- Pharo-dev - lists.pharo.org

Re: [Pharo-project] Fwd: [ANN] 10254

by Adrian Lienhard

Probably this was an override by Polymorph, and removing an overridden method in one package, as far as I know, reverts it to the method version provided by the other package. Overrides should be avoided, they just cause troubles. Adrian On Mar 19, 2009, at 13:46 , Henrik Johansen wrote: > I know, which is why I thought it was strange, the .cs contained: > !BorderedMorph removeSelector: #areasRemainingToFill:! > which when I file it in, results in no areasRemainingToFill: method > for > BorderedMorph in the image. > > I really have no idea how the update process makes it revert to an > older > version instead, nor how one would go about fixing it... :/ > > Cheers, > Henrik > > StÃ©phane Ducasse wrote: >> >> >> henrik I just integrated the file you gave me. >> If you have a new version that fixes the problem I will do a new >> version. >> >> Stef >> >> On Mar 19, 2009, at 12:20 PM, Henrik Johansen wrote: >> >>> It seems BorderedMorph's areasRemainingToFill: method got reverted >>> to >>> the version in use prior to the override by PolyMorph (which has now >>> been promoted to Morph), instead of removed completely. >>> (This is the state my image is in after updating a released 248, at >>> least) >>> >>> Cheers, >>> Henry >>> >>> StÃ©phane Ducasse wrote: >>>> Issue 655: Overlapping Graphics on World do not Properly Repaint >>>> Issue 658: Preferences resetHaloSpecifications >>>> Issue 664: unimplemented errorCantGoBack >>>> Closed 665 >>>> >>>> now we will focus on the closures. >>>> >>>> Stef >>>> >>>> >>>> >>>> >>>> >>>> _______________________________________________ >>>> Pharo-project mailing list >>>> Pharo-project(a)lists.gforge.inria.fr >>>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >>>> >>>> >>>> >>> >>> _______________________________________________ >>> Pharo-project mailing list >>> Pharo-project(a)lists.gforge.inria.fr >>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >>> >> >> ------------------------------------------------------------------------ >> >> _______________________________________________ >> Pharo-project mailing list >> Pharo-project(a)lists.gforge.inria.fr >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

March 19, 2009

[Pharo-project] Composing UI layout question

by astares＠gmx.de

Hi Igor, I like the fraction/offset model from SmalltalkMT. I once ported it to Squeak. This gives you full control on absolute and relative positioning within the bounds of the parent and controls resizing behavior. Only 4 parameters for anything you need ... Example: self addMorph: startButton to: pane offset: #(0 0 101 20) fractions: #(0 0 0 0). means the button starts top left at x1=0, y1=0 with the bottom right corner at x2=101 and y2=20. There is no fraction given - so we have absolute positioning. If you give a fraction you tell the system that for instance a subpane should always start at 30% from the side of a parents box. -> relative positions with resizing. If you additionally give an offset you can define that there is an additional absolute margin after the 30%. With a negative offset you can absolute position the control/submorph on the other side of the 30% virtual line. With fractions: #(0 0 1 1 ) which means 100% the control will fill the parent and will fully resize with it. Smalltalk MT used 1000 for 100% of the parents area, but thats only a definition. Have a look at the project "Sandcastle" on Squeaksource, especially Sandcastle-tbn.9.mcz. SCApplicationWindow (a subclass of SystemWindow here) provides a few methods to support this dynamic layout and its subclass SCWelcomeWindow shows how you can use it. Hope this helps Bye T. -- Aufgepasst: Sind Ihre Daten beim Online-Banking auch optimal geschÃ¼tzt? Jetzt absichern: https://homebanking.gmx.net/?mc=mail@footer.hb

March 19, 2009

[Pharo-project] [update] #10255

by Marcus Denker

Issue 662: more general beep Issue 659: Convenience Methods on Pragma and Additional Tests Issue 666: Removes ScriptLoader>>buildPharoDev Issue 665: Adds missing Trait>>withAllSubclasses -- Marcus Denker -- denker(a)acm.org http://2denker.de

March 19, 2009

Re: [Pharo-project] Fwd: [ANN] 10254

by Henrik Johansen

I know, which is why I thought it was strange, the .cs contained: !BorderedMorph removeSelector: #areasRemainingToFill:! which when I file it in, results in no areasRemainingToFill: method for BorderedMorph in the image. I really have no idea how the update process makes it revert to an older version instead, nor how one would go about fixing it... :/ Cheers, Henrik StÃ©phane Ducasse wrote: > > > henrik I just integrated the file you gave me. > If you have a new version that fixes the problem I will do a new version. > > Stef > > On Mar 19, 2009, at 12:20 PM, Henrik Johansen wrote: > >> It seems BorderedMorph's areasRemainingToFill: method got reverted to >> the version in use prior to the override by PolyMorph (which has now >> been promoted to Morph), instead of removed completely. >> (This is the state my image is in after updating a released 248, at >> least) >> >> Cheers, >> Henry >> >> StÃ©phane Ducasse wrote: >>> Issue 655: Overlapping Graphics on World do not Properly Repaint >>> Issue 658: Preferences resetHaloSpecifications >>> Issue 664: unimplemented errorCantGoBack >>> Closed 665 >>> >>> now we will focus on the closures. >>> >>> Stef >>> >>> >>> >>> >>> >>> _______________________________________________ >>> Pharo-project mailing list >>> Pharo-project(a)lists.gforge.inria.fr >>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >>> >>> >>> >> >> _______________________________________________ >> Pharo-project mailing list >> Pharo-project(a)lists.gforge.inria.fr >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >> > > ------------------------------------------------------------------------ > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

March 19, 2009

Re: [Pharo-project] Fwd: [ANN] 10254

by StÃ©phane Ducasse

henrik I just integrated the file you gave me. If you have a new version that fixes the problem I will do a new version. Stef On Mar 19, 2009, at 12:20 PM, Henrik Johansen wrote: > It seems BorderedMorph's areasRemainingToFill: method got reverted to > the version in use prior to the override by PolyMorph (which has now > been promoted to Morph), instead of removed completely. > (This is the state my image is in after updating a released 248, at > least) > > Cheers, > Henry > > StÃ©phane Ducasse wrote: >> Issue 655: Overlapping Graphics on World do not Properly Repaint >> Issue 658: Preferences resetHaloSpecifications >> Issue 664: unimplemented errorCantGoBack >> Closed 665 >> >> now we will focus on the closures. >> >> Stef >> >> >> >> >> >> _______________________________________________ >> Pharo-project mailing list >> Pharo-project(a)lists.gforge.inria.fr >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >> >> >> > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >

March 19, 2009

Re: [Pharo-project] Fwd: [ANN] 10254

by Henrik Johansen

It seems BorderedMorph's areasRemainingToFill: method got reverted to the version in use prior to the override by PolyMorph (which has now been promoted to Morph), instead of removed completely. (This is the state my image is in after updating a released 248, at least) Cheers, Henry StÃ©phane Ducasse wrote: > Issue 655: Overlapping Graphics on World do not Properly Repaint > Issue 658: Preferences resetHaloSpecifications > Issue 664: unimplemented errorCantGoBack > Closed 665 > > now we will focus on the closures. > > Stef > > > > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > >

March 19, 2009

[Pharo-project] http://goran.krampe.se/blog/Squeak/Tirade.rdoc

by StÃ©phane Ducasse

http://goran.krampe.se/blog/Squeak/Tirade.rdoc In my revived work on Deltastreams in Squeak I ended up facing the choice of native file format for Deltas. Matthew has made an advanced format calledInterleavedChangeset which manages to squeeze a binary representation of a Delta into a Changeset file (which is in Smalltalk chunk format). An impressive feat, and it has the advantage of being backwards compatible in the sense that a Delta in this format can be filed in as a plain old Changeset into an old Squeak image. .....

March 19, 2009

[Pharo-project] Composing UI layout question

by Igor Stasenko

Hi folks, suppose you having a visual UI editing tool, where you can place morphs/controls and define their position(s). I have a model in mind, where coordinates can be defined by formal description, i.e. morph left: morph parent left + some offset morph width: morph parent width / 2 etc. but i wonder, in what way such relations can be represented by UI editor/design tool. I want that user could easily specify following: suppose i placing a button in form, and say that its corner (bottom/right) should be aligned to its parent corner, with some negative offset (-5 @ -5). And button width/height should be calculated by taking a button label text width & height. So, user by specifying two ordinates - corner and width & height, is fully describing a button visual layout. The problem is that button's label is submorph, and i not sure how to formally encode such relation i.e. parent width & height (or any other parent coordinate) depends on its submorph's fixed width/height. -- Best regards, Igor Stasenko AKA sig.

March 19, 2009

Re: [Pharo-project] Drawing graphics with OB

by David RÃ¶thlisberger

Hi Alex, > Each OBNode understand the message definition. In the first version of > OB, this definition method is supposed to return an instance of > OBDefinition or one of its subclass. > I saw a way to have a graphical object instead of the text pane. What > is the trick to achieve this? What I did is to use OBClosableDefinitionPanel that has an #exchangeContentWith: aMorph method where you can pass any morph that gets displayed instead of the text that is usually there. We could extend that to be able to also use "visualization definitions" directly in OBNode, but I haven't yet done this. The visualizations I added get displayed on the execution of a command, by default there is always a text definition shown that can get exchanged with a visualization at runtime. David

March 19, 2009

[Pharo-project] Fwd: [ANN] 10254

by StÃ©phane Ducasse

Issue 655: Overlapping Graphics on World do not Properly Repaint Issue 658: Preferences resetHaloSpecifications Issue 664: unimplemented errorCantGoBack Closed 665 now we will focus on the closures. Stef

March 19, 2009