December 2013 - Pharo-dev - lists.pharo.org

Re: [Pharo-dev] pharo 3.0 image

by HernÃ¡n Morales Durand

Try this: wget -O- get.pharo.org/30+vm | bash Pharo3.0 Latest update: #30664 2013/12/29 Johan Brichau <johan(a)inceptive.be> > What is the latest âworkingâ pharo 3 image? > > Every recent build I try to download opens with a blank screen and I > cannot do anything. > The latest build I downloaded some time ago was 30526 and that one opens > fineâ¦ > > Ideas? > > Johan >

Dec. 29, 2013

Tagging #catalogContactInfo?

by HernÃ¡n Morales Durand

Hi Stef, Is there any format for defining #catalogContactInfo contents? For example ^ '[[Author:Walter White]] [[e-mail: blabla(a)bla.com]] [[website: http://www.www.www]]' Cheers, HernÃ¡n

Dec. 29, 2013

pharo 3.0 image

by Johan Brichau

What is the latest âworkingâ pharo 3 image? Every recent build I try to download opens with a blank screen and I cannot do anything. The latest build I downloaded some time ago was 30526 and that one opens fineâ¦ Ideas? Johan

Dec. 29, 2013

Re: [Pharo-dev] A small patch for Zinc (related to XMLParser)

by Sven Van Caekenberghe

Hi Monty, On 29 Dec 2013, at 07:47, monty <monty.os(a)engineer.com> wrote: > Hi Sven. your ZnMimeType>>isBinary method is wrongly treating application/xml-dtd and other XML application subtypes like xml-external-parsed-entity as binary instead of text. To be consistent, if you're going to treat one applicaiton/xml type as text, you should treat them all that way. I attached a patch that fixes it by treating "-" as a subtype separator too. More info: http://www.ietf.org/rfc/rfc3023.txt Excellent extension, I integrated it in Zinc upstream, bleedingEdge === Name: Zinc-Resource-Meta-Core-SvenVanCaekenberghe.30 Author: SvenVanCaekenberghe Time: 29 December 2013, 12:05:01.614278 pm UUID: 18ce8118-2e25-43f1-8524-0d6b817be8cb Ancestors: Zinc-Resource-Meta-Core-SvenVanCaekenberghe.29 Extend ZnMimeType>>#isBinary to recognize application/xml-dtd (and more from http://www.ietf.org/rfc/rfc3023.txt) as non-binary by using both + and - as subtype separators when looking for xml (thanks monty); Extend ZnMimeTypeTests>>testIsBinary to cover the new functionality === Name: Zinc-Resource-Meta-Tests-SvenVanCaekenberghe.18 Author: SvenVanCaekenberghe Time: 29 December 2013, 12:05:30.83878 pm UUID: 4c2e4a78-c496-41b7-9f5c-cd90b982a9d6 Ancestors: Zinc-Resource-Meta-Tests-SvenVanCaekenberghe.17 Extend ZnMimeType>>#isBinary to recognize application/xml-dtd (and more from http://www.ietf.org/rfc/rfc3023.txt) as non-binary by using both + and - as subtype separators when looking for xml (thanks monty); Extend ZnMimeTypeTests>>testIsBinary to cover the new functionality === > Also, my lack of commit messages is because if I'm just making some small changes to private classes/methods or refactoring, I don't see the point in detailing it for users who don't (and shouldn't) care. They should just focus on the public API (which I haven't changed much). Hmm, I think that for both the community and for yourself in half a year it is important to write commit messages like the one I wrote above. It does not have to go into much detail, something like ârefactored the DOM parser for entitiesâ is enough - it will help hunting for regressions and other problems. The writeup of your changes that was in the email is already very helpful. BTW, thanks for working on XML parser, it is an important tool. > <ZnMimeType-isBinary.st> Thanks again for the patch ! Regards, Sven -- Sven Van Caekenberghe Proudly supporting Pharo http://pharo.org http://association.pharo.org http://consortium.pharo.org

Dec. 29, 2013

Re: [Pharo-dev] Error in loading libreries

by StÃ©phane Ducasse

you should pay attention that you added the right kind of repository. Because repository kinds (ss, ss3, smalltalkhub) check that they got the right format. Stef > Hi, > > every time I try to load a package from Monticello, I receive the message: > > ERROR: MCHttpRepository expected. > > Can anybody tell me why and suggest me a workeround? > > Many thanks and seasonâs greetings > > Lorenzo > >

Dec. 28, 2013

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

by HernÃ¡n Morales Durand

Hi Tudor. This is clearer now. Let me rewrite my concern so you can have a better understanding. I have been reviewing the Roassal code and resulted very clear to me how to get a visualization by scripting. But I am afraid that you would get better feedback from scientists using MATLAB, R and similar software every day. Most of them, I presume, didn't hear about Smalltalk or Moose. Having a more directed bridge to attract them would pay off. See for example what people is doing behind the University of Florida http://www2.research.att.com/~yifanhu/gallery/indexAll.html (first hit for "visualizing big matrices" in Google). They're using MATLAB, Rutherford/Boeing and MatrixMarket formats http://www.cise.ufl.edu/research/sparse/matrices/ Please don't get me wrong, I am not asking you or anyone to implement such layer. I am saying it would be nice to have more Smalltalkers to build these missing pieces. What can we do about it? BTW I've just found another inspirational picture for the relation tower: http://biopython.org/DIST/docs/tutorial/images/tRNA_chrom.png Cheers, HernÃ¡n 2013/12/27 Tudor Girba <tudor(a)tudorgirba.com> > 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. 28, 2013

Re: [Pharo-dev] [ANN] Ephestos

by btc＠openinworld.com

kilon alios wrote: > Still long way to go until its really useful but following your advice > I decided to release it to public and announce it instead of having it > hidden. > > Ladies , gentleman and hybrids I present to you Ephestos, my visual > coding project for graphics and sound. > > This is my very first project with Pharo and Smalltalk and my second > coding project. > > Just released a version 0.1 so as you can imagine you wont see > anything impressive yet but none the less it included my first > Ephestos tool, Hyperion a vector graphics editor. Right now it can > only edit a single vector straight line and add new lines. > > I am working now on version 0.2 that will allow you to edit other > lines , moving them around with mouse and add support for bezier curves. > > So its moving .... slowly .... very slowly... but moving none the less :D > > You can find Ephestos here along with installation instructions and > how to use Hyperion. Suffice to say you will need Pharo 3 to run it > > http://www.smalltalkhub.com/#!/~kilon/Ephestos > <http://www.smalltalkhub.com/#%21/%7Ekilon/Ephestos> > > I hope I will have Version 1 in January 2015 that will set out the > base functionality of Hyperion which will be a complete very basic > vector editor with features you will expect to find in a vector editor > like moving lines around, editing them, adding new control points to > them, making gradients and basic layer support. Will also introduce > Prometheas a new documentation system . > > I would like to thank all people who make Pharo so amazing and cool, > my problems with this project have more to do with my understanding on > how to code this thing properly than pharo and its libraries. So I > definitely did not expected it to be such a smooth ride. That looks like a nice start. I could move the initial line around by its handles. Some thoughts (which are probably on your radar anyway)... * Tooltips for the buttons would be nice * Add a superclass to the buttons to site between them and Morph, to hold common methods. Interesting to see your use of Athens, since I haven't used this yet. For future announcments you may want to include pharo-users mail list, since that is a good place to discuss user apps. cheers -ben

Dec. 28, 2013

Re: [Pharo-dev] Are #explorerContents and #hasContentsInExplorer obsolete ?

by Tudor Girba

Thanks for raising this :). I now added a simple implementation in the GTInspector to have Pointers available for all objects. Perhaps it is overdoing it to have it by default on all objects, but it was cheap to implement in a crude way. I still do not eliminate all inspector related objects that point to the current object, but it's a start (see attached the pointers to Smalltalk). Object>>gtInspectorPointersIn: composite <gtInspectorPresentationOrder: 20> composite list title: 'Pointers'; display: [ (self pointersToExcept: { self }) reject: [ :each | each pointsOnlyWeaklyTo: self ] ]; format: [ :each | GTObjectPrinter new asTruncatedTextFrom: each ]; showOnly: 50 [image: Inline image 1] Cheers, Doru On Wed, Dec 25, 2013 at 12:04 PM, Marcus Denker <marcus.denker(a)inria.fr>wrote: > > On 21 Dec 2013, at 15:03, Sven Van Caekenberghe <sven(a)stfx.eu> wrote: > > > Hi, > > > > It seems that #explorerContents and #hasContentsInExplorer are no longer > used. > > > > Implementing a new Eye inspector with #addSpecialFields: seems to affect > both the Inspector view as well as the Tree Inspector. > > > > So, should they be removed ? > > > > Yes, but we need to first find a solution for the Pointer Explorer: right > now this is a subclass of the old Explorere, therefore this is > still in the image. > > https://pharo.fogbugz.com/f/cases/11709/remove-old-ObjectExplorer > > Marcus > > > -- www.tudorgirba.com "Every thing has its own flow"

Dec. 28, 2013

Re: [Pharo-dev] [fun] Editing graphic vectorial content

by Tudor Girba

Excellent! It would be great to build on this. Cheers, Doru On Sat, Dec 28, 2013 at 12:51 PM, Alexandre Bergel <alexandre.bergel(a)me.com>wrote: > Just to share a 10 min coding session, just before lunch. > > Result here: > > https://www.facebook.com/photo.php?fbid=564129843673617&set=a.3411893793009… > > code: > -=-=-=-=-=-=-=-= > | view stack selected statusBar | > stack := ROViewStack new. > view := ROView new. > selected := nil. > statusBar := nil. > > view on: ROMouseClick do: [ :evt | > selected == #line ifTrue: [ > | ex1 ex2 line | > ex1 := (ROBox new size: 10; color: (Color purple alpha: > 0.2)) element. > ex2 := (ROBox new size: 10; color: (Color purple alpha: > 0.2)) element. > ex1 translateTo: evt position. > ex2 translateTo: evt position + (100 @ 60). > ex1 @ ROLightlyHighlightable. > ex2 @ ROLightlyHighlightable. > ex1 @ RODraggable. > ex2 @ RODraggable. > line := (ROLine new width: 3; color: (Color blue alpha: > 0.2)) elementFrom: ex1 to: ex2. > view addAll: { ex1 . ex2 . line }. > ]. > > selected == #circle ifTrue: [ > | ellipse | > ellipse := (ROEllipse new size: 40; color: (Color yellow > alpha: 0.2)) element. > ellipse translateTo: evt position. > ellipse @ RODraggable. > view add: ellipse. > ]. > > > selected == #rectangle ifTrue: [ > | box | > box := (ROBox new width: 40; height: 30; color: (Color > blue alpha: 0.2)) element. > box translateTo: evt position. > box @ RODraggable. > view add: box. > ]. > view signalUpdate. > ]. > > > > view @ RODraggable. > > stack zoomInButton; zoomOutButton. > stack addMenu: '+ line' callBack: [ :evt | selected := #line. statusBar > model: 'line' ]. > stack addMenu: '+ rectangle' callBack: [ :evt | selected := #rectangle. > statusBar model: 'rectangle' ]. > stack addMenu: '+ circle' callBack: [ :evt | selected := #circle. > statusBar model: 'circle' ]. > stack addMenu: 'Info' callBack: [ :evt | UIManager default inform: > 'Vectorial editing is hyper cool' ]. > > stack addView: view. > stack title: 'Vectorial edition'. > > "Status bar at the bottom of the window" > statusBar := (ROElement new) + ROLabel + ROBox white. > stack add: statusBar. > ROConstraint stickAtTheBottomOfTheWindow: statusBar. > > stack open > -=-=-=-=-=-=-=-= > > Cheers, > Alexandre > -- > _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: > Alexandre Bergel http://www.bergel.eu > ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. > > > > > -- www.tudorgirba.com "Every thing has its own flow"

Dec. 28, 2013

[fun] Editing graphic vectorial content

by Alexandre Bergel

Just to share a 10 min coding session, just before lunch. Result here: https://www.facebook.com/photo.php?fbid=564129843673617&set=a.3411893793009… code: -=-=-=-=-=-=-=-= | view stack selected statusBar | stack := ROViewStack new. view := ROView new. selected := nil. statusBar := nil. view on: ROMouseClick do: [ :evt | selected == #line ifTrue: [ | ex1 ex2 line | ex1 := (ROBox new size: 10; color: (Color purple alpha: 0.2)) element. ex2 := (ROBox new size: 10; color: (Color purple alpha: 0.2)) element. ex1 translateTo: evt position. ex2 translateTo: evt position + (100 @ 60). ex1 @ ROLightlyHighlightable. ex2 @ ROLightlyHighlightable. ex1 @ RODraggable. ex2 @ RODraggable. line := (ROLine new width: 3; color: (Color blue alpha: 0.2)) elementFrom: ex1 to: ex2. view addAll: { ex1 . ex2 . line }. ]. selected == #circle ifTrue: [ | ellipse | ellipse := (ROEllipse new size: 40; color: (Color yellow alpha: 0.2)) element. ellipse translateTo: evt position. ellipse @ RODraggable. view add: ellipse. ]. selected == #rectangle ifTrue: [ | box | box := (ROBox new width: 40; height: 30; color: (Color blue alpha: 0.2)) element. box translateTo: evt position. box @ RODraggable. view add: box. ]. view signalUpdate. ]. view @ RODraggable. stack zoomInButton; zoomOutButton. stack addMenu: '+ line' callBack: [ :evt | selected := #line. statusBar model: 'line' ]. stack addMenu: '+ rectangle' callBack: [ :evt | selected := #rectangle. statusBar model: 'rectangle' ]. stack addMenu: '+ circle' callBack: [ :evt | selected := #circle. statusBar model: 'circle' ]. stack addMenu: 'Info' callBack: [ :evt | UIManager default inform: 'Vectorial editing is hyper cool' ]. stack addView: view. stack title: 'Vectorial edition'. "Status bar at the bottom of the window" statusBar := (ROElement new) + ROLabel + ROBox white. stack add: statusBar. ROConstraint stickAtTheBottomOfTheWindow: statusBar. stack open -=-=-=-=-=-=-=-= Cheers, Alexandre -- _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: Alexandre Bergel http://www.bergel.eu ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.

Dec. 28, 2013