- Pharo-dev - lists.pharo.org

Re: [Pharo-dev] installing gtoolkit

by phil＠highoctane.be

On Tue, Mar 25, 2014 at 8:17 AM, Tudor Girba <tudor(a)tudorgirba.com> wrote: > Hi, > > If you want to install GToolkit (the project that contains GTInspector, > GTPlayground and GTDebugger) in your image, you can follow one of the > options as described here: > http://www.humane-assessment.com/blog/installing-gtoolkit > > If you give it a try, please let us know how it works for you. > Looks *very* nice. Takes a while to get used to. Now, the gtoolkit image I used didn't respected the Pharo3 keybindings that allow one to open a workspace (Alt-o Alt-w / Cmd-o Cmd-w). It would be nice to have them so that opening the Playground is done in a snap. Question: if one has opened a Miller column too far, how can we close that one? It was annoying to have to use to precisely use the scrollbar down there to get back to where I was. And if the column is not out of view, it keeps on updating and moves the thing to the right. Kind of ping-pong style. I'll load it in my current development project and see how it goes (that superwhite theme of Moose is just an eye killer for me). Phil > > Cheers, > Doru > > -- > www.tudorgirba.com > > "Every thing has its own flow" >

March 25, 2014

Re: [Pharo-dev] Using Spec based UI in Nautilus Plugins

by Benjamin

You create your own subclass of AbstractNautilusPlugin in where the display method will look like display ^ mySpec buildWithSpec This should do the trick Ben On 25 Mar 2014, at 08:41, phil(a)highoctane.be wrote: > I wanted to use a MethodToolbar as part of a Nautilus plugin. > > Ok, I can use the widget in with an openWithSpec but how can I put this into a widget that Nautilus will accept as a plugin? > > > | b | > b := MethodToolbar new. > b openWithSpec. > b method: <method from plugin...> > > > Phil >

March 25, 2014

Re: [Pharo-dev] Crazy Smalltalk code snippets

by phil＠highoctane.be

On Tue, Mar 25, 2014 at 8:36 AM, Norbert Hartl <norbert(a)hartl.name> wrote: > > > Am 25.03.2014 um 00:33 schrieb "phil(a)highoctane.be" <phil(a)highoctane.be>: > > On Tue, Mar 25, 2014 at 12:11 AM, Eliot Miranda <eliot.miranda(a)gmail.com>wrote: > >> >> >> >> On Mon, Mar 24, 2014 at 4:02 PM, phil(a)highoctane.be <phil(a)highoctane.be>wrote: >> >>> I am curious. >>> >>> Maybe on the first one, substitute something when a DNU is encountered. >>> Like logging undefined receivers. >>> >>> I wonder how the next one behaves on JITted methods. I fear that offset >>> errors may lead to weird errors. >>> >> >> It *should* "just work" :-) (provided the jump distance of -12 is >> correct). The VM traps assignments to variables of contexts, and converts >> them to vanilla contexts. Therefore the jump doesn't occur in JITTED code >> but back in normal interpreted bytecode. >> >> > How magical! > > >> On Mon, Mar 24, 2014 at 11:51 PM, Pavel Krivanek < >>> pavel.krivanek(a)gmail.com> wrote: >>> >>>> Who can find the most useful usage of this? >>>> >>>> thisContext instVarNamed: #receiver put: 42. >>>> self factorial >>>> >>>> GOTO statement in Pharo: >>>> >>>> FileStream stdout nextPutAll: 'Hello world'; lf. >>>> thisContext jump: -12. >>>> Let's collect the next ones :-) >>>> >>> >> this is my favourite, and would lock-up a strict blue book VM. (I once >> locked up Allen Wirfs-Brock's 4404 with this and he wasn't best pleased >> [forgive me Allen]) >> >> >> | a | >> a := Array with: #perform:withArguments: with: nil. >> a at: 2 put: a. >> a perform: a first withArguments: a >> >> ;-) >> >> > > The infinite is near. > > Not as blocking, but still... (Alt-. interrupts this one) > > Crazy>>run > Continuation currentDo: [ :cc | here := cc ]. > here value: true. > > Crazy new run > > BTW, is there any support for partial continuations in Pharo? > Continuations like this one look like full continuations and that's *huge*. > > If you mean delimited continuations have a look here: > > > http://www.lshift.net/blog/2011/04/20/direct-implementation-of-shiftreset-i… > > Thanks. I saw a mention of this but not that article. Now, I loaded the code in Pharo3 and there are missing methods in BlockContext... ControlTests>>testNestedShiftResetNest (I guess other things would say the same) is broken testNestedShiftResetsNest | innerCont outerCont | [1 + [ :k | outerCont := k. "stack reified as [ :x | 1 + x ]" k value: [ 2 + [ :kk | innerCont := kk. "stack reified as [ :x | 2 + x ]" kk value: 0 ] shift ] reset ] shift ] reset. self assert: 3 equals: (innerCont value: 1). self assert: 2 equals: (outerCont value: 1). self assert: 5 equals: ((outerCont compose: [:x | x * 2]) value: 2) due to MethodContext's DNU: copyTo:bottomContextDo: [image: Inline image 1] There is a ContextPart: aContext blocks: dict but but looking like the same thing. Composition tests do work. Gurus welcome... If we put continuations in the Pharo image, why not put this as well? Phil > Norbert > > >>>> Cheers, >>>> -- Pavel >>>> >>> >>> >> >> >> -- >> best, >> Eliot >> > >

March 25, 2014

Re: [Pharo-dev] Trait1 Trait2 Trait3 in Smalltalk globals allClassesAndTraits

by Camille Teruel

On 24 mars 2014, at 23:30, phil(a)highoctane.be wrote: > On a fresh 3.0, there are Trait1, Trait2 and Trait3 in Smalltalk globals allClassesAndTraits > > These 3 look like they are equivalents to Trait. What are these for? That's just some test fixtures. > > Phil > >

March 25, 2014

Re: [Pharo-dev] Crazy Smalltalk code snippets

by Norbert Hartl

Am 24.03.2014 um 23:51 schrieb Pavel Krivanek <pavel.krivanek(a)gmail.com>: > Who can find the most useful usage of this? > > thisContext instVarNamed: #receiver put: 42. > self factorial > > GOTO statement in Pharo: > > FileStream stdout nextPutAll: 'Hello world'; lf. > thisContext jump: -12. > > Let's collect the next ones :-) > | injectedContext returnValue | injectedContext := [ returnValue := self dummySendToSkip. block value. returnValue ] asContext. injectedContext pc: injectedContext startpc + 2. injectedContext swapSender: (self targetContext swapSender: injectedContext) I used this once. It is hackish but I still like it because I donât know how to solver that better :) Norbert

March 25, 2014

Using Spec based UI in Nautilus Plugins

by phil＠highoctane.be

I wanted to use a MethodToolbar as part of a Nautilus plugin. Ok, I can use the widget in with an openWithSpec but how can I put this into a widget that Nautilus will accept as a plugin? | b | b := MethodToolbar new. b openWithSpec. b method: <method from plugin...> Phil

March 25, 2014

Re: [Pharo-dev] Crazy Smalltalk code snippets

by Norbert Hartl

> Am 25.03.2014 um 00:33 schrieb "phil(a)highoctane.be" <phil(a)highoctane.be>: > >> On Tue, Mar 25, 2014 at 12:11 AM, Eliot Miranda <eliot.miranda(a)gmail.com> wrote: > >> >> >> >>> On Mon, Mar 24, 2014 at 4:02 PM, phil(a)highoctane.be <phil(a)highoctane.be> wrote: >>> I am curious. >>> >>> Maybe on the first one, substitute something when a DNU is encountered. >>> Like logging undefined receivers. >>> >>> I wonder how the next one behaves on JITted methods. I fear that offset errors may lead to weird errors. >> >> It *should* "just work" :-) (provided the jump distance of -12 is correct). The VM traps assignments to variables of contexts, and converts them to vanilla contexts. Therefore the jump doesn't occur in JITTED code but back in normal interpreted bytecode. > > How magical! > >>>> On Mon, Mar 24, 2014 at 11:51 PM, Pavel Krivanek <pavel.krivanek(a)gmail.com> wrote: >>>> Who can find the most useful usage of this? >>>> >>>> thisContext instVarNamed: #receiver put: 42. >>>> self factorial >>>> >>>> GOTO statement in Pharo: >>>> >>>> FileStream stdout nextPutAll: 'Hello world'; lf. >>>> thisContext jump: -12. >>>> >>>> Let's collect the next ones :-) >> >> this is my favourite, and would lock-up a strict blue book VM. (I once locked up Allen Wirfs-Brock's 4404 with this and he wasn't best pleased [forgive me Allen]) >> >> >> | a | >> a := Array with: #perform:withArguments: with: nil. >> a at: 2 put: a. >> a perform: a first withArguments: a >> >> ;-) > > The infinite is near. > > Not as blocking, but still... (Alt-. interrupts this one) > > Crazy>>run > Continuation currentDo: [ :cc | here := cc ]. > here value: true. > > Crazy new run > > BTW, is there any support for partial continuations in Pharo? Continuations like this one look like full continuations and that's *huge*. > If you mean delimited continuations have a look here: http://www.lshift.net/blog/2011/04/20/direct-implementation-of-shiftreset-i… Norbert > >>>> >>>> Cheers, >>>> -- Pavel >> >> >> >> -- >> best, >> Eliot >

March 25, 2014

installing gtoolkit

by Tudor Girba

Hi, If you want to install GToolkit (the project that contains GTInspector, GTPlayground and GTDebugger) in your image, you can follow one of the options as described here: http://www.humane-assessment.com/blog/installing-gtoolkit If you give it a try, please let us know how it works for you. Cheers, Doru -- www.tudorgirba.com "Every thing has its own flow"

March 25, 2014

TWM and shortcuts

by phil＠highoctane.be

In Pharo3, there seems to be new shortcuts going on. These do interfere with TWM ones (or the other way around). What to do to get this to work again? TWMKeyMapping>>buildKeyMappingOn: aBuilder <keymap> "See shortucts accessor to add/change a shortcut " self shortcutsActionAndKeysDo: [:action :keys| (aBuilder shortcut: action) category: self shortcutCategory default: keys do: [ self windowManager perform: action ]. ]. aBuilder attachShortcutCategory: self shortcutCategory to: PasteUpMorph.. TWMKeyMapping class>>shortcuts ^ { #minimizeAllWindows -> 'Minimize all windows' -> [$r command, $m command]. #maximizeAllWindows -> 'Maximize all windows' -> [$r command, $m shift command]. #tileVisibleWindows -> 'Tile visible windows' -> [$r command, $t command]. #tileLastUsedWindows -> 'Tile last used windows' -> [$r command, $l command]. #tileBrowsers -> 'Tile browsers' -> [$r command, $b command]. #tileWorkspaces -> 'Tile workspaces' -> [$r command, $z command]. #tileTesters -> 'Tile testers' -> [$r command, $u command]. #tileMonticellos -> 'Tile monticellos' -> [$r command, $c command]. #tileInspectors -> 'Tile inspectors' -> [$r command, $i command]. #tileDebuggers -> 'Tile debuggers' -> [$r command, $d command]. #tileChanges -> 'Tile changes' -> [$r command, $c shift command]. #tileTools -> 'Tile tools' -> [$r command, $t shift command]. #tileSystem -> 'Tile system windows' -> [$r command, $y command]. #snapshotWorld -> 'Snapshot current world state' -> [$r command, $s command]. #worldBackward -> 'Move to previous world state' -> [$r command, $p command]. #worldForward -> 'Move to next world state' -> [$r command, $n command]. #deleteCurrentSnapshot -> 'Delete current world state' -> [$r command, $d shift command]. #topWindowMoveLeft -> 'Move window to left area' -> [$r command, $1]. #topWindowMoveRight -> 'Move window to right area' -> [$r command, $2]. #topWindowMoveTop -> 'Move window to top area' -> [$r command, $3]. #topWindowMoveBottom -> 'Move window to bottom area' -> [$r command, $4]. #topWindowMoveTopLeft -> 'Move window to top left area' -> [$r command, $5]. #topWindowMoveTopRight -> 'Move window to top right area' -> [$r command, $6]. #topWindowMoveBottomLeft -> 'Move window to bottom left area' -> [$r command, $7]. #topWindowMoveBottomRight -> 'Move window to bottom right area' -> [$r command, $8]. #cycleClockwise -> 'Last visible window on stack come to top' -> [$r command, $>]. #cycleCounterClockwise -> 'First visible window on stack go to bottom' -> [$r command, $<]. #switchToNextWorld -> 'Switch to next world' -> [$r command, $r command] } Phil

March 25, 2014

Dev Art from Google

by S Krish

Moose art visualizations if not more from Phratch et als can make a mark here. https://devart.withgoogle.com/?utm_source=email&utm_medium=gsp&utm_campaign…

March 25, 2014