you > are behind a firewall. Could they be moved to a regular port? > > > Thierry > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside > From thierry.thelliez.tech at gmail.com Thu Aug 6 21:31:43 2009 From: thierry.thelliez.tech at gmail.com (Thierry Thelliez) Date: Thu Aug 6 21:31:46 2009 Subject: [Seaside] Seaside + JQuery example page on port 9090? In-Reply-To: References: <42f5c4430908061237h3ab5cbf6vb75bddc63756e671@mail.gmail.com> Message-ID: <42f5c4430908061431x4eaae4a2xba777e305e449135@mail.gmail.com> I was referring to the links on http://www.seaside.st/about/examples. The ones pointing to jQuery, jQueryUI and Scriptaculous. Thierry From merlyn at stonehenge.com Thu Aug 6 21:43:57 2009 From: merlyn at stonehenge.com (Randal L. Schwartz) Date: Thu Aug 6 21:44:00 2009 Subject: [Seaside] Seaside + JQuery example page on port 9090? In-Reply-To: <42f5c4430908061431x4eaae4a2xba777e305e449135@mail.gmail.com> (Thierry Thelliez's message of "Thu, 6 Aug 2009 15:31:43 -0600") References: <42f5c4430908061237h3ab5cbf6vb75bddc63756e671@mail.gmail.com> <42f5c4430908061431x4eaae4a2xba777e305e449135@mail.gmail.com> Message-ID: <86d478sjmq.fsf@blue.stonehenge.com> >>>>> "Thierry" == Thierry Thelliez writes: Thierry> I was referring to the links on http://www.seaside.st/about/examples. Thierry> The ones pointing to jQuery, jQueryUI and Scriptaculous. Those examples are also in the one-click image, as I recall. Just download the image, and run it locally, if you're on a needlessly impaired internet. -- Randal L. Schwartz - Stonehenge Consulting Services, Inc. - +1 503 777 0095 Smalltalk/Perl/Unix consulting, Technical writing, Comedy, etc. etc. See http://methodsandmessages.vox.com/ for Smalltalk and Seaside discussion From thierry.thelliez.tech at gmail.com Thu Aug 6 22:48:06 2009 From: thierry.thelliez.tech at gmail.com (Thierry Thelliez) Date: Thu Aug 6 22:48:09 2009 Subject: [Seaside] Seaside + JQuery example page on port 9090? In-Reply-To: <86d478sjmq.fsf@blue.stonehenge.com> References: <42f5c4430908061237h3ab5cbf6vb75bddc63756e671@mail.gmail.com> <42f5c4430908061431x4eaae4a2xba777e305e449135@mail.gmail.com> <86d478sjmq.fsf@blue.stonehenge.com> Message-ID: <42f5c4430908061548g43068dd8xc3d6a8b37e1cb275@mail.gmail.com> The one-click image contains some scriptaculous examples (http://localhost:8080/seaside/tests/scriptaculous). But I could not find any jQuery ones. Thierry From steve.hayes at cogentconsulting.com.au Thu Aug 6 22:53:15 2009 From: steve.hayes at cogentconsulting.com.au (Steve Hayes) Date: Thu Aug 6 22:53:41 2009 Subject: [Seaside] Seaside + JQuery example page on port 9090? In-Reply-To: <42f5c4430908061548g43068dd8xc3d6a8b37e1cb275@mail.gmail.com> References: <42f5c4430908061237h3ab5cbf6vb75bddc63756e671@mail.gmail.com> <42f5c4430908061431x4eaae4a2xba777e305e449135@mail.gmail.com> <86d478sjmq.fsf@blue.stonehenge.com> <42f5c4430908061548g43068dd8xc3d6a8b37e1cb275@mail.gmail.com> Message-ID: <010DA3F6-981E-4D11-A022-A150244DC71F@cogentconsulting.com.au> On 07/08/2009, at 8:48 AM, Thierry Thelliez wrote: > The one-click image contains some scriptaculous examples > (http://localhost:8080/seaside/tests/scriptaculous). But I could not > find any jQuery ones. > > Thierry I think the one-click image is for Seaside 2.8.4 and the JQuery examples are in Seaside 2.9 ----- Steve Hayes Cogent Consulting http://www.cogentconsulting.com.au http://iridescenturchin.blogspot.com/ twitter: steve_hayes GTalk: steve.e.hayes Yahoo: steve_e_hayes Mobile: 0403 902 431 SkypeIn: +61 3 9005 6695 Measure you code quality: http://iridescenturchin.blogspot.com/2009/07/what-would-you-like-in-code-quality.html Have more things to do than you have time to do them? Use http://www.runwayapp.com . Runway - where your life takes off. From pdebruic at gmail.com Thu Aug 6 23:02:47 2009 From: pdebruic at gmail.com (Paul DeBruicker) Date: Thu Aug 6 23:03:14 2009 Subject: [Seaside] Seaside + JQuery example page on port 9090? In-Reply-To: <010DA3F6-981E-4D11-A022-A150244DC71F@cogentconsulting.com.au> References: <42f5c4430908061237h3ab5cbf6vb75bddc63756e671@mail.gmail.com> <42f5c4430908061431x4eaae4a2xba777e305e449135@mail.gmail.com> <86d478sjmq.fsf@blue.stonehenge.com> <42f5c4430908061548g43068dd8xc3d6a8b37e1cb275@mail.gmail.com> <010DA3F6-981E-4D11-A022-A150244DC71F@cogentconsulting.com.au> Message-ID: <0A74FBBD-3D0E-4837-BE87-E24372749976@gmail.com> So to load seaside 2.9 get a fresh Pharos beta image and then go to http://builder.seaside.st and download a Monticello load script, configured how you want it, and drag the *.cs file onto the Pharo image and then choose filein. After it's completed you should find the jquery bits you want under localhost:9090/seaside/tests On Aug 6, 2009, at 6:53 PM, Steve Hayes wrote: > > On 07/08/2009, at 8:48 AM, Thierry Thelliez wrote: > >> The one-click image contains some scriptaculous examples >> (http://localhost:8080/seaside/tests/scriptaculous). But I could not >> find any jQuery ones. >> >> Thierry > > I think the one-click image is for Seaside 2.8.4 and the JQuery > examples are in Seaside 2.9 > > ----- > Steve Hayes > > Cogent Consulting > http://www.cogentconsulting.com.au > http://iridescenturchin.blogspot.com/ > twitter: steve_hayes > GTalk: steve.e.hayes > Yahoo: steve_e_hayes > Mobile: 0403 902 431 > SkypeIn: +61 3 9005 6695 > > Measure you code quality: http://iridescenturchin.blogspot.com/2009/07/what-would-you-like-in-code-quality.html > > Have more things to do than you have time to do them? Use http://www.runwayapp.com > . > Runway - where your life takes off. > > > > > > > > > > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside From aldreds at velocitynet.com.au Thu Aug 6 23:35:51 2009 From: aldreds at velocitynet.com.au (Steve Aldred) Date: Thu Aug 6 23:33:42 2009 Subject: [Seaside] Seaside + JQuery example page on port 9090? In-Reply-To: <86d478sjmq.fsf@blue.stonehenge.com> References: <42f5c4430908061237h3ab5cbf6vb75bddc63756e671@mail.gmail.com> <42f5c4430908061431x4eaae4a2xba777e305e449135@mail.gmail.com> <86d478sjmq.fsf@blue.stonehenge.com> Message-ID: <4A7B68D7.4060203@velocitynet.com.au> Randal L. Schwartz wrote: >>>>>> "Thierry" == Thierry Thelliez writes: >>>>>> > > Thierry> I was referring to the links on http://www.seaside.st/about/examples. > Thierry> The ones pointing to jQuery, jQueryUI and Scriptaculous. > > Those examples are also in the one-click image, as I recall. Just download > the image, and run it locally, if you're on a needlessly impaired internet. > Needlessly or not if you're living in a bureaucratic organization with a high level of paranoia filters for other than the most common ports exist. If we want Seaside to be widely accessible, all of the demonstration/example code should be available on port 80 and not needlessly moved to some other port. Let the 'experts' hosting the examples map the request however they like but let those in restricted environments be free to explore all that is out there. Anything that makes it more difficult for people to access information in other than the most obvious way makes it more likely they will give up. cheers Steve From boris at deepcovelabs.com Thu Aug 6 23:36:19 2009 From: boris at deepcovelabs.com (Boris Popov) Date: Thu Aug 6 23:36:25 2009 Subject: [Seaside] Seaside + JQuery example page on port 9090? In-Reply-To: <4A7B68D7.4060203@velocitynet.com.au> References: <42f5c4430908061237h3ab5cbf6vb75bddc63756e671@mail.gmail.com> <42f5c4430908061431x4eaae4a2xba777e305e449135@mail.gmail.com><86d478sjmq.fsf@blue.stonehenge.com> <4A7B68D7.4060203@velocitynet.com.au> Message-ID: I agree. In fact many personal firewall products also block uncommon ports such as this one and only leave things like 80, 443, 25 etc by default. -Boris -- +1.604.689.0322 DeepCove Labs Ltd. 4th floor 595 Howe Street Vancouver, Canada V6C 2T5 http://tinyurl.com/r7uw4 -----Original Message----- From: seaside-bounces@lists.squeakfoundation.org [mailto:seaside-bounces@lists.squeakfoundation.org] On Behalf Of Steve Aldred Sent: Thursday, August 06, 2009 4:36 PM To: Seaside - general discussion Subject: Re: [Seaside] Seaside + JQuery example page on port 9090? Randal L. Schwartz wrote: >>>>>> "Thierry" == Thierry Thelliez writes: >>>>>> > > Thierry> I was referring to the links on http://www.seaside.st/about/examples. > Thierry> The ones pointing to jQuery, jQueryUI and Scriptaculous. > > Those examples are also in the one-click image, as I recall. Just download > the image, and run it locally, if you're on a needlessly impaired internet. > Needlessly or not if you're living in a bureaucratic organization with a high level of paranoia filters for other than the most common ports exist. If we want Seaside to be widely accessible, all of the demonstration/example code should be available on port 80 and not needlessly moved to some other port. Let the 'experts' hosting the examples map the request however they like but let those in restricted environments be free to explore all that is out there. Anything that makes it more difficult for people to access information in other than the most obvious way makes it more likely they will give up. cheers Steve _______________________________________________ seaside mailing list seaside@lists.squeakfoundation.org http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside From merlyn at stonehenge.com Fri Aug 7 00:05:51 2009 From: merlyn at stonehenge.com (Randal L. Schwartz) Date: Fri Aug 7 00:05:55 2009 Subject: [Seaside] Seaside + JQuery example page on port 9090? In-Reply-To: <4A7B68D7.4060203@velocitynet.com.au> (Steve Aldred's message of "Fri, 07 Aug 2009 09:35:51 +1000") References: <42f5c4430908061237h3ab5cbf6vb75bddc63756e671@mail.gmail.com> <42f5c4430908061431x4eaae4a2xba777e305e449135@mail.gmail.com> <86d478sjmq.fsf@blue.stonehenge.com> <4A7B68D7.4060203@velocitynet.com.au> Message-ID: <868whwsd28.fsf@blue.stonehenge.com> >>>>> "Steve" == Steve Aldred writes: Steve> If we want Seaside to be widely accessible, all of the Steve> demonstration/example code should be available on port 80 and not Steve> needlessly moved to some other port. Let the 'experts' hosting the Steve> examples map the request however they like but let those in restricted Steve> environments be free to explore all that is out there. Or, just download an image that lets it all be available locally, as has been suggested at least three times in this thread now. -- Randal L. Schwartz - Stonehenge Consulting Services, Inc. - +1 503 777 0095 Smalltalk/Perl/Unix consulting, Technical writing, Comedy, etc. etc. See http://methodsandmessages.vox.com/ for Smalltalk and Seaside discussion From merlyn at stonehenge.com Fri Aug 7 00:06:46 2009 From: merlyn at stonehenge.com (Randal L. Schwartz) Date: Fri Aug 7 00:06:48 2009 Subject: [Seaside] Seaside + JQuery example page on port 9090? In-Reply-To: (Boris Popov's message of "Thu, 6 Aug 2009 16:36:19 -0700") References: <42f5c4430908061237h3ab5cbf6vb75bddc63756e671@mail.gmail.com> <42f5c4430908061431x4eaae4a2xba777e305e449135@mail.gmail.com> <86d478sjmq.fsf@blue.stonehenge.com> <4A7B68D7.4060203@velocitynet.com.au> Message-ID: <864osksd0p.fsf@blue.stonehenge.com> >>>>> "Boris" == Boris Popov writes: Boris> I agree. In fact many personal firewall products also block uncommon Boris> ports such as this one and only leave things like 80, 443, 25 etc by Boris> default. That's completely nuts. There's nothing more "dangerous" at port 9090 than at port 80. Anyone who constructs such a firewall is hopelessly naive, and if you're using such a product, turn it off, and demand a refund. -- Randal L. Schwartz - Stonehenge Consulting Services, Inc. - +1 503 777 0095 Smalltalk/Perl/Unix consulting, Technical writing, Comedy, etc. etc. See http://methodsandmessages.vox.com/ for Smalltalk and Seaside discussion From dave.bauer at gmail.com Fri Aug 7 00:10:45 2009 From: dave.bauer at gmail.com (Dave Bauer) Date: Fri Aug 7 00:10:48 2009 Subject: [Seaside] Seaside + JQuery example page on port 9090? In-Reply-To: <864osksd0p.fsf@blue.stonehenge.com> References: <42f5c4430908061237h3ab5cbf6vb75bddc63756e671@mail.gmail.com> <42f5c4430908061431x4eaae4a2xba777e305e449135@mail.gmail.com> <86d478sjmq.fsf@blue.stonehenge.com> <4A7B68D7.4060203@velocitynet.com.au> <864osksd0p.fsf@blue.stonehenge.com> Message-ID: <6c4e07720908061710k7b492cebod7eccb6df06af5ce@mail.gmail.com> On Thu, Aug 6, 2009 at 8:06 PM, Randal L. Schwartz wrote: >>>>>> "Boris" == Boris Popov writes: > > Boris> I agree. In fact many personal firewall products also block uncommon > Boris> ports such as this one and only leave things like 80, 443, 25 etc by > Boris> default. > > That's completely nuts. > > There's nothing more "dangerous" at port 9090 than at port 80. > > Anyone who constructs such a firewall is hopelessly naive, and if > you're using such a product, turn it off, and demand a refund. > That is totally besides the point. Unfortunately the Seaside community regularly tries to scare away those interested in the product by giving unhelpful advice. Everyone expects publically available web sites to be on port 80. Anyone who is trying to market a web site product should understand that people will expect stuff to be available on port 80, and also understand the technique to make it available that way. Dave > -- > Randal L. Schwartz - Stonehenge Consulting Services, Inc. - +1 503 777 0095 > > Smalltalk/Perl/Unix consulting, Technical writing, Comedy, etc. etc. > See http://methodsandmessages.vox.com/ for Smalltalk and Seaside discussion > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside > -- Dave Bauer dave@solutiongrove.com http://www.solutiongrove.com From boris at deepcovelabs.com Fri Aug 7 00:20:31 2009 From: boris at deepcovelabs.com (Boris Popov) Date: Fri Aug 7 00:20:39 2009 Subject: [Seaside] Seaside + JQuery example page on port 9090? In-Reply-To: <868whwsd28.fsf@blue.stonehenge.com> References: <42f5c4430908061237h3ab5cbf6vb75bddc63756e671@mail.gmail.com><42f5c4430908061431x4eaae4a2xba777e305e449135@mail.gmail.com><86d478sjmq.fsf@blue.stonehenge.com><4A7B68D7.4060203@velocitynet.com.au> <868whwsd28.fsf@blue.stonehenge.com> Message-ID: Why such fierce opposition to such simple suggestion, I wonder? If it helps attract a single person to Seaside who happens to work in a corporate environment (or using public hotspot, or lame personal firewall, pick one) that ultimately blocks 9090, why not? Online demos exist solely for the purpose of making quick evaluation easier, as soon as you ask someone to download anything, many will run away quicker than you can say "installer". -Boris -- +1.604.689.0322 DeepCove Labs Ltd. 4th floor 595 Howe Street Vancouver, Canada V6C 2T5 http://tinyurl.com/r7uw4 -----Original Message----- From: seaside-bounces@lists.squeakfoundation.org [mailto:seaside-bounces@lists.squeakfoundation.org] On Behalf Of Randal L. Schwartz Sent: Thursday, August 06, 2009 5:06 PM To: Steve Aldred Cc: Seaside - general discussion Subject: Re: [Seaside] Seaside + JQuery example page on port 9090? >>>>> "Steve" == Steve Aldred writes: Steve> If we want Seaside to be widely accessible, all of the Steve> demonstration/example code should be available on port 80 and not Steve> needlessly moved to some other port. Let the 'experts' hosting the Steve> examples map the request however they like but let those in restricted Steve> environments be free to explore all that is out there. Or, just download an image that lets it all be available locally, as has been suggested at least three times in this thread now. -- Randal L. Schwartz - Stonehenge Consulting Services, Inc. - +1 503 777 0095 Smalltalk/Perl/Unix consulting, Technical writing, Comedy, etc. etc. See http://methodsandmessages.vox.com/ for Smalltalk and Seaside discussion _______________________________________________ seaside mailing list seaside@lists.squeakfoundation.org http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside From miguel.coba at gmail.com Fri Aug 7 00:25:27 2009 From: miguel.coba at gmail.com (Miguel Enrique =?ISO-8859-1?Q?Cob=E1?= Martinez) Date: Fri Aug 7 00:25:38 2009 Subject: [Seaside] Seaside + JQuery example page on port 9090? In-Reply-To: <6c4e07720908061710k7b492cebod7eccb6df06af5ce@mail.gmail.com> References: <42f5c4430908061237h3ab5cbf6vb75bddc63756e671@mail.gmail.com> <42f5c4430908061431x4eaae4a2xba777e305e449135@mail.gmail.com> <86d478sjmq.fsf@blue.stonehenge.com> <4A7B68D7.4060203@velocitynet.com.au> <864osksd0p.fsf@blue.stonehenge.com> <6c4e07720908061710k7b492cebod7eccb6df06af5ce@mail.gmail.com> Message-ID: <1249604727.4271.10.camel@laptop.localdomain> El jue, 06-08-2009 a las 20:10 -0400, Dave Bauer escribi?: > On Thu, Aug 6, 2009 at 8:06 PM, Randal L. Schwartz wrote: > >>>>>> "Boris" == Boris Popov writes: > > > > Boris> I agree. In fact many personal firewall products also block uncommon > > Boris> ports such as this one and only leave things like 80, 443, 25 etc by > > Boris> default. > > > > That's completely nuts. > > > > There's nothing more "dangerous" at port 9090 than at port 80. > > > > Anyone who constructs such a firewall is hopelessly naive, and if > > you're using such a product, turn it off, and demand a refund. > > > > That is totally besides the point. Unfortunately the Seaside community > regularly tries to scare away those interested in the product by > giving unhelpful advice. If I am developing web applications and am not capable of configure or find the cause of my net issues I have a deeper problem that I think and it is not fault on the Seaside community. > > Everyone expects publically available web sites to be on port 80. > Anyone who is trying to market a web site product should understand > that people will expect stuff to be available on port 80, and also > understand the technique to make it available that way. > First, the "everyone" your are refering not even know what a port is. they just write a url on the browser or better, go to google, find something and click on a link. NO ports in the process. Second, the fact that an webapps runs in a specific port doesn't says nothing about the quality of the product. Rails runs on port 3000, Tomcat in 8180. That is just a simple random port. When you buy a domain, put your server in production then you can omit the port by running on a standard port (80 for http) but even so, if you are on unix/mac you can't run a server on a port below 1024 so you must use Apache/Lighttpd to listen on 80 and proxy to the other port (3000, 8180, 8080, 9090, etc). To the user the service is running on "port 80" but actually isn't. So don't blame the comunnity, they are trying to help but not resolve *all* your problems. At the end, everyone must do their homework. Miguel Cob? > Dave > > > -- > > Randal L. Schwartz - Stonehenge Consulting Services, Inc. - +1 503 777 0095 > > > > Smalltalk/Perl/Unix consulting, Technical writing, Comedy, etc. etc. > > See http://methodsandmessages.vox.com/ for Smalltalk and Seaside discussion > > _______________________________________________ > > seaside mailing list > > seaside@lists.squeakfoundation.org > > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside > > > > > From marianopeck at gmail.com Fri Aug 7 00:43:29 2009 From: marianopeck at gmail.com (Mariano Martinez Peck) Date: Fri Aug 7 00:43:31 2009 Subject: [Seaside] Questions about double (or more) request In-Reply-To: <66666f210908061322y51c924e0x4b53d49613e3333d@mail.gmail.com> References: <66666f210908052216p63546c83mf1362d2250a2db91@mail.gmail.com> <66666f210908061322y51c924e0x4b53d49613e3333d@mail.gmail.com> Message-ID: Excellent. Now I get it. Thank you so much for your answers. Best, Mariano On Thu, Aug 6, 2009 at 5:22 PM, Philippe Marschall < philippe.marschall@gmail.com> wrote: > 2009/8/6 Mariano Martinez Peck : > > > > > > On Thu, Aug 6, 2009 at 4:16 AM, Philippe Marschall > > wrote: > >> > >> 2009/8/5 Mariano Martinez Peck : > >> > Hi! I am trying to understand how seaside handle double or more, > >> > request. I > >> > started with the simpler example: WACounter, which has: > >> > > >> > renderContentOn: html > >> > Transcript show: 'renderContentOn'; cr. > >> > html heading: count. > >> > html anchor > >> > callback: [ self increase ]; > >> > with: '++'. > >> > html space. > >> > html anchor > >> > callback: [ self decrease ]; > >> > with: '--' > >> > > >> > > >> > increase > >> > Transcript show: 'increase'; cr. > >> > count := count + 1. > >> > (Delay forSeconds: 3) wait. > >> > > >> > > >> > Now, suppose the counter is in 0. I click one time on '++' and wait > and > >> > I > >> > see in the transcipt: > >> > > >> > increase > >> > renderContentOn > >> > > >> > Ok, perfect. New session, 0 again. Now I click on '++' but before it > >> > finish, > >> > I click 3 times more (very quickly). I can see this in the transcipt: > >> > > >> > increase > >> > increase > >> > increase > >> > increase > >> > renderContentOn > >> > > >> > So, as I can see: > >> > > >> > 1) the callback (self increase in this case) is called for every > click. > >> > 2) the renderContentOn isn't call until all of the self increase of > that > >> > session finish. > >> > 3) when renderContentOn is called, the webpage shows the number 1 > >> > (which is > >> > correct instead of 4). > >> > > >> > Now the questions are: > >> > > >> > a) how does seaside can do that? every request has its own variables? > I > >> > think this has to be with continuations but I want to be sure. > >> > >> I order of you observations: > >> 1) You (your browser) makes three callback requests and discards the > >> redirect to render for the first two > > > > Ok, but which is the "condition" to be discarded? if the callback request > is > > finished but there are others callbacks request then it is discarded? > > Seaside doesn't discard anything, the browser has to. Here's a rough > sketch of what happens: > - You initiate a content exchange for the "callbacks page". > - The browser sends a request. > - The Seaside callback machinery kicks in. > - Before Seaside can send a response you initiate three more requests > resulting in three more requests. > > Seaside doesn't really have a problem with this, it will process these > requests one after another, meaning "executing" the same "callback > page" several times and send the redirect response to the render page > for each one. However the browser has, he has just one pane where he > can render the resonse. So he'll have to discard all but one of the > responses. The least bad thing is to discard all but the last one. > > >> > >> 2) There is a lock that allows only one request per session at a time, > >> otherwise your components would have to be thread safe > >> 3) state backtracking, before every request the state of the counter > >> is reverted to 0 > > > > that's because there was the method states ? > > Yes > > >> > >> > b) which of all of the self increase is the one that is finally > >> > rendered? > >> > the first one? > >> > >> That should be the last one. The browser should discard processing the > >> previous requests. Disable backtracking to be sure. > > > > what do you mean with Disable backtracking ? remove the method states > from > > the component? > > Yes > > Cheers > Philippe > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.squeakfoundation.org/pipermail/seaside/attachments/20090806/a248760c/attachment-0001.htm From boris at deepcovelabs.com Fri Aug 7 00:54:52 2009 From: boris at deepcovelabs.com (Boris Popov) Date: Fri Aug 7 00:54:59 2009 Subject: [Seaside] Seaside + JQuery example page on port 9090? In-Reply-To: <1249604727.4271.10.camel@laptop.localdomain> References: <42f5c4430908061237h3ab5cbf6vb75bddc63756e671@mail.gmail.com><42f5c4430908061431x4eaae4a2xba777e305e449135@mail.gmail.com><86d478sjmq.fsf@blue.stonehenge.com><4A7B68D7.4060203@velocitynet.com.au><864osksd0p.fsf@blue.stonehenge.com><6c4e07720908061710k7b492cebod7eccb6df06af5ce@mail.gmail.com> <1249604727.4271.10.camel@laptop.localdomain> Message-ID: Okay, something has definitely gone amiss in this thread ;) The original report was simply that the public demo may not be accessible from certain environments, which is akin to reporting a dead link IMHO. The maintainer of that specific demo site has two options: a) ignore it or b) do something about it. It is well known and documented what that person needs to do to expose Seaside services via port 80 should they choose option b) and wasn't the subject of this thread at all. -Boris -- +1.604.689.0322 DeepCove Labs Ltd. 4th floor 595 Howe Street Vancouver, Canada V6C 2T5 http://tinyurl.com/r7uw4 -----Original Message----- From: seaside-bounces@lists.squeakfoundation.org [mailto:seaside-bounces@lists.squeakfoundation.org] On Behalf Of Miguel Enrique Cob? Martinez Sent: Thursday, August 06, 2009 5:25 PM To: dave.bauer@gmail.com; Seaside - general discussion Subject: Re: [Seaside] Seaside + JQuery example page on port 9090? El jue, 06-08-2009 a las 20:10 -0400, Dave Bauer escribi?: > On Thu, Aug 6, 2009 at 8:06 PM, Randal L. Schwartz wrote: > >>>>>> "Boris" == Boris Popov writes: > > > > Boris> I agree. In fact many personal firewall products also block uncommon > > Boris> ports such as this one and only leave things like 80, 443, 25 etc by > > Boris> default. > > > > That's completely nuts. > > > > There's nothing more "dangerous" at port 9090 than at port 80. > > > > Anyone who constructs such a firewall is hopelessly naive, and if > > you're using such a product, turn it off, and demand a refund. > > > > That is totally besides the point. Unfortunately the Seaside community > regularly tries to scare away those interested in the product by > giving unhelpful advice. If I am developing web applications and am not capable of configure or find the cause of my net issues I have a deeper problem that I think and it is not fault on the Seaside community. > > Everyone expects publically available web sites to be on port 80. > Anyone who is trying to market a web site product should understand > that people will expect stuff to be available on port 80, and also > understand the technique to make it available that way. > First, the "everyone" your are refering not even know what a port is. they just write a url on the browser or better, go to google, find something and click on a link. NO ports in the process. Second, the fact that an webapps runs in a specific port doesn't says nothing about the quality of the product. Rails runs on port 3000, Tomcat in 8180. That is just a simple random port. When you buy a domain, put your server in production then you can omit the port by running on a standard port (80 for http) but even so, if you are on unix/mac you can't run a server on a port below 1024 so you must use Apache/Lighttpd to listen on 80 and proxy to the other port (3000, 8180, 8080, 9090, etc). To the user the service is running on "port 80" but actually isn't. So don't blame the comunnity, they are trying to help but not resolve *all* your problems. At the end, everyone must do their homework. Miguel Cob? > Dave > > > -- > > Randal L. Schwartz - Stonehenge Consulting Services, Inc. - +1 503 777 0095 > > > > Smalltalk/Perl/Unix consulting, Technical writing, Comedy, etc. etc. > > See http://methodsandmessages.vox.com/ for Smalltalk and Seaside discussion > > _______________________________________________ > > seaside mailing list > > seaside@lists.squeakfoundation.org > > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside > > > > > _______________________________________________ seaside mailing list seaside@lists.squeakfoundation.org http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside From aldreds at velocitynet.com.au Fri Aug 7 01:03:12 2009 From: aldreds at velocitynet.com.au (Steve Aldred) Date: Fri Aug 7 01:01:04 2009 Subject: [Seaside] Seaside + JQuery example page on port 9090? In-Reply-To: <1249604727.4271.10.camel@laptop.localdomain> References: <42f5c4430908061237h3ab5cbf6vb75bddc63756e671@mail.gmail.com> <42f5c4430908061431x4eaae4a2xba777e305e449135@mail.gmail.com> <86d478sjmq.fsf@blue.stonehenge.com> <4A7B68D7.4060203@velocitynet.com.au> <864osksd0p.fsf@blue.stonehenge.com> <6c4e07720908061710k7b492cebod7eccb6df06af5ce@mail.gmail.com> <1249604727.4271.10.camel@laptop.localdomain> Message-ID: <4A7B7D50.7060208@velocitynet.com.au> Miguel Enrique Cob? Martinez wrote: > El jue, 06-08-2009 a las 20:10 -0400, Dave Bauer escribi?: > >> On Thu, Aug 6, 2009 at 8:06 PM, Randal L. Schwartz wrote: >> >>>>>>>> "Boris" == Boris Popov writes: >>>>>>>> >>> Boris> I agree. In fact many personal firewall products also block uncommon >>> Boris> ports such as this one and only leave things like 80, 443, 25 etc by >>> Boris> default. >>> >>> That's completely nuts. >>> >>> There's nothing more "dangerous" at port 9090 than at port 80. >>> >>> Anyone who constructs such a firewall is hopelessly naive, and if >>> you're using such a product, turn it off, and demand a refund. >>> >>> >> That is totally besides the point. Unfortunately the Seaside community >> regularly tries to scare away those interested in the product by >> giving unhelpful advice. >> > > If I am developing web applications and am not capable of configure or > find the cause of my net issues I have a deeper problem that I think and > it is not fault on the Seaside community. You and Randal are assuming that you have control of the firewall, that was not where this discussion started. In many large organizations those who do control the firewalls will not change their policies regardless of how inane they are. Why exclude interested people just because some other decision makers in their organizations are idiots? Steve From merlyn at stonehenge.com Fri Aug 7 02:10:23 2009 From: merlyn at stonehenge.com (Randal L. Schwartz) Date: Fri Aug 7 02:10:26 2009 Subject: [Seaside] Seaside + JQuery example page on port 9090? In-Reply-To: <4A7B7D50.7060208@velocitynet.com.au> (Steve Aldred's message of "Fri, 07 Aug 2009 11:03:12 +1000") References: <42f5c4430908061237h3ab5cbf6vb75bddc63756e671@mail.gmail.com> <42f5c4430908061431x4eaae4a2xba777e305e449135@mail.gmail.com> <86d478sjmq.fsf@blue.stonehenge.com> <4A7B68D7.4060203@velocitynet.com.au> <864osksd0p.fsf@blue.stonehenge.com> <6c4e07720908061710k7b492cebod7eccb6df06af5ce@mail.gmail.com> <1249604727.4271.10.camel@laptop.localdomain> <4A7B7D50.7060208@velocitynet.com.au> Message-ID: <86zlacqsq8.fsf@blue.stonehenge.com> >>>>> "Steve" == Steve Aldred writes: Steve> Why exclude interested people just because some other decision makers Steve> in their organizations are idiots? And I'm not excluding anyone. If you're behind a stupidly configured firewall, just download the one-click demo. But you'd be better off also getting your stupidly configured firewall working better, so as not to just solve this problem, but many others. -- Randal L. Schwartz - Stonehenge Consulting Services, Inc. - +1 503 777 0095 Smalltalk/Perl/Unix consulting, Technical writing, Comedy, etc. etc. See http://methodsandmessages.vox.com/ for Smalltalk and Seaside discussion From steve.hayes at cogentconsulting.com.au Fri Aug 7 02:28:44 2009 From: steve.hayes at cogentconsulting.com.au (Steve Hayes) Date: Fri Aug 7 02:29:11 2009 Subject: [Seaside] Seaside + JQuery example page on port 9090? In-Reply-To: <86zlacqsq8.fsf@blue.stonehenge.com> References: <42f5c4430908061237h3ab5cbf6vb75bddc63756e671@mail.gmail.com> <42f5c4430908061431x4eaae4a2xba777e305e449135@mail.gmail.com> <86d478sjmq.fsf@blue.stonehenge.com> <4A7B68D7.4060203@velocitynet.com.au> <864osksd0p.fsf@blue.stonehenge.com> <6c4e07720908061710k7b492cebod7eccb6df06af5ce@mail.gmail.com> <1249604727.4271.10.camel@laptop.localdomain> <4A7B7D50.7060208@velocitynet.com.au> <86zlacqsq8.fsf@blue.stonehenge.com> Message-ID: On 07/08/2009, at 12:10 PM, Randal L. Schwartz wrote: >>>>>> "Steve" == Steve Aldred writes: > > Steve> Why exclude interested people just because some other > decision makers > Steve> in their organizations are idiots? > > And I'm not excluding anyone. > > If you're behind a stupidly configured firewall, just download the > one-click > demo. But you'd be better off also getting your stupidly configured > firewall > working better, so as not to just solve this problem, but many others. > If someone just wants to see if something works, they may very well not want to go to the bother of convincing another group to change their corporate firewall policies and/or download and install an application. They may decide to look at another technology/solution instead. The Seaside community can decide if it wants to include those people (make an evaluation easier for them) or not. There's no inherent right or wrong. I think we've established that you can't use the one-click demo to look at JQuery examples because it's 2.8.4. I will say that I'm fleeing the Ruby on Rails community because of my perception of the arrogance and uncooperative behaviour of the core group at a technical level. My impression to date was that the Seaside community was more interested in things that "just work" and on being inclusive and accessible. ----- Steve Hayes Cogent Consulting http://www.cogentconsulting.com.au http://iridescenturchin.blogspot.com/ twitter: steve_hayes GTalk: steve.e.hayes Yahoo: steve_e_hayes Mobile: 0403 902 431 SkypeIn: +61 3 9005 6695 Measure you code quality: http://iridescenturchin.blogspot.com/2009/07/what-would-you-like-in-code-quality.html Have more things to do than you have time to do them? Use http://www.runwayapp.com . Runway - where your life takes off. From philippe.marschall at gmail.com Fri Aug 7 05:03:14 2009 From: philippe.marschall at gmail.com (Philippe Marschall) Date: Fri Aug 7 05:03:17 2009 Subject: [Seaside] Seaside + JQuery example page on port 9090? In-Reply-To: <4A7B68D7.4060203@velocitynet.com.au> References: <42f5c4430908061237h3ab5cbf6vb75bddc63756e671@mail.gmail.com> <42f5c4430908061431x4eaae4a2xba777e305e449135@mail.gmail.com> <86d478sjmq.fsf@blue.stonehenge.com> <4A7B68D7.4060203@velocitynet.com.au> Message-ID: <66666f210908062203l2e86cad5wd13d02c7e80784bc@mail.gmail.com> 2009/8/7 Steve Aldred : > Randal L. Schwartz wrote: >>>>>>> >>>>>>> "Thierry" == Thierry Thelliez >>>>>>> writes: >>>>>>> >> >> Thierry> I was referring to the links on >> http://www.seaside.st/about/examples. >> Thierry> The ones pointing to jQuery, jQueryUI and Scriptaculous. >> >> Those examples are also in the one-click image, as I recall. ?Just >> download >> the image, and run it locally, if you're on a needlessly impaired >> internet. >> > > Needlessly or not if you're living in a bureaucratic organization with a > high level of paranoia filters for other than the most common ports exist. > > If we want Seaside to be widely accessible, all of the demonstration/example > code should be available on port 80 and not needlessly moved to some other > port. Let the 'experts' hosting the examples map the request however they > like but let those in restricted environments be free to explore all that is > out there. Port 80 requires root privileges. We can't do that. Cheers Philippe From renggli at gmail.com Fri Aug 7 07:52:24 2009 From: renggli at gmail.com (Lukas Renggli) Date: Fri Aug 7 07:52:27 2009 Subject: [Seaside] Seaside + JQuery example page on port 9090? In-Reply-To: <42f5c4430908061237h3ab5cbf6vb75bddc63756e671@mail.gmail.com> References: <42f5c4430908061237h3ab5cbf6vb75bddc63756e671@mail.gmail.com> Message-ID: <67628d690908070052h13ec67d3gabe68447437d5d50@mail.gmail.com> Hi Thierry > Having these examples on port 9090 is making access difficult when you > are behind a firewall. Could they be moved to a regular port? Thank you for reporting and your interest in Seaside. To me it looks like the server for the JQuery and Scriptaculous demos is simply down. The examples linked from are provided by individual members of the community, we don't have control over these installations. Maybe it would make sense to run the demos on itself? Originally we decided not to run such code on , because we prefer to improve the Scriptaculous and JQuery libraries instead of updating the examples on the server every now and then ;-) Anyway, I've created a bug report and I will look into this as soon as possible: . Cheers, Lukas -- Lukas Renggli http://www.lukas-renggli.ch From jfitzell at gmail.com Fri Aug 7 08:01:52 2009 From: jfitzell at gmail.com (Julian Fitzell) Date: Fri Aug 7 08:01:56 2009 Subject: [Seaside] Seaside + JQuery example page on port 9090? In-Reply-To: <66666f210908062203l2e86cad5wd13d02c7e80784bc@mail.gmail.com> References: <42f5c4430908061237h3ab5cbf6vb75bddc63756e671@mail.gmail.com> <42f5c4430908061431x4eaae4a2xba777e305e449135@mail.gmail.com> <86d478sjmq.fsf@blue.stonehenge.com> <4A7B68D7.4060203@velocitynet.com.au> <66666f210908062203l2e86cad5wd13d02c7e80784bc@mail.gmail.com> Message-ID: <4a5dbd430908070101y65a839fcr3a9cdd86d8963201@mail.gmail.com> On Thu, Aug 6, 2009 at 10:03 PM, Philippe Marschall wrote: > 2009/8/7 Steve Aldred : >> Randal L. Schwartz wrote: >>>>>>>> >>>>>>>> "Thierry" == Thierry Thelliez >>>>>>>> writes: >>>>>>>> >>> >>> Thierry> I was referring to the links on >>> http://www.seaside.st/about/examples. >>> Thierry> The ones pointing to jQuery, jQueryUI and Scriptaculous. >>> >>> Those examples are also in the one-click image, as I recall. ?Just >>> download >>> the image, and run it locally, if you're on a needlessly impaired >>> internet. >>> >> >> Needlessly or not if you're living in a bureaucratic organization with a >> high level of paranoia filters for other than the most common ports exist. >> >> If we want Seaside to be widely accessible, all of the demonstration/example >> code should be available on port 80 and not needlessly moved to some other >> port. Let the 'experts' hosting the examples map the request however they >> like but let those in restricted environments be free to explore all that is >> out there. > > Port 80 requires root privileges. We can't do that. Erm... The squeak image doesn't have to be running on port 80 - apache can forward requests to an image on another port same as usual. But that's not really the point. My observations are that: a) this demo is run by someone else, not the Seaside team, so there's little we can do about what port it's on b) the server seems to be simply down (at least right now) c) jQuery is only in Seaside 2.9, which is under heavy development and still in alpha d) Once it goes beta, it would probably be a good idea to run the demos on seaside.st I don't think any of us are interested in making things difficult for a few potential users just to prove a point about their firewalls. Is this why so many sites have big warnings saying "this link goes to an external site"? Julian From damien.cassou at gmail.com Fri Aug 7 08:49:47 2009 From: damien.cassou at gmail.com (Damien Cassou) Date: Fri Aug 7 08:50:10 2009 Subject: [Seaside] The rendering context XHTML document is nil on page refresh In-Reply-To: <6ac749c10908040203x10f42948g3655917e6b010a0e@mail.gmail.com> References: <6ac749c10908020420l697679b1mc5107ca7df0a74e7@mail.gmail.com> <4a5dbd430908021804r59bfe00bj77c520381eff912f@mail.gmail.com> <6ac749c10908030200g252a2228y7a53403333b044c7@mail.gmail.com> <4a5dbd430908030934k9c40c8ew9bf04dccf73fd60f@mail.gmail.com> <6ac749c10908040203x10f42948g3655917e6b010a0e@mail.gmail.com> Message-ID: <6ac749c10908070149h1b4e009o785b6dfad47bdf3c@mail.gmail.com> Hi, On Tue, Aug 4, 2009 at 11:03 AM, Damien Cassou wrote: > On Mon, Aug 3, 2009 at 6:34 PM, Julian Fitzell wrote: >> I don't see an obvious problem, I'm afraid. As Karsten mentions, a >> RequestContext has its variables nil'ed out when it is finished and >> its response is returned, and that's the most likely cause of the >> document being nil, but I certainly don't see anything in the code you >> posted that would explain that. Usually that is caused by somebody >> storing the canvas and using it again in another request (like, an >> ajax rendering callback). > > Thank you and Karsten for you explanations. I will come back to you > later with the solution or a small piece of code. I finally found the problem: part of my code was changing state and manipulating cookies outside callbacks. Thank you very much -- Damien Cassou http://damiencassou.seasidehosting.st "Lambdas are relegated to relative obscurity until Java makes them popular by not having them." James Iry From diogenes.moreira at gmail.com Fri Aug 7 09:17:34 2009 From: diogenes.moreira at gmail.com (Diogenes Alberto Moreira) Date: Fri Aug 7 09:17:43 2009 Subject: [Seaside] Seaside + JQuery example page on port 9090? In-Reply-To: <42f5c4430908061237h3ab5cbf6vb75bddc63756e671@mail.gmail.com> References: <42f5c4430908061237h3ab5cbf6vb75bddc63756e671@mail.gmail.com> Message-ID: In a workspace you must run. WAEncoded stop. WAEncoded startOn:80. Please check if you dont have any http server running in you machine. Di?genes A. Moreira www.smallworks.com.ar diogenes.moreira@smallworks.com.ar skype: diogenes.moreira El 06/08/2009, a las 16:37, Thierry Thelliez escribi?: > Having these examples on port 9090 is making access difficult when you > are behind a firewall. Could they be moved to a regular port? > > > Thierry > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside From renggli at gmail.com Fri Aug 7 10:48:23 2009 From: renggli at gmail.com (Lukas Renggli) Date: Fri Aug 7 10:48:24 2009 Subject: [Seaside] ignore Code Critics rules per class. (Was Re: Slime in one click image) In-Reply-To: <24798870.post@talk.nabble.com> References: <24736487.post@talk.nabble.com> <4a5dbd430907300912r449f9c73w5521c223195124f9@mail.gmail.com> <102261990907310000x1562970fic70e635c632e9b29@mail.gmail.com> <67628d690907310412t5a8b19f0sb2901125c28ee0d1@mail.gmail.com> <24798870.post@talk.nabble.com> Message-ID: <67628d690908070348j65e39337va61651baa47f05a7@mail.gmail.com> > Hi, When I subclass WASession, I get 'Class not referenced' from code > critics; there's one instance of the problem for each method in the class. > I've tried to set a pragma to ignore the rule as per: > http://www.lukas-renggli.ch/blog/ignoring-lint-rules , but can't get it to > ignore: > > currentNeatSheet > > ? ? ? ? configured by Seaside, accessed by self session' author: 'Stan Shepherd'> > > ? ? ? ?^ currentNeatSheet > > or > > currentNeatSheet > > ? ? ? ? configured by Seaside, accessed by self session' author: 'Stan Shepherd'> > > ? ? ? ?^ currentNeatSheet > > or > > currentNeatSheet > > ? ? ? ? Seaside, accessed by self session' author: 'Stan Shepherd'> > > ? ? ? ?^ currentNeatSheet All your examples should actually work. The reasons they don't is because the filters only work on SelectorEnviornments. The SelectorEnvironment is the object that is used internally to collect the matches if the result are individual methods. In the case of the RBClassNotReferencedRule a class environment is used, as the result are individual classes. I tried to fix the problem in Refactoring-Core-lr.53. It might not be perfect yet, but it should solve your problem. Lukas -- Lukas Renggli http://www.lukas-renggli.ch From stan.shepherd414 at gmail.com Fri Aug 7 12:49:18 2009 From: stan.shepherd414 at gmail.com (stan shepherd) Date: Fri Aug 7 12:49:25 2009 Subject: [Seaside] ignore Code Critics rules per class. In-Reply-To: <67628d690908070348j65e39337va61651baa47f05a7@mail.gmail.com> References: <24736487.post@talk.nabble.com> <4a5dbd430907300912r449f9c73w5521c223195124f9@mail.gmail.com> <102261990907310000x1562970fic70e635c632e9b29@mail.gmail.com> <67628d690907310412t5a8b19f0sb2901125c28ee0d1@mail.gmail.com> <24798870.post@talk.nabble.com> <67628d690908070348j65e39337va61651baa47f05a7@mail.gmail.com> Message-ID: <24864271.post@talk.nabble.com> Thanks, that's got it. So I put the pragma in any one method for a class rule, correct? Is there a place to attach the pragma to the class itself? Thanks, ...Stan Lukas Renggli wrote: > >> Hi, When I subclass WASession, I get 'Class not referenced' from code >> critics; there's one instance of the problem for each method in the >> class. >> I've tried to set a pragma to ignore the rule as per: >> http://www.lukas-renggli.ch/blog/ignoring-lint-rules , but can't get it >> to >> ignore: >> >> currentNeatSheet >> >> ? ? ? ?> configured by Seaside, accessed by self session' author: 'Stan Shepherd'> >> >> ? ? ? ?^ currentNeatSheet >> >> or >> >> currentNeatSheet >> >> ? ? ? ?> class >> configured by Seaside, accessed by self session' author: 'Stan Shepherd'> >> >> ? ? ? ?^ currentNeatSheet >> >> or >> >> currentNeatSheet >> >> ? ? ? ?> configured by >> Seaside, accessed by self session' author: 'Stan Shepherd'> >> >> ? ? ? ?^ currentNeatSheet > > All your examples should actually work. The reasons they don't is > because the filters only work on SelectorEnviornments. The > SelectorEnvironment is the object that is used internally to collect > the matches if the result are individual methods. In the case of the > RBClassNotReferencedRule a class environment is used, as the result > are individual classes. > > I tried to fix the problem in Refactoring-Core-lr.53. It might not be > perfect yet, but it should solve your problem. > > Lukas > > -- > Lukas Renggli > http://www.lukas-renggli.ch > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside > > -- View this message in context: http://www.nabble.com/Slime-in-one-click-image-tp24736487p24864271.html Sent from the Squeak - Seaside mailing list archive at Nabble.com. From renggli at gmail.com Fri Aug 7 14:08:21 2009 From: renggli at gmail.com (Lukas Renggli) Date: Fri Aug 7 14:08:24 2009 Subject: [Seaside] ignore Code Critics rules per class. In-Reply-To: <24864271.post@talk.nabble.com> References: <24736487.post@talk.nabble.com> <4a5dbd430907300912r449f9c73w5521c223195124f9@mail.gmail.com> <102261990907310000x1562970fic70e635c632e9b29@mail.gmail.com> <67628d690907310412t5a8b19f0sb2901125c28ee0d1@mail.gmail.com> <24798870.post@talk.nabble.com> <67628d690908070348j65e39337va61651baa47f05a7@mail.gmail.com> <24864271.post@talk.nabble.com> Message-ID: <67628d690908070708n6edb0056jba1b0acd2bc9b597@mail.gmail.com> > Thanks, that's got it. So I put the pragma in any one method for a class > rule, correct? Yes, any method will do. Unfortunately Squeak/Pharo has no possibility to annotate other entities than methods. Lukas > Is there a place to attach the pragma to the class itself? > > Thanks, ...Stan > > > > Lukas Renggli wrote: >> >>> Hi, When I subclass WASession, I get 'Class not referenced' from code >>> critics; there's one instance of the problem for each method in the >>> class. >>> I've tried to set a pragma to ignore the rule as per: >>> http://www.lukas-renggli.ch/blog/ignoring-lint-rules , but can't get it >>> to >>> ignore: >>> >>> currentNeatSheet >>> >>> ? ? ? ?>> configured by Seaside, accessed by self session' author: 'Stan Shepherd'> >>> >>> ? ? ? ?^ currentNeatSheet >>> >>> or >>> >>> currentNeatSheet >>> >>> ? ? ? ?>> class >>> configured by Seaside, accessed by self session' author: 'Stan Shepherd'> >>> >>> ? ? ? ?^ currentNeatSheet >>> >>> or >>> >>> currentNeatSheet >>> >>> ? ? ? ?>> configured by >>> Seaside, accessed by self session' author: 'Stan Shepherd'> >>> >>> ? ? ? ?^ currentNeatSheet >> >> All your examples should actually work. The reasons they don't is >> because the filters only work on SelectorEnviornments. The >> SelectorEnvironment is the object that is used internally to collect >> the matches if the result are individual methods. In the case of the >> RBClassNotReferencedRule a class environment is used, as the result >> are individual classes. >> >> I tried to fix the problem in Refactoring-Core-lr.53. It might not be >> perfect yet, but it should solve your problem. >> >> Lukas >> >> -- >> Lukas Renggli >> http://www.lukas-renggli.ch >> _______________________________________________ >> seaside mailing list >> seaside@lists.squeakfoundation.org >> http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside >> >> > > -- > View this message in context: http://www.nabble.com/Slime-in-one-click-image-tp24736487p24864271.html > Sent from the Squeak - Seaside mailing list archive at Nabble.com. > > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside > -- Lukas Renggli http://www.lukas-renggli.ch From thierry.thelliez.tech at gmail.com Fri Aug 7 15:05:39 2009 From: thierry.thelliez.tech at gmail.com (Thierry Thelliez) Date: Fri Aug 7 15:05:42 2009 Subject: [Seaside] Seaside + JQuery example page on port 9090? In-Reply-To: References: <42f5c4430908061237h3ab5cbf6vb75bddc63756e671@mail.gmail.com> Message-ID: <42f5c4430908070805m32812f5fqdeaed79c052f48eb@mail.gmail.com> Sorry about the noise generated by this thread. The real issue was that I was in a user group meeting comparing different web frameworks. The room was filled with various PHP, Java, Ruby developers. During a JQuery discussion, I suggested that they have a look at Seaside. The access for the examples from the local public Wifi did not work. I had to come up with vague excuses, since all the other sites seemed to work. I suspected that port 9090 was the issue. Not untypical for a public wifi. It could have been that the server was simply down. That would have been unfortunate. Maybe the server(s) should be monitored? But again the issue is that a quick access to examples is needed. Yes I could have preloaded an image. But in that case I needed a quick easy access. Thierry From merlyn at stonehenge.com Fri Aug 7 15:31:22 2009 From: merlyn at stonehenge.com (Randal L. Schwartz) Date: Fri Aug 7 15:31:24 2009 Subject: [Seaside] Seaside + JQuery example page on port 9090? In-Reply-To: <42f5c4430908070805m32812f5fqdeaed79c052f48eb@mail.gmail.com> (Thierry Thelliez's message of "Fri, 7 Aug 2009 09:05:39 -0600") References: <42f5c4430908061237h3ab5cbf6vb75bddc63756e671@mail.gmail.com> <42f5c4430908070805m32812f5fqdeaed79c052f48eb@mail.gmail.com> Message-ID: <86iqgzr67p.fsf@blue.stonehenge.com> >>>>> "Thierry" == Thierry Thelliez writes: Thierry> I suspected that port 9090 was the issue. Not untypical for a public Thierry> wifi. I use free wifi all around the world. I can't recall when *any* of those blocked port 9090. The only time I've had trouble getting to a port 80/443 was when I was on a "guest" wireless in a classroom where I was teaching, and some local IT guy had misunderstood non-standard ports as being "more dangerous" than standard ports. This is part of why I'm bringing this up in this conversation... it *isn't*. Thierry> It could have been that the server was simply down. That would have Thierry> been unfortunate. This is *far* more likely. In fact, somewhere else in this thread, that was already noted. Thierry> Maybe the server(s) should be monitored? It's all about volunteer resources. Are you volunteering? :) -- Randal L. Schwartz - Stonehenge Consulting Services, Inc. - +1 503 777 0095 Smalltalk/Perl/Unix consulting, Technical writing, Comedy, etc. etc. See http://methodsandmessages.vox.com/ for Smalltalk and Seaside discussion From jfitzell at gmail.com Fri Aug 7 15:34:33 2009 From: jfitzell at gmail.com (Julian Fitzell) Date: Fri Aug 7 15:34:36 2009 Subject: [Seaside] The rendering context XHTML document is nil on page refresh In-Reply-To: <6ac749c10908070149h1b4e009o785b6dfad47bdf3c@mail.gmail.com> References: <6ac749c10908020420l697679b1mc5107ca7df0a74e7@mail.gmail.com> <4a5dbd430908021804r59bfe00bj77c520381eff912f@mail.gmail.com> <6ac749c10908030200g252a2228y7a53403333b044c7@mail.gmail.com> <4a5dbd430908030934k9c40c8ew9bf04dccf73fd60f@mail.gmail.com> <6ac749c10908040203x10f42948g3655917e6b010a0e@mail.gmail.com> <6ac749c10908070149h1b4e009o785b6dfad47bdf3c@mail.gmail.com> Message-ID: <4a5dbd430908070834j1bc2f27cvc528e367d0f2fc09@mail.gmail.com> On Fri, Aug 7, 2009 at 1:49 AM, Damien Cassou wrote: > Hi, > > On Tue, Aug 4, 2009 at 11:03 AM, Damien Cassou wrote: >> On Mon, Aug 3, 2009 at 6:34 PM, Julian Fitzell wrote: >>> I don't see an obvious problem, I'm afraid. As Karsten mentions, a >>> RequestContext has its variables nil'ed out when it is finished and >>> its response is returned, and that's the most likely cause of the >>> document being nil, but I certainly don't see anything in the code you >>> posted that would explain that. Usually that is caused by somebody >>> storing the canvas and using it again in another request (like, an >>> ajax rendering callback). >> >> Thank you and Karsten for you explanations. I will come back to you >> later with the solution or a small piece of code. > > I finally found the problem: part of my code was changing state and > manipulating cookies outside callbacks. > > Thank you very much Aha - good, glad you found it. Julian From thierry.thelliez.tech at gmail.com Fri Aug 7 15:57:55 2009 From: thierry.thelliez.tech at gmail.com (Thierry Thelliez) Date: Fri Aug 7 15:57:58 2009 Subject: [Seaside] Seaside + JQuery example page on port 9090? In-Reply-To: <86iqgzr67p.fsf@blue.stonehenge.com> References: <42f5c4430908061237h3ab5cbf6vb75bddc63756e671@mail.gmail.com> <42f5c4430908070805m32812f5fqdeaed79c052f48eb@mail.gmail.com> <86iqgzr67p.fsf@blue.stonehenge.com> Message-ID: <42f5c4430908070857y58b4343p67095d590ac36d4a@mail.gmail.com> Randal, The point of this thread is more marketing than technical. I am just pointing an issue with the public facing seaside web site. I thought that the seaside leaders would be interested/concerned. To monitor a site, you can register in few seconds fro free at site24x7.com for example. It would make more sense for the site admin to do that. Thierry On Fri, Aug 7, 2009 at 9:31 AM, Randal L. Schwartz wrote: >>>>>> "Thierry" == Thierry Thelliez writes: > > Thierry> I suspected that port 9090 was the issue. Not untypical for a public > Thierry> wifi. > > I use free wifi all around the world. ?I can't recall when *any* of those > blocked port 9090. ?The only time I've had trouble getting to a port 80/443 > was when I was on a "guest" wireless in a classroom where I was teaching, > and some local IT guy had misunderstood non-standard ports as being "more > dangerous" than standard ports. ?This is part of why I'm bringing this > up in this conversation... it *isn't*. > > Thierry> It could have been that the server was simply down. That would have > Thierry> been unfortunate. > > This is *far* more likely. ?In fact, somewhere else in this thread, that was > already noted. > > Thierry> Maybe the server(s) should be monitored? > > It's all about volunteer resources. ?Are you volunteering? :) > > -- > Randal L. Schwartz - Stonehenge Consulting Services, Inc. - +1 503 777 0095 > > Smalltalk/Perl/Unix consulting, Technical writing, Comedy, etc. etc. > See http://methodsandmessages.vox.com/ for Smalltalk and Seaside discussion > From merlyn at stonehenge.com Fri Aug 7 16:15:47 2009 From: merlyn at stonehenge.com (Randal L. Schwartz) Date: Fri Aug 7 16:15:49 2009 Subject: [Seaside] Seaside + JQuery example page on port 9090? In-Reply-To: <42f5c4430908070857y58b4343p67095d590ac36d4a@mail.gmail.com> (Thierry Thelliez's message of "Fri, 7 Aug 2009 09:57:55 -0600") References: <42f5c4430908061237h3ab5cbf6vb75bddc63756e671@mail.gmail.com> <42f5c4430908070805m32812f5fqdeaed79c052f48eb@mail.gmail.com> <86iqgzr67p.fsf@blue.stonehenge.com> <42f5c4430908070857y58b4343p67095d590ac36d4a@mail.gmail.com> Message-ID: <8663czr45o.fsf@blue.stonehenge.com> >>>>> "Thierry" == Thierry Thelliez writes: Thierry> The point of this thread is more marketing than technical. I am just Thierry> pointing an issue with the public facing seaside web site. I thought Thierry> that the seaside leaders would be interested/concerned. Yes, but we've now established that your original complaint had two pieces: 1) you couldn't reach the site, and that was bad (we all agree on this) 2) you *thought* it was because it was port 9090 so, you wanted a change to the port 9090 issue. Can we now agree that the 9090 port was a red herring, and not related to your initial complaint? Please? -- Randal L. Schwartz - Stonehenge Consulting Services, Inc. - +1 503 777 0095 Smalltalk/Perl/Unix consulting, Technical writing, Comedy, etc. etc. See http://methodsandmessages.vox.com/ for Smalltalk and Seaside discussion From thierry.thelliez.tech at gmail.com Fri Aug 7 16:44:35 2009 From: thierry.thelliez.tech at gmail.com (Thierry Thelliez) Date: Fri Aug 7 16:44:38 2009 Subject: [Seaside] Seaside + JQuery example page on port 9090? In-Reply-To: <8663czr45o.fsf@blue.stonehenge.com> References: <42f5c4430908061237h3ab5cbf6vb75bddc63756e671@mail.gmail.com> <42f5c4430908070805m32812f5fqdeaed79c052f48eb@mail.gmail.com> <86iqgzr67p.fsf@blue.stonehenge.com> <42f5c4430908070857y58b4343p67095d590ac36d4a@mail.gmail.com> <8663czr45o.fsf@blue.stonehenge.com> Message-ID: <42f5c4430908070944g4f236f38ke98cc077abefa324@mail.gmail.com> Randal, I live in a (production) world where everything but http, https and ssh is closed. The ':9090' misguided me. I had no way to debug that in the public meeting I was in. Sorry. But please do not shoot the messenger! I was just trying to spread the seaside joy! Thierry From garduino at gmail.com Sat Aug 8 16:36:03 2009 From: garduino at gmail.com (=?ISO-8859-1?Q?Germ=E1n_Arduino?=) Date: Sat Aug 8 16:36:06 2009 Subject: [Seaside] Socket(Object)>>primitiveFailed Message-ID: Hi: I'm trying to start an image in the form "completepath/vmname completpath/imagename completpath/startupscript arg1" were arg1 is the a port number to start WAKom. As not worked headless I tried showing the main Squeak window and I get the subject error with not much more info. I researched the lists but not found nothing related. I'm using "Squeak 4.1.1beta2U.app" with a 3.10.2 image running on Mac OS X Server 10.5.5. I tried on a linux box with a unix vm and get the same error at socket level (Socket>>listenOn: / TcpListener>>pvtOldListenLoop:). Any help will be appreciated. Thanks. -- Germ?n S. Arduino http://www.arsol.biz http://www.arsol.net From diogenes.moreira at gmail.com Sat Aug 8 16:44:19 2009 From: diogenes.moreira at gmail.com (Diogenes Alberto Moreira) Date: Sat Aug 8 16:44:30 2009 Subject: [Seaside] Socket(Object)>>primitiveFailed In-Reply-To: References: Message-ID: Hi German Are you sure what port is free?, that is the tipical error when the port is used by other process.. In Addition, We are start the image with the port in the script like this WAKomEncoded stop. WAKomEncoded startOn: 8080. Im not sure if VM manage more parameters then tipicals. Best Regards Di?genes A. Moreira www.smallworks.com.ar diogenes.moreira@smallworks.com.ar skype: diogenes.moreira El 08/08/2009, a las 13:36, Germ?n Arduino escribi?: > Hi: > > I'm trying to start an image in the form "completepath/vmname > completpath/imagename completpath/startupscript arg1" were arg1 is the > a port number to start WAKom. > > As not worked headless I tried showing the main Squeak window and I > get the subject error with not much more info. I researched the lists > but not found nothing related. > > I'm using "Squeak 4.1.1beta2U.app" with a 3.10.2 image running on Mac > OS X Server 10.5.5. I tried on a linux box with a unix vm and get the > same error at socket level (Socket>>listenOn: / > TcpListener>>pvtOldListenLoop:). > > Any help will be appreciated. > > Thanks. > > > -- > Germ?n S. Arduino > http://www.arsol.biz > http://www.arsol.net > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside From garduino at gmail.com Sat Aug 8 16:48:09 2009 From: garduino at gmail.com (=?ISO-8859-1?Q?Germ=E1n_Arduino?=) Date: Sat Aug 8 16:48:16 2009 Subject: [Seaside] Socket(Object)>>primitiveFailed In-Reply-To: References:

Message-ID: Hi Di?genes: Yes, the port is free. My script only read the port, as follow: | port | port := SmalltalkImage current getSystemAttribute: 3. WAKom stop. WAKom startOn: port. The script will be more complex in the future, now is at this way only to try if works. Cheers. 2009/8/8 Diogenes Alberto Moreira : > Hi German > > Are you sure what port is free?, that is the tipical error when the port is > used by other process.. > > In Addition, We are start the image with the port in the script like this > > > WAKomEncoded stop. > WAKomEncoded startOn: 8080. > > Im not sure if VM manage more parameters then tipicals. > > Best Regards > > > Di?genes A. Moreira > www.smallworks.com.ar > diogenes.moreira@smallworks.com.ar > skype: diogenes.moreira > > > > El 08/08/2009, a las 13:36, Germ?n Arduino escribi?: > >> Hi: >> >> I'm trying to start an image in the form "completepath/vmname >> completpath/imagename completpath/startupscript arg1" were arg1 is the >> a port number to start WAKom. >> >> As not worked headless I tried showing the main Squeak window and I >> get the subject error with not much more info. I researched the lists >> but not found nothing related. >> >> I'm using "Squeak 4.1.1beta2U.app" with a 3.10.2 image running on Mac >> OS X Server 10.5.5. I tried on a linux box with a unix vm and get the >> same error at socket level (Socket>>listenOn: / >> TcpListener>>pvtOldListenLoop:). >> >> Any help will be appreciated. >> >> Thanks. >> >> >> -- >> Germ?n S. Arduino >> http://www.arsol.biz >> http://www.arsol.net >> _______________________________________________ >> seaside mailing list >> seaside@lists.squeakfoundation.org >> http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside > > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside > From garduino at gmail.com Sat Aug 8 16:59:40 2009 From: garduino at gmail.com (=?ISO-8859-1?Q?Germ=E1n_Arduino?=) Date: Sat Aug 8 16:59:45 2009 Subject: [Seaside] Socket(Object)>>primitiveFailed In-Reply-To: References:

Message-ID: Trying a bit more on Linux and debugging, the error is a bit different: Socket(Object)>>doesNotUnderstand: #waitForConnectionUntil: On the debugger, on TcpListener I can see that port is ok, and the handler is "seaside [running] port: 8888", the port that I passed, but.....the image don't respond on such port. And the debugger is here :( See you. 2009/8/8 Germ?n Arduino : > Hi Di?genes: > > Yes, the port is free. > > My script only read the port, as follow: > > | port | > port := SmalltalkImage current getSystemAttribute: 3. > WAKom stop. > > WAKom startOn: port. > > The script will be more complex in the future, now is at this way only > to try if works. > > Cheers. > > > 2009/8/8 Diogenes Alberto Moreira : >> Hi German >> >> Are you sure what port is free?, that is the tipical error when the port is >> used by other process.. >> >> In Addition, We are start the image with the port in the script like this >> >> >> WAKomEncoded stop. >> WAKomEncoded startOn: 8080. >> >> Im not sure if VM manage more parameters then tipicals. >> >> Best Regards >> >> >> Di?genes A. Moreira >> www.smallworks.com.ar >> diogenes.moreira@smallworks.com.ar >> skype: diogenes.moreira >> >> >> >> El 08/08/2009, a las 13:36, Germ?n Arduino escribi?: >> >>> Hi: >>> >>> I'm trying to start an image in the form "completepath/vmname >>> completpath/imagename completpath/startupscript arg1" were arg1 is the >>> a port number to start WAKom. >>> >>> As not worked headless I tried showing the main Squeak window and I >>> get the subject error with not much more info. I researched the lists >>> but not found nothing related. >>> >>> I'm using "Squeak 4.1.1beta2U.app" with a 3.10.2 image running on Mac >>> OS X Server 10.5.5. I tried on a linux box with a unix vm and get the >>> same error at socket level (Socket>>listenOn: / >>> TcpListener>>pvtOldListenLoop:). >>> >>> Any help will be appreciated. >>> >>> Thanks. >>> >>> >>> -- >>> Germ?n S. Arduino >>> http://www.arsol.biz >>> http://www.arsol.net >>> _______________________________________________ >>> seaside mailing list >>> seaside@lists.squeakfoundation.org >>> http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside >> >> _______________________________________________ >> seaside mailing list >> seaside@lists.squeakfoundation.org >> http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside >> From astares at gmx.de Sat Aug 8 18:42:37 2009 From: astares at gmx.de (Torsten Bergmann) Date: Sat Aug 8 18:42:41 2009 Subject: [Seaside] Seaside + JQuery example page on port 9090? Message-ID: <20090808184237.208970@gmx.net> There are tools to access "non standard" ports if you are behind a closed (8080/80) firewall. But I agree with Thierry - ports like 9090 make life harder if you quickly want to show a (non local) seaside demo. So having all demos on a reliable server with standard port would make it more comfortable. Bye T. -- Jetzt kostenlos herunterladen: Internet Explorer 8 und Mozilla Firefox 3 - sicherer, schneller und einfacher! http://portal.gmx.net/de/go/atbrowser From johnmci at smalltalkconsulting.com Sat Aug 8 22:12:14 2009 From: johnmci at smalltalkconsulting.com (John M McIntosh) Date: Sat Aug 8 22:12:24 2009 Subject: [Seaside] Socket(Object)>>primitiveFailed In-Reply-To: References: Message-ID: <81C9414E-6DEB-41D3-AEB7-4A3ADA035A55@smalltalkconsulting.com> Btw the Squeak 4.1.1beta2U.app and the Unix version of Squeak.app use the same source code for the virtual machine Unix Socket support since both systems are Unix *like* operating systems, so you shouldn't find a difference in behaviour at the primitive layer. But given you get "Socket(Object)>>doesNotUnderstand: #waitForConnectionUntil:" where did the image & VM come from? On 8-Aug-09, at 9:36 AM, Germ?n Arduino wrote: > I'm using "Squeak 4.1.1beta2U.app" with a 3.10.2 image running on Mac > OS X Server 10.5.5. I tried on a linux box with a unix vm and get the > same error at socket level (Socket>>listenOn: / > TcpListener>>pvtOldListenLoop:) -- = = = ======================================================================== John M. McIntosh Twitter: squeaker68882 Corporate Smalltalk Consulting Ltd. http://www.smalltalkconsulting.com = = = ======================================================================== From johnmci at smalltalkconsulting.com Sat Aug 8 23:05:11 2009 From: johnmci at smalltalkconsulting.com (John M McIntosh) Date: Sat Aug 8 23:06:03 2009 Subject: [Seaside] wikiserver 1.4.4 promo codes Message-ID: <91F31A40-606D-4D3E-842F-1C62AA284B6B@smalltalkconsulting.com> Apple approved WikiServer 1.4.4 for sale on the App store yesterday, as usual they have handed us another 50 USA promo codes, so if there are people on the list that would like one or two, please email me, not the list, a request. I note these are for USA itunes customers, but there are ways for people to create a new dummy itunes USA account with out supplying credit card information to redeem the promo code then install the app on your device for a month of free usage. Since we now support the ability to export the wiki data and run on a desktop version it's easy to preserve any data for later importing into a purchased version. -- = = = ======================================================================== John M. McIntosh Twitter: squeaker68882 Corporate Smalltalk Consulting Ltd. http://www.smalltalkconsulting.com = = = ======================================================================== From johnmci at smalltalkconsulting.com Sun Aug 9 04:12:44 2009 From: johnmci at smalltalkconsulting.com (John M McIntosh) Date: Sun Aug 9 04:12:53 2009 Subject: [Seaside] Re: [squeak-dev] wikiserver 1.4.4 promo codes In-Reply-To: <4A7E3259.1000100@cox.net> References: <91F31A40-606D-4D3E-842F-1C62AA284B6B@smalltalkconsulting.com> <4A7E3259.1000100@cox.net> Message-ID: On 8-Aug-09, at 7:20 PM, Lawson English wrote: > > John have you considered cashing in on the iPhone craze by writing a > book on how to program the iphone using Squeak? > L Ah, my friends who write books tell me it's a tedious process and generally results in zero or negative income. There are a few I know who have been wildly successful, but I"m afraid the topic won't sell much copy. However I will talk about the process at ESUG in Sept, http://www.esug.org/Conferences/2009/Current+List+of+Talks/Smalltalk+on+iPhone and again at the XTC group in London the week after http://xpday-london.editme.com/eXtremeTuesdayClub -- = = = ======================================================================== John M. McIntosh Twitter: squeaker68882 Corporate Smalltalk Consulting Ltd. http://www.smalltalkconsulting.com = = = ======================================================================== From garduino at gmail.com Sun Aug 9 14:10:45 2009 From: garduino at gmail.com (=?ISO-8859-1?Q?Germ=E1n_Arduino?=) Date: Sun Aug 9 14:10:51 2009 Subject: [Seaside] Socket(Object)>>primitiveFailed In-Reply-To: <81C9414E-6DEB-41D3-AEB7-4A3ADA035A55@smalltalkconsulting.com> References: <81C9414E-6DEB-41D3-AEB7-4A3ADA035A55@smalltalkconsulting.com> Message-ID: Hi John: The image is the same in both cases (a 3.10.2 that I use to develop. I installed seaside and friends and a few dev tools packages). The MAC vm is the version I mentioned (downloaded from your page I think) and the Linux version is Squeak-3.10-6.i686-pc-linux-gnu.tar.gz (from http://www.squeakvm.org/unix/). Cheers. 2009/8/8 John M McIntosh : > Btw the Squeak 4.1.1beta2U.app and the Unix version of Squeak.app ?use the > same source code for the virtual machine > Unix Socket support since both systems are Unix *like* operating systems, so > you shouldn't find a difference in behaviour > at the primitive layer. > > But given you get "Socket(Object)>>doesNotUnderstand: > #waitForConnectionUntil:" > where did the image & VM come from? > > > On 8-Aug-09, at 9:36 AM, Germ?n Arduino wrote: > >> I'm using "Squeak 4.1.1beta2U.app" with a 3.10.2 image running on Mac >> OS X Server 10.5.5. I tried on a linux box with a unix vm and get the >> same error at socket level (Socket>>listenOn: / >> TcpListener>>pvtOldListenLoop:) > > -- > =========================================================================== > John M. McIntosh ? Twitter: ?squeaker68882 > Corporate Smalltalk Consulting Ltd. ?http://www.smalltalkconsulting.com > =========================================================================== > > > > > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside > From ssastre at seaswork.com Sun Aug 9 19:29:53 2009 From: ssastre at seaswork.com (Sebastian Sastre) Date: Sun Aug 9 19:34:27 2009 Subject: [Seaside] two inputs one updater Message-ID: Hey guys, anyone knows how to: - update 2 inputs - using one updater? having that: - there are other inputs in the same page - only one big (tag) form wrapping all When I saw the implementation of triggerFormElement: I've though two parameters can came serialized in the POST so I've tried this: updater callback:[:h| self renderThoseTwoInputsOn: h]; triggerFormElement: idInput1; triggerFormElement: idInput2; yourself. but with that I only get the model of input1 updated. As workaround I'm posting the whole form but is really not ideal. What could be interfering with the update of the second input? maybe the serialized value of the second input is not hitting its callback? fixable? hints? thanks, sebastian From ramiro at diaztrepat.name Sun Aug 9 21:24:32 2009 From: ramiro at diaztrepat.name (Ramiro Diaz Trepat) Date: Sun Aug 9 21:24:35 2009 Subject: [Seaside] Magma + Seaside (+Seaside28Jetsam?) Message-ID: <313dbe090908091424i69349550v190c3ffffd5d07da@mail.gmail.com> Hi,I am trying to get a working image that includes Seaside and Magma. I am using the latest Pharo web as my start point. I have published a couple of Monticellos with the extension migrations from BlockContext to BlockClosure. After this, Magma seems to work fine in Pharo. But, to install the seaside modules, the wiki says that I must first install Seaside28Jetsam Which seems to break Seaside (2.8). This was surprising since the latest version seems to be very recent (7th of July). Any clues on how to get this stack working will be greatly appreciated. Cheers r. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.squeakfoundation.org/pipermail/seaside/attachments/20090809/d7531109/attachment.htm From keith_hodges at yahoo.co.uk Sun Aug 9 22:53:22 2009 From: keith_hodges at yahoo.co.uk (Keith Hodges) Date: Sun Aug 9 22:53:25 2009 Subject: [Seaside] Magma + Seaside (+Seaside28Jetsam?) In-Reply-To: <313dbe090908091424i69349550v190c3ffffd5d07da@mail.gmail.com> References: <313dbe090908091424i69349550v190c3ffffd5d07da@mail.gmail.com> Message-ID: <4A7F5362.3000705@yahoo.co.uk> Ramiro Diaz Trepat wrote: > Hi, > I am trying to get a working image that includes Seaside and Magma. > I am using the latest Pharo web as my start point. > I have published a couple of Monticellos with the extension migrations > from BlockContext to BlockClosure. After this, Magma seems to work > fine in Pharo. > But, to install the seaside modules, the wiki says that I must first > install > Seaside28Jetsam > Which seems to break Seaside (2.8). This was surprising since the > latest version seems to be very recent (7th of July). > Any clues on how to get this stack working will be greatly appreciated. > Cheers > > r. I liitle more info on what's "broken" would be good Keith From ramiro at diaztrepat.name Sun Aug 9 23:02:45 2009 From: ramiro at diaztrepat.name (Ramiro Diaz Trepat) Date: Sun Aug 9 23:02:48 2009 Subject: [Seaside] Magma + Seaside (+Seaside28Jetsam?) In-Reply-To: <4A7F5362.3000705@yahoo.co.uk> References: <313dbe090908091424i69349550v190c3ffffd5d07da@mail.gmail.com> <4A7F5362.3000705@yahoo.co.uk> Message-ID: <313dbe090908091602l277265b4wffdaede41249c62d@mail.gmail.com> Sorry for the little info Keith.Here's a bit more, but if you load Seaside28Jetsam.kph.79.mcz in the latest Pharo web, you will see that just accessing http://localhost:8080/seaside you get: MessageNotUnderstood: WAWalkback>>updateRootLibraries:WASession>>updateRoot: WARenderContinuation(WASessionContinuation)>>updateRoot: WARenderContinuation>>newHtmlRoot WARenderContinuation>>processRendering: [] in WARenderContinuation>>render WARenderContinuation(WASessionContinuation)>>respond: WARenderContinuation>>render [] in WARenderContinuation>>run BlockClosure>>on:do: WARenderContinuation>>withNotificationHandler: WARenderContinuation>>run WARenderLoop>>run [] in WARenderLoop>>call:withToolFrame: AnswerContinuation class(Continuation class)>>currentDo: WARenderLoop>>call:withToolFrame: WAWalkbackErrorHandler>>open: WAWalkbackErrorHandler>>handleError: WAWalkbackErrorHandler class(WAErrorHandler class)>>handleError: [] in [] in WASession>>withErrorHandler: BlockClosure>>valueWithPossibleArgs: If you load version 77, you can access the home but you can't configure the application's options, you see something like: ConfigurationGeneralDeployment Mode*false*fixedset by clientError Handler* WAWalkbackErrorHandler*fixedset by clientMain Class*WARenderLoopMain*fixedset by clientRedirect Continuation Class*WARedirectContinuation*fixedset by client I hope this helps. Cheers r. On Sun, Aug 9, 2009 at 11:53 PM, Keith Hodges wrote: > Ramiro Diaz Trepat wrote: > > Hi, > > I am trying to get a working image that includes Seaside and Magma. > > I am using the latest Pharo web as my start point. > > I have published a couple of Monticellos with the extension migrations > > from BlockContext to BlockClosure. After this, Magma seems to work > > fine in Pharo. > > But, to install the seaside modules, the wiki says that I must first > > install > > Seaside28Jetsam > > Which seems to break Seaside (2.8). This was surprising since the > > latest version seems to be very recent (7th of July). > > Any clues on how to get this stack working will be greatly appreciated. > > Cheers > > > > r. > I liitle more info on what's "broken" would be good > > Keith > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.squeakfoundation.org/pipermail/seaside/attachments/20090810/777605ee/attachment.htm From keith_hodges at yahoo.co.uk Sun Aug 9 23:40:19 2009 From: keith_hodges at yahoo.co.uk (Keith Hodges) Date: Sun Aug 9 23:40:22 2009 Subject: [Seaside] Magma + Seaside (+Seaside28Jetsam?) In-Reply-To: <313dbe090908091602l277265b4wffdaede41249c62d@mail.gmail.com> References: <313dbe090908091424i69349550v190c3ffffd5d07da@mail.gmail.com> <4A7F5362.3000705@yahoo.co.uk> <313dbe090908091602l277265b4wffdaede41249c62d@mail.gmail.com> Message-ID: <4A7F5E63.2010201@yahoo.co.uk> Ramiro Diaz Trepat wrote: > Sorry for the little info Keith. > Here's a bit more, but if you load Seaside28Jetsam.kph.79.mcz in the > latest Pharo web, you will see that just accessing > http://localhost:8080/seaside you get: > > > MessageNotUnderstood: WAWalkback>>updateRootLibraries: > > WASession>>updateRoot: > WARenderContinuation(WASessionContinuation)>>updateRoot: Ok fixed that issue... this missing method was in pier-jetsam (I usually have both loaded) Keith From ramiro at diaztrepat.name Mon Aug 10 07:13:31 2009 From: ramiro at diaztrepat.name (Ramiro Diaz Trepat) Date: Mon Aug 10 07:13:33 2009 Subject: [Seaside] Magma + Seaside (+Seaside28Jetsam?) In-Reply-To: <4A7F5E63.2010201@yahoo.co.uk> References: <313dbe090908091424i69349550v190c3ffffd5d07da@mail.gmail.com> <4A7F5362.3000705@yahoo.co.uk> <313dbe090908091602l277265b4wffdaede41249c62d@mail.gmail.com> <4A7F5E63.2010201@yahoo.co.uk> Message-ID: <313dbe090908100013u720f5ba9j369c222161a3643f@mail.gmail.com> Hi Keith, Thanks for looking at this. The error still persists after loading version 80 though. Cheers r. On Mon, Aug 10, 2009 at 12:40 AM, Keith Hodges wrote: > Ramiro Diaz Trepat wrote: > > Sorry for the little info Keith. > > Here's a bit more, but if you load Seaside28Jetsam.kph.79.mcz in the > > latest Pharo web, you will see that just accessing > > http://localhost:8080/seaside you get: > > > > > > MessageNotUnderstood: WAWalkback>>updateRootLibraries: > > > > WASession>>updateRoot: > > WARenderContinuation(WASessionContinuation)>>updateRoot: > Ok fixed that issue... this missing method was in pier-jetsam (I > usually have both loaded) > > Keith > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.squeakfoundation.org/pipermail/seaside/attachments/20090810/43faeef4/attachment-0001.htm From renggli at gmail.com Mon Aug 10 07:56:43 2009 From: renggli at gmail.com (Lukas Renggli) Date: Mon Aug 10 07:56:45 2009 Subject: [Seaside] two inputs one updater In-Reply-To: References: Message-ID: <67628d690908100056q4817fec7h2ff16e2cdb94e6ab@mail.gmail.com> > When I saw the implementation of triggerFormElement: I've though two parameters > can came serialized in the POST so I've tried this: > > ? ? ? ?updater > ? ? ? ? ? ? ? ?callback:[:h| > ? ? ? ? ? ? ? ? ? ? ? ?self renderThoseTwoInputsOn: h]; > ? ? ? ? ? ? ? ?triggerFormElement: idInput1; > ? ? ? ? ? ? ? ?triggerFormElement: idInput2; > ? ? ? ? ? ? ? ?yourself. Sure, that's how it is ment to be used. What kind of input fields are you using? Be sure to carefully consider what is written in the method comment of #triggerFormElement: if any of the input fields is a check-box or a multi-select list. Lukas -- Lukas Renggli http://www.lukas-renggli.ch From ssastre at seaswork.com Mon Aug 10 11:41:30 2009 From: ssastre at seaswork.com (Sebastian Sastre) Date: Mon Aug 10 11:46:10 2009 Subject: [Seaside] two inputs one updater In-Reply-To: <67628d690908100056q4817fec7h2ff16e2cdb94e6ab@mail.gmail.com> Message-ID: Ah.. mm ok, but then I don't get why is not the second model being updated. I was aware of checkboxes and such. This inputs are just text inputs. I mean both. I'll give a look with the debugger when the parameters arrive to see if I get more info about it sebastian > -----Mensaje original----- > De: seaside-bounces@lists.squeakfoundation.org > [mailto:seaside-bounces@lists.squeakfoundation.org] En nombre > de Lukas Renggli > Enviado el: Monday, August 10, 2009 04:57 > Para: Seaside - general discussion > Asunto: Re: [Seaside] two inputs one updater > > > When I saw the implementation of triggerFormElement: I've > though two parameters > > can came serialized in the POST so I've tried this: > > > > ? ? ? ?updater > > ? ? ? ? ? ? ? ?callback:[:h| > > ? ? ? ? ? ? ? ? ? ? ? ?self renderThoseTwoInputsOn: h]; > > ? ? ? ? ? ? ? ?triggerFormElement: idInput1; > > ? ? ? ? ? ? ? ?triggerFormElement: idInput2; > > ? ? ? ? ? ? ? ?yourself. > > Sure, that's how it is ment to be used. > > What kind of input fields are you using? Be sure to carefully consider > what is written in the method comment of #triggerFormElement: if any > of the input fields is a check-box or a multi-select list. > > Lukas > > -- > Lukas Renggli > http://www.lukas-renggli.ch > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside From john.chludzinski at gmail.com Mon Aug 10 18:26:05 2009 From: john.chludzinski at gmail.com (John Chludzinski) Date: Mon Aug 10 18:26:07 2009 Subject: [Seaside] Re: Problems with Seaside 2.9 libraries In-Reply-To: <27a48dd60908020043v1975137el241c5413b402945f@mail.gmail.com> References: <27a48dd60908020043v1975137el241c5413b402945f@mail.gmail.com> Message-ID: <27a48dd60908101126sc3dfdd8i92f6f4d28236ec58@mail.gmail.com> > JQuery and Prototype/Scriptaculous do not work together, unless you > take extra precautions > . If you are > porting an existing application you presumably only use > Prototype/Scriptaculous, so you should not include > JQDevelopmentLibrary; similar if you use JQuery you don't want to > include PTDevelopmentLibrary, SUComponentLibrary and > SUDevelopmentLibrary. > > Lukas I have an app which is a suite of "applets" (simple applications subclassed from WAComponent). One applet allows users to view his buddies as a hierarchy of groups (e.g., football>pro>team or football>college>team). This uses Scriptaculous (SUTreeReport). Another applet allows a group of users to concurrently view a presentation (of slides). This uses Comet. If I include PTDevelopmentLibrary to make the Scriptaculous work, then Comet fails. If I exclude PTDevelopmentLibrary, then Scriptaculous fails. Is there a way to re-architect this suite of applets into a suite of apps, each of which can include whatever libraries it needs (or exclude libraries it is in conflict with)? ---John -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.squeakfoundation.org/pipermail/seaside/attachments/20090810/5a7018ed/attachment.htm From renggli at gmail.com Mon Aug 10 18:59:04 2009 From: renggli at gmail.com (Lukas Renggli) Date: Mon Aug 10 18:59:07 2009 Subject: [Seaside] Re: Problems with Seaside 2.9 libraries In-Reply-To: <27a48dd60908101126sc3dfdd8i92f6f4d28236ec58@mail.gmail.com> References: <27a48dd60908020043v1975137el241c5413b402945f@mail.gmail.com> <27a48dd60908101126sc3dfdd8i92f6f4d28236ec58@mail.gmail.com> Message-ID: <67628d690908101159kb77a784g546a2d6feb100bc2@mail.gmail.com> > ?This uses Scriptaculous (SUTreeReport). ?Another applet allows a group of > users to concurrently view a presentation (of slides). ?This uses Comet. ?If > I include PTDevelopmentLibrary to make the Scriptaculous work, then Comet > fails. ?If I exclude PTDevelopmentLibrary, then Scriptaculous fails. The Comet Javascript support code in Seaside 2.9 does not depend or conflict with Prototype or JQuery. It is useable with either of the two, or any other Javascript library of your choice. The examples that currently come with the Comet package use JQuery, but this is just a personal choice of mine. In Seaside 2.8 Scriptaculous was used for the same demos. Lukas -- Lukas Renggli http://www.lukas-renggli.ch From watchlala at hotmail.com Tue Aug 11 04:38:19 2009 From: watchlala at hotmail.com (Robert Sirois) Date: Tue Aug 11 04:38:26 2009 Subject: [Seaside] More jQuery questions!!! Rendering on dialogs... In-Reply-To: References: <67628d690908100056q4817fec7h2ff16e2cdb94e6ab@mail.gmail.com> Message-ID: Here's the case in question: A dialog is rendered on the root component, and pretty much just hangs out. My master plan involves rendering links and whatnot on it so it acts like a little floating menu. So, changing pages re-renders the dialog contents with new links. I've messed around with a few different ways of doing this and sorta settled on announcements within ajax callbacks (or regular callbacks, doesn't matter too much at this juncture). My problem lies in sending the information within an announcements payload. WAComponents work fine, but I was hoping I could just use a method such as the following (this oughta show my Seaside knowledge heh): self session announce: (TOCAnnounceUpdateActionMenu on: (newWindow menuContent: html)). menuContent: html html text: 'blah'. I get this error: MessageNotUnderstood: receiver of "print:" is nil I have to pass in a real object, huh? Thanks! RS _________________________________________________________________ Get back to school stuff for them and cashback for you. http://www.bing.com/cashback?form=MSHYCB&publ=WLHMTAG&crea=TEXT_MSHYCB_BackToSchool_Cashback_BTSCashback_1x1 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.squeakfoundation.org/pipermail/seaside/attachments/20090810/fa31d925/attachment.htm From renggli at gmail.com Tue Aug 11 09:55:32 2009 From: renggli at gmail.com (Lukas Renggli) Date: Tue Aug 11 09:55:34 2009 Subject: [Seaside] Seaside + JQuery example page on port 9090? In-Reply-To: <67628d690908070052h13ec67d3gabe68447437d5d50@mail.gmail.com> References: <42f5c4430908061237h3ab5cbf6vb75bddc63756e671@mail.gmail.com> <67628d690908070052h13ec67d3gabe68447437d5d50@mail.gmail.com> Message-ID: <67628d690908110255u55c93954o69f8eff559e35234@mail.gmail.com> > I've created a bug report and I will look into this as soon as > possible: . The issue has been resolved. Lukas -- Lukas Renggli http://www.lukas-renggli.ch From cdrick65 at gmail.com Tue Aug 11 11:53:50 2009 From: cdrick65 at gmail.com (=?UTF-8?B?Q8OpZHJpY2sgQsOpbGVy?=) Date: Tue Aug 11 11:53:57 2009 Subject: [Seaside] Re: [squeak-dev] wikiserver 1.4.4 promo codes In-Reply-To: References: <91F31A40-606D-4D3E-842F-1C62AA284B6B@smalltalkconsulting.com> <4A7E3259.1000100@cox.net> Message-ID: > However I will talk about the process at ESUG in Sept, > http://www.esug.org/Conferences/2009/Current+List+of+Talks/Smalltalk+on+iPhone excellent ! I would love a workshop too. Is it possible ? I know there are ObjectiveC bindings but are there bindings for the iphone API ? Maybe a camp project is possible in relation to iphone dev in smalltalk... Thanks > and again at the XTC group in London the week after > http://xpday-london.editme.com/eXtremeTuesdayClub > > -- > =========================================================================== > John M. McIntosh ? Twitter: ?squeaker68882 > Corporate Smalltalk Consulting Ltd. ?http://www.smalltalkconsulting.com > =========================================================================== > > > > > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside > -- C?drick From john.chludzinski at gmail.com Tue Aug 11 16:02:49 2009 From: john.chludzinski at gmail.com (John Chludzinski) Date: Tue Aug 11 16:03:04 2009 Subject: [Seaside] Re: Problems with Seaside 2.9 libraries In-Reply-To: <27a48dd60908101126sc3dfdd8i92f6f4d28236ec58@mail.gmail.com> References: <27a48dd60908020043v1975137el241c5413b402945f@mail.gmail.com> <27a48dd60908101126sc3dfdd8i92f6f4d28236ec58@mail.gmail.com> Message-ID: <27a48dd60908110902w6ebcc647qbd054b290b0fca4@mail.gmail.com> Thanks Lukas for the clues. I went back to my previous implementation, for example: self pusher javascript: [ :script | ... script element id: #WWW; writeAttribute: 'src' value: MMAFileLibrary / self getCurrentPageFile. versus: self pusher javascript: [ :script | ... script << (script jQuery: #WWW) attributeAt: 'src' put: MMAFileLibrary / self getCurrentPageFile. Removed all references to jQuery from my code, deleted JQDevelopmentLibrary, kept PTDevelopmentLibrary, and all now works. ---John -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.squeakfoundation.org/pipermail/seaside/attachments/20090811/d10cfb82/attachment.htm From johnmci at smalltalkconsulting.com Tue Aug 11 23:03:22 2009 From: johnmci at smalltalkconsulting.com (John M McIntosh) Date: Tue Aug 11 23:03:30 2009 Subject: [Seaside] Re: [squeak-dev] wikiserver 1.4.4 promo codes In-Reply-To: References: <91F31A40-606D-4D3E-842F-1C62AA284B6B@smalltalkconsulting.com> <4A7E3259.1000100@cox.net> Message-ID: On 11-Aug-09, at 4:53 AM, C?drick B?ler wrote: >> However I will talk about the process at ESUG in Sept, >> http://www.esug.org/Conferences/2009/Current+List+of+Talks/Smalltalk+on+iPhone > > excellent ! > > I would love a workshop too. Is it possible ? > > I know there are ObjectiveC bindings but are there bindings for the > iphone API ? Maybe a camp project is possible in relation to iphone > dev in smalltalk... > The iPhone API is obj-c , I don't have an FFI api for the iPhone so core foundation procedures get wrapped in the application's delegate obj-c api when I need to deal with them. Take a listen to James Robertson Industry Misinterpretations 140: OBjective Seas http://www.cincomsmalltalk.com/audio/2009/industry_misinterpretations140.mp3 The objective-c api use hangs a proxy ObjectiveCObject off of ProtoObject which was a *lot* of fun... Then DNU is dynamically handled at run time to resolve into the best method of calling the method. There are many ways to make the call in the objective-c runtime so we resolve which to use based on the parms (if any) and return type (if any). This does not preclude bulding subclasses, like say ObjectiveCNSMethodSignature which provide more static binding to the api. However the problem with building the static binding is someone has to maintain it when Apple changes the API version, and how do you test it all etc. Obviously if someone wants to fund that? I note I think it was Avi who did an earlier objective-c bridge and did provide I'd guess generated classes to do binding to much of the core obj-c classes for os-x. After you listen to the MP3 then you can let us which type of objective-c bridge I have implemented according to CinCom -- = = = ======================================================================== John M. McIntosh Twitter: squeaker68882 Corporate Smalltalk Consulting Ltd. http://www.smalltalkconsulting.com = = = ======================================================================== From emaringolo at gmail.com Wed Aug 12 13:46:41 2009 From: emaringolo at gmail.com (Esteban A. Maringolo) Date: Wed Aug 12 13:47:03 2009 Subject: [Seaside] Questions about double (or more) request In-Reply-To: <66666f210908061322y51c924e0x4b53d49613e3333d@mail.gmail.com> References: <66666f210908052216p63546c83mf1362d2250a2db91@mail.gmail.com> <66666f210908061322y51c924e0x4b53d49613e3333d@mail.gmail.com> Message-ID: <4eca49da0908120646l5919c7x20bc9552a430f306@mail.gmail.com> Hi, 2009/8/6 Philippe Marschall : > 2009/8/6 Mariano Martinez Peck : >> On Thu, Aug 6, 2009 at 4:16 AM, Philippe Marschall >> wrote: >>> 2009/8/5 Mariano Martinez Peck : >>> > Hi! I am trying to understand how seaside handle double or more, >>> > request. I > Seaside doesn't discard anything, the browser has to. Here's a rough > sketch of what happens: > - You initiate a content exchange for the "callbacks page". > - The browser sends a request. > - The Seaside callback machinery kicks in. > - Before Seaside can send a response you initiate three more requests > resulting in three more requests. > Seaside doesn't really have a problem with this, it will process these > requests one after another, meaning "executing" the same "callback > page" several times and send the redirect response to the render page > for each one. However the browser has, he has just one pane where he > can render the resonse. So he'll have to discard all but one of the > responses. The least bad thing is to discard all but the last one. Wouldn't it be useful to have one-time callbacks? Once it has been executed (block activation, message send, etc), it can't be run again. Raising an exception if so. This can be optional on certain, user defined, callbacks. I know #isolate: can do the trick, but being a decoration, it works at component level as a whole, at a #call: level (flow). If I have an action callback, executing something via ajax, I would not want to execute it twice. Regards! Esteban A. Maringolo From philippe.marschall at gmail.com Wed Aug 12 13:50:34 2009 From: philippe.marschall at gmail.com (Philippe Marschall) Date: Wed Aug 12 13:50:37 2009 Subject: [Seaside] Questions about double (or more) request In-Reply-To: <4eca49da0908120646l5919c7x20bc9552a430f306@mail.gmail.com> References: <66666f210908052216p63546c83mf1362d2250a2db91@mail.gmail.com> <66666f210908061322y51c924e0x4b53d49613e3333d@mail.gmail.com> <4eca49da0908120646l5919c7x20bc9552a430f306@mail.gmail.com> Message-ID: <66666f210908120650x5470cdc5q694a7b36fceed147@mail.gmail.com> 2009/8/12 Esteban A. Maringolo : > Hi, > > > 2009/8/6 Philippe Marschall : >> 2009/8/6 Mariano Martinez Peck : >>> On Thu, Aug 6, 2009 at 4:16 AM, Philippe Marschall >>> wrote: >>>> 2009/8/5 Mariano Martinez Peck : >>>> > Hi! I am trying to understand how seaside handle double or more, >>>> > request. I > >> Seaside doesn't discard anything, the browser has to. Here's a rough >> sketch of what happens: >> - You initiate a content exchange for the "callbacks page". >> - The browser sends a request. >> - The Seaside callback machinery kicks in. >> - Before Seaside can send a response you initiate three more requests >> resulting in three more requests. > >> Seaside doesn't really have a problem with this, it will process these >> requests one after another, meaning "executing" the same "callback >> page" several times and send the redirect response to the render page >> for each one. However the browser has, he has just one pane where he >> can render the resonse. So he'll have to discard all but one of the >> responses. The least bad thing is to discard all but the last one. > > Wouldn't it be useful to have one-time callbacks? > > Once it has been executed (block activation, message send, etc), it > can't be run again. Raising an exception if so. > This can be optional on certain, user defined, callbacks. > > I know #isolate: can do the trick, but being a decoration, it works at > component level as a whole, at a #call: level (flow). > If I have an action callback, executing something via ajax, I would > not want to execute it twice. Yes: http://code.google.com/p/seaside/issues/detail?id=157 Cheers Philippe From emaringolo at gmail.com Wed Aug 12 14:04:03 2009 From: emaringolo at gmail.com (Esteban A. Maringolo) Date: Wed Aug 12 14:04:30 2009 Subject: [Seaside] Questions about double (or more) request In-Reply-To: <66666f210908120650x5470cdc5q694a7b36fceed147@mail.gmail.com> References: <66666f210908052216p63546c83mf1362d2250a2db91@mail.gmail.com> <66666f210908061322y51c924e0x4b53d49613e3333d@mail.gmail.com> <4eca49da0908120646l5919c7x20bc9552a430f306@mail.gmail.com> <66666f210908120650x5470cdc5q694a7b36fceed147@mail.gmail.com> Message-ID: <4eca49da0908120704x254f17e8mb0401b65ef0c7558@mail.gmail.com> 2009/8/12 Philippe Marschall : > 2009/8/12 Esteban A. Maringolo : >> Wouldn't it be useful to have one-time callbacks? > Yes: > http://code.google.com/p/seaside/issues/detail?id=157 Great :) And the proposed solution, a wrapper around the callback, is the same I thought. Maybe a reification of the callback itself would work too. Best regards, -- Esteban From lenglish5 at cox.net Thu Aug 13 06:00:12 2009 From: lenglish5 at cox.net (Lawson English) Date: Thu Aug 13 06:00:20 2009 Subject: [Seaside] Startup glitches Message-ID: <4A83ABEC.4040408@cox.net> I consistently get this error the first time I start up the Pharo + seaside image: use instead: Author fullNamePerSe Select Proceed to continue, or close this window to cancel the operation. I'm using: Pharo1.0beta Latest update: #10413 Also, I'm trying to work through the first seaside tutorial: http://www.seaside.st/documentation/tutorials/A+Walk+on+the+Seaside The first discrepancy is that there's no "eye" icon, but a magnifying glass icon after I click "Toggle Halos". The second is that when I try to modify the increase method, I get the same error I get at startup: [ Seaside Walkback Deprecation: use instead: Author fullNamePerSe [...] So I'm stuck on the first page... :-/ Lawson From renggli at gmail.com Thu Aug 13 06:46:13 2009 From: renggli at gmail.com (Lukas Renggli) Date: Thu Aug 13 06:46:16 2009 Subject: [Seaside] Startup glitches In-Reply-To: <4A83ABEC.4040408@cox.net> References: <4A83ABEC.4040408@cox.net> Message-ID: <67628d690908122346n39d60a54s1a962dc9943a593f@mail.gmail.com> > I consistently get this error the first time I start up the Pharo + seaside > image: Please use the One-Click image provided on the Seaside website. > I'm trying to work through the first seaside tutorial: > http://www.seaside.st/documentation/tutorials/A+Walk+on+the+Seaside > > The first discrepancy is that there's no "eye" icon, but a magnifying glass > icon after I click "Toggle Halos". There are tooltips on the halo buttons that give a descriptive label of the action being performed. I also created a bug report that the tutorial should be updated to the latest Seaside: . > The second is that when I try to modify > the increase method, I get the same error I get at startup: > [ > Seaside Walkback > > Deprecation: use instead: Author fullNamePerSe Pharo is in an early beta stage. If you don't have prior Pharo experience it is better if you stick with the ready-made image provided on seaside.st. This image is also Pharo-based, but we've used it for a while to make sure that basic functionality is there. Please keep on asking questions. Cheers, Lukas -- Lukas Renggli http://www.lukas-renggli.ch From lenglish5 at cox.net Thu Aug 13 08:01:18 2009 From: lenglish5 at cox.net (Lawson English) Date: Thu Aug 13 08:01:23 2009 Subject: [Seaside] Startup glitches In-Reply-To: <67628d690908122346n39d60a54s1a962dc9943a593f@mail.gmail.com> References: <4A83ABEC.4040408@cox.net> <67628d690908122346n39d60a54s1a962dc9943a593f@mail.gmail.com> Message-ID: <4A83C84E.1070702@cox.net> Lukas Renggli wrote: >> I consistently get this error the first time I start up the Pharo + seaside >> image: >> > > Please use the One-Click image provided on the Seaside website. > > That one is about 60 updates out of date, however, and selecting Software Update yields several errors over the period of the hour or so it takes to update the package (actually takes 3-ish iterations of "Software Update" to stop the cycle). >> I'm trying to work through the first seaside tutorial: >> http://www.seaside.st/documentation/tutorials/A+Walk+on+the+Seaside >> >> The first discrepancy is that there's no "eye" icon, but a magnifying glass >> icon after I click "Toggle Halos". >> > > There are tooltips on the halo buttons that give a descriptive label > of the action being performed. > > I also created a bug report that the tutorial should be updated to the > latest Seaside: > . > > >> The second is that when I try to modify >> the increase method, I get the same error I get at startup: >> [ >> Seaside Walkback >> >> Deprecation: use instead: Author fullNamePerSe >> > > Pharo is in an early beta stage. If you don't have prior Pharo > experience it is better if you stick with the ready-made image > provided on seaside.st. This image is also Pharo-based, but we've used > it for a while to make sure that basic functionality is there. > > Please keep on asking questions. > > K. I'll revert to using the one-click, it seems to me that the reason why I went with the latest image = was because I was having other issues with the one-click isntall and I thought perhaps that was because I didn't have the latest image. LOL.. Thanks for responding so fast. I'll let you know how it goes. Lawson From sebnozzi at googlemail.com Thu Aug 13 13:20:59 2009 From: sebnozzi at googlemail.com (Sebastian Nozzi) Date: Thu Aug 13 13:21:22 2009 Subject: [Seaside] What to write in the "onDrop:" ? Message-ID: <1e8ee5cb0908130620m50ad5f08s6ed1ea2c645624d0@mail.gmail.com> Hello Seasiders, I am having trouble understanding how to make the Javascript (Scriptaculous) part talk to the Smalltalk one. Using Seaside 2.8. What I want to have are mouse-movable DIVs (which represend model objects) that remember their position the next time they are rendered. For that, I would like to know how to pass the coordinates (offset) from a dropped DIV back to the Smalltalk side and store that property in one model object. I've seen that SUElement provides something like this, but I'm clueless as to what to write in the #onDrop: part of my droppable to make the connection. It should extract the information via... Javascript? DOM? and send it back as a ajax-request to the server/smalltalk side... I am correct? The modified snippets would look like these: "Render the drop-container" html div script: (html droppable accept: 'myDraggableClass'; onDrop: "..." ); with:[ ... "Render the DIVs to be dragged inside container" myElements do:[:e | html div passenger: e; class: 'myDraggableClass'; script:(html draggable); with:'dragMe']. ...]. Any hints? Are there examples on the use of SUElement and co.? Thanks in advance! Sebastian. From Richard.Durr at gmail.com Thu Aug 13 13:44:33 2009 From: Richard.Durr at gmail.com (Richard.Durr@gmail.com) Date: Thu Aug 13 13:44:36 2009 Subject: [Seaside] What to write in the "onDrop:" ? In-Reply-To: <1e8ee5cb0908130620m50ad5f08s6ed1ea2c645624d0@mail.gmail.com> References: <1e8ee5cb0908130620m50ad5f08s6ed1ea2c645624d0@mail.gmail.com> Message-ID: <75D0B1B3-84FC-4181-B68C-55653DD01F7A@gmail.com> Hey Sebastian, in my Bachelor Thesis, I do something like that this way: html div id: myID; onMouseOver: (html draggable onEnd: (html request callback: [:value | self updatePositionOf: aPassenger with: value.] value: (html element id: myID) positionedOffset)); where value is a String in the form (x,y). Am 13.08.2009 um 15:20 schrieb Sebastian Nozzi: > Hello Seasiders, > > I am having trouble understanding how to make the Javascript > (Scriptaculous) part talk to the Smalltalk one. Using Seaside 2.8. > > What I want to have are mouse-movable DIVs (which represend model > objects) that remember their position the next time they are rendered. > > For that, I would like to know how to pass the coordinates (offset) > from a dropped DIV back to the Smalltalk side and store that property > in one model object. I've seen that SUElement provides something like > this, but I'm clueless as to what to write in the #onDrop: part of my > droppable to make the connection. > > It should extract the information via... Javascript? DOM? and send it > back as a ajax-request to the server/smalltalk side... I am correct? > > The modified snippets would look like these: > > "Render the drop-container" > html div script: (html droppable accept: 'myDraggableClass'; > onDrop: "..." ); with:[ ... > > "Render the DIVs to be dragged inside container" > myElements do:[:e | html div passenger: e; > class: 'myDraggableClass'; > script:(html draggable); > with:'dragMe']. > > ...]. > > Any hints? Are there examples on the use of SUElement and co.? > > Thanks in advance! > > Sebastian. > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside From sebnozzi at googlemail.com Thu Aug 13 13:59:06 2009 From: sebnozzi at googlemail.com (Sebastian Nozzi) Date: Thu Aug 13 13:59:28 2009 Subject: [Seaside] What to write in the "onDrop:" ? In-Reply-To: <75D0B1B3-84FC-4181-B68C-55653DD01F7A@gmail.com> References: <1e8ee5cb0908130620m50ad5f08s6ed1ea2c645624d0@mail.gmail.com> <75D0B1B3-84FC-4181-B68C-55653DD01F7A@gmail.com> Message-ID: <1e8ee5cb0908130659t7600109csbd707746e5b17d98@mail.gmail.com> Hello Richard, this is great! Thanks for your answer. I'll try it out as soon as I can. I didn't consider the approach to do something on the draggable, instead of the droppable... Also, I might have misunderstood or overseen how to use callback:value: Is it always used like that? (the value part "extracting" or executing Javascript info) Thanks again, Sebastian 2009/8/13 Richard.Durr@gmail.com : > Hey Sebastian, > > in my Bachelor Thesis, I do something like that this way: > > html div id: myID; > ?onMouseOver: (html draggable > ? ?onEnd: (html request > ? ? ?callback: [:value | > ? ? ? ?self updatePositionOf: aPassenger with: value.] > ? ? ?value: > ? ? ? ?(html element id: myID) positionedOffset)); > > where value is a String in the form (x,y). From Richard.Durr at gmail.com Thu Aug 13 15:21:01 2009 From: Richard.Durr at gmail.com (Richard Durr) Date: Thu Aug 13 15:21:04 2009 Subject: [Seaside] What to write in the "onDrop:" ? In-Reply-To: <1e8ee5cb0908130659t7600109csbd707746e5b17d98@mail.gmail.com> References: <1e8ee5cb0908130620m50ad5f08s6ed1ea2c645624d0@mail.gmail.com> <75D0B1B3-84FC-4181-B68C-55653DD01F7A@gmail.com> <1e8ee5cb0908130659t7600109csbd707746e5b17d98@mail.gmail.com> Message-ID: <570A3951-75BE-4F90-A5D3-63D536A51443@gmail.com> Happy to help ^^ I think it is always used like that since the comment says: "Register aBlock as a secondary callback. anObject is interpreted as JavaScript on the client-side, the result will be sent back to the server and passed into aBlock. Multiple secondary callbacks can be defined with one receiver." RD Am 13.08.2009 um 15:59 schrieb Sebastian Nozzi: > Hello Richard, > > this is great! Thanks for your answer. I'll try it out as soon as I > can. > > I didn't consider the approach to do something on the draggable, > instead of the droppable... > > Also, I might have misunderstood or overseen how to use > callback:value: > Is it always used like that? (the value part "extracting" or executing > Javascript info) > > > Thanks again, > > Sebastian > > > 2009/8/13 Richard.Durr@gmail.com : >> Hey Sebastian, >> >> in my Bachelor Thesis, I do something like that this way: >> >> html div id: myID; >> onMouseOver: (html draggable >> onEnd: (html request >> callback: [:value | >> self updatePositionOf: aPassenger with: value.] >> value: >> (html element id: myID) positionedOffset)); >> >> where value is a String in the form (x,y). > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside > From Richard.Durr at gmail.com Thu Aug 13 15:30:29 2009 From: Richard.Durr at gmail.com (Richard Durr) Date: Thu Aug 13 15:30:32 2009 Subject: [Seaside] Where to find WAListenerEncoded? Message-ID: <0758F062-7EDB-4DEF-8401-B5E08F690872@gmail.com> Hello ^^ I have a problem with unicode in my Comet-based application. Whenever I use Javascript to update some html in the browser the special characters seem fine. But when I reload the page and the html is rendered directly, unicode characters don't work / are displayed as some diamond with a questionmark inside. I can only see them again, when I switch my browser to ISO-Latin-1. Additionally, some chars, such as "?" make the browser download the page instead of displaying it. I tried to copy methods from WAKomEncoded, but this does not seem to work, unfortunatelly. Is there any way to force everything to be unicode and "just work"? RD From lenglish5 at cox.net Thu Aug 13 16:13:44 2009 From: lenglish5 at cox.net (Lawson English) Date: Thu Aug 13 16:13:47 2009 Subject: [Seaside] Startup glitches In-Reply-To: <67628d690908122346n39d60a54s1a962dc9943a593f@mail.gmail.com> References: <4A83ABEC.4040408@cox.net> <67628d690908122346n39d60a54s1a962dc9943a593f@mail.gmail.com> Message-ID: <4A843BB8.8010603@cox.net> Lukas Renggli wrote: >> I consistently get this error the first time I start up the Pharo + seaside >> image: >> > > Please use the One-Click image provided on the Seaside website. > > >> I'm trying to work through the first seaside tutorial: >> http://www.seaside.st/documentation/tutorials/A+Walk+on+the+Seaside >> >> The first discrepancy is that there's no "eye" icon, but a magnifying glass >> icon after I click "Toggle Halos". >> > > There are tooltips on the halo buttons that give a descriptive label > of the action being performed. > > I also created a bug report that the tutorial should be updated to the > latest Seaside: > . > > >> The second is that when I try to modify >> the increase method, I get the same error I get at startup: >> [ >> Seaside Walkback >> >> Deprecation: use instead: Author fullNamePerSe >> > > Pharo is in an early beta stage. If you don't have prior Pharo > experience it is better if you stick with the ready-made image > provided on seaside.st. This image is also Pharo-based, but we've used > it for a while to make sure that basic functionality is there. > > Thanks. I re-installed the one-click seaside install and only did one iteration of software updates and it still seems to be stable. Lawson From thierry.thelliez.tech at gmail.com Thu Aug 13 16:20:44 2009 From: thierry.thelliez.tech at gmail.com (Thierry Thelliez) Date: Thu Aug 13 16:20:45 2009 Subject: [Seaside] www.seaside.st down? Message-ID: <42f5c4430908130920w108a3f3eqe1449e005875d310@mail.gmail.com> I can access other web sites, but www.seaside.st appears to be down from here. Thierry From miguel.coba at gmail.com Thu Aug 13 16:23:54 2009 From: miguel.coba at gmail.com (Miguel Enrique =?ISO-8859-1?Q?Cob=E1?= Martinez) Date: Thu Aug 13 16:24:18 2009 Subject: [Seaside] www.seaside.st down? In-Reply-To: <42f5c4430908130920w108a3f3eqe1449e005875d310@mail.gmail.com> References: <42f5c4430908130920w108a3f3eqe1449e005875d310@mail.gmail.com> Message-ID: <1250180635.4135.37.camel@laptop.localdomain> El jue, 13-08-2009 a las 10:20 -0600, Thierry Thelliez escribi?: > I can access other web sites, but www.seaside.st appears to be down from here. > > +1 > Thierry > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside -- Miguel Cob? http://miguel.leugim.com.mx From lenglish5 at cox.net Thu Aug 13 16:31:41 2009 From: lenglish5 at cox.net (Lawson English) Date: Thu Aug 13 16:31:46 2009 Subject: [Seaside] www.seaside.st down? In-Reply-To: <42f5c4430908130920w108a3f3eqe1449e005875d310@mail.gmail.com> References: <42f5c4430908130920w108a3f3eqe1449e005875d310@mail.gmail.com> Message-ID: <4A843FED.6090404@cox.net> Thierry Thelliez wrote: > I can access other web sites, but www.seaside.st appears to be down from here. > > > Thierry > > Proxy Error The proxy server received an invalid response from an upstream server. The proxy server could not handle the request GET /. Reason: Error reading from remote server Is what I get back... Lawson From jfitzell at gmail.com Thu Aug 13 17:24:05 2009 From: jfitzell at gmail.com (Julian Fitzell) Date: Thu Aug 13 17:24:07 2009 Subject: [Seaside] Where to find WAListenerEncoded? In-Reply-To: <0758F062-7EDB-4DEF-8401-B5E08F690872@gmail.com> References: <0758F062-7EDB-4DEF-8401-B5E08F690872@gmail.com> Message-ID: <4a5dbd430908131024m127ce062j510fcdb03bd220b6@mail.gmail.com> This may not be helpful to you, but in 2.9, all adaptors have a configurable encoding so I'm guessing this would work better there, though I haven't specifically tried it myself. WAKom and WAKomEncoded only exist in 2.9 for compatibility and simply create appropriately configured instances of WAComancheAdaptor. Julian On Thu, Aug 13, 2009 at 8:30 AM, Richard Durr wrote: > Hello ^^ > > I have a problem with unicode in my Comet-based application. Whenever I use > Javascript to update some html in the browser the special characters seem > fine. But when I reload the page and the html is rendered directly, unicode > characters don't work / are displayed as some diamond with a questionmark > inside. > > I can only see them again, when I switch my browser to ISO-Latin-1. > Additionally, some chars, such as "?" make the browser download the page > instead of displaying it. > > I tried to copy methods from WAKomEncoded, but this does not seem to work, > unfortunatelly. Is there any way to force everything to be unicode and "just > work"? > > RD > > > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside > From lenglish5 at cox.net Thu Aug 13 18:30:22 2009 From: lenglish5 at cox.net (Lawson English) Date: Thu Aug 13 18:30:25 2009 Subject: [Seaside] simple example for a simple man? Message-ID: <4A845BBE.1050701@cox.net> Hey all, I've been trying to go through the various materials available, but I'm having a hard time finding a simple example of the kind of thing I want to do with seaside, at least as a helloworld project, which is basically POST arbitrary data to an arbitrary url. I can find plenty of examples of GUI-related stuff, but not much that shows how to create and use a simple instance of the HTTP request stuff. If there's some easy example I've missed, or a mailing list entry that I haven't found with googling site:lists.squeakfoundation.org, I'd much appreciate a pointer or 3. Thanks Lawson From boris at deepcovelabs.com Thu Aug 13 18:36:06 2009 From: boris at deepcovelabs.com (Boris Popov) Date: Thu Aug 13 18:36:12 2009 Subject: [Seaside] simple example for a simple man? In-Reply-To: <4A845BBE.1050701@cox.net> References: <4A845BBE.1050701@cox.net> Message-ID: If you're using VisualWorks, (HttpRequest post: 'http://mydomain.com/post-handler.php') formData: (Array with: 'hello' -> 'world'); execute -Boris -- +1.604.689.0322 DeepCove Labs Ltd. 4th floor 595 Howe Street Vancouver, Canada V6C 2T5 http://tinyurl.com/r7uw4 -----Original Message----- From: seaside-bounces@lists.squeakfoundation.org [mailto:seaside-bounces@lists.squeakfoundation.org] On Behalf Of Lawson English Sent: Thursday, August 13, 2009 11:30 AM To: Seaside - general discussion Subject: [Seaside] simple example for a simple man? Hey all, I've been trying to go through the various materials available, but I'm having a hard time finding a simple example of the kind of thing I want to do with seaside, at least as a helloworld project, which is basically POST arbitrary data to an arbitrary url. I can find plenty of examples of GUI-related stuff, but not much that shows how to create and use a simple instance of the HTTP request stuff. If there's some easy example I've missed, or a mailing list entry that I haven't found with googling site:lists.squeakfoundation.org, I'd much appreciate a pointer or 3. Thanks Lawson _______________________________________________ seaside mailing list seaside@lists.squeakfoundation.org http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside From lenglish5 at cox.net Thu Aug 13 18:39:21 2009 From: lenglish5 at cox.net (Lawson English) Date: Thu Aug 13 18:39:23 2009 Subject: [Seaside] simple example for a simple man? In-Reply-To: References: <4A845BBE.1050701@cox.net> Message-ID: <4A845DD9.2070004@cox.net> Boris Popov wrote: > If you're using VisualWorks, > > (HttpRequest post: 'http://mydomain.com/post-handler.php') > formData: (Array with: 'hello' -> 'world'); > execute > > -Boris > > AH thanks, definitely at my level, but I'm using pharo/squeak... Lawson From jfitzell at gmail.com Thu Aug 13 18:44:27 2009 From: jfitzell at gmail.com (Julian Fitzell) Date: Thu Aug 13 18:44:29 2009 Subject: [Seaside] simple example for a simple man? In-Reply-To: <4A845DD9.2070004@cox.net> References: <4A845BBE.1050701@cox.net> <4A845DD9.2070004@cox.net> Message-ID: <4a5dbd430908131144j561d9d0fsa47dedd180375d86@mail.gmail.com> On Thu, Aug 13, 2009 at 11:39 AM, Lawson English wrote: > Boris Popov wrote: >> >> If you're using VisualWorks, >> >> (HttpRequest post: 'http://mydomain.com/post-handler.php') >> ? formData: (Array with: 'hello' -> 'world'); >> ? execute >> >> -Boris >> >> > > > AH thanks, definitely at my level, but I'm using pharo/squeak... If you're trying to create an outgoing HTTP connection, Seaside doesn't deal with that directly. In Squeak, you could use HTTPSocket (see class-side methods such as #httpPost:args:accept:). I don't love the implementation, though, and you might also want to look at the Curl plugin if your needs are at all complex. Julian From lenglish5 at cox.net Thu Aug 13 18:53:20 2009 From: lenglish5 at cox.net (Lawson English) Date: Thu Aug 13 18:53:47 2009 Subject: [Seaside] simple example for a simple man? In-Reply-To: <4a5dbd430908131144j561d9d0fsa47dedd180375d86@mail.gmail.com> References: <4A845BBE.1050701@cox.net> <4A845DD9.2070004@cox.net> <4a5dbd430908131144j561d9d0fsa47dedd180375d86@mail.gmail.com> Message-ID: <4A846120.8070606@cox.net> Julian Fitzell wrote: > On Thu, Aug 13, 2009 at 11:39 AM, Lawson English wrote: > >> Boris Popov wrote: >> >>> If you're using VisualWorks, >>> >>> (HttpRequest post: 'http://mydomain.com/post-handler.php') >>> formData: (Array with: 'hello' -> 'world'); >>> execute >>> >>> -Boris >>> >>> >>> >> AH thanks, definitely at my level, but I'm using pharo/squeak... >> > > If you're trying to create an outgoing HTTP connection, Seaside > doesn't deal with that directly. In Squeak, you could use HTTPSocket > (see class-side methods such as #httpPost:args:accept:). I don't love > the implementation, though, and you might also want to look at the > Curl plugin if your needs are at all complex. > > Julian > > Thanks. At this point, my needs are exceedingly simple. But, its not obvious to me how to get started. At the simplest level, I simply want to create one or more webpages of arbitrary GUI complexity that can POST input to some external website from a page served in localhost. Lawson From svanlacke at caesarsystems.com Fri Aug 14 13:32:05 2009 From: svanlacke at caesarsystems.com (Sebastian Van Lacke) Date: Fri Aug 14 13:32:09 2009 Subject: [Seaside] Isolate doubt Message-ID: <001c01ca1ce3$9d8c1e00$4801a8c0@Pythagoras> Hi, I need to isolate a transaction on a registration task. I don't know why the application is redirecting me to an empty page and not to the last component rendered in the task. Here is my task RegistrationTask>>go | participant | self isolate: [ participant := self call: STsRegistrationForm new. self call: STsHumanValidator new. ]. self call: (STsRegistrationResult new participant: participant ). I need to go to the result page when the user goes back with the browser. I can't find the problem. I appreciate your help. Sebastian -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.squeakfoundation.org/pipermail/seaside/attachments/20090814/d1f2414a/attachment.htm From stan.shepherd414 at gmail.com Sat Aug 15 11:28:57 2009 From: stan.shepherd414 at gmail.com (stan shepherd) Date: Sat Aug 15 11:29:03 2009 Subject: [Seaside] Seaside Testing in Seaside one click image Message-ID: <102261990908150428hd4ae9ebrabff27566085bcd4@mail.gmail.com> Hi David, I ran up your most recent Seaside Testing in the one click experience. It works fine, but one of the tests failed, because WAAllTests has been changed. The attached version passes all tests green. Let me know if it requires explanation. Cheers, Stan -------------- next part -------------- A non-text attachment was scrubbed... Name: SeasideTesting-R1.5-stan_shepherd.27.mcz Type: application/octet-stream Size: 50127 bytes Desc: not available Url : http://lists.squeakfoundation.org/pipermail/seaside/attachments/20090815/4771a555/SeasideTesting-R1.5-stan_shepherd.27-0001.obj From lenglish5 at cox.net Sat Aug 15 17:56:06 2009 From: lenglish5 at cox.net (Lawson English) Date: Sat Aug 15 17:56:16 2009 Subject: [Seaside] Clarifying error Message-ID: <4A86F6B6.4090105@cox.net> To clarify what I get with various versions of the seaside one-click, if I install the one-click seaside image, and run software update twice, it throws deprecation warning initialsPerSe - please use fullNamePerSe If I run software update only once on the one-click seaside image, which updates to 10343, it doesn't show that error. Lawson From Richard.Durr at gmail.com Sat Aug 15 19:04:11 2009 From: Richard.Durr at gmail.com (Richard Durr) Date: Sat Aug 15 19:04:14 2009 Subject: [Seaside] So why exactly isn't there a seaside wiki? Message-ID: Hello Everybody, I currently try to figure out how to work with scriptaculous in Seaside and what I mostly miss are examples, examples, examples. Yes sure, there is scriptaculous.seasidehosting... but that shows only a part of what can be done and is especially silent on working with Element etc. (I still don't know how to create a script that uses Element to get some value and than print that value, sigh). On the other hand, there are some things that I have already learnt and would like to contribute back. But appart from setting up my own blog, there is no official and no concentrated way of doing that, I believe. A community edited wiki can akkumulate much knowledge, examples and so on. (Vgl. c2.com) I therefore think it would be cool if an official seaside wiki existed. RD From renggli at gmail.com Sat Aug 15 19:04:30 2009 From: renggli at gmail.com (Lukas Renggli) Date: Sat Aug 15 19:04:34 2009 Subject: [Seaside] Clarifying error In-Reply-To: <4A86F6B6.4090105@cox.net> References: <4A86F6B6.4090105@cox.net> Message-ID: <67628d690908151204r2d817a07o5ec6440d631ec648@mail.gmail.com> > To clarify what I get with various versions of the seaside one-click, > if I install the one-click seaside image, and run software update twice, it > throws > > deprecation warning ?initialsPerSe - please use fullNamePerSe This is a (known) Pharo issue. Please report such problems on the Pharo list. > If I run software update only once on the one-click seaside image, which > updates to 10343, it doesn't show that error. I don't think you should run software update in non-core images, e.g. images that have other packages loaded such as the Seaside one-click. Uupdating the base system influences all loaded code and might break the base system or the loaded packages in various subtle ways. Some people reported that updating worked for them, but I think in recent versions of Pharo the menu item has been hidden. Cheers, Lukas -- Lukas Renggli http://www.lukas-renggli.ch From renggli at gmail.com Sat Aug 15 19:15:25 2009 From: renggli at gmail.com (Lukas Renggli) Date: Sat Aug 15 19:15:28 2009 Subject: [Seaside] So why exactly isn't there a seaside wiki? In-Reply-To: References: Message-ID: <67628d690908151215s536d7fbo719aa5f64efc3fb2@mail.gmail.com> > A community edited wiki can akkumulate much knowledge, examples and so on. > (Vgl. c2.com) We had a wiki several years ago, but it died (http://wiki.squeak.org/squeak/2272). It is simply too hard to keep it in sync with the code. You can see the same on the script.aculo.us website, their documentation wiki died too. They have integrated the documentation in code now. > I therefore think it would be cool if an official seaside wiki existed. Personally I would prefer if you could provide simple examples in code, that could be integrated with the existing examples. That would ensure that the examples are actually in sync with the framework. Another possibility would be to write a tutorial that could be integrated on seaside.st, however the site rots quickly too and already today requires major cleanup (http://code.google.com/p/seaside/issues/detail?id=432, http://code.google.com/p/seaside/issues/detail?id=402). Lukas -- Lukas Renggli http://www.lukas-renggli.ch From marianopeck at gmail.com Sat Aug 15 19:17:24 2009 From: marianopeck at gmail.com (Mariano Martinez Peck) Date: Sat Aug 15 19:17:27 2009 Subject: [Seaside] Clarifying error In-Reply-To: <67628d690908151204r2d817a07o5ec6440d631ec648@mail.gmail.com> References: <4A86F6B6.4090105@cox.net> <67628d690908151204r2d817a07o5ec6440d631ec648@mail.gmail.com> Message-ID: On Sat, Aug 15, 2009 at 4:04 PM, Lukas Renggli wrote: > > To clarify what I get with various versions of the seaside one-click, > > if I install the one-click seaside image, and run software update twice, > it > > throws > > > > deprecation warning initialsPerSe - please use fullNamePerSe > > This is a (known) Pharo issue. Please report such problems on the Pharo > list. > Yes, see http://code.google.com/p/pharo/issues/detail?can=1&q=1070&colspec=ID%20Type%20Status%20Summary%20Milestone&id=1070 > > > If I run software update only once on the one-click seaside image, which > > updates to 10343, it doesn't show that error. > > I don't think you should run software update in non-core images, e.g. > images that have other packages loaded such as the Seaside one-click. > Uupdating the base system influences all loaded code and might break > the base system or the loaded packages in various subtle ways. Some > people reported that updating worked for them, but I think in recent > versions of Pharo the menu item has been hidden. > > Cheers, > Lukas > > -- > Lukas Renggli > http://www.lukas-renggli.ch > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.squeakfoundation.org/pipermail/seaside/attachments/20090815/5f9487a6/attachment.htm From lenglish5 at cox.net Tue Aug 18 16:47:25 2009 From: lenglish5 at cox.net (Lawson English) Date: Tue Aug 18 16:47:31 2009 Subject: [Seaside] Second LIfe plugin API Message-ID: <4A8ADB1D.9030105@cox.net> As part of my on-going work on finding synergy, someone, somewhere, might want to look at this: http://wiki.secondlife.com/wiki/Media_Rendering_Plugin_Framework_Technical_Overview and consider what might be done with Squeak/Phara/Croquet/Cobalt/Seaside in a Second Life context. Lawson From lenglish5 at cox.net Tue Aug 18 17:30:42 2009 From: lenglish5 at cox.net (Lawson English) Date: Tue Aug 18 17:30:51 2009 Subject: [Seaside] Re: [croquet-dev] Re: [squeak-dev] Second LIfe plugin API In-Reply-To: References: <4A8ADB1D.9030105@cox.net> Message-ID: <4A8AE542.2050407@cox.net> Howard Stearns wrote: > Cool. What do you have in mind? > > Do you have any experience with the QuickTime media plugin? How is it > in general? How is it for streaming sources? > > The model for coordinating plugin activity across clients was not > clear to me. E.g., if I fire up QuickTime for a movie, how do all > other clients in the sim (that happen to have the plugin installed) > get coordinated to the same place in the movie? > > > Howard Stearns > Qwaq: +1-650-331-1437 > mobile: +1-608-658-2419 > > The way it works right now (and in the future unless they change things), a land-owner can set a "parcel media" URL which is the URL to a streaming media server that will play on any local object that displays a certain texture. The plugin API extends that to allow for custom plugins to run, but only in the context of the parcel media URL texture. I can imagine many cases where a media plugin served from localhost might be useful. It should be possible to specify localhost as a valid url for media-on-a-prim (HUD at least) that is settable separately from, or as an alternative to, the parcel media URL. IOW, if I install a custom HUD with a localhost url as the parcel media url, I should be able to interact with a private plugin running/streaming on my own comp. E.G. a seaside server on localhost doing all sorts of spiffy things that I haven't thought of yet. The plugin API makes it easy to have a plugin be part of the second life world, as far as GUI is concerned. I'm trying to convince people to create another message/event mechanism to allow more direct interaction with the SL viewer. IE, use Squeak as a scripting plugin on the viewer side. One could use the media API to provide a GUI and use the events API to send commands for generic scripting of the SL viewer. Lawson From lenglish5 at cox.net Wed Aug 19 00:15:05 2009 From: lenglish5 at cox.net (Lawson English) Date: Wed Aug 19 00:15:11 2009 Subject: [Seaside] Re: [croquet-dev] Re: [squeak-dev] Second LIfe plugin API In-Reply-To: References: <4A8ADB1D.9030105@cox.net> Message-ID: <4A8B4409.3030209@cox.net> https://svn.secondlife.com/svn/linden/branches/2009/plugin-api/indra/media_plugins/ Is the source for the plugin samples. That's as much info as I can get past what little is on the wiki... http://wiki.secondlife.com/wiki/Media_Rendering_Plugin_Framework Lawson From lenglish5 at cox.net Wed Aug 19 01:14:10 2009 From: lenglish5 at cox.net (Lawson English) Date: Wed Aug 19 01:14:15 2009 Subject: [Seaside] Re: [squeak-dev] Second LIfe plugin API In-Reply-To: <4A8ADB1D.9030105@cox.net> References: <4A8ADB1D.9030105@cox.net> Message-ID: <4A8B51E2.1040409@cox.net> Lawson English wrote: > As part of my on-going work on finding synergy, someone, somewhere, > might want to look at this: > > http://wiki.secondlife.com/wiki/Media_Rendering_Plugin_Framework_Technical_Overview > > > Apparently the SL plugin test app already handles this: http://www.canvaspaint.org/ So the sky should be the limit for what can be done... Lawson From estebanlm at gmail.com Wed Aug 19 02:23:38 2009 From: estebanlm at gmail.com (Esteban Lorenzano) Date: Wed Aug 19 02:24:06 2009 Subject: [Seaside] How to trigger an element from jQuery, in seaside 2.9? Message-ID: Hi, I wonder how I should trigger a form element using jQuery instead prototype... with prototype I'd use: html prototype request triggerFormElement: id. There is a way to do it using JQuery? Thanks in advance, Esteban From estebanlm at gmail.com Wed Aug 19 02:34:11 2009 From: estebanlm at gmail.com (Esteban Lorenzano) Date: Wed Aug 19 02:34:38 2009 Subject: [Seaside] Re: How to trigger an element from jQuery, in seaside 2.9? References: Message-ID: Forget I ask... I found it: html jQuery ajax serializeThis On 2009-08-18 23:23:38 -0300, Esteban Lorenzano said: > Hi, > I wonder how I should trigger a form element using jQuery instead > prototype... with prototype I'd use: html prototype request > triggerFormElement: id. > There is a way to do it using JQuery? > > Thanks in advance, > Esteban From astares at gmx.de Wed Aug 19 06:45:55 2009 From: astares at gmx.de (Torsten Bergmann) Date: Wed Aug 19 06:45:58 2009 Subject: [Seaside] Getting started with Seaside 2.9 + pharo Message-ID: <20090819064555.23720@gmx.net> Pat Maddox wrote on Pharo-project list >So far getting set up has been quite frustrating and now I'm looking >for some help...I want to start developing Seaside 2.9 on Pharo. How >do I do that? I crosspost this to squeak-dev and seaside too since this question comes up from time to time: If you want to load and use all the latest seaside stuff (Pharo and seaside29) then you should download the newest beta image (non-web) from the pharo download site, see [1]. If you are on win32 you can alternatively use the installer exe to get the full package (including latest VM, fonts, ...), see [2]. If you want you can update the image (according to Damien [3] it is now not only possible to update the core image but also the pharo main image with all the dev-tools, thats the one on the pharo download site) However, in a recent Pharo you can evaluate ScriptLoader loadSeaside29 This internally loads the "LoadOrderTest" (check the class ScriptLoader how its done) from the seaside squeaksource repo ([4]). The load order test is typically updated by the gemstone/seaside developers with the latest seaside 29 packages in the correct order - so it loads the latest seaside 2.9 stuff. Acccording to planetmisc feed it was updated on 17th August (see [5]). so you should get all new packages from Seaside29. Click "OK" when asked to load, NO to run the tests and after loading select "Tools -> More -> Seaside Control Panel" from the world menu. See also [6] This should bring up the new control panel and you can use the context menu to add an adapter on a specific port and run it. If you want you can open Monticello browser on the seaside 29 repo and load newer packages. Have fun Torsten [1] http://pharo-project.org/pharo-download [2] http://lists.gforge.inria.fr/pipermail/pharo-project/2009-August/012100.html [3] http://lists.gforge.inria.fr/pipermail/pharo-project/2009-August/012048.html [3] http://www.squeaksource.com/Seaside29.html [4] http://planetmisc.squeak.org/ [5] http://lists.gforge.inria.fr/pipermail/pharo-project/2009-August/011942.html -- Neu: GMX Doppel-FLAT mit Internet-Flatrate + Telefon-Flatrate f?r nur 19,99 Euro/mtl.!* http://portal.gmx.net/de/go/dsl02 From Das.Linux at gmx.de Wed Aug 19 13:59:39 2009 From: Das.Linux at gmx.de (Tobias Pape) Date: Wed Aug 19 13:59:47 2009 Subject: [Seaside] [ANN] AwesomeAtom -- FriendlyFeeds Message-ID: Hello ?, In order to support Atom Feeds in Seaside 2.9, I took Ramon Leons AtomFeed package and rewrote it accroding to RSRSS in Seaside 2.9; AwesomeAtom is in alpha stage and available on SqueakMap/SqueakSource/Universe. Additionally, there is a Seaside Decoration and Configuration for inserting Feed references () (RSS and Atom) into Applications, needs Seaside 2.9, RSRSS, and AwesomeAtom; FriendlyFeeds is in alpha stage and available on SqueakMap/SqueakSource/Universe. Have fun with these Packages. So Long, -Tobias Pape -------------- next part -------------- A non-text attachment was scrubbed... Name: PGP.sig Type: application/pgp-signature Size: 194 bytes Desc: Signierter Teil der Nachricht Url : http://lists.squeakfoundation.org/pipermail/seaside/attachments/20090819/530f15b8/PGP.pgp From damien.cassou at gmail.com Wed Aug 19 14:36:40 2009 From: damien.cassou at gmail.com (Damien Cassou) Date: Wed Aug 19 14:37:03 2009 Subject: [Seaside] Re: [Pharo-project] Seaside Testing in Seaside one click image In-Reply-To: <102261990908150428hd4ae9ebrabff27566085bcd4@mail.gmail.com> References: <102261990908150428hd4ae9ebrabff27566085bcd4@mail.gmail.com> Message-ID: <6ac749c10908190736p227e5b5cw15160e303da955ed@mail.gmail.com> Hi Stan, 2009/8/15 stan shepherd : > Hi David, I ran up your most recent Seaside Testing in the one click > experience. It works fine, but one of the tests failed, because > WAAllTests has been changed. The attached version passes all tests > green. this is related to Seaside and not to Pharo. Please post to the Seaside mailing list: http://seaside.st/community/mailinglist -- Damien Cassou http://damiencassou.seasidehosting.st "Lambdas are relegated to relative obscurity until Java makes them popular by not having them." James Iry From ssastre at seaswork.com Thu Aug 20 00:12:00 2009 From: ssastre at seaswork.com (Sebastian Sastre) Date: Thu Aug 20 00:16:41 2009 Subject: [Seaside] [ANN] AwesomeAtom -- FriendlyFeeds In-Reply-To: Message-ID: more standards, cool thanks for sharing sebastian > -----Mensaje original----- > De: seaside-bounces@lists.squeakfoundation.org > [mailto:seaside-bounces@lists.squeakfoundation.org] En nombre > de Tobias Pape > Enviado el: Wednesday, August 19, 2009 11:00 > Para: Seaside - general discussion > Asunto: [Seaside] [ANN] AwesomeAtom -- FriendlyFeeds > > Hello ?, > > In order to support Atom Feeds in Seaside 2.9, I took Ramon Leons > AtomFeed package and rewrote it accroding to RSRSS in Seaside 2.9; > AwesomeAtom is in alpha stage and available on > SqueakMap/SqueakSource/Universe. > Additionally, there is a Seaside Decoration and Configuration for > inserting Feed references () (RSS and Atom) into Applications, > needs Seaside 2.9, RSRSS, and AwesomeAtom; > FriendlyFeeds is in alpha stage and available on > SqueakMap/SqueakSource/Universe. > > Have fun with these Packages. > > So Long, > -Tobias Pape From astares at gmx.de Thu Aug 20 06:48:02 2009 From: astares at gmx.de (Torsten Bergmann) Date: Thu Aug 20 06:48:06 2009 Subject: [Seaside] =?utf-8?q?=22Learning_Seaside=E2=80=9D_Videos?= Message-ID: <20090820064802.316020@gmx.net> 3 new videos done by James Foster: http://programminggems.wordpress.com/2009/08/19/videos/ you can also vote for it on http://www.dzone.com/links/getting_started_with_the_seaside_web_application.html Have fun T. -- Neu: GMX Doppel-FLAT mit Internet-Flatrate + Telefon-Flatrate f?r nur 19,99 Euro/mtl.!* http://portal.gmx.net/de/go/dsl02 From renggli at gmail.com Thu Aug 20 07:16:51 2009 From: renggli at gmail.com (Lukas Renggli) Date: Thu Aug 20 07:16:55 2009 Subject: =?windows-1252?Q?Re=3A_=5BSeaside=5D_=22Learning_Seaside=94_Videos?= In-Reply-To: <20090820064802.316020@gmx.net> References: <20090820064802.316020@gmx.net> Message-ID: <67628d690908200016q321844ffsc6dac32e2e4a9e49@mail.gmail.com> These are great videos! Thank you James for sharing. I cleaned up the page on videos on the website and put the links there: http://www.seaside.st/documentation/videos Cheers, Lukas 2009/8/20 Torsten Bergmann : > 3 new videos done by James Foster: > > http://programminggems.wordpress.com/2009/08/19/videos/ > > you can also vote for it on > http://www.dzone.com/links/getting_started_with_the_seaside_web_application.html > > Have fun > T. > -- > Neu: GMX Doppel-FLAT mit Internet-Flatrate + Telefon-Flatrate > f?r nur 19,99 Euro/mtl.!* http://portal.gmx.net/de/go/dsl02 > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside > From goran at krampe.se Thu Aug 20 21:27:12 2009 From: goran at krampe.se (=?ISO-8859-1?Q?G=F6ran_Krampe?=) Date: Thu Aug 20 21:27:38 2009 Subject: [Seaside] simple example for a simple man? In-Reply-To: <4A846120.8070606@cox.net> References: <4A845BBE.1050701@cox.net> <4A845DD9.2070004@cox.net> <4a5dbd430908131144j561d9d0fsa47dedd180375d86@mail.gmail.com> <4A846120.8070606@cox.net> Message-ID: <4A8DBFB0.3090003@krampe.se> Lawson English wrote: > Thanks. At this point, my needs are exceedingly simple. But, its not > obvious to me how to get started. At the simplest level, I simply want > to create one or more webpages of arbitrary GUI complexity that can POST > input to some external website from a page served in localhost. The Seaside GUI bit I presume you can easily pick up in Seaside tutorials (the book) or whatever. Making POSTs from Squeak can be done using various means - Curl (you need the plugin though) and HTTPSocket (old "standard" stuff) probably all can do it - but you also have HttpClient: http://map.squeak.org/packagebyname/httpclient I have used that package to build a rather advanced HTTP robot for a customer and it does POSTS etc quite handily - even through HTTPS using stunnel as a relay. regards, G?ran From johnmci at smalltalkconsulting.com Fri Aug 21 00:10:03 2009 From: johnmci at smalltalkconsulting.com (John M McIntosh) Date: Fri Aug 21 00:10:15 2009 Subject: [Seaside] Mac carbon VM 4.2.1Beta1U posted Message-ID: <3D008C96-63BA-4F66-A2C1-F60E6135C6A8@smalltalkconsulting.com> I've posted a macintosh carbon VM 4.2.1Beta1U to the usual places found via http://www.smalltalkconsulting.com/squeak.html If there are no immediate problems then people should consider using this VM for their one-click offerings and ensuring this is the current default macintosh VM on Squeak.org. Early adopters should email the list right away in case of problems. As a reminder the 4.x series of macintosh VM have clean 32bit address space support, IPV6, full UTF-8 for file names and clipboard, Closure support when activated by the image, and BitBlt fixes to enable beautiful fonts via Squeak on the macintosh, plus the VM plugins folder has all the latest plugins. If you *must* run a 3.8.x flavor of VM (why?) then you should use the latest fixed version which is 3.8.21b1 3.8.20 & 3.8.19 both have flaws causing image save failures, and 3.8.17 or earlier are not 32bit clean hence more probable to crash on you. A warning about BitBltPlugin.bundle distributed as "BitBltPlugin.v1.0.0b1.FreeTypePlus.bundle.zip" This was an override to provide some better font behaviour rendering that was done in early 2007. If you have this on your computer in a Plugins folder or think you need it, beware it will override the NEW fixed bitblt and you wonder why your fonts don't render correctly in the last font efforts for Squeak. Therefore you should delete all/any copies of BitBltPlugin.bundle Changes: There are some fixes to enable this VM to run a 64bit based Image, this is still a 32bit VM btw, and now the VM comes with a macintosh QuickLook plugin that renders the thumbnail out of newer eToys projects. Plus it has the Aug fixes for rendering antialiased StrikeFonts. Other fixes detailed below: 4.2.1b1 We update our VM to VMMaker-dtl.135 Fixes Reference Mantis 7260: some Float comparison primitives don't handle Nan correctly Enable VM to run 64 bit images as a 32 bit VM Reference Mantis 5238: Integer bug on 64bit image/32bit VM Reference Mantis 7376: Bug in BitBlt (rgbMul sets alpha to 0) Reference Mantis 7377: VM Crashes when setting port on destroyed socket (Win32) Reference Mantis 7384: GeniePlugin does not compile on closure- enabled images Added eToys suggestion for a read only resource directory when sandboxed. Info.plist gets a UTI and the SqueakResourceDirectory path Via Bert Freudenberg added QuickLook plugin for rendering eToys project image Thumbnail -- = = = ======================================================================== John M. McIntosh Twitter: squeaker68882 Corporate Smalltalk Consulting Ltd. http://www.smalltalkconsulting.com = = = ======================================================================== From philippe.marschall at gmail.com Fri Aug 21 05:08:19 2009 From: philippe.marschall at gmail.com (Philippe Marschall) Date: Fri Aug 21 05:08:21 2009 Subject: [Seaside] simple example for a simple man? In-Reply-To: <4A8DBFB0.3090003@krampe.se> References: <4A845BBE.1050701@cox.net> <4A845DD9.2070004@cox.net> <4a5dbd430908131144j561d9d0fsa47dedd180375d86@mail.gmail.com> <4A846120.8070606@cox.net> <4A8DBFB0.3090003@krampe.se> Message-ID: <66666f210908202208u65e5eef1rfabdc3afa0bea4f3@mail.gmail.com> 2009/8/20 G?ran Krampe : > Lawson English wrote: >> >> Thanks. At this point, my needs are exceedingly simple. But, its not >> obvious to me how to get started. At the simplest level, I simply want to >> create one or more webpages of arbitrary GUI complexity that can POST input >> to some external website from a page served in localhost. > > The Seaside GUI bit I presume you can easily pick up in Seaside tutorials > (the book) or whatever. > > Making POSTs from Squeak can be done using various means - Curl (you need > the plugin though) and HTTPSocket (old "standard" stuff) probably all can do > it - but you also have HttpClient: > > http://map.squeak.org/packagebyname/httpclient Download link is broken since a long time ago. Cheers Philippe From goran at krampe.se Fri Aug 21 07:40:01 2009 From: goran at krampe.se (=?UTF-8?B?R8O2cmFuIEtyYW1wZQ==?=) Date: Fri Aug 21 07:40:14 2009 Subject: [Seaside] simple example for a simple man? In-Reply-To: <66666f210908202208u65e5eef1rfabdc3afa0bea4f3@mail.gmail.com> References: <4A845BBE.1050701@cox.net> <4A845DD9.2070004@cox.net> <4a5dbd430908131144j561d9d0fsa47dedd180375d86@mail.gmail.com> <4A846120.8070606@cox.net> <4A8DBFB0.3090003@krampe.se> <66666f210908202208u65e5eef1rfabdc3afa0bea4f3@mail.gmail.com> Message-ID: <4A8E4F51.3050608@krampe.se> Philippe Marschall wrote: > 2009/8/20 G?ran Krampe : >> Lawson English wrote: >>> Thanks. At this point, my needs are exceedingly simple. But, its not >>> obvious to me how to get started. At the simplest level, I simply want to >>> create one or more webpages of arbitrary GUI complexity that can POST input >>> to some external website from a page served in localhost. >> The Seaside GUI bit I presume you can easily pick up in Seaside tutorials >> (the book) or whatever. >> >> Making POSTs from Squeak can be done using various means - Curl (you need >> the plugin though) and HTTPSocket (old "standard" stuff) probably all can do >> it - but you also have HttpClient: >> >> http://map.squeak.org/packagebyname/httpclient > > Download link is broken since a long time ago. Aha! But the SqueakMap cache has it, so installing/downloading using the SqueakMap Package Loader should work. Otherwise, download this URL (lesser known SM trick) and rename to proper filename: http://map.squeak.org/packagebyname/httpclient/autoversion/1/cache regards, G?ran PS. I might have some patches for it in that project I built... Ask for them and I can dig. :) From john.chludzinski at gmail.com Fri Aug 21 15:20:31 2009 From: john.chludzinski at gmail.com (John Chludzinski) Date: Fri Aug 21 15:20:34 2009 Subject: [Seaside] Seaside + Croquet/Cobalt/Jabber question? Message-ID: <27a48dd60908210820l152db517tea38b21cf0c6af57@mail.gmail.com> I'm currently developing a suite of web apps using Croquet/Cobalt + Seaside (? la [Nikolay Suslov] http://nsuslovi.blogspot.com/2009/07/seaside-29-and-comet-to-control.html). And I want to see how I might exploit Croquet to improve my Seaside apps. My current effort to exploit Cobalt is to develop a Jabber based Seaside chat app. I've tried first to successively connect and send a message via some Jabber server. Using my jabber.org account I tried JabberBasicTests>>test03ConnectAndMessage. This calls JabberBasicTests>>setupJabberOrgService: and JabberBasicTests>>newJabberIdentity. I added passwd: to newJabberIdentity. This doesn't appear to work. What am I doing wrong? ---John -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.squeakfoundation.org/pipermail/seaside/attachments/20090821/537f9101/attachment.htm From ssastre at seaswork.com Fri Aug 21 18:00:42 2009 From: ssastre at seaswork.com (Sebastian Sastre) Date: Fri Aug 21 18:05:25 2009 Subject: [Seaside] callbacks of disabled submit buttons Message-ID: Hi there, fact: when you disable a submit button it's not going to be serialized. problem: you want to disable the button after the click but you also want to execute its callback. This is often needed. For instance a login form. You want the user to login and wait for the result. The disable of the button is just to make it idempotent until the result is received preventing silly server load. Another example is a "confirm purchase" kind of button where a capture of money is made. Any ideas? sebastian From renggli at gmail.com Fri Aug 21 19:01:56 2009 From: renggli at gmail.com (Lukas Renggli) Date: Fri Aug 21 19:01:58 2009 Subject: [Seaside] callbacks of disabled submit buttons In-Reply-To: References: Message-ID: <67628d690908211201w55534a19s2b556fa79a9cef60@mail.gmail.com> > fact: when you disable a submit button it's not going to be serialized. Indeed, see http://bit.ly/2J2Oma. > This is often needed. For instance a login form. You want the user to login and > wait for the result. The disable of the button is just to make it idempotent > until the result is received preventing silly server load. You are looking for read-only, I guess. The difference between disabled and read-only is described a bit above: http://bit.ly/4xG24Y. > Another example is a "confirm purchase" kind of button where a capture of money > is made. I wouldn't count on the we client, to make it secure you *have to* prevent a double confirmation on the server, for example by checking in your model or using #isolate: Cheers, Lukas -- Lukas Renggli http://www.lukas-renggli.ch From richard.durr at googlemail.com Fri Aug 21 19:35:36 2009 From: richard.durr at googlemail.com (Richard Durr) Date: Fri Aug 21 19:35:38 2009 Subject: [Seaside] Uploading an image using AJAX and iframes Message-ID: <6fcb8b1b0908211235o56d1b84ft5876fa32c6e65f9f@mail.gmail.com> Hello List, I am currently implementing an ImageChooser Component, where I like users to upload an image by submtiting the form containing the fileUpload using an iframe like this: html form id: (self formId); attributeAt: #target put: (self frameName); multipart; method: #post; with: [ html fileUpload id: self uploadInputId; accept: 'png jpg jpeg gif'; on: #image of: self. html div with: [ html anchor onClick: (html javascript factory form id: self formId; submit); with: 'Upload Image'. ]. ]. html iframe name: self frameName; style: 'display: none; widht: 0; height: 0; border: 0;'. This works ? but it loads the entire page into the iframe, again. I would rather just load an html stub containing a javascript which indicates if the upload was successful or not and executes in the context of the original html. I tried to return a response during the setting of my component's image instance-variable but this does not work. Has anybody an Idea? RD -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.squeakfoundation.org/pipermail/seaside/attachments/20090821/49c2f650/attachment.htm From ssastre at seaswork.com Fri Aug 21 19:32:14 2009 From: ssastre at seaswork.com (Sebastian Sastre) Date: Fri Aug 21 19:37:24 2009 Subject: [Seaside] callbacks of disabled submit buttons In-Reply-To: <67628d690908211201w55534a19s2b556fa79a9cef60@mail.gmail.com> Message-ID: > make it idempotent > > until the result is received preventing silly server load. > > You are looking for read-only, I guess. The difference between > disabled and read-only is described a bit above: http://bit.ly/4xG24Y. > That sounds about right. Figuring out how... > > Another example is a "confirm purchase" kind of button > where a capture of money > > is made. > > I wouldn't count on the we client, to make it secure you *have to* > prevent a double confirmation on the server, for example by checking > in your model or using #isolate: > > Cheers, > Lukas sure. What I mean is both. Actually is decided by the server but as an extra measure I want additionally to prevent UI side excess of clicks. Never bad to remember it anyway. thanks for the pointer sebastian From renggli at gmail.com Fri Aug 21 19:57:38 2009 From: renggli at gmail.com (Lukas Renggli) Date: Fri Aug 21 19:57:40 2009 Subject: [Seaside] Uploading an image using AJAX and iframes In-Reply-To: <6fcb8b1b0908211235o56d1b84ft5876fa32c6e65f9f@mail.gmail.com> References: <6fcb8b1b0908211235o56d1b84ft5876fa32c6e65f9f@mail.gmail.com> Message-ID: <67628d690908211257i42dc6187l81b2f505732a325a@mail.gmail.com> > This works ? but it loads the entire page into the iframe, again. I would > rather just load an html stub containing a javascript which indicates if the > upload was successful or not and executes in the context of the original > html. I tried to return a response during the setting of my component's > image instance-variable but this does not work. > Has anybody an Idea? You need to start a new render loop (to avoid that the parent page is displayed within the frame) and display some component. Try to add the following code at the end of the method #image: ... WARenderLoop new show: (WAComponent new addMessage: 'File successfully uploaded'; yourself) withToolFrame: false Replace the expression within brackets with your own component that displays the confirmation. Cheers, Lukas -- Lukas Renggli http://www.lukas-renggli.ch From ssastre at seaswork.com Fri Aug 21 20:10:18 2009 From: ssastre at seaswork.com (Sebastian Sastre) Date: Fri Aug 21 20:15:01 2009 Subject: [Seaside] callbacks of disabled submit buttons In-Reply-To: Message-ID: Ok, here I go.. for the record: f* buttons won't work as supossed with readonly. Is nothing to do with Seaside. They should, but they won't. Don't ask why. So.. a way I've found to (almost) achieve this feature (without using ajax) is to let the button enabled and observe the submit event. If the submit event was already received (you need to store that state at DOM) then react stopping it to prevent the extra POST's. peace & love sebastian > -----Mensaje original----- > De: seaside-bounces@lists.squeakfoundation.org > [mailto:seaside-bounces@lists.squeakfoundation.org] En nombre > de Sebastian Sastre > Enviado el: Friday, August 21, 2009 16:32 > Para: 'Seaside - general discussion' > Asunto: RE: [Seaside] callbacks of disabled submit buttons > > > make it idempotent > > > until the result is received preventing silly server load. > > > > You are looking for read-only, I guess. The difference between > > disabled and read-only is described a bit above: > http://bit.ly/4xG24Y. > > > That sounds about right. Figuring out how... > > > > > Another example is a "confirm purchase" kind of button > > where a capture of money > > > is made. > > > > I wouldn't count on the we client, to make it secure you *have to* > > prevent a double confirmation on the server, for example by checking > > in your model or using #isolate: > > > > Cheers, > > Lukas > > sure. What I mean is both. Actually is decided by the server > but as an extra > measure I want additionally to prevent UI side excess of clicks. > > Never bad to remember it anyway. > > thanks for the pointer > sebastian > > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside From squeakman at gmail.com Sat Aug 22 00:21:19 2009 From: squeakman at gmail.com (Squeaker) Date: Sat Aug 22 00:21:49 2009 Subject: [Seaside] Wicket Message-ID: Greetings, A colleague just introduced me to Wicket http://wicket.apache.org/ I have spent about 30 minutes browsing their docs. Just wondering if anyone has any experience with Wicket? How does it compare with Seaside? I will be looking into Wicket over the next week or so, if there is any interest, I will share what I learn. Cheers, Frank From philippe.marschall at gmail.com Sat Aug 22 08:06:51 2009 From: philippe.marschall at gmail.com (Philippe Marschall) Date: Sat Aug 22 08:06:53 2009 Subject: [Seaside] Wicket In-Reply-To: References: Message-ID: <66666f210908220106t57907c11pbac56e340d063757@mail.gmail.com> 2009/8/22 Squeaker : > Greetings, > > A colleague just introduced me to Wicket http://wicket.apache.org/ > I have spent about 30 minutes browsing their docs. > > Just wondering if anyone has any experience with Wicket? Yes. > How does it compare with Seaside? The tempting is IMHO the biggest difference. In Wicket there's an HTML template associated with each (regular) component whereas Seaside does programmatic HTML rendering. AJAX is different as well. In the part that I saw you tell the response which components should be updated and Wicket renders them and takes care of generating ids. JavaScript library integration is available as well though. The HTML element objects in Wicket seem closer to AIDA elements than Seaside tags. They can be augmented with behaviors which I haven't fully grokked yet. Wicket has by default ugly URLs as Seaside but you have better control over URL generation in Wicket. Wicket makes a difference between "page" components and "normal" components. Cheers Philippe From richard.durr at googlemail.com Sat Aug 22 10:17:10 2009 From: richard.durr at googlemail.com (Richard Durr) Date: Sat Aug 22 10:17:13 2009 Subject: [Seaside] Uploading an image using AJAX and iframes In-Reply-To: <67628d690908211257i42dc6187l81b2f505732a325a@mail.gmail.com> References: <6fcb8b1b0908211235o56d1b84ft5876fa32c6e65f9f@mail.gmail.com> <67628d690908211257i42dc6187l81b2f505732a325a@mail.gmail.com> Message-ID: <6fcb8b1b0908220317i78a01a90vad2cfd58a92dd839@mail.gmail.com> Works great, thank you!I have to learn so much more, there is still so much that I haven't fully groked, sigh. ^^ RD On Fri, Aug 21, 2009 at 9:57 PM, Lukas Renggli wrote: > > This works ? but it loads the entire page into the iframe, again. I would > > rather just load an html stub containing a javascript which indicates if > the > > upload was successful or not and executes in the context of the original > > html. I tried to return a response during the setting of my component's > > image instance-variable but this does not work. > > Has anybody an Idea? > > You need to start a new render loop (to avoid that the parent page is > displayed within the frame) and display some component. Try to add the > following code at the end of the method #image: > > ... > WARenderLoop new > show: (WAComponent new > addMessage: 'File successfully uploaded'; > yourself) > withToolFrame: false > > Replace the expression within brackets with your own component that > displays the confirmation. > > Cheers, > Lukas > > -- > Lukas Renggli > http://www.lukas-renggli.ch > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.squeakfoundation.org/pipermail/seaside/attachments/20090822/851cf382/attachment.htm From timothy at churchmilitant.org Thu Aug 27 16:02:08 2009 From: timothy at churchmilitant.org (Timothy James Ziebart) Date: Thu Aug 27 16:02:12 2009 Subject: [Seaside] Updating non Smallfaces component. Message-ID: <4A96AE00.9020403@churchmilitant.org> Hi All, I have an object SearchInterface, a subclass of WASmallFacesComposite. On SearchInterface there are rendered a number of SFListBoxes and SFButtons. Included are three listboxes that are cascaded and the control was created like this: createRegionListBox |pane panes list| list := self session findCodeByParent: 'region'. panes := OrderedCollection with: ( self paneNamed: 'nhoodList') with: ( self paneNamed: 'cityList'). pane := SFListBox new. pane name: 'regionList'; items: list; lines: 5; printSelector: #asString; owner: self; when: #changed: send: #regionSelectionChanged: to: self refresh: panes. ^pane The listboxes work correctly. What I need to do is have a GoogleMap rendered on SearchInterfaces as well. The map will not be a WAComponent as I need to render it as new whenever the listbox contents change. But not before an Update button is pressed. Such as: createUpdateButton | pane panes | panes := OrderedCollection with: (*self paneNamed: 'mapDisplay')*. pane := SFButton new. pane label: 'Update Map'; name: 'updateButton'; when: #clicked send: #updateButtonPressed to: self refresh: panes. ^pane The question is how do I create *mapDisplay*? mapDisplay is rendered like this everytime the update button is pressed. map ^LBCGoogleMap new setCenter: model coordinates zoom: model zoom; addControl: GControl largeMapControl; setUIToDefault; yourself. Can I place it on a SmallFace component that will render as I desire or do I replace it with non SmallFace components? Thank you for any assistance. Tim From grshiplett2 at yahoo.com Thu Aug 27 19:14:29 2009 From: grshiplett2 at yahoo.com (Robert Shiplett) Date: Thu Aug 27 19:14:33 2009 Subject: [Seaside] non-HTML output from Seaside 2.9 Message-ID: <862733.83290.qm@web51607.mail.re2.yahoo.com> Yesterday I succeeded in running the Seaside Counter demo in Curl (www.curl.com) in Squeak 3.10 using Alpha 4 without any XML or HTML output.Doing so required a great many changes to Seaside: MIME type defaults and hard-coded class references. ?Seaside should be markup language neutral ( the Curl RTE is in use at about 30% of the global Fortune 100 and in one major government )The poblem most difficult to detect was affecting WAApplication as there is a preferences default set to MIME 'text/html' in the configuration dict up at WAEntryPoint. ?Even when all other problems were corrected, the Curl MIME type was ignored for my builder in its render context.Any generic method used in rendering which hard-codes html document type or html content type or names an Html class remains problematic. ?In many cases I could only proceed by creating my own subclass and hardcoding "Curl" in place of "Html" in the offending methods. ?At a few palces more generic behavior is actually in the HTml subclass than in the supposed generic parent class (a simpel refactoring issue)It was not possible to produce a package for Curl web content as an alternative to an Html package or XML package or XHTML package due to these issues: my package only runs with some 14 classes hacked to obviate hard-coded HTML bias: so I have a working IMAGE and not a working PACKAGE.Some Mime problems violated the encapsulation of the MIME classes by hard-coding 'text' and 'html' as main and sub values. ?This should not be exposed without good reason. ?Mime types should be parsed, not hacked.WABuilder has bias in many methods that interfered with creating a Curl subclassHttpResponse has numerous methods hard-coded to contentTypeHtmlWAPrettyPrintedDocument has HTML bias in a few methodsWAUrl has bias in printON: aStreamWARequestHandlingConfiguration has bias in >> describeOn: config:WARequestContext has HTML bias in several methodsWAREnderPhaseContinuation has a bias in >>processRenderingWhen I found WASlime >>findBrushSelectors I did not know how to proceed, so I removed the compteing HTML subclasses and left only my Curl subclassesWARoot is worrosime of the various methods in WAUpdateRootVisitorWASession has methods such as beXHTMLString which have to be overridden whenever they are meaningless for non-XML markupIt was difficutl to ascertain why the WARenderContext of the Canvas was always holding a WAHtmlDocument.WriteStream has a default contentType tied to MIMEDocument outside SeasideObject has a default contetType the sameI -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.squeakfoundation.org/pipermail/seaside/attachments/20090827/5b63fbbb/attachment.htm From juan.burella at gmail.com Thu Aug 27 20:01:41 2009 From: juan.burella at gmail.com (Burella Juan M.) Date: Thu Aug 27 20:01:45 2009 Subject: [Seaside] Updating non Smallfaces component. In-Reply-To: <4A96AE00.9020403@churchmilitant.org> References: <4A96AE00.9020403@churchmilitant.org> Message-ID: <60481d60908271301v2d8ab726j1c02af6acc82e17c@mail.gmail.com> Hi Tim, Let me see if I've understood what are you doing. You need to: -Create (or modify?) a LBCGoogleMap when a ListBox changes -Render or update the LBCGoogleMap when a Button is clicked. You said LBCGoogleMap is not a WAComponent, what it is? If it's a SFObject, then you can add to your SearcherInterface sending as argument of #addSubpane: message, as already knows how to answer #name (remember, it must answer #name properly, that is, not the #name method in Object). Then you can have: >>initialize .... self addSubpane: self map ... >>map ^LBCGoogleMap new name: 'mapDisplay'; " configure the mapDisplay's name" setCenter: model coordinates zoom: model zoom; addControl: GControl largeMapControl; setUIToDefault; yourself. If when the list changes, you have to re-create the mapDisplay, in the #regionSelectionChanged "callback" (the message sent when the list is #changed) you should replace it as follows: oldMap := self paneNamed: 'mapDisplay'. self replacePane: oldMap with: self map And to refresh properly the mapDisplay it should be rendered inside a tag with id equals to its styleId, like this: LBCGoogleMap >>renderContentOn: aRenderer aRenderer div id: self styleId; with: [ "render your component" ] and must answer #refreshOn: (please see #when:send:to:refresh: section in http://wiki.squeak.org/squeak/6093). You can implement it this way: refreshOn: aRenderer "Support for the events" self renderContentOn: aRenderer You will need to preserve the same #styleId, to do this simply send the message styleId: 'anyId' to the mapDisplay. Another way to solve this is adding a SFPanel which contains the mapDisplay and refresh it when the button is clicked. Regards Juan M. On Thu, Aug 27, 2009 at 1:02 PM, Timothy James Ziebart < timothy@churchmilitant.org> wrote: > Hi All, > > I have an object SearchInterface, a subclass of WASmallFacesComposite. On > SearchInterface there are rendered a number of SFListBoxes and SFButtons. > Included are three listboxes that are cascaded and the control was created > like this: > > createRegionListBox > |pane panes list| > list := self session findCodeByParent: 'region'. > panes := OrderedCollection > with: ( self paneNamed: 'nhoodList') > with: ( self paneNamed: 'cityList'). > pane := SFListBox new. > pane > name: 'regionList'; > items: list; > lines: 5; > printSelector: #asString; > owner: self; > when: #changed: > send: #regionSelectionChanged: > to: self > refresh: panes. > ^pane > > The listboxes work correctly. What I need to do is have a GoogleMap > rendered on SearchInterfaces as well. The map will not be a WAComponent as > I need to render it as new whenever the listbox contents change. But not > before an Update button is pressed. Such as: > > createUpdateButton > > | pane panes | > panes := OrderedCollection with: (*self paneNamed: 'mapDisplay')*. > pane := SFButton new. > pane > label: 'Update Map'; > name: 'updateButton'; > when: #clicked > send: #updateButtonPressed > to: self > refresh: panes. > ^pane > > The question is how do I create *mapDisplay*? mapDisplay is rendered like > this everytime the update button is pressed. > > map > > ^LBCGoogleMap new > setCenter: model coordinates zoom: model zoom; > addControl: GControl largeMapControl; > setUIToDefault; > yourself. > > Can I place it on a SmallFace component that will render as I desire or do > I replace it with non SmallFace components? Thank you for any assistance. > > Tim > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.squeakfoundation.org/pipermail/seaside/attachments/20090827/e70cc2f8/attachment.htm From grshiplett2 at yahoo.com Thu Aug 27 20:10:11 2009 From: grshiplett2 at yahoo.com (Robert Shiplett) Date: Thu Aug 27 20:10:14 2009 Subject: [Seaside] Seaside versus HttpView 2 for non-HTML output Message-ID: <263778.95853.qm@web51606.mail.re2.yahoo.com> I think it may be worth noting that Seaside 2.9 Alpha 4 took about 10 hours of conversion and debugging to generate a single page of nn-XML and non-HTML content using ints Builder ?and rendering framework (i.e. not plain-text output, but a working "counter" demo aplI also took about 10 hours to achieve the same in HttpView2 (both on the same Squeal 3.10.2 base )The problems in HttpView2 may appear obvious: at least one problem in Seaside was not obvious ( preferences dict in configuration of WAEntry Point )This task is documetned at develoeprs.curl.com, http://eclectic-pencil.blogspot.com/ and?http://aule-browser.blogspot.com/I have posted one reason to take Curl seriously over at??http://aule-browser.blogspot.com/2009/08/poetry-of-curl-macros.htmlfora nyone unfamiliar with the power of a homoiconic wweb content langauge such as Curl 7.0 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.squeakfoundation.org/pipermail/seaside/attachments/20090827/75edc65c/attachment.htm From timothy at churchmilitant.org Thu Aug 27 20:12:40 2009 From: timothy at churchmilitant.org (Timothy James Ziebart) Date: Thu Aug 27 20:12:44 2009 Subject: [Seaside] Updating non Smallfaces component. In-Reply-To: <60481d60908271301v2d8ab726j1c02af6acc82e17c@mail.gmail.com> References: <4A96AE00.9020403@churchmilitant.org> <60481d60908271301v2d8ab726j1c02af6acc82e17c@mail.gmail.com> Message-ID: <4A96E8B8.6000304@churchmilitant.org> Hi Juan, thanks for responding. The google map is an object that implements a renderOn method. It is the GoogleMaps package by James Foster. http://seaside.gemstone.com/ss/GoogleMaps. I have created WAComponent and non WAComponent versions. I do not have a SmallFace equivalent widget. Maybe that is my question - is it possible to build a SmallFaces widget that will take advantage of the features available? Tim Burella Juan M. wrote: > Hi Tim, > Let me see if I've understood what are you doing. You need to: > > -Create (or modify?) a LBCGoogleMap when a ListBox changes > -Render or update the LBCGoogleMap when a Button is clicked. > > You said LBCGoogleMap is not a WAComponent, what it is? If it's a > SFObject, then you can add to your SearcherInterface sending as > argument of #addSubpane: message, as already knows how to answer > #name (remember, it must answer #name properly, that is, not the #name > method in Object). > > Then you can have: > > >>initialize > .... > self addSubpane: self map > ... > > > >>map > ^LBCGoogleMap new > name: 'mapDisplay'; " configure the mapDisplay's name" > setCenter: model coordinates zoom: model zoom; > addControl: GControl largeMapControl; > setUIToDefault; > yourself. > > If when the list changes, you have to re-create the mapDisplay, in the > #regionSelectionChanged "callback" (the message sent when the list is > #changed) you should replace it as follows: > > oldMap := self paneNamed: 'mapDisplay'. > self replacePane: oldMap with: self map > > And to refresh properly the mapDisplay it should be rendered inside a > tag with id equals to its styleId, like this: > > LBCGoogleMap >>renderContentOn: aRenderer > > aRenderer div > id: self styleId; > with: [ "render your component" ] > > and must answer #refreshOn: (please see #when:send:to:refresh: section > in http://wiki.squeak.org/squeak/6093). You can implement it this way: > > refreshOn: aRenderer > "Support for the events" > > self renderContentOn: aRenderer > > > You will need to preserve the same #styleId, to do this simply send > the message styleId: 'anyId' to the mapDisplay. Another way to solve > this is adding a SFPanel which contains the mapDisplay and refresh it > when the button is clicked. > > Regards > Juan M. > > On Thu, Aug 27, 2009 at 1:02 PM, Timothy James Ziebart > > wrote: > > Hi All, > > I have an object SearchInterface, a subclass of > WASmallFacesComposite. On SearchInterface there are rendered a > number of SFListBoxes and SFButtons. Included are three listboxes > that are cascaded and the control was created like this: > > createRegionListBox > |pane panes list| > list := self session findCodeByParent: 'region'. > panes := OrderedCollection > with: ( self paneNamed: 'nhoodList') > with: ( self paneNamed: 'cityList'). > pane := SFListBox new. > pane > name: 'regionList'; > items: list; > lines: 5; > printSelector: #asString; > owner: self; > when: #changed: > send: #regionSelectionChanged: > to: self > refresh: panes. > ^pane > > The listboxes work correctly. What I need to do is have a > GoogleMap rendered on SearchInterfaces as well. The map will not > be a WAComponent as I need to render it as new whenever the > listbox contents change. But not before an Update button is > pressed. Such as: > > createUpdateButton > > | pane panes | > panes := OrderedCollection with: (*self paneNamed: 'mapDisplay')*. > pane := SFButton new. > pane > label: 'Update Map'; > name: 'updateButton'; > when: #clicked > send: #updateButtonPressed > to: self > refresh: panes. > ^pane > > The question is how do I create *mapDisplay*? mapDisplay is > rendered like this everytime the update button is pressed. > > map > > ^LBCGoogleMap new > setCenter: model coordinates zoom: model zoom; > addControl: GControl largeMapControl; > setUIToDefault; > yourself. > > Can I place it on a SmallFace component that will render as I > desire or do I replace it with non SmallFace components? Thank > you for any assistance. > > Tim > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside > > > ------------------------------------------------------------------------ > > _______________________________________________ > seaside mailing list > seaside@lists.squeakfoundation.org > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside > From philippe.marschall at gmail.com Thu Aug 27 20:41:43 2009 From: philippe.marschall at gmail.com (Philippe Marschall) Date: Thu Aug 27 20:41:45 2009 Subject: [Seaside] non-HTML output from Seaside 2.9 In-Reply-To: <862733.83290.qm@web51607.mail.re2.yahoo.com> References: <862733.83290.qm@web51607.mail.re2.yahoo.com> Message-ID: <66666f210908271341v140e591rcff7a6f8d6e9aade@mail.gmail.com> Thanks for you feedback, we're very much interested in this. You're the fist one to do this so you're unfortunately the first one finding all these problems. It would help us a lot if you have any test cases. That would help us find the bugs, fix them and make sure they stay fixed. Also, can you give us a short overview over the Curl syntax, document structure and escaping? At least I am not familiar with it. 2009/8/27 Robert Shiplett : > Yesterday I succeeded in running the Seaside Counter demo in Curl > (www.curl.com) in Squeak 3.10 using Alpha 4 without any XML or HTML output. > > Doing so required a great many changes to Seaside: MIME type defaults and > hard-coded class references. ?Seaside should be markup language neutral ( > the Curl RTE is in use at about 30% of the global Fortune 100 and in one > major government ) > > The poblem most difficult to detect was affecting WAApplication as there is > a preferences default set to MIME 'text/html' in the configuration dict up > at WAEntryPoint. ?Even when all other problems were corrected, the Curl MIME > type was ignored for my builder in its render context. This setting only affects the Content-Type HTTP header. When you create a builder you must make sure it has the correct renderer. The easiest way is to send #bilder to your renderer class. > Any generic method used in rendering which hard-codes html document type or > html content type or names an Html class remains problematic. ?In many cases > I could only proceed by creating my own subclass and hardcoding "Curl" in > place of "Html" in the offending methods. ?At a few palces more generic > behavior is actually in the HTml subclass than in the supposed generic > parent class (a simpel refactoring issue) > > It was not possible to produce a package for Curl web content as an > alternative to an Html package or XML package or XHTML package due to these > issues: my package only runs with some 14 classes hacked to obviate > hard-coded HTML bias: so I have a working IMAGE and not a working PACKAGE. > > Some Mime problems violated the encapsulation of the MIME classes by > hard-coding 'text' and 'html' as main and sub values. ?This should not be > exposed without good reason. ?Mime types should be parsed, not hacked. Agreed, can you give examples? > WABuilder has bias in many methods that interfered with creating a Curl > subclass Can you give examples? > HttpResponse has numerous methods hard-coded to contentTypeHtml That comes from Comanche. > WAPrettyPrintedDocument has HTML bias in a few methods Indeed but it is a subclass of WAHtmlDocument. It is used for the halos to pretty print HTML. I don't think this class is useful in your context. > WAUrl has bias in printON: aStream That is true. Do you use #printOn: or do you use #encodeOn: through your document class? > WARequestHandlingConfiguration has bias in >> describeOn: config: Do you mean the default document class? > WARequestContext has HTML bias in several methods There are the class extensions from the canvas package. They should cause not problem as long as you don't send them. Then there are various defaults if no entry point is in the handler chain. As long as your entry point implements the correct methods they shouldn't be used. > WAREnderPhaseContinuation has a bias in >>processRendering True, but that is configurable in your application with the renderPhaseContinuationClass property. > When I found WASlime >>findBrushSelectors I did not know how to proceed, so > I removed the compteing HTML subclasses and left only my Curl subclasses That's just a code checking utility. > WARoot is worrosime of the various methods in WAUpdateRootVisitor Can you give examples? > WASession has methods such as beXHTMLString which have to be overridden > whenever they are meaningless for non-XML markup I don't have that method. > It was difficutl to ascertain why the WARenderContext of the Canvas was > always holding a WAHtmlDocument. It should only hold on to a WADocument. Can you give examples (sorry, it's kinda repeating)? > WriteStream has a default contentType tied to MIMEDocument outside Seaside That comes from Comanche. > Object has a default contetType the same That comes from Comanche. Cheers Philippe From philippe.marschall at gmail.com Thu Aug 27 20:55:16 2009 From: philippe.marschall at gmail.com (Philippe Marschall) Date: Thu Aug 27 20:55:22 2009 Subject: [Seaside] non-HTML output from Seaside 2.9 In-Reply-To: <862733.83290.qm@web51607.mail.re2.yahoo.com> References: <862733.83290.qm@web51607.mail.re2.yahoo.com> Message-ID: <66666f210908271355l687b3b17q5bfb7efd7d4d3b84@mail.gmail.com> 2009/8/27 Robert Shiplett : > Yesterday I succeeded in running the Seaside Counter demo in Curl > (www.curl.com) in Squeak 3.10 using Alpha 4 without any XML or HTML output. > > Doing so required a great many changes to Seaside: MIME type defaults and > hard-coded class references. ?Seaside should be markup language neutral ( > the Curl RTE is in use at about 30% of the global Fortune 100 and in one > major government ) > > The poblem most difficult to detect was affecting WAApplication as there is > a preferences default set to MIME 'text/html' in the configuration dict up > at WAEntryPoint. ?Even when all other problems were corrected, the Curl MIME > type was ignored for my builder in its render context. > > Any generic method used in rendering which hard-codes html document type or > html content type or names an Html class remains problematic. ?In many cases > I could only proceed by creating my own subclass and hardcoding "Curl" in > place of "Html" in the offending methods. ?At a few palces more generic > behavior is actually in the HTml subclass than in the supposed generic > parent class (a simpel refactoring issue) > > It was not possible to produce a package for Curl web content as an > alternative to an Html package or XML package or XHTML package due to these > issues: my package only runs with some 14 classes hacked to obviate > hard-coded HTML bias: so I have a working IMAGE and not a working PACKAGE. Sorry I'm from Berne. Couldn't you just save the dirty Seaside packages as Monticello versions? We could then extract the diff. Cheers Philippe From grshiplett2 at yahoo.com Thu Aug 27 21:04:43 2009 From: grshiplett2 at yahoo.com (Robert Shiplett) Date: Thu Aug 27 21:04:46 2009 Subject: [Seaside] Seaside 2.8 versus 2.9 Alph4: non-HTML ouput Message-ID: <185163.84399.qm@web51612.mail.re2.yahoo.com> In the VW port of 2.8 I found such obvious problems as WAResponse>>beXML contentType := contentType copyReplaceAll: '/html' with: '/xml'. But in 2.9 Alpha 4 some problems remain which ?are just as obvious: WASession>>updateRoot: anHtmlRoot self application updateRoot: anHtmlRoot. anHtmlRoot meta contentType: self application contentType. anHtmlRoot meta contentScriptType: WAMimeType textHtml. anHtmlRoot htmlAttributes at: 'name' put: 'updateRoot'. anHtmlRoot beXhtml10Strict. anHtmlRoot title: 'Seaside' " on a first pass I simply flip 'textHtml' method to a new textCurl' method while trying to track down why all contetn remained HTML even after removing the HTML classes (that was the preferences dict in WAEntryPont ) " My class WACurlRoot is forced to implement a truly meaningless message?? >>beXhtml10Strictas an empty stub " if you think any and all output can be XML you have drunk too much of that X-Kool-Aid ?" The far more serious issue is attributes: this will imapct sem web in Curl output. Semweb for XHTML probably means RDFa or rdf in attributes. ?Curl has always had metadata for some macros, but other procedures have variable number options. ?Options in Curl are not XML attributes. ?They are sometimes very similar.
{br} title{link target = "_blank", href = {url "some URL}, title} One of the first problems was a method that was forcing tags to uppercase. ?Ok in XHTML; fatal in ?Curl. Notice the comma delimiter in Curl for options.

header

{heading level = 1, header}

has no couner-part in vanilla Curl. has no counter-part in vanilla Curl. Curl has no need of