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

Re: [Pharo-project] Backward Compatibility
by Henrik Sperre Johansen Nov. 23, 2011

Nov. 23, 2011

On 23.11.2011 21:08, Henrik Sperre Johansen wrote: > On 23.11.2011 20:20, Sean P. DeNigris wrote: >> The Pharo Manifesto (http://code.google.com/p/pharo/) lists "Not >> backward >> compatible" as one of its 4 tenants. Yet, while working through the >> issue >> tracker, I've noticed some adding/keeping complexity to keep things >> backward >> compatible - most recently >> http://code.google.com/p/pharo/issues/detail?id=5007. >> >> I just wanted to start a discussion because obviously the more mature >> Pharo >> gets, the harder it will be to remove things... >> >> Sean >> >> -- >> View this message in context: >> http://forum.world.st/Backward-Compatibility-tp4101111p4101111.html >> Sent from the Pharo Smalltalk mailing list archive at Nabble.com. >> > I don't get it, I don't see how ctrl-s does anyhing in the original > case... > You click OK without accepting the contents first, and it still > returns the modified text. > Why not just remove the yellow "this editor has unaccepted contents" > marker for this type of dialog? > > Altering ctrl-s to do an accept of the input isn't really what I'd > expect from an input dialog... > > Cheers, > Henry > And igor thinks it should, so I guess opinions differ :) My point was I'd expect ok/cancel dialogs to act consistently, and afaik, the rest do not accept on a cmd-s, do they? Either way, I don't think it's a big enough difference to bother with a setting. Then again, it's a bikeshed, so I'll shut up from now on. Cheers, Henry

3 2

[Pharo-project] Backward Compatibility
by Sean P. DeNigris Nov. 23, 2011

Nov. 23, 2011

The Pharo Manifesto (http://code.google.com/p/pharo/) lists "Not backward compatible" as one of its 4 tenants. Yet, while working through the issue tracker, I've noticed some adding/keeping complexity to keep things backward compatible - most recently http://code.google.com/p/pharo/issues/detail?id=5007. I just wanted to start a discussion because obviously the more mature Pharo gets, the harder it will be to remove things... Sean -- View this message in context: http://forum.world.st/Backward-Compatibility-tp4101111p4101111.html Sent from the Pharo Smalltalk mailing list archive at Nabble.com.

4 3

[Pharo-project] [TODO 1.3] Open Issues in 1.3
by Marcus Denker Nov. 23, 2011

Nov. 23, 2011

Just 6. Three of them have suggested solutions that just need to be checked: http://code.google.com/p/pharo/issues/list?can=2&q=milestone=1.3 -- Marcus Denker -- http://marcusdenker.de

1 0

[Pharo-project] Call for feedback: assessing the test coverage with Hapao
by Alexandre Bergel Nov. 23, 2011

Nov. 23, 2011

Dear All, As you may know, Vanessa and I have been working on a visual tool to help people increase the test coverage. The website contains screenshots, documents and screencast: http://hapao.dcc.uchile.cl/ Hapao is easy to install: -=-=-=--=-=-=--=-=-=--=-=-=- Gofer new squeaksource: 'Spy'; package: 'ConfigurationOfSpy'; load. (Smalltalk at: #ConfigurationOfSpy) project lastVersion load -=-=-=--=-=-=--=-=-=--=-=-=- And then get it from the world menu. We crucially need feedback. This is necessary to target serious conferences. It does not take long to try. If you are developing right now, then you can let Hapao help you focus where you write new unit tests. You can leave your feedback on: https://docs.google.com/spreadsheet/viewform?hl=en_US&formkey=dDRKMzh4VzZBZ… We need feedback before the end of the month. Thanks to all! Cheers, Alexandre -- _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: Alexandre Bergel http://www.bergel.eu ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.

1 0

[Pharo-project] [TODO 1.4] Open Issues in 1.4
by Marcus Denker Nov. 23, 2011

Nov. 23, 2011

http://code.google.com/p/pharo/issues/list?can=2&q=milestone=1.4 -- Marcus Denker -- http://marcusdenker.de

1 0

[Pharo-project] OB Command to Send unary class-side methods
by Sean P. DeNigris Nov. 23, 2011

Nov. 23, 2011

I was adding a traditional "self myUnaryClassSideSelector" comment that was missing in Polymorph, and decided that would be the last time, so I added a command to OB: Gofer it squeaksource: 'SPDPlayground'; package: 'OBExtensions'; load. To use it: 1. Open OB (it seems only windows open after the file-in will have the new functionality) 2. Browse a class with an interesting unary class-side method (e.g. RecentMessageList, which has #openInWorld) 3. Select the method in the method pane 4. Bring up the menu and choose "Send Message" (or hit Cmd-s). Enjoy! Sean p.s. Lukas, this seems generally applicable. Will you add it to OB? -- View this message in context: http://forum.world.st/OB-Command-to-Send-unary-class-side-methods-tp4101082… Sent from the Pharo Smalltalk mailing list archive at Nabble.com.

1 0

[Pharo-project] [update 1.4] #14234
by Marcus Denker Nov. 23, 2011

Nov. 23, 2011

14234 ----- Issue 5009: Workspace method protocol clean-up http://code.google.com/p/pharo/issues/detail?id=5009 Issue 5007: On TextEditorDialogWindow http://code.google.com/p/pharo/issues/detail?id=5007 Issue 4974: better UUID comparison http://code.google.com/p/pharo/issues/detail?id=4974 Issue 5008: FixDebuggerWhenSelectorIs>> http://code.google.com/p/pharo/issues/detail?id=5008 -- Marcus Denker -- http://marcusdenker.de

1 0

[Pharo-project] Java team looking into Smalltalk and Seaside
by Sean P. DeNigris Nov. 23, 2011

Nov. 23, 2011

Maybe someone with experience using Smalltalk on a team can answer this: http://stackoverflow.com/questions/8244502/version-control-for-smalltalk-se… -- View this message in context: http://forum.world.st/Java-team-looking-into-Smalltalk-and-Seaside-tp410026… Sent from the Pharo Smalltalk mailing list archive at Nabble.com.

1 0

Re: [Pharo-project] Athens work in progress: couple screenshots
by Fernando Olivero Nov. 23, 2011

Nov. 23, 2011

Nice work Igor! I believe that Paragraph has too much responsibility right now, for example it takes care of drawing itself, besides the composition of lines to a fit a certain extent. For example, (VIEW) Paragraph>>showCaret: (MODEL) Paragraph>>compose: t style: ts from: startingIndex in: textContainer Looks like some of the behavior should be moved to the TextMorph. Fernando On Tue, Nov 22, 2011 at 3:25 PM, Igor Stasenko <siguctua(a)gmail.com> wrote: > On 22 November 2011 13:55, Tudor Girba <tudor(a)tudorgirba.com> wrote: >> Is there a way to load these? > > Yes. > on SqS/Athens > i just uploaded last snapshot. > but be warned, there's a lot of trash and cruft, which i will clean up > eventually :) > >> >> Doru >> >> On Tue, Nov 22, 2011 at 1:40 PM, Alexandre Bergel >> <alexandre.bergel(a)me.com> wrote: >>>> Stephane, asked me to show this :) >>> >>> Gorgeous! >>> Go go go Igor! >>> >>> Alexandre >>> >>> >>>> >>>> Nothing fancy there, just a same window, rendered with different >>>> scale. The code which doing this, is following: >>>> >>>> testWindowRender >>>> Â Â Â | surf | >>>> Â Â Â surf := AthensBalloonSurface new form: (Form extent: Display extent depth: 32). >>>> >>>> Â Â Â surf drawDuring: [ :canvas | >>>> >>>> Â Â Â Â Â Â Â canvas pathTransform scaleBy: 0.5. Â " or 2 " >>>> Â Â Â Â Â Â Â canvas fullDrawMorph: self testWindow. >>>> >>>> Â Â Â ]. >>>> >>>> Â Â Â Display getCanvas translucentImage: surf form at: 0@0 >>>> >>>> >>>> There are problems of course: >>>> - i have to use different Paragraph class, >>>> and therefore i have to use different TextMorph subclass, because >>>> otherwise it will use hardcoded Paragraph class. >>>> >>>> The paragraph model is a bit messy. It doesn't clear to me, what role it has. >>>> I thinking that it is a cached state for Text, to draw it on screen. >>>> But apparently it is more than a cache , and there are couple things >>>> which simply don't fit with it. >>>> I am thinking about doing some overhaul with TextMorph/Paragraph >>>> later, to clean up unused stuff (believe me there's a lot of it), >>>> and to make clear separation of roles. >>>> >>>> In fact, this stuff could live by itself. You can render text without >>>> using paragraph(s) mess. >>>> But i want to integrate that with Morphic, so we can switch to Athens >>>> at some point. That's why i need to deal with this. :) >>>> >>>> -- >>>> Best regards, >>>> Igor Stasenko. >>>> <Screen shot 2011-11-22 at 11.43.32 AM.png><Screen shot 2011-11-22 at 11.44.07 AM.png> >>> >>> -- >>> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: >>> Alexandre Bergel Â http://www.bergel.eu >>> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. >>> >>> >>> >>> >>> >>> >>> >> >> >> >> -- >> www.tudorgirba.com >> "Every thing has its own flow" >> >> > > > > -- > Best regards, > Igor Stasenko. > >

4 7

Re: [Pharo-project] Athens work in progress: couple screenshots
by Fernando Olivero Nov. 23, 2011

Nov. 23, 2011

Igor, what do i need to properly run the example? AthensBalloonSurface is using NB ? or a there's a plugin? Can i use a Mac VM to use Athens? Fernando On Wed, Nov 23, 2011 at 11:45 AM, Igor Stasenko <siguctua(a)gmail.com> wrote: > On 23 November 2011 11:21, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> wrote: >> Igor >> >> prepare the following: >> Â Â Â Â - a class showing an example >> Â Â Â Â Â Â Â Â with the transformation stuff (because it is unclear) > > There is one: > AthensBalloonSurfaceExamples > > >> >> Â Â Â Â - a tester canvas so that people can to >> >> (CanvasTester testClass: CircleMorph) open. >> > err.. what is it? > > To test morphic display on Athens you can do following: > > | surf | > Â Â Â Â surf := AthensBalloonSurface new form: (Form extent: Display extent depth: 32). > > Â Â Â Â surf drawDuring: [ :canvas | > > Â Â Â Â Â Â Â Â canvas fullDrawMorph: yourMorph. > ]. > > Display getCanvas translucentImage: surf form at: 0@0 > > > > >> People could then extend the Athens package with the method in the morph. >> >>> It would be great to help adapting all morphs for Athens. >>> Morphs, that having own draw method(s), should also have own >>> #drawOnAthensCanvas: >>> >>> I didn't covered all of them yet, so a little help with it would be nice. >>> Also, it is good for checking the missing functionality in Athens API >>> as well as learning it. >>> >>> There's not much to learn yet :) >>> >>> basicly what you should know is that you have coordinate system: >>> >>> canvas pathTransform >>> Â (instance of AthensAffineTransform) >>> >>> which you can manipulate to achieve desirable results. >>> >>> And actual drawing: >>> >>> canvas setPaint: somePaintObject. ( color, gradient, or anything which >>> conforms to AthensPaint protocol ). >>> canvas drawShape: Â (rect or path or anything , which responds to: >>> 'paintFillsUsing: aPaint on: anAthensCanvas' ). >>> canvas draw: (anything , which responds to #drawOnAthensCanvas: ) >> >> think that we are dummies willing to help and do the two things I mention above. >> I'm then sure you will get contributions. >> >> Stef >> > > > > -- > Best regards, > Igor Stasenko. > >

2 1