- Pharo-dev - lists.pharo.org

Re: [Pharo-dev] Call for fixing the selection in the debugger

by Nicolai Hess

On problem with the current solution (from 12144) is that TextModels readSelectionHolder is not updated when setSelection: is called. So, if you read the selection from the PluggableTextMorph it has always the same value (from its initialization) and not the actual selection (if you use the mouse to set a new selection). Compare TextModel>>setSelection: in Pharo3 and Pharo2 ( keep in mind that Pharo3 uses a MorphicTextAdapter between the model (TextModel) and the view (PluggableTextMorph). But I don't like this version (12144 with correct readSelectionHolder updating). I find it pretty confusing how textselection is handled in TextModel, MorphicTextAdapter, PluggableTextMorph and TextEditor. And look at issue 12487 <https://pharo.fogbugz.com/default.asp?12487> and my last comment in 12389 <https://pharo.fogbugz.com/default.asp?12389>(this is about textselection initializating from building the TextMorph with spec, TextModels spec does not handle the current selection. And I couldn't find a way to tell SpecInterpreter how to use the actual selection (if it is of kind Interval)) Nicolai 2013/12/27 Andrei Chis <chisvasileandrei(a)gmail.com> > PluggableTextMorph maintains its own selectionInterval. > > When the PluggableTextMorph >>selectionInterval: it is called it sets the > selection in the model. > This will trigger MorphicTextAdapter>>setSelectionFromModel: that should > adapt the selection in the morph: > MorphicTextAdapter >>setSelectionFromModel: aSelection > self widget ifNotNil: [:w | > w selectionInterval ~= aSelection ifTrue: [ > w setSelection: aSelection ]]. > > However, the second test (w selectionInterval ~= aSelection) will always > return false. This is because > PluggableTextMorph>>selectionInterval has the following implementation > '^ textMorph editor selectionInterval' > > And in the editor the selection is already correct (even > before PluggableTextMorph >>selectionInterval: is called). > > Actually most of the methods from PluggableTextMorph that change the > selection also explicitly update the > selectionInterval instance variable. > > So the previous solution should work. > > @Doru this is related with Spec, so we should not have it in Glamour. > > > > On Fri, Dec 27, 2013 at 5:07 PM, Benjamin < > Benjamin.VanRyseghem.Pharo(a)gmail.com> wrote: > >> I think that when the model is notified, it will set it in the morph >> So it will happen twice >> >> Ben >> >> On 27 Dec 2013, at 15:52, Andrei Chis <chisvasileandrei(a)gmail.com> wrote: >> >> Wouldn't then explicitly setting the selection of the morph before >> setting it in the model be ok? >> >> PluggableTextMorph>>selectionInterval: sel >> selectionInterval := sel. >> setSelectionSelector >> ifNotNil: [ self model perform:setSelectionSelector with: sel ] >> >> >> On Fri, Dec 27, 2013 at 4:28 PM, Benjamin < >> Benjamin.VanRyseghem.Pharo(a)gmail.com> wrote: >> >>> ok :( >>> Then I run out of brilliant idea :P >>> >>> >>> Ben >>> >>> On 27 Dec 2013, at 15:25, Andrei Chis <chisvasileandrei(a)gmail.com> >>> wrote: >>> >>> Adding 'self changed' at the end of the current PluggableTextMorph>>selectionInterval: >>> doesn't solve the problem. >>> >>> >>> On Fri, Dec 27, 2013 at 4:15 PM, Benjamin < >>> Benjamin.VanRyseghem.Pharo(a)gmail.com> wrote: >>> >>>> Because when the focus changes, the morph is proposed to redraw itself >>>> Try with a >>>> >>>> self changed >>>> >>>> at the end >>>> >>>> Ben >>>> >>>> On 27 Dec 2013, at 14:44, Andrei Chis <chisvasileandrei(a)gmail.com> >>>> wrote: >>>> >>>> >>>> The change that seems to have caused this problem is in >>>> PluggableTextMorph>>selectionInterval: The selectionInterval instance >>>> variable of the morph is not set any more; only the model is updated, which >>>> doesn't seem to update the morph. >>>> >>>> PluggableTextMorph>>selectionInterval: sel >>>> setSelectionSelector >>>> ifNil: [ selectionInterval := sel ] >>>> ifNotNil: [ self model perform:setSelectionSelector with: sel ] >>>> >>>> This could be solved like bellow, though I'm not sure it is the best >>>> solution. >>>> If I put a halt in PluggableTextMorph >>inspectIt and hit proceed the >>>> the selection is inspected correctly, without this modification. >>>> >>>> PluggableTextMorph>>selectionInterval: sel >>>> selectionInterval := sel. >>>> setSelectionSelector >>>> ifNotNil: [ self model perform:setSelectionSelector with: sel ] >>>> >>>> >>>> >>>> >>>> >>>> On Fri, Dec 27, 2013 at 3:11 PM, Marcus Denker <marcus.denker(a)inria.fr>wrote: >>>> >>>>> >>>>> On 27 Dec 2013, at 13:01, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> >>>>> wrote: >>>>> >>>>> > Hi guys >>>>> > >>>>> > we should not let the system in such state. We cannot inspect print >>>>> expression in the debugger. >>>>> > Does anybody have an idea of the change that broke it? >>>>> > >>>>> Yes: >>>>> >>>>> https://pharo.fogbugz.com/f/cases/12144/TextModel-getSelection >>>>> >>>>> it was added in 655, so for reverting the easiest is to dl 654, do a >>>>> merge and look at all the changes >>>>> (there are just 4-5 that are not re-catorizations). >>>>> >>>>> This is on my TODO nextâ¦ >>>>> >>>>> Marcus >>>>> >>>>> >>>>> >>>> >>>> >>> >>> >> >> >

Dec. 27, 2013

Re: [Pharo-dev] Thank you for Pharo

by Alexis Parseghian

VB>> Thank you for Pharo â¦. I'd like to add my voice to this. I've been away from Pharo for several months, although I kept reading the lists. The amount of stuff happening since before 2.0 was finalized is just amazing. VB>> - 64 Bit VM and MultiCore Support SD> first we will really integrate all the work of eliot. This too is important to me. I've found multi-arch in debian sid (amd64) to be broken at best, which leaves me with just an old notebook (sid i386) to run Pharo on. Or full-blown i386 virtual machines, which is also far from ideal. Beyond that, considering its history I like to think of Smalltalk as inherently multi-platform, and expect (long-term) Pharo to follow in those steps and be able to run on any system, any arch. (Yes, I'm aware of the limited resources of the team :) ) On a personal level, beyond debian/amd64 I'd love to be able to run it on (debian, openbsd) SPARC64... I'm afraid hacking the VM is beyond me at this point, but I'd like to help.

Dec. 27, 2013

Compiling chapter of "Pharo for the Enterprise"

by Natalia Tymchuk

Hello. Merry Christmas. I write a chapter to "Pharo for the Enterprise". And I want to compile it, but when I executing ./compile.sh DSL/DSL.pier, like it is writen here: https://github.com/SquareBracketAssociates/PharoForTheEnterprise-english/bl… , I get No yet implemented. I just start this chapter and it isn't ready to be added to the PFTE.tex. Should I add it to PFTE.tex or there are another way? What can you recomend me? Thanks. Best regards, Natalia

Dec. 27, 2013

Re: [Pharo-dev] problem while parsing xml

by StÃ©phane Ducasse

> In any case, it is strange that the items in the CharacterSet cannot be printed by the EyeInspector. GTInspector shows them properly (see attached). > > It gets even worse. Try doing this in a fresh Pharo 3.0 image: > > CharacterSet allCharacters explore probably because the explorer is using at: something. but CharacterSet allCharacters inspect works with the problems I mentioned. > > and you will get a DNU that 'Instances of CharacterSet are not indexable'. > > Cheers, > Doru > > <MessageNotUnderstood: SmallInteger>>asciiValue.png> > > > On Fri, Dec 27, 2013 at 6:47 PM, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> wrote: > I read a bit about internal public and other dtd and it looks like this is ok > > Ahhh. > > The bug was showing a parser problemâ¦. so I could not get that it was related to the network :) > >> Because monty's code downloads the DTD for validation from >> >> http://www.musicxml.org/dtds/partwise.dtd >> >> I should check if is the "right thing to do" because it would increase parsing time significantly. At least it should be an option. >> >> Cheers, >> >> Hernan >> >> >> 2013/12/27 StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> >> XMLDOMParser parse: '<?xml version="1.0" encoding="UTF-8" standalone="no"?> >> <!DOCTYPE score-partwise PUBLIC >> "-//Recordare//DTD MusicXML 3.0 Partwise//EN" >> "http://www.musicxml.org/dtds/partwise.dtd"> >> <score-partwise version="3.0"> >> <part-list> >> <score-part id="P1"> >> <part-name>Music</part-name> >> </score-part> >> </part-list> >> </score-partwise> >> ' readStream. >> >> produces the error >> while >> >> XMLDOMParser parse: '<?xml version="1.0" encoding="UTF-8" standalone="no"?> >> <score-partwise version="3.0"> >> <part-list> >> <score-part id="P1"> >> <part-name>Music</part-name> >> </score-part> >> </part-list> >> </score-partwise> >> ' readStream. >> >> does not. >> >> Stef >> >> > > > > > -- > www.tudorgirba.com > > "Every thing has its own flow"

Dec. 27, 2013

Re: [Pharo-dev] Relation tower is coming for Roassal

by Tudor Girba

Hi, I think there is a little confusion here. I meant to say that Roassal allows you to map your whatever objects onto a visualization. How you do this mapping is completely up to you. If you have a million objects, you will have to do some preprocessing yourself. That is all. There really is no magic involved. I just like to use "transformation" as a term in order to point out that visualization is no different than any other object manipulation. Did I address your concern? Cheers, Doru On Thu, Dec 26, 2013 at 9:48 PM, HernÃ¡n Morales Durand < hernan.morales(a)gmail.com> wrote: > Hi Tudor, > Thanks for the clarification. > Can you comment or point a link for the mechanism to transform my data for > Roassal? Because if you want to visualize big data like population or > association studies input is sometimes millions of rows. Of course I am not > asking you to write my parsers :) > > Cheers, > > HernÃ¡n > > > > > 2013/12/26 Tudor Girba <tudor(a)tudorgirba.com> > >> Hi HernÃ¡n, >> >> The input should be independent from the visualization engine. Roassal >> offers you a way of transforming your objects into a visualization and it >> should not be responsible for the input formats. >> >> Cheers, >> Doru >> >> >> On Thu, Dec 26, 2013 at 9:28 PM, HernÃ¡n Morales Durand < >> hernan.morales(a)gmail.com> wrote: >> >>> Hi Alexandre and Ricardo, >>> >>> I am glad to see your advances in Roassal visualization! >>> I wonder which input formats do you plan to support? >>> >>> Your relation tower can be used to visualize the crossover percentages >>> between haplotype blocks. For example, here are some Haplotypes drawings >>> from Haploview (http://en.wikipedia.org/wiki/Haploview) which may >>> interest you: >>> >>> >>> http://www.jnsbm.org/viewimage.asp?img=JNatScBiolMed_2012_3_2_139_101887_f1… >>> from this paper >>> http://www.jnsbm.org/temp/JNatScBiolMed32139-5367486_145434.pdf >>> >>> http://www.mailund.dk/wp-content/uploads/2010/04/haplotypes.png >>> >>> http://www.mailund.dk/wp-content/uploads/2010/04/journal.pone_.0010207.g005… >>> >>> I can send you the exported haplotypes from a study if you are >>> interested. It contains 131 blocks which is somewhat big to attach to this >>> list. The only problem I see is that a parser is needed to parse the >>> Haplotype text output file, although the format should not be difficult to >>> parse. >>> >>> Cheers, >>> >>> HernÃ¡n >>> >>> >>> >>> >>> >>> >>> >>> 2013/12/17 Alexandre Bergel <alexandre.bergel(a)me.com> >>> >>>> Hi! >>>> >>>> A new builder for Roassal is comingâ¦ The goal of Relation tower is to >>>> easily visualize relations between group of elements. We had the >>>> inspiration from the following website http://peoplemov.in/#! >>>> >>>> Here are some screenshots: >>>> >>>> https://www.facebook.com/media/set/?set=a.559114290841839.1073741837.340543… >>>> >>>> If you like these screenshots, then you know where the like button is >>>> :-) >>>> >>>> We will have this visualization on the web, thanks to our smooth and >>>> nice Pharo2Amber migration process. >>>> >>>> Any idea on how to extend/apply Relation tower? Any set of data you >>>> want us to play with? Your data? >>>> Feedback are very welcome! >>>> >>>> Cheers, >>>> Ricardo & Alexandre >>>> >>>> -- >>>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: >>>> Alexandre Bergel http://www.bergel.eu >>>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. >>>> >>>> >>>> >>> >> >> >> -- >> www.tudorgirba.com >> >> "Every thing has its own flow" >> > > -- www.tudorgirba.com "Every thing has its own flow"

Dec. 27, 2013

Re: [Pharo-dev] problem while parsing xml

by Tudor Girba

In any case, it is strange that the items in the CharacterSet cannot be printed by the EyeInspector. GTInspector shows them properly (see attached). It gets even worse. Try doing this in a fresh Pharo 3.0 image: CharacterSet allCharacters explore and you will get a DNU that 'Instances of CharacterSet are not indexable'. Cheers, Doru [image: Inline image 1] On Fri, Dec 27, 2013 at 6:47 PM, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr > wrote: > I read a bit about internal public and other dtd and it looks like this is > ok > > Ahhh. > > The bug was showing a parser problemâ¦. so I could not get that it was > related to the network :) > > Because monty's code downloads the DTD for validation from > > http://www.musicxml.org/dtds/partwise.dtd > > I should check if is the "right thing to do" because it would increase > parsing time significantly. At least it should be an option. > > Cheers, > > Hernan > > > 2013/12/27 StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> > >> XMLDOMParser parse: '<?xml version="1.0" encoding="UTF-8" >> standalone="no"?> >> <!DOCTYPE score-partwise PUBLIC >> "-//Recordare//DTD MusicXML 3.0 Partwise//EN" >> "http://www.musicxml.org/dtds/partwise.dtd"> >> <score-partwise version="3.0"> >> <part-list> >> <score-part id="P1"> >> <part-name>Music</part-name> >> </score-part> >> </part-list> >> </score-partwise> >> ' readStream. >> >> produces the error >> while >> >> XMLDOMParser parse: '<?xml version="1.0" encoding="UTF-8" >> standalone="no"?> >> <score-partwise version="3.0"> >> <part-list> >> <score-part id="P1"> >> <part-name>Music</part-name> >> </score-part> >> </part-list> >> </score-partwise> >> ' readStream. >> >> does not. >> >> Stef >> >> > > -- www.tudorgirba.com "Every thing has its own flow"

Dec. 27, 2013

Re: [Pharo-dev] [ANN] JetStorm

by jannik.laval

Argh, I saw the project too late. It is interesting, I will se that. Cheers, Jannik On Dec 20, 2013, at 12:53 PM, Sebastian Sastre <sebastian(a)flowingconcept.com> wrote: > Fantastic work. > > Maybe you should try talking with these guys: > http://www.kickstarter.com/projects/1039674461/primo-teaching-programming-l… > > Sounds like there is space for an intermediate product to ease the learning curve there > > sebastian > > o/ > > > > > > On Dec 18, 2013, at 7:09 AM, jannik.laval <jannik.laval(a)gmail.com> wrote: > >> Hi all, >> Another announcement for Christmas. >> >> Do you know Lego MindStorms ? The last one is the Ev3 serie (http://www.lego.com/fr-fr/mindstorms/) >> One particularity of this version compared to the previous ones is the possibility to plug a Wifi key and connect via TCP. >> >> So, if you have this material (one Mindstorms Ev3, one compatible Wifi key), a wireshark to obtain the TCP port to connect to Ev3 (See the process here, only to understand why we need a wireshark to know the port: http://www.monobrick.dk/guides/how-to-establish-a-wifi-connection-with-the-…) >> >> Now, you can control your robot with Pharo ! >> Just load these lines: >> â >> Gofer it >> url: 'http://smalltalkhub.com/mc/JLaval/JetStorm/main' >> username: '' >> password: ''; >> package: 'ConfigurationOfJetStorm'; >> load. >> (Smalltalk at: #ConfigurationOfJetStorm) loadBleedingEdge. >> â >> >> I know also that for Christmas, it is not possible to learn a new API, we all have a lot of other things to do ! So, you can play with it into Phratch. >> For that, just load Phratch and the package EV3Phratch as following. >> >> â >> Gofer it >> url: 'http://smalltalkhub.com/mc/JLaval/Phratch/main' >> username: '' >> password: ''; >> package: 'ConfigurationOfPhratch'; >> load. >> (Smalltalk at: #ConfigurationOfPhratch) loadBleedingEdge. >> >> Gofer it >> url: 'http://smalltalkhub.com/mc/JLaval/JetStorm/main' >> username: '' >> password: ''; >> package: 'EV3Phratch'; >> load. >> â >> >> Have fun ! >> Jannik >> >> PS: for now, this package in prototype, it can change a lot and can have bugs. Do not hesitate to give feedbacks. >> PS2: I hope to write a doc for it ! >> <ev3.png> >

Dec. 27, 2013

Re: [Pharo-dev] Nautilus: tab key change focus

by Juraj Kubelka

Thanks for the advice. I have just tried to add my own key stroke handling and I face interesting issue. The first time I enter the package filter input field, everything works. When I press down-key, it change focus to package list. But since then whenever I enter again the filter, any key stroke change the focus back to package list. I do not see any glue right now. The code is like this: buildPackageSearch "It is an search dialog which filters package list" searchWidget := SearchMorph new model: self; updateSelector: #packageSearchUpdate:; ghostText: 'Type: Pkg1|^Pkg2|Pk.*Core$'; list: SharedValueHolder instance; useSelectionIndex: false; setIndexSelector: #packageSearchUpdate:; keystrokeSelector: #packageSearchKeyStroke:; yourself. ^ searchWidget packageSearchKeyStroke: anEvent (anEvent keyCharacter = Character arrowDown) ifTrue: [ self giveFocusToPackage. ^ true ]. ^ false Thanks. Jura El 26-12-2013, a las 19:35, GOUBIER Thierry <thierry.goubier(a)cea.fr> escribiÃ³: > Have you tried Ctrl+Tab in the filter? I remember that text fields would catch the tab for themselves, and probably shift+tab as well, but we could force that filter to switch by using an onKeyCombination: Character tab asKeyCombination do: [ "setFocusToPackageListMorph" ] on the filter. > > With this asKeyCombination: do: , we could also change the order for the Package list and the class list (or just make sure the filter is added to the main window before the package list). If you change the order by hand, make sure you also correct the reverse order as well (Shift+tab and Ctrl+Shift+tab). > > I can't have a look at the code right now :( > > Thierry > > > ________________________________________ > De : Pharo-dev [pharo-dev-bounces(a)lists.pharo.org] de la part de Juraj Kubelka [juraj.kubelka(a)gmail.com] > Date d'envoi : jeudi 26 dÃ©cembre 2013 20:42 > Ã : Pharo Development List > Objet : Re: [Pharo-dev] Nautilus: tab key change focus > > Hi Thierry, > > thank you. Actually I have started with an issue with package filter input. When it has a focus, there is no way to change a focus with key press. Like tab, or down arrow key, or anything. And I also find out the tab order is strange. When package list has focus, tab change focus to the filter. I think it should change to class list. > > For some reasons it is possible to escape the filter input by pressing shift+tab, but not by pressing tab. So I wanted to try to change the âtab" order. > > Thanks, > Jura > > El 26-12-2013, a las 16:34, GOUBIER Thierry <thierry.goubier(a)cea.fr> escribiÃ³: > >> Hi Jura, >> >> it's implemented at the Morphic level; it follow the chain of owners / submorphs and should correctly tab to all morphs in the window, except when going through text fields which will catch the tab for themselves (but Ctrl+tab should work). >> >> To change the order, you may have a look at the order in which the morphs are added to the window. >> >> Another thing which may be of interest to you, Spec has the necessary code to override the default Morphic tab navigation and force an application defined order. >> >> What is the issue you are facing ? >> >> Thierry >> ________________________________________ >> De : Pharo-dev [pharo-dev-bounces(a)lists.pharo.org] de la part de Juraj Kubelka [juraj.kubelka(a)gmail.com] >> Date d'envoi : jeudi 26 dÃ©cembre 2013 17:09 >> Ã : Pharo Development List >> Objet : [Pharo-dev] Nautilus: tab key change focus >> >> Hi, >> >> Can someone give me any glue how tab-key-order and handling works in Nautilus? I am trying to fix the tab order in Nautilus but I do not see how it is implemented. >> >> Thanks, >> Jura >> > > >

Dec. 27, 2013

Re: [Pharo-dev] problem while parsing xml

by StÃ©phane Ducasse

I read a bit about internal public and other dtd and it looks like this is ok Ahhh. The bug was showing a parser problemâ¦. so I could not get that it was related to the network :) > Because monty's code downloads the DTD for validation from > > http://www.musicxml.org/dtds/partwise.dtd > > I should check if is the "right thing to do" because it would increase parsing time significantly. At least it should be an option. > > Cheers, > > Hernan > > > 2013/12/27 StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> > XMLDOMParser parse: '<?xml version="1.0" encoding="UTF-8" standalone="no"?> > <!DOCTYPE score-partwise PUBLIC > "-//Recordare//DTD MusicXML 3.0 Partwise//EN" > "http://www.musicxml.org/dtds/partwise.dtd"> > <score-partwise version="3.0"> > <part-list> > <score-part id="P1"> > <part-name>Music</part-name> > </score-part> > </part-list> > </score-partwise> > ' readStream. > > produces the error > while > > XMLDOMParser parse: '<?xml version="1.0" encoding="UTF-8" standalone="no"?> > <score-partwise version="3.0"> > <part-list> > <score-part id="P1"> > <part-name>Music</part-name> > </score-part> > </part-list> > </score-partwise> > ' readStream. > > does not. > > Stef > >

Dec. 27, 2013

Re: [Pharo-dev] problem while parsing xml

by HernÃ¡n Morales Durand

Because monty's code downloads the DTD for validation from http://www.musicxml.org/dtds/partwise.dtd I should check if is the "right thing to do" because it would increase parsing time significantly. At least it should be an option. Cheers, Hernan 2013/12/27 StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> > XMLDOMParser parse: '<?xml version="1.0" encoding="UTF-8" standalone="no"?> > <!DOCTYPE score-partwise PUBLIC > "-//Recordare//DTD MusicXML 3.0 Partwise//EN" > "http://www.musicxml.org/dtds/partwise.dtd"> > <score-partwise version="3.0"> > <part-list> > <score-part id="P1"> > <part-name>Music</part-name> > </score-part> > </part-list> > </score-partwise> > ' readStream. > > produces the error > while > > XMLDOMParser parse: '<?xml version="1.0" encoding="UTF-8" standalone="no"?> > <score-partwise version="3.0"> > <part-list> > <score-part id="P1"> > <part-name>Music</part-name> > </score-part> > </part-list> > </score-partwise> > ' readStream. > > does not. > > Stef > >

Dec. 27, 2013