February 2011 - Pharo-dev - lists.pharo.org

[Pharo-project] [BUG]: XMLRPC 1.0-alpha4 calls with -/+ 1 arg fail

by Sean P. DeNigris

It seems that 1.0-alpha4 only handles calls with one argument. The problem seems to be in following statement from XMLRPCServerRequest>>methodResponse: ret := ((srv key) perform: (srv value) with: (self params)). #perform:with: only expects one argument (per the comment) and fails if otherwise A solution is to use #perform:withArguments: like so: ret := ((srv key) perform: (srv value) withArguments: self params asArray). n.b. The parameters are stored in an OrderedCollection, which must be converted to an array Changeset attached: http://forum.world.st/file/n3328882/XMLRPCServerRequest-methodResponse.st XMLRPCServerRequest-methodResponse.st HTH, Sean -- View this message in context: http://forum.world.st/BUG-XMLRPC-1-0-alpha4-calls-with-1-arg-fail-tp3328882… Sent from the Pharo Smalltalk mailing list archive at Nabble.com.

Feb. 28, 2011

Re: [Pharo-project] Fwd: [ANN] pharo focused sprint - bern, feb 26

by Tudor Girba

Great! Let's give it a try :) Cheers, Doru On 28 Feb 2011, at 22:58, Esteban Lorenzano wrote: > Hi, > Yes I'm willing to join you to another "glamour work day" :) > Let's say next saturday? > > best, > Esteban > > El 27/02/2011, a las 11:36a.m., Tudor Girba escribiÃ³: > >> Hi, >> >> On 27 Feb 2011, at 12:04, StÃ©phane Ducasse wrote: >> >>>>>> - Lukas went over the WeakAnnouncements implementation to get the WeakMessageSend working. The code is in PharoTaskForces >>>>> >>>>> should we take action on this one? >>>>> I mean integrating it. >>>> >>>> It can be integrated, because the base code works in the same way as before. >>>> >>>> Currently, the solution is implemented through a WeakAnnouncer which subclasses Announcer. I wanted the weak thing to be merged, because the clients of announcements should benefit from the Weak implementation, but Lukas says that he does not trust the Weak support in Pharo and that if you have many WeakAnnouncers it somehow does not scale. >>> >>> ok so ...? >> >> So, we need to investigate more :). It would be cool if someone with more knowledge would take a look, too. Igor, would you have a bit of time to look into it? >> >> I will probably also give it a look to see how it works in the context of Glamour. I hope that Esteban will join me again. Esteban, what do you say :)? >> >> >>>> The problem with the current approach is that models have to take an explicit stand of whether they want weak or non weak when instantiating the announcer. Unfortunately, I do have enough technical knowledge for this. It would be cool if someone with better knowledge in this area would take a look. >>>> >>>>>> - Lukas worked on getting OB work with the changes in TextMorph from Pharo 1.2. >>>>>> >>>>>> - Adrian, Jorge and Toon worked on implementing a Debugger on top of Glamour. They got a first version working that is able to do basic actions: step, restart, step into. The challenge was to get to understand the model, and to figure out that <primitive: 19> is used for marking that a method should not be debugged. In any case, working with Glamour seemed to work quite smoothly. The code is available in the Glamorous Toolkit. Adrian, did I get it right? :) >>>>>> >>>>>> - Mircea built a new widget for searching multiple categories of items in the same time, similar to spotlight where you enter one string and you get multiple hits from multiple categories. The implementation was spawned from the OBCompletionDialog / OBCompletionRequest and currently it simply provides multiple lists one below the other. The idea is to use this kind of widget to search simultaneously for classes / methods / packages. The code is available in Glamour (Glamour-Morphic-Renderer). The widget can definitely be improved visually, but Mircea already did a nice job. >>>>> >>>>> is the widget only for glamour because I think that a widget with search list integrated should be used everywhere? >>>>> We were discussing with alain to have a look at the one in OB and push it/rewrite it as a core widget >>>> >>>> Neither the OB nor the Glamour widgets depend on Glamour, so they can be used independently. You just need to take the Request and the Dialog together. >>> >>> >>> ok where is the code? >> >> >> OBCompletionDialog / OBCompletionRequest come with the OB from Pharo 1.2. The way to use it is like: >> >> OBCompletionRequest new >> prompt: 'Find Class'; >> searchBlock: [ :string | OBCmdFindClass new findClassIn: Smalltalk pattern: string ]; >> labelBlock: [ :class | class name ]; >> iconBlock: [ :class | class browserIcon ]; >> signal. >> >> The GLMSpotterRequest / GLMMorphicSpotterDialog are in Glamour-Morphic-Renderer. You can use them like >> >> | composite classRequest methodRequest thirdRequest | >> composite := GLMSpotterRequest new. >> >> classRequest := GLMSingleSpotterRequest new >> prompt: 'Find Class'; >> searchBlock: [ :string | OBCmdFindClass new findClassIn: Smalltalk pattern: string ]; >> labelBlock: [ :class | class name ]; >> iconBlock: [ :class | class browserIcon ]. >> >> methodRequest := GLMSingleSpotterRequest new >> prompt: 'Object Selectors'; >> searchBlock: [ :string | Object selectors select: [:e| string, '*' match: e asString] ]; >> labelBlock: [ :e | e ]. >> >> thirdRequest := GLMSingleSpotterRequest new >> prompt: 'Other Selectors'; >> searchBlock: [ :string | Class selectors select: [:e| string, '*' match: e asString] ]; >> labelBlock: [ :e | e ]. >> >> composite >> add: classRequest; >> add: methodRequest; >> add: thirdRequest; >> signal >> >> >> Cheers, >> Doru >> >> >> >> -- >> www.tudorgirba.com >> >> "Reasonable is what we are accustomed with." >> >> > -- www.tudorgirba.com "Every thing should have the right to be different."

Feb. 28, 2011

Re: [Pharo-project] Fwd: [ANN] pharo focused sprint - bern, feb 26

by Esteban Lorenzano

Hi, Yes I'm willing to join you to another "glamour work day" :) Let's say next saturday? best, Esteban El 27/02/2011, a las 11:36a.m., Tudor Girba escribiÃ³: > Hi, > > On 27 Feb 2011, at 12:04, StÃ©phane Ducasse wrote: > >>>>> - Lukas went over the WeakAnnouncements implementation to get the WeakMessageSend working. The code is in PharoTaskForces >>>> >>>> should we take action on this one? >>>> I mean integrating it. >>> >>> It can be integrated, because the base code works in the same way as before. >>> >>> Currently, the solution is implemented through a WeakAnnouncer which subclasses Announcer. I wanted the weak thing to be merged, because the clients of announcements should benefit from the Weak implementation, but Lukas says that he does not trust the Weak support in Pharo and that if you have many WeakAnnouncers it somehow does not scale. >> >> ok so ...? > > So, we need to investigate more :). It would be cool if someone with more knowledge would take a look, too. Igor, would you have a bit of time to look into it? > > I will probably also give it a look to see how it works in the context of Glamour. I hope that Esteban will join me again. Esteban, what do you say :)? > > >>> The problem with the current approach is that models have to take an explicit stand of whether they want weak or non weak when instantiating the announcer. Unfortunately, I do have enough technical knowledge for this. It would be cool if someone with better knowledge in this area would take a look. >>> >>>>> - Lukas worked on getting OB work with the changes in TextMorph from Pharo 1.2. >>>>> >>>>> - Adrian, Jorge and Toon worked on implementing a Debugger on top of Glamour. They got a first version working that is able to do basic actions: step, restart, step into. The challenge was to get to understand the model, and to figure out that <primitive: 19> is used for marking that a method should not be debugged. In any case, working with Glamour seemed to work quite smoothly. The code is available in the Glamorous Toolkit. Adrian, did I get it right? :) >>>>> >>>>> - Mircea built a new widget for searching multiple categories of items in the same time, similar to spotlight where you enter one string and you get multiple hits from multiple categories. The implementation was spawned from the OBCompletionDialog / OBCompletionRequest and currently it simply provides multiple lists one below the other. The idea is to use this kind of widget to search simultaneously for classes / methods / packages. The code is available in Glamour (Glamour-Morphic-Renderer). The widget can definitely be improved visually, but Mircea already did a nice job. >>>> >>>> is the widget only for glamour because I think that a widget with search list integrated should be used everywhere? >>>> We were discussing with alain to have a look at the one in OB and push it/rewrite it as a core widget >>> >>> Neither the OB nor the Glamour widgets depend on Glamour, so they can be used independently. You just need to take the Request and the Dialog together. >> >> >> ok where is the code? > > > OBCompletionDialog / OBCompletionRequest come with the OB from Pharo 1.2. The way to use it is like: > > OBCompletionRequest new > prompt: 'Find Class'; > searchBlock: [ :string | OBCmdFindClass new findClassIn: Smalltalk pattern: string ]; > labelBlock: [ :class | class name ]; > iconBlock: [ :class | class browserIcon ]; > signal. > > The GLMSpotterRequest / GLMMorphicSpotterDialog are in Glamour-Morphic-Renderer. You can use them like > > | composite classRequest methodRequest thirdRequest | > composite := GLMSpotterRequest new. > > classRequest := GLMSingleSpotterRequest new > prompt: 'Find Class'; > searchBlock: [ :string | OBCmdFindClass new findClassIn: Smalltalk pattern: string ]; > labelBlock: [ :class | class name ]; > iconBlock: [ :class | class browserIcon ]. > > methodRequest := GLMSingleSpotterRequest new > prompt: 'Object Selectors'; > searchBlock: [ :string | Object selectors select: [:e| string, '*' match: e asString] ]; > labelBlock: [ :e | e ]. > > thirdRequest := GLMSingleSpotterRequest new > prompt: 'Other Selectors'; > searchBlock: [ :string | Class selectors select: [:e| string, '*' match: e asString] ]; > labelBlock: [ :e | e ]. > > composite > add: classRequest; > add: methodRequest; > add: thirdRequest; > signal > > > Cheers, > Doru > > > > -- > www.tudorgirba.com > > "Reasonable is what we are accustomed with." > >

Feb. 28, 2011

Re: [Pharo-project] Startup weekend success

by StÃ©phane Ducasse

>> > > Stephan and me are already working on some news entry... we will post it the next days. excellent!!! stef

Feb. 28, 2011

Re: [Pharo-project] Startup weekend success

by Marcus Denker

On Feb 28, 2011, at 9:13 PM, StÃ©phane Ducasse wrote: > Stephan > > do you have a web link so that we announce it on the pharo web site? > Stephan and me are already working on some news entry... we will post it the next days. > Stef > > On Feb 28, 2011, at 3:33 PM, Stephan Eggermont wrote: > >> In the weekend of 20 Feb. Diego Lont and I participated in Startup Weekend Eindhoven. As we expected, the rapid development made possible by the combination of Seaside on Pharo made it possible for us to develop a convincing demonstration in two days. The "Parent Support" app, an idea presented on friday night by Liesbeth Vriens, was converted by us into a combination iPhone app and web application in the two following days. This together with the business plan developed with Dolf WittkÃ¤mper were presented by Liesbeth on sunday evening in an award-winning 3-minute pitch. The jury awarded us the innovation price. >> >> Stephan Eggermont > > -- Marcus Denker -- http://www.marcusdenker.de INRIA Lille -- Nord Europe. Team RMoD.

Feb. 28, 2011

Re: [Pharo-project] Issue 3732 in pharo: default omni browser size bigger

by pharo＠googlecode.com

Comment #2 on issue 3732 by snoob...(a)yahoo.ie: default omni browser size bigger http://code.google.com/p/pharo/issues/detail?id=3732 I was thinking of making it relative rather than fixed. So in example below it would be 20% bigger depending on size of window.... initialExtent ^ RealEstateAgent standardWindowExtent * 1.2 min: RealEstateAgent maximumUsableArea extent.

Feb. 28, 2011

Re: [Pharo-project] Issue 3761 in pharo: overlappingPairsCollect: fails on OrderedCollections

by pharo＠googlecode.com

Comment #2 on issue 3761 by snoob...(a)yahoo.ie: overlappingPairsCollect: fails on OrderedCollections http://code.google.com/p/pharo/issues/detail?id=3761 Need to change #at:put: to rather be #add:. The reason for this is explained in comment of OrderedCollection>>at:put: "Put anObject at element index anInteger. at:put: cannot be used to append, front or back, to an ordered collection; it is used by a knowledgeable client to replace an element." Change could be as given below... overlappingPairsCollect: aBlock "Answer the result of evaluating aBlock with all of the overlapping pairs of my elements." | retval | retval := self species new: self size - 1. 1 to: self size - 1 do: [:i | retval add: (aBlock value: (self at: i) value: (self at: i + 1)) ]. ^retval

Feb. 28, 2011

Re: [Pharo-project] Startup weekend success

by StÃ©phane Ducasse

Stephan do you have a web link so that we announce it on the pharo web site? Stef On Feb 28, 2011, at 3:33 PM, Stephan Eggermont wrote: > In the weekend of 20 Feb. Diego Lont and I participated in Startup Weekend Eindhoven. As we expected, the rapid development made possible by the combination of Seaside on Pharo made it possible for us to develop a convincing demonstration in two days. The "Parent Support" app, an idea presented on friday night by Liesbeth Vriens, was converted by us into a combination iPhone app and web application in the two following days. This together with the business plan developed with Dolf WittkÃ¤mper were presented by Liesbeth on sunday evening in an award-winning 3-minute pitch. The jury awarded us the innovation price. > > Stephan Eggermont

Feb. 28, 2011

Re: [Pharo-project] [Hudson] News

by Tudor Girba

Great work! Doru On 28 Feb 2011, at 18:05, Mariano Martinez Peck wrote: > excellent!!! > > On Mon, Feb 28, 2011 at 2:33 PM, Marcus Denker <marcus.denker(a)inria.fr> wrote: > Hi, > > Slow but steady progress on the hudson infrastructure... > > 1) We have a Mac slave! > Alive and kicking, but not doing anything other then an hourly ping > https://pharo-ic.lille.inria.fr/hudson/job/Check%20Builder%20Mac/ > > 2) Cog Unix (both JIT and Stack VM) build. Fully automatic from repository to binary > using Pharo 1.3 as the host system. > https://pharo-ic.lille.inria.fr/hudson/job/Cog%20Unix%201.3/ > https://pharo-ic.lille.inria.fr/hudson/job/Stack%20VM%20Unix/ > > lots to do... mac build, windows build. Build debug version, build latest stable > *and* latest development, run tests... use the generated VM as the base VM > for all 1.3 builds, use generated VM for 1.3 One-Click... and so on. > > 3) There is a build project for Project Sista... Image side for now, work in progress. > https://pharo-ic.lille.inria.fr/hudson/job/Sista/ (broken right now) > (TODO: We need to build cog-sista) > > > > Marcus > > > -- > Marcus Denker -- http://www.marcusdenker.de > INRIA Lille -- Nord Europe. Team RMoD. > > > -- www.tudorgirba.com "Not knowing how to do something is not an argument for how it cannot be done."

Feb. 28, 2011

Re: [Pharo-project] Metacello Configuration Pharo1.2 solo sprint ... status

by Dale Henrichs

Miguel, Thanks for checking! Dale On 02/28/2011 04:58 AM, Miguel CobÃ¡ wrote: > Loads correctly on: > > Pharo1.2rc2 > Latest update: #12336 > > Test results: > > 10761 run, 10691 passes, 65 expected failures, 3 failures, 1 errors, 1 > unexpected passes > > OS: > > Debian GNU/Linux 6.0 64 bits. > > VM: > 4.0.3-2202 #1 XShm Tue Apr 13 11:56:47 PDT 2010 gcc 4.3.2 > Linux vps2.piumarta.com 2.6.18-028stab053.10-ent #1 SMP Thu Feb 28 > 20:34:08 MSK 2008 i686 GNU/Linux > plugin path: /opt/Squeak-4.0.3.2202-linux_i386/lib/squeak/4.0.3-2202/ > [default: /opt/Squeak-4.0.3.2202-linux_i386/lib/squeak/4.0.3-2202/] > > > Thanks Dale > > > El lun, 28-02-2011 a las 08:12 +0100, StÃ©phane Ducasse escribiÃ³: >> thanks! >> >> On Feb 28, 2011, at 6:51 AM, Dale Henrichs wrote: >> >>> I've just finished copying the cleaned up configurations up to MetacelloRepository. >>> >>> All of the configurations that are referenced in ConfigurationOfPharo validate clean ... >>> >>> I loaded the config into a PHaroCore image, but had a number of failed tests, but I am running on a mac and I seem to remember that there were still some failing tests on the mac ... >>> >>> I'll check on the hudson results first thing tomorrow morning to double check .. >>> >>> Dale >>> >> >> >

Feb. 28, 2011