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

Re: [Pharo-project] what do we do with SoundService?

by stephane ducasse

Sounds ok for me. We should do that for 3.0alpha :) Now I do not care about Beeper removal. I will split sound in different packages because there are sound and score and these are different domains. Stef On Mar 10, 2013, at 11:38 AM, Pavel Krivanek <pavel.krivanek(a)gmail.com> wrote: > Hi, > > removal of this class is proposed in > https://code.google.com/p/pharo/issues/detail?id=6996 > > -- Pavel > > > On Sun, Mar 10, 2013 at 11:31 AM, StÃ©phane Ducasse > <stephane.ducasse(a)inria.fr> wrote: >> Hi guy >> >> Sound has an AbstractSoundSystem and a Base and Dummy. Then in addition there is SounService. >> >> SoundService is not doing much >> >> soundEnabled >> ^ self defaultSoundEnabled >> >> defaultSoundEnabled >> ^DefaultSoundEnabled ifNil: [false] >> >> toggleSoundEnabling >> self soundEnabled: self soundEnabled not >> >> >> >> So we could just keep AbstractSoundSystem and deprecate SoundService. >> >> Stef >

March 10, 2013

Re: [Pharo-project] [Lsehub-staff] Quite a saturday night!

by Herby VojÄÃk

Marcus Denker wrote: > On Mar 10, 2013, at 11:13 AM, Herby VojÄÃk<herby(a)mailbox.sk> wrote: > >> >> Marcus Denker wrote: >>> Could Helios be used to edit Pharo code? >> Ha! If taken broader, it can be GSoC idea (and can help to more or less standardize server and client protocols between running Smalltalk and Smalltalk IDE; this can bring, in the future, for example, a Smalltalk IDE by JetBrains (pipedreaming again))! > > A) Compatibility across many dialects will result in something that can evolve maximally as fast > as the slowest evolving Smalltalk dialect. Remote Amber and plain Pharo are enough or the server side. Others can use the protocol if they will. > B) The IDE written in Smalltak is an asset, outsourcing that to Java is the wrong way to go. You asked if Helios could edit Pharo code. Helios is, more or less, JavaScript app, so not any more a truly pure Smalltalk IDE. Maybe that protocol would _de_facto_ need a Smalltalk at the other side, but why taking it categorically aside? And of course, Smalltalk IDE written in Smalltalk itself _is_ an asset (not that I used anything from its malleableness yet, maximum is Method Finder); but I get used to good IDEs from other vendors as well; I in fact don't care what WebStorm is written in (yes, I know it is Java, as you mentioned), I like how it helps me). > Marcus Herby

March 10, 2013

Re: [Pharo-project] what do we do with SoundService?

by Pavel Krivanek

Hi, removal of this class is proposed in https://code.google.com/p/pharo/issues/detail?id=6996 -- Pavel On Sun, Mar 10, 2013 at 11:31 AM, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> wrote: > Hi guy > > Sound has an AbstractSoundSystem and a Base and Dummy. Then in addition there is SounService. > > SoundService is not doing much > > soundEnabled > ^ self defaultSoundEnabled > > defaultSoundEnabled > ^DefaultSoundEnabled ifNil: [false] > > toggleSoundEnabling > self soundEnabled: self soundEnabled not > > > > So we could just keep AbstractSoundSystem and deprecate SoundService. > > Stef

March 10, 2013

Re: [Pharo-project] [Lsehub-staff] Quite a saturday night!

by Marcus Denker

On Mar 10, 2013, at 11:13 AM, Herby VojÄÃk <herby(a)mailbox.sk> wrote: > > > Marcus Denker wrote: >> Could Helios be used to edit Pharo code? > > Ha! If taken broader, it can be GSoC idea (and can help to more or less standardize server and client protocols between running Smalltalk and Smalltalk IDE; this can bring, in the future, for example, a Smalltalk IDE by JetBrains (pipedreaming again))! A) Compatibility across many dialects will result in something that can evolve maximally as fast as the slowest evolving Smalltalk dialect. B) The IDE written in Smalltak is an asset, outsourcing that to Java is the wrong way to go. Marcus

March 10, 2013

Re: [Pharo-project] FMSound bass1 play

by stephane ducasse

In fact I get really unpredictable behavior. Sometimes I can produce a sound sometimes not. SoundService default soundEnabled: true I'm puzzled Stef On Mar 9, 2013, at 10:21 PM, stephane ducasse <stephane.ducasse(a)free.fr> wrote: > Hi guys > > I'm migrating the package Sound and MorphicSound to Pharo2.0 and SmalltalkHub. > But I cannot make > > FMSound bass1 play > produce a sound. > > Phil if i remember correctly you worked on that a couple of months ago. > Can you summarize your findings? > > Stef

March 10, 2013

Re: [Pharo-project] [amber-lang] Quite a saturday night!

by Marcus Denker

On Mar 10, 2013, at 11:11 AM, Herby VojÄÃk <herby(a)mailbox.sk> wrote: > Wow! > > (and I learned what pomodoro is, except the tomato, of course :-) ) > http://www.pomodorotechnique.com The inventor of the Pomodoro (Francesco Cirillo) actually is a long time Smalltalker and an active Pharo user. Marcus

March 10, 2013

[Pharo-project] what do we do with SoundService?

by StÃ©phane Ducasse

Hi guy Sound has an AbstractSoundSystem and a Base and Dummy. Then in addition there is SounService. SoundService is not doing much soundEnabled ^ self defaultSoundEnabled defaultSoundEnabled ^DefaultSoundEnabled ifNil: [false] toggleSoundEnabling self soundEnabled: self soundEnabled not So we could just keep AbstractSoundSystem and deprecate SoundService. Stef

March 10, 2013

Re: [Pharo-project] [Lsehub-staff] Quite a saturday night!

by Herby VojÄÃk

Marcus Denker wrote: > Could Helios be used to edit Pharo code? Ha! If taken broader, it can be GSoC idea (and can help to more or less standardize server and client protocols between running Smalltalk and Smalltalk IDE; this can bring, in the future, for example, a Smalltalk IDE by JetBrains (pipedreaming again))! Herby > On Mar 10, 2013, at 5:16 AM, Nicolas Petton<petton.nicolas(a)gmail.com> wrote: > >> Hey guys, >> >> Together with Ben we decided to take the evening and kick Helios[1]' ass! >> Here's the result, after 16 pomodoros: >> >> helios speed, 2013-03-09 18:22:26 +0000, 25, 0, 0 >> helios announcement fix, 2013-03-09 18:54:51 +0000, 25, 0, 0 >> helios announcement fix, 2013-03-09 19:24:53 +0000, 25, 0, 0 >> helios announcement fix 2, 2013-03-09 19:55:41 +0000, 25, 0, 0 >> helios selectors cache, 2013-03-09 20:40:53 +0000, 25, 0, 0 >> helios selectors cache 2, 2013-03-09 22:19:47 +0000, 25, 0, 0 >> helios selectors cache 3, 2013-03-09 22:50:05 +0000, 25, 0, 0 >> helios selectors cache 4, 2013-03-09 23:21:42 +0000, 25, 0, 0 >> helios selectors cache 5, 2013-03-10 00:01:45 +0000, 25, 0, 0 >> helios selectors cache 6, 2013-03-10 00:31:47 +0000, 25, 0, 0 >> Helios Sytem Announcements, 2013-03-10 01:02:05 +0000, 25, 0, 0 >> Helios Sytem Announcements, 2013-03-10 01:32:10 +0000, 25, 0, 0 >> Helios Sytem Announcements 2, 2013-03-10 02:13:40 +0000, 25, 0, 0 >> Helios Sytem Announcements 2, 2013-03-10 02:45:35 +0000, 25, 0, 0 >> Helios Sytem Announcements 3 (Protocol), 2013-03-10 03:18:10 +0000, 25, 0, 0 >> Helios Sytem Announcements 4, 2013-03-10 04:07:26 +0000, 25, 0, 0 >> >> As a result, Helios listens to all system announcements and it's something like 10 times faster. >> >> Cheers, >> Nico >> >> [1] The next IDE of Amber > >

March 10, 2013

Re: [Pharo-project] [amber-lang] Quite a saturday night!

by Herby VojÄÃk

Wow! (and I learned what pomodoro is, except the tomato, of course :-) ) Nicolas Petton wrote: > Hey guys, > > Together with Ben we decided to take the evening and kick Helios[1]' ass! > Here's the result, after 16 pomodoros: > > helios speed, 2013-03-09 18:22:26 +0000, 25, 0, 0 > helios announcement fix, 2013-03-09 18:54:51 +0000, 25, 0, 0 > helios announcement fix, 2013-03-09 19:24:53 +0000, 25, 0, 0 > helios announcement fix 2, 2013-03-09 19:55:41 +0000, 25, 0, 0 > helios selectors cache, 2013-03-09 20:40:53 +0000, 25, 0, 0 > helios selectors cache 2, 2013-03-09 22:19:47 +0000, 25, 0, 0 > helios selectors cache 3, 2013-03-09 22:50:05 +0000, 25, 0, 0 > helios selectors cache 4, 2013-03-09 23:21:42 +0000, 25, 0, 0 > helios selectors cache 5, 2013-03-10 00:01:45 +0000, 25, 0, 0 > helios selectors cache 6, 2013-03-10 00:31:47 +0000, 25, 0, 0 > Helios Sytem Announcements, 2013-03-10 01:02:05 +0000, 25, 0, 0 > Helios Sytem Announcements, 2013-03-10 01:32:10 +0000, 25, 0, 0 > H elios Sytem Announcements 2, 2013-03-10 02:13:40 +0000, 25, 0, 0 > Helios Sytem Announcements 2, 2013-03-10 02:45:35 +0000, 25, 0, 0 > Helios Sytem Announcements 3 (Protocol), 2013-03-10 03:18:10 +0000, 25, 0, 0 > Helios Sytem Announcements 4, 2013-03-10 04:07:26 +0000, 25, 0, 0 > > As a result, Helios listens to all system announcements and it's something like 10 times faster. > > Cheers, > Nico > > [1] The next IDE of Amber

March 10, 2013

[Pharo-project] MorphicEventHandler behavior.

by StÃ©phane Ducasse

Hi fernando May be you know more than me on MorphicEventHandler. Gofer new squeaksource: 'PharoNonCorePackages'; package: 'Sound'; package: 'SoundMorphicUserInterface'; load. While trying to make PianoKeyboardMorph new openInWorld (which works in 1.4 and does not in 20). It breaks here: ------------- mouseUpPitch: pitch event: event noteMorph: noteMorph noteMorph color: ((#(0 1 3 5 6 8 10) includes: pitch \\12) ifTrue: [whiteKeyColor] ifFalse: [blackKeyColor]). soundPlaying ifNotNil: [soundPlaying stopGracefully]. pitch is nil because notify:from: uses valueParameter which is nil. ------------- mouseUpPitch is invoked by notify: anEvent from: sourceMorph which is invoked by notifyMorphsOfEvent: anEvent ofType: eventType from: sourceMorph notify: anEvent from: sourceMorph | arity | recipient ifNil: [^ self]. arity := selector numArgs. arity = 0 ifTrue: [^ recipient perform: selector]. arity = 1 ifTrue: [^ recipient perform: selector with: anEvent]. arity = 2 ifTrue: [^ recipient perform: selector with: anEvent with: sourceMorph]. arity = 3 ifTrue: [^ recipient perform: selector with: valueParameter with: anEvent with: sourceMorph]. ^^^^^^^^^^^^^^^^^^^^^^^ self error: 'Event handling selectors must be Symbols and take 0-3 arguments' notifyMorphsOfEvent: anEvent ofType: eventType from: sourceMorph | result | result := false. ((subscriptions includesKey: eventType) not or: [ (subscriptions at: eventType) isEmpty ]) ifTrue: [ ^false ]. (subscriptions at: eventType) do: [ :s | result := result | ((s notify: anEvent from: sourceMorph) == true) ]. ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^result The idea is that morphs get notified when a certain event occurs so when we look at the morph creation and event registry we get: The buildKeyboard create morph and define event with a given value that will be used on: #mouseDown send: #mouseDownPitch:event:noteMorph: to: self withValue: i-1*12 + (#(2 4 7 9 11) at: j) below: buildKeyboard | wtWid bkWid keyRect octavePt nWhite nBlack | self removeAllMorphs. wtWid := 8. bkWid := 5. self extent: 10@10. 1 to: nOctaves+1 do: [:i | i <= nOctaves ifTrue: [nWhite := 7. nBlack := 5] ifFalse: [nWhite := 1. nBlack := 0 "High C"]. octavePt := self innerBounds topLeft + ((7*wtWid*(i-1)-1) @ -1). 1 to: nWhite do: [:j | keyRect := octavePt + (j-1*wtWid@0) extent: (wtWid+1)@36. self addMorph: ((RectangleMorph newBounds: keyRect color: whiteKeyColor) borderWidth: 1; on: #mouseDown send: #mouseDownPitch:event:noteMorph: to: self withValue: i-1*12 + (#(1 3 5 6 8 10 12) at: j))]. 1 to: nBlack do: [:j | keyRect := octavePt + ((#(6 15 29 38 47) at: j)@1) extent: bkWid@21. self addMorph: ((Morph newBounds: keyRect color: blackKeyColor) on: #mouseDown send: #mouseDownPitch:event:noteMorph: to: self withValue: i-1*12 + (#(2 4 7 9 11) at: j))]]. self submorphsDo: [:m | m on: #mouseMove send: #mouseMovePitch:event:noteMorph: to: self; on: #mouseUp send: #mouseUpPitch:event:noteMorph: to: self; on: #mouseEnterDragging send: #mouseDownPitch:event:noteMorph: to: self; on: #mouseLeaveDragging send: #mouseUpPitch:event:noteMorph: to: self]. self extent: (self fullBounds extent + borderWidth - 1) Now I can understand that #mouseDownPitch:event:noteMorph: get notified with a value because this is what is written. But I do not get why this would be the case with #mouseUpPitch:event:noteMorph: since we simply get on: #mouseUp send: #mouseUpPitch:event:noteMorph: to: self; And I do not get why this was working in 1.4

March 10, 2013