May 2009 - Pharo-dev - lists.pharo.org

Re: [Pharo-project] UIManager default informUserDuring: -> DNU (10300)

by Gary Chambers

The method in question incorporates some of Keith Hodges fun things. Valid values are: #current (return bar size) #label (return current label) #setMax (allegedly change the max value) #setMin (allegedly change the min value) #stripe (use animated stripey background) #increment (add one to current value) #decrement (subtract one from current value) #refresh (force redraw) a String (set the label) a Number (set the value) I'll update the comment example in PSUIManager to: UIManager default informUserDuring: [:bar| #('one' 'two' 'three') do: [:info| bar value: info. 1 to: 100 do: [:v | bar value: v. (Delay forMilliseconds: 20) wait]]] Regards, Gary ----- Original Message ----- From: "StÃ©phane Ducasse" <stephane.ducasse(a)inria.fr> To: "Pharo Development" <Pharo-project(a)lists.gforge.inria.fr> Sent: Tuesday, May 05, 2009 9:06 PM Subject: [Pharo-project] UIManager default informUserDuring: -> DNU (10300) > UIManager default informUserDuring: > [:bar | > #(one two three) > do: [ :info | > bar value: info. > (Delay forSeconds: 1)]] > > leads to an error. > Apparently barValue does not like to be a symbol. > > Stef > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

May 12, 2009

Re: [Pharo-project] [ANN] OCompletion improves the accuracy of eCompletion

by StÃ©phane Ducasse

Thanks a lot romain. Does OCompletion take into account that the receiver is a class (to look for class method only)? Stef On May 12, 2009, at 10:07 AM, Romain Robbes wrote: > Dear Pharo users, > > OCompletion is an extension to ECompletion. Like ECompletion, it > completes > the names of methods as you write them in order to save you some > typing. > The O in OCompletion stands for Omniscient and Omnipresent. > > OCompletion is much more accurate than ECompletion. We tested it on > the > development history of several projects > (see http://www.inf.unisi.ch/phd/robbes/papers/ASE2008- > completion.pdf). > > Across several hundred thousands tests of the completion, > OCompletion's > algorithm scores a 70 (out of 100), while the eCompletion's algorithm > scores a 12. To achieve this level of accuracy, it learns from your > usage > of squeak and proposes to you methods you are more likely to use. > > OCompletion shows a small, non-instrusive completion menu with > potential > completions as you type method names. You don't need to call it > explicitely > with tab. Just press tab to insert a match, or the down arrow to > select the next > match. > > To install it on Squeak or Pharo, simply execute the following line in > your image: > > Installer squeaksource project: 'OCompletion'; install: 'OCLoader' > > Cheers, > Romain > > -- > Romain Robbes > http://www.inf.unisi.ch/phd/robbes > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >

May 12, 2009

[Pharo-project] Alien question

by StÃ©phane Ducasse

Hi I'm trying to make sure that we can load Alien in Pharo. I'm moving some examples out of Alien-Core. Now I have some questions: - are these DynSuperSend #LdOuter #LdExplicitOuter class variable needed? - then we get a problem because the addition of classVariables should be done before parsing so I will move that to the preLoadingPackage. Stef initialize "ParseNode initialize. VariableNode initialize" #(#DynSuperSend #LdOuter #LdExplicitOuter) do: [:c | [self addClassVarName: c] ifError: []]. DynSuperSend := 126. LdInstType := 1. LdTempType := 2. LdLitType := 3. LdLitIndType := 4. SendType := 5. CodeBases := #(0 16 32 64 208 ). CodeLimits := #(16 16 32 32 16 ). LdSelf := 112. LdTrue := 113. LdFalse := 114. LdNil := 115. LdMinus1 := 116. LdOuter := 127. LdExplicitOuter := 139. LoadLong := 128. Store := 129. StorePop := 130. ShortStoP := 96. SendLong := 131. DblExtDoAll := 132. SendLong2 := 134. LdSuper := 133. Pop := 135. Dup := 136. LdThisContext := 137. EndMethod := 124. EndRemote := 125. Jmp := 144. Bfp := 152. JmpLimit := 8. JmpLong := 164. "code for jmp 0" BtpLong := 168. SendPlus := 176. Send := 208. SendLimit := 16

May 12, 2009

Re: [Pharo-project] Alien

by StÃ©phane Ducasse

On May 12, 2009, at 11:11 AM, Torsten Bergmann wrote: >> We are trying to package it to load in the latest pharo. > > Good - but whats the state of support for the major platforms? > (AFAIK John had it working on Mac). So far we cannot load it on mac. I'm moving the not loading examples into another packages so that we can at least load. > For instance on Windows a good pharo foreign interface is crucial > for commercial value (accessing Win32API/COM/.NET/third party). Well, it depends on windows people :). Now I wish and open that there will be support > > > Whats the plan regarding Alien? Intended for the 1.0 release or > later? Loadable package or included within the standard image? for now loadable package > Will it be a merge of FFI and Alien, or is Alien good enough? I do not know but from eliot told me Alien is good enough. People may build on top of it better support to have higher level to bridge with dll (a bit like dllcc in VW). > > > Thx > Torsten > > > > -- > Neu: GMX FreeDSL Komplettanschluss mit DSL 6.000 Flatrate + > Telefonanschluss fÃ¼r nur 17,95 Euro/mtl.!* http://dslspecial.gmx.de/freedsl-surfflat/?ac=OM.AD.PD003K11308T4569a > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >

May 12, 2009

[Pharo-project] Alien

by Torsten Bergmann

>We are trying to package it to load in the latest pharo. Good - but whats the state of support for the major platforms? (AFAIK John had it working on Mac). For instance on Windows a good pharo foreign interface is crucial for commercial value (accessing Win32API/COM/.NET/third party). Whats the plan regarding Alien? Intended for the 1.0 release or later? Loadable package or included within the standard image? Will it be a merge of FFI and Alien, or is Alien good enough? Thx Torsten -- Neu: GMX FreeDSL Komplettanschluss mit DSL 6.000 Flatrate + Telefonanschluss fÃ¼r nur 17,95 Euro/mtl.!* http://dslspecial.gmx.de/freedsl-surfflat/?ac=OM.AD.PD003K11308T4569a

May 12, 2009

Re: [Pharo-project] Alien

by StÃ©phane Ducasse

We are trying to package it to load in the latest pharo. Because right now we get an error when loading. Stef On May 12, 2009, at 10:38 AM, Torsten Bergmann wrote: > Any new from the "Alien vs. FFI" front ;) > Is Alien already available for various platforms like Win32? > > Thx > Torsten > -- > Neu: GMX FreeDSL Komplettanschluss mit DSL 6.000 Flatrate + > Telefonanschluss fÃ¼r nur 17,95 Euro/mtl.!* http://dslspecial.gmx.de/freedsl-surfflat/?ac=OM.AD.PD003K11308T4569a > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >

May 12, 2009

[Pharo-project] Wiresong and OB

by Torsten Bergmann

Looks like wiresong.ca is not responding (or is it only me). Why not move OB project code to squeaksource - this would also allow for easy browsing the code using the webbrowser ... Thx Torsten -- Neu: GMX FreeDSL Komplettanschluss mit DSL 6.000 Flatrate + Telefonanschluss fÃ¼r nur 17,95 Euro/mtl.!* http://dslspecial.gmx.de/freedsl-surfflat/?ac=OM.AD.PD003K11308T4569a

May 12, 2009

[Pharo-project] Alien

by Torsten Bergmann

Any new from the "Alien vs. FFI" front ;) Is Alien already available for various platforms like Win32? Thx Torsten -- Neu: GMX FreeDSL Komplettanschluss mit DSL 6.000 Flatrate + Telefonanschluss fÃ¼r nur 17,95 Euro/mtl.!* http://dslspecial.gmx.de/freedsl-surfflat/?ac=OM.AD.PD003K11308T4569a

May 12, 2009

[Pharo-project] [ANN] OCompletion improves the accuracy of eCompletion

by Romain Robbes

Dear Pharo users, OCompletion is an extension to ECompletion. Like ECompletion, it completes the names of methods as you write them in order to save you some typing. The O in OCompletion stands for Omniscient and Omnipresent. OCompletion is much more accurate than ECompletion. We tested it on the development history of several projects (see http://www.inf.unisi.ch/phd/robbes/papers/ASE2008-completion.pdf) Across several hundred thousands tests of the completion, OCompletion's algorithm scores a 70 (out of 100), while the eCompletion's algorithm scores a 12. To achieve this level of accuracy, it learns from your usage of squeak and proposes to you methods you are more likely to use. OCompletion shows a small, non-instrusive completion menu with potential completions as you type method names. You don't need to call it explicitely with tab. Just press tab to insert a match, or the down arrow to select the next match. To install it on Squeak or Pharo, simply execute the following line in your image: Installer squeaksource project: 'OCompletion'; install: 'OCLoader' Cheers, Romain -- Romain Robbes http://www.inf.unisi.ch/phd/robbes

May 12, 2009

Re: [Pharo-project] Long delay after loading 10301+10302

by StÃ©phane Ducasse

On May 12, 2009, at 12:37 AM, Nicolas Cellier wrote: > indeed, I think it's got nothing to do with change contents per se. > It must be kind of bad luck with finalizationProcess... > See also http://bugs.squeak.org/view.php?id=7350 : WeakRegistry > behaves as bad as WeakSet. > It probably has more to do with bug encountered with loadOBalpha. This > one was reproduceable: maybe I should not have pushed a workaround > that soon ;) yes probably may be we could have a cs to revert the fixaround Stef > > > Nicolas > > 2009/5/12 StÃ©phane Ducasse <stephane.ducasse(a)inria.fr>: >> Hi nicolas >> >> this is strange because 10301 was a bunch for collection test and a >> merge fix for traits. >> then the 302 were mainly your fixes on numbers. So I have no idea >> about >> what is happening. >> >> Did that happened during 10303 because there the changes were the one >> below + InputSensor initalization. >> >> >> >> >> >> Stef >> >> >> 'From Pharo0.1 of 16 May 2008 [Latest update: #10300] on 11 May 2009 >> at 8:00:50 pm'! >> >> !HostSystemMenusMenuItem class methodsFor: 'handlers' stamp: >> 'michael.rueger 5/11/2009 19:57'! >> fakeKeyboardEventBlockasciiActual: anAsciiValue unicode: >> aUnicodeValue >> event: evt >> |event | >> event := Array new: 8. >> event at: 1 put: 2 "EventTypeKeyboard"; >> at: 2 put: Time millisecondClockValue; >> at: 3 put: aUnicodeValue asInteger; >> at: 4 put: 1; "key down" >> at: 5 put: 8; "modifier keys (CmmandKeyBit)" >> at: 8 put: (evt at: 8). >> Sensor handleEvent: event. >> >> event := Array new: 8. >> event at: 1 put: 2 "EventTypeKeyboard"; >> at: 2 put: Time millisecondClockValue; >> at: 3 put: anAsciiValue asInteger; >> at: 4 put: 0; "key char" >> at: 5 put: 8; "modifier keys (CmmandKeyBit)" >> at: 8 put: (evt at: 8). >> Sensor handleEvent: event. >> >> event := Array new: 8. >> event at: 1 put: 2 "EventTypeKeyboard"; >> at: 2 put: Time millisecondClockValue; >> at: 3 put: aUnicodeValue asInteger; >> at: 4 put: 2; "key press/release" >> at: 5 put: 64; "modifier keys (CmmandKeyBit)" >> at: 8 put: (evt at: 8). >> Sensor handleEvent: event! ! >> >> !HostSystemMenusMenuItem class methodsFor: 'handlers' stamp: >> 'michael.rueger 5/11/2009 19:57'! >> fakeKeyboardEventBlockasciiActual: anAsciiValue unicode: >> aUnicodeValue >> event: evt virtualKey: aVirtualKeyValue >> |event | >> >> event := Array new: 8. >> event at: 1 put: 2 "EventTypeKeyboard"; >> at: 2 put: Time millisecondClockValue; >> at: 3 put: aVirtualKeyValue asInteger; >> at: 4 put: 1; "key down" >> at: 5 put: 8; "modifier keys (CmmandKeyBit)" >> at: 8 put: (evt at: 8). >> Sensor handleEvent: event. >> >> event := Array new: 8. >> event at: 1 put: 2 "EventTypeKeyboard"; >> at: 2 put: Time millisecondClockValue; >> at: 3 put: anAsciiValue asInteger; >> at: 4 put: 0; "key char" >> at: 5 put: 8; "modifier keys (CmmandKeyBit)" >> at: 6 put: aUnicodeValue asInteger; "virtual key >> code" >> at: 8 put: (evt at: 8). >> Sensor handleEvent: event. >> >> event := Array new: 8. >> event at: 1 put: 2 "EventTypeKeyboard"; >> at: 2 put: Time millisecondClockValue; >> at: 3 put: aVirtualKeyValue asInteger; >> at: 4 put: 2; "key press/release" >> at: 5 put: 64; "modifier keys (CmmandKeyBit)" >> at: 8 put: (evt at: 8). >> Sensor handleEvent: event! ! >>> >> >>> Hmm, a high priority process taking CPU for a long time, that >>> smells... >> >> >> _______________________________________________ >> 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 >

May 12, 2009