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

Re: [Pharo-project] Call for Pharo Support **important**

by StÃ©phane Ducasse

On Feb 18, 2011, at 10:50 AM, Norbert Hartl wrote: > Of course I'm willing to sign, too. But to be honest I don't understand the meaning of "...in my capacity of.." what are you in the company kind of Inria gave me doc from another consortium :) so I thought it was ok. Stef

Feb. 18, 2011

Re: [Pharo-project] Pharo on LinkedIn ...

by StÃ©phane Ducasse

cool I hope that we will get more :). Stef >> 63 and counting... >> > > 74 and counting .... > -- > View this message in context: http://forum.world.st/Pharo-on-LinkedIn-tp3312460p3312759.html > Sent from the Pharo Smalltalk mailing list archive at Nabble.com. >

Feb. 18, 2011

Re: [Pharo-project] Issue 3733 in pharo: Float>>roundUpTo: wrong results

by pharo＠googlecode.com

Updates: Status: WontFix Comment #1 on issue 3733 by nicolas....(a)gmail.com: Float>>roundUpTo: wrong results http://code.google.com/p/pharo/issues/detail?id=3733 Don't put too many expectations on Float, they are inexact: 560.19 > 560.19s2 -> true 500.19 < 500.19s2 -> true QED Please use ScaledDecimal. 560.19s2 roundUpTo: 0.01s2. 560.19s2 500.19s2 roundUpTo: 0.01s2. 500.19s2 560.19s2 roundUpTo: 0.1s2. 560.20s2 500.19s2 roundUpTo: 0.1s2. 500.20s2

Feb. 18, 2011

Re: [Pharo-project] [squeak-dev] Re: Cog VM -- Thanks and Performance / Optimization Questions

by Juan Vuletich

Hi Eliot, Eliot Miranda wrote: > Hi John, > > good questions. > > On Thu, Feb 17, 2011 at 6:21 AM, John B Thiel <jbthiel(a)gmail.com > <mailto:jbthiel@gmail.com>> wrote: > > Cog VM -- Thanks and Performance / Optimization Questions > > ... > (Also, any notoriously slow subsystems? For example, Transcript > writing is glacial.) > > > Someone should replace the Transcript's reliance on (I think) some > kind of FormMorph which moved huge numbers of bits on each write. But > this is not a VM issue. It's a Smalltalk issue. Whoever did this > would instantly become a hero. There are other problems with Transcript besides performance. It is easy to break the UI if called at the wrong times. In those cases, it is usually needed to kill Squeak from the OS. Besides, it is not thread safe. Instantly becoming a hero sounds cool, so this is my attempt at it :) I only tried this in Cuis, but I believe it should work on Squeak and Pharo without much trouble. After loading this code, you can evaluate 'Smalltalk at: #Transcript put: NewTranscript'. Some good properties of this are: - Very fast. - Thread safe! - Doesn't use Morphic at all. Can be used to debug Morphic itself. - Doesn't do delayed execution (no events, no forked processes). Immediate visual feedback. - Can show itself on Display, log to File, both, or none. - Can be used for headless or remote images (for example, web servers). Some possible downsides are: - Output only. Not a text editor. We have the Workspace and the Transcripter for that. - No seamless Morphic integration. I'll do one for Cuis, though. If no serious problems are found, I'll remove TranscriptStream from Cuis, and rename this class as Transcript. Cheers, Juan Vuletich 'From Cuis 3.0 of 31 January 2011 [latest update: #790] on 18 February 2011 at 2:36:31 pm'! !classDefinition: #NewTranscript category: #'System-Support'! Object subclass: #NewTranscript instanceVariableNames: '' classVariableNames: '' poolDictionaries: '' category: 'System-Support'! !NewTranscript commentStamp: '<historical>' prior: 0! A new implementation of Transcript. - Thread safe. - Very fast. - Independent of Morphic or any other UI framework. - Inmediate feedback. - Can log to file. - Not an editor. Only used for output. - All protocol is on the Class side! !classDefinition: 'NewTranscript class' category: nil! NewTranscript class instanceVariableNames: 'entries firstIndex lastIndex accessSemaphore unfinishedEntry logToFile showOnDisplay innerRectangle lastDisplayTime'! !DateAndTime methodsFor: 'squeak protocol' stamp: 'jmv 2/18/2011 12:57'! printWithMsOn: aStream "Print with millisecond resolution, no leading space, no offset." | ps | self printYMDOn: aStream withLeadingSpace: false. aStream nextPut: $T. self printHMSOn: aStream. ps _ (self nanoSecond // 1000000) printString padded: #left to: 3 with: $0. aStream nextPut: $.. aStream nextPutAll: ps! ! !NewTranscript class methodsFor: 'preferred protocol' stamp: 'jmv 2/18/2011 14:16'! clear | stream | accessSemaphore critical: [ "Having at least one entry simplifies handling of the entries circular collection" firstIndex _ 1. lastIndex _ 1. entries at: 1 put: 'Transcript'. unfinishedEntry reset. logToFile ifTrue: [ stream _ StandardFileStream forceNewFileNamed: self filename. [ stream nextPutAll: 'Transcript log started: '. DateAndTime now printOn: stream. stream lf; nextPutAll: '------------------------------------------------------------------------'; lf ] ensure: [ stream close ]]]. self display! ! !NewTranscript class methodsFor: 'preferred protocol' stamp: 'jmv 2/18/2011 14:16'! log: aString self addEntry: aString. self display! ! !NewTranscript class methodsFor: 'preferred protocol' stamp: 'jmv 2/18/2011 11:49'! logToFile " self logToFile " logToFile _ true! ! !NewTranscript class methodsFor: 'old Transcript compatibility' stamp: 'jmv 2/18/2011 11:20'! cr "WriteStream protocol. In the older TranscriptStream, it added a CR character. Now, finish the current incomplete entry." self finishEntry! ! !NewTranscript class methodsFor: 'old Transcript compatibility' stamp: 'jmv 2/18/2011 11:20'! crtab "WriteStream protocol. End the current entry, and start a new one starting with a single tab character." self cr; tab! ! !NewTranscript class methodsFor: 'old Transcript compatibility' stamp: 'jmv 2/18/2011 11:21'! endEntry "For compatibility with old TranscriptStream. nop here"! ! !NewTranscript class methodsFor: 'old Transcript compatibility' stamp: 'jmv 2/18/2011 11:21'! flush "For compatibility with old TranscriptStream. nop here"! ! !NewTranscript class methodsFor: 'old Transcript compatibility' stamp: 'jmv 2/18/2011 14:17'! nextPut: aCharacter "WriteStream protocol. Append aCharacter to the unfinishedEntry. cr characters sent with this message do NOT finish the current unfinishedEntry." unfinishedEntry nextPut: aCharacter. self displayUnfinishedEntry! ! !NewTranscript class methodsFor: 'old Transcript compatibility' stamp: 'jmv 2/18/2011 14:17'! nextPutAll: aString "WriteStream protocol. Append aString to the unfinishedEntry. cr characters sent with this message do NOT finish the current unfinishedEntry." unfinishedEntry nextPutAll: aString. self displayUnfinishedEntry! ! !NewTranscript class methodsFor: 'old Transcript compatibility' stamp: 'jmv 2/18/2011 11:20'! print: anObject "Stream protocol" anObject printOn: self! ! !NewTranscript class methodsFor: 'old Transcript compatibility' stamp: 'jmv 2/18/2011 11:22'! show: anObject "Old TranscriptStream protocol." self nextPutAll: anObject asString! ! !NewTranscript class methodsFor: 'old Transcript compatibility' stamp: 'jmv 2/18/2011 11:22'! space "WriteStream protocol. Append a space character to the receiver." self nextPut: Character space! ! !NewTranscript class methodsFor: 'old Transcript compatibility' stamp: 'jmv 2/18/2011 11:22'! tab "WriteStream protocol. Append a tab character to the receiver." self nextPut: Character tab! ! !NewTranscript class methodsFor: 'private' stamp: 'jmv 2/18/2011 12:59'! addEntry: aString "Add a new entrie to the entries circular list. If full, a new entry will replace the oldest one." | msg now | logToFile ifTrue: [ now _ DateAndTime now. msg _ String streamContents: [ :strm | now printWithMsOn: strm. strm nextPutAll: ' process:'; nextPutAll: Processor activeProcess priority printString; nextPut: $ ; nextPutAll: Processor activeProcess hash printString; nextPut: $ ; nextPutAll: aString; lf ]]. self addEntry: aString logToFile: msg! ! !NewTranscript class methodsFor: 'private' stamp: 'jmv 2/18/2011 12:34'! addEntry: aString logToFile: otherString "Add a new entrie to the entries circular list. If full, a new entry will replace the oldest one." | stream | accessSemaphore critical: [ "Internal circular collection" lastIndex _ lastIndex \\ self maxEntries + 1. firstIndex = lastIndex ifTrue: [ firstIndex _ firstIndex \\ self maxEntries + 1 ]. entries at: lastIndex put: aString. "external file" otherString ifNotNil: [ [ stream _ StandardFileStream fileNamed: self filename. stream setToEnd; nextPutAll: otherString; flush] ensure: [ stream close ] ] ]! ! !NewTranscript class methodsFor: 'private' stamp: 'jmv 2/18/2011 14:16'! finishEntry | newEntry | newEntry _ unfinishedEntry contents. unfinishedEntry reset. self addEntry: newEntry. self display! ! !NewTranscript class methodsFor: 'displaying' stamp: 'jmv 2/18/2011 14:20'! display showOnDisplay ifTrue: [ self displayOn: Display. lastDisplayTime _ DateAndTime now ]! ! !NewTranscript class methodsFor: 'displaying' stamp: 'jmv 2/18/2011 14:09'! displayOn: aForm " experimentos NewTranscript displayOn: Display " | font port count i string x y fh f | aForm border: (innerRectangle outsetBy: 3) width: 3. aForm fill: innerRectangle fillColor: Color white. port _ BitBlt toForm: aForm. port clipWidth: innerRectangle right. font _ StrikeFont default. font installOn: port foregroundColor: Color black. fh _ font height. count _ innerRectangle height // fh-1. x _ innerRectangle left. y _ innerRectangle top. f _ firstIndex-1. firstIndex > lastIndex ifTrue: [ f _ f - self maxEntries ]. i _ (lastIndex - count max: f) \\ self maxEntries + 1. [ string _ entries at: i. port displayString: string from: 1 to: string size at: x@y strikeFont: font kern: font baseKern negated. y _ y + fh. i = lastIndex ] whileFalse: [ i _ i \\ self maxEntries + 1 ]. string _ unfinishedEntry contents. port displayString: string from: 1 to: string size at: x@y strikeFont: font kern: font baseKern negated.! ! !NewTranscript class methodsFor: 'displaying' stamp: 'jmv 2/18/2011 14:23'! displayUnfinishedEntry showOnDisplay ifTrue: [ (lastDisplayTime isNil or: [ (DateAndTime now - lastDisplayTime) totalSeconds > 1 ]) ifTrue: [ ^self display ]. self displayUnfinishedEntryOn: Display ]! ! !NewTranscript class methodsFor: 'displaying' stamp: 'jmv 2/18/2011 14:14'! displayUnfinishedEntryOn: aForm | font port count string x y fh | port _ BitBlt toForm: aForm. port clipWidth: innerRectangle right. font _ StrikeFont default. font installOn: port foregroundColor: Color black. fh _ font height. count _ innerRectangle height // fh-1. x _ innerRectangle left. string _ unfinishedEntry contents. y _ ((lastIndex - firstIndex \\ self maxEntries) min: count-1) + 1 * font height + innerRectangle top. port displayString: string from: 1 to: string size at: x@y strikeFont: font kern: font baseKern negated.! ! !NewTranscript class methodsFor: 'class initialization' stamp: 'jmv 2/18/2011 13:13'! initialize " self initialize " showOnDisplay _ true. innerRectangle _ 20@20 extent: 300@500. logToFile _ false. entries _ Array new: self maxEntries. unfinishedEntry _ '' writeStream. accessSemaphore _ Semaphore forMutualExclusion. self clear! ! !NewTranscript class methodsFor: 'constants' stamp: 'jmv 2/18/2011 12:33'! filename ^'transcript.txt'! ! !NewTranscript class methodsFor: 'constants' stamp: 'jmv 2/18/2011 12:59'! maxEntries ^1000! ! NewTranscript initialize!

Feb. 18, 2011

Re: [Pharo-project] Issue 3694 in pharo: Balloon message should be opaque for better readability

by pharo＠googlecode.com

Comment #2 on issue 3694 by jbth...(a)gmail.com: Balloon message should be opaque for better readability http://code.google.com/p/pharo/issues/detail?id=3694 I agree, the new burnt orange balloon color is too transparent and ugly, near-illegible, IMO. The former cyan color+transparency was very nice, why not keep that. (hmm, In the 1.1.1 image, there does not seem to be an initializer for the BalloonMorph BalloonColor classvar, and there are no senders of BalloonMorph class>>setBalloonColorTo: So possibly the initializer got lost a while back and we've been running on an image constant for some time.) I found the former nice cyan color is rgb = 977210367 alpha = 242 TranslucentColor r:0.91 g:0.94 b: 1.0 alpha: 0.949 (note: slight glitch, that r:g:b:alpha: expression actually yields alpha=241 rgb=977210367 It's constructed in the Squeak 4.1 BlueSmallLandColorTheme, as follows balloonColor ^ (self light: 1) twiceLighter alpha: 0.95 where (self light: 1) in this case is: (Color fromArray: #(0.85 0.9 1.0 )) Squeak 4.1 looks to have a reasonably well developed Color system with different themes. It's not clear that needs to be alternately re-implemented, if that's what's happening here. At least we should keep some of the good themes. Please lets restore the legacy cyan balloon color for now, and I agree with making this a themable SystemSetting. The quick fix patch for 1.2rc2 is BalloonMorph class>>defaultBalloonColor ^ (Color fromArray: #(0.85 0.9 1.0 )) twiceLighter alpha: 0.95

Feb. 18, 2011

Re: [Pharo-project] minimalism (formerly case-statement-stuff)

by Igor Stasenko

On 18 February 2011 17:42, Craig Latta <craig(a)netjam.org> wrote: > >> But the first image must have a compiler, only the derived (aggregated >> by transfer) can avoid a compiler. > > Â Â Sure, any image can load the compiler when needed (without needing > the compiler to do that :), and unload it at will. The point is that it > need not be part of the kernel image. > > For developers like me, loading the code is a bit less of value than being able to create new code (through compiling of course). :) But of course for production systems, one may choose to not have compiler, which is perfectly fine and even helps to close some security gaps. > -C > > -- > Craig Latta > www.netjam.org/resume > +31 Â 06 2757 7177 > + 1 415 Â 287 3547 > -- Best regards, Igor Stasenko AKA sig.

Feb. 18, 2011

Re: [Pharo-project] minimalism (formerly case-statement-stuff)

by Craig Latta

> But the first image must have a compiler, only the derived (aggregated > by transfer) can avoid a compiler. Sure, any image can load the compiler when needed (without needing the compiler to do that :), and unload it at will. The point is that it need not be part of the kernel image. -C -- Craig Latta www.netjam.org/resume +31 06 2757 7177 + 1 415 287 3547

Feb. 18, 2011

Re: [Pharo-project] minimalism (formerly case-statement-stuff)

by Miguel CobÃ¡

But the first image must have a compiler, only the derived (aggregated by transfer) can avoid a compiler. Unless that every derived image uses a central compiler enabled image to process its new source code and get back the compiled version. Cheers El vie, 18-02-2011 a las 10:09 +0100, Craig Latta escribiÃ³: > Hi-- > > If you can transfer compiled methods directly, then you can make > the entire compiler an optional module... > > As the Americans are fond of saying, "just sayin'". :) > > > -C > > -- > Craig Latta > www.netjam.org/resume > +31 06 2757 7177 > + 1 415 287 3547 > > > -- Miguel CobÃ¡ http://twitter.com/MiguelCobaMtz http://miguel.leugim.com.mx

Feb. 18, 2011

[Pharo-project] Mailing list weekly digest #2

by Francois Stephany

Hi there, I've posted the second weekly digest. If I missed something or if you have feedback please tell me ! This digest thing will probably be hosted on the main pharo website in the (near?) future. If you want to help, you're more than welcome. http://articles.tulipemoutarde.be/pharo-mailing-list-weekly-digest Have a good weekend, Francois

Feb. 18, 2011

Re: [Pharo-project] MNU MIMETYPE class>>forURIReturnSingleMimeTypeOrDefault:

by Diogenes Moreira

Well I find the solution in the 2797 issue.<http://code.google.com/p/pharo/issues/detail?id=2797> [1] thanks anyway [1] http://code.google.com/p/pharo/issues/detail?id=2797 On Fri, Feb 18, 2011 at 11:57 AM, Diogenes Moreira < diogenes.moreira(a)gmail.com> wrote: > Hi.. > > that is a StandardFileStream bug or I'm loosing some thing... > may be a versions problem.. > > somebody can say where is implement that method. > > > Best >

Feb. 18, 2011