December 2018 - Pharo-users

Iceberg repository import

by Trussardi Dario Romano

Ciao, i work with Pharo 7.0.0 build 34. in the past, into another pharo environment, i have clone the git Seaside project ( master 9566cb4 ) Now in the new image for import the Seaside project, i do the Iceberg command: Import from existing clone the Repository window report the entry: Seaside Master No Project Found The relative Package window is empty. Because ? what i need to do? In parallel i work with Magritte project, i import it from existing clone without problem. Some considerations ? Thanks, Dario

Dec. 15, 2018

Re: [Pharo-users] Pharo 7 and Zero conf problem

by Esteban Lorenzano

You can ignore it (but is something we need to fix, indeed) > On 15 Dec 2018, at 00:53, Vitor Medina Cruz <vitormcruz(a)gmail.com> wrote: > > Hello, > > Under debian, loading > > curl https://get.pharo.org/64/70+vm <https://get.pharo.org/64/70+vm> | bash > > And then executing: > > ./pharo Pharo.image eval "Class allInstances" > > I got: > > self recordInitializationSuccess ] in LGitLibrary>>initializeLibGit2 in Block: [ self libgit2_init.... > BlockClosure>>on:do: > LGitLibrary>>initializeLibGit2 > LGitLibrary class>>startUp: > ClassSessionHandler>>startup: > [ :each | each startup: isImageStarting ] in WorkingSession>>runStartup: in Block: [ :each | each startup: isImageStarting ] > [ aBlock value: each ] in [ :each | > [ aBlock value: each ] > on: Exception > do: [ :error | self errorHandler handleError: error ] ] in WorkingSession>>runList:do: in Block: [ aBlock value: each ] > BlockClosure>>on:do: > [ :each | > [ aBlock value: each ] > on: Exception > do: [ :error | self errorHandler handleError: error ] ] in WorkingSession>>runList:do: in Block: [ :each | ... > Array(SequenceableCollection)>>do: > WorkingSession>>runList:do: > WorkingSession>>runStartup: > WorkingSession>>start: > SessionManager>>launchSnapshot:andQuit: > [ isImageStarting := self launchSnapshot: save andQuit: quit. > wait signal ] in SessionManager>>snapshot:andQuit: in Block: [ isImageStarting := self launchSnapshot: save and...etc... > [ self value. > Processor terminateActive ] in BlockClosure>>newProcess in Block: [ self value.... > #() > > If I use pharo 6 this error don't happened and I got #() > > Can I ignore this error? Should I use Pharo 6? Something missed? > > Regards, > Vitor > > <http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_cam…> Livre de vÃrus. www.avg.com <http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_cam…>. > <x-msg://13/#m_5683544748734822669_DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>

Dec. 15, 2018

Re: [Pharo-users] how to rewrite this to use a stream

by Roelof Wobben

Dec. 15, 2018

Re: [Pharo-users] how to rewrite this to use a stream

by Roelof Wobben

Dec. 15, 2018

Re: [Pharo-users] how to rewrite this to use a stream

by phil＠highoctane.be

Thx for those reviews. And interesting paper. Phil On Sat, Dec 15, 2018, 04:20 Richard O'Keefe <raoknz(a)gmail.com wrote: > Starting from the top:\ > (0) It's not *wrong* to start a selector with a capital letter, but > it is certainly unusual. > (1) Intention-revealing names are good. That's why this name is bad. > It does not *delete* anything. > (2) The comment again says "delete" but nothing is deleted. Comments > that mislead are worse than no comments. > (3) An issue I have seen over and over in student code (in other > programming languages) is writing regular expressions as string > literals where they are used, so that every time you try a match > the regular expression gets recompiled, over and over again. > You're not exactly doing that, but you *are* rebuilding patterns > over and over again. It would make more sense to hold a > forbiddenPatterns collection. > (4) You are saying "(coll reject: [..test..]) notEmpty", building an > entire collection just to ask whether it has any members. What > you want to know is "are there any elements of coll that do not > satisfy the test", which can be expressed as > (coll allSatisfy: [..test..]) not. (Sadly there is no > #notAllSatisfy:, although #notAnySatisfy: *does* exist under the > name #noneSatisfy:. > (5) #match: is a highly risky way to look for a substring. What if > you want to look for a literal # or * ? What if you need to > look for 'Boring' but not 'boring'? ("Boring" is a perfectly > good surname.) > (6) You really don't need the "result" variable. > > allowedWords > "Answer those words which do not contain any forbiddenWord > as a substring." > ^words select: [:each | > forbiddenWords noneSatisfy: [:forbidden | > <<each includes forbidden as a substring>>]] > > (7) You can find out how to do the <<part>> by looking at the methods > for String. You will find several candidate methods. > (8) Of course, if there are W words and F forbidden words, in the > worst case you'll do W*F substring inclusion tests. This is not > good. "Multiple pattern matching" is a much-studied problem, > see for example Wu and Manber's paper > > http://web.archive.org/web/20130203031412/http://www.inf.fu-berlin.de/inst/… > There is more recent work also. > > Does anyone know whether there's a multiple pattern algorithm for > Pharo? One could always fall back on ternary search trees... > > > > On Fri, 14 Dec 2018 at 05:13, Roelof Wobben <r.wobben(a)home.nl> wrote: > >> Hello, >> >> I have this code : >> >> FindAndDeleteWordsWithForbiddenParts >> "deletes all the words that contain forbidden parts" >> >> | result | >> result := words >> select: [ :word | >> [ (forbiddenWords reject: [ :forbidden | '*' , forbidden , '*' match: word ]) isNotEmpty ] >> on: NotFound >> do: [ false ] ]. >> ^ result >> >> >> but I see warnings that I have to use a stream instead of string concentation. >> >> Anyone hints how to do so ? >> >> Roelof >> >> >>

Dec. 15, 2018

Re: [Pharo-users] StCAD is open source

by askoh

Hi: I am preparing to port StCAD from VisualWorks to Pharo. Please advise how this can be done most efficiently. I am not familiar with Pharo GUI development. StCAD uses Jun which uses OpenGL and GLUT. I am open to using newer graphics frameworks and CUDA or OpenCL. Thanks, Aik-Siong Koh -- Sent from: http://forum.world.st/Pharo-Smalltalk-Users-f1310670.html

Dec. 15, 2018

Re: [Pharo-users] how to rewrite this to use a stream

by Richard O'Keefe

Starting from the top:\ (0) It's not *wrong* to start a selector with a capital letter, but it is certainly unusual. (1) Intention-revealing names are good. That's why this name is bad. It does not *delete* anything. (2) The comment again says "delete" but nothing is deleted. Comments that mislead are worse than no comments. (3) An issue I have seen over and over in student code (in other programming languages) is writing regular expressions as string literals where they are used, so that every time you try a match the regular expression gets recompiled, over and over again. You're not exactly doing that, but you *are* rebuilding patterns over and over again. It would make more sense to hold a forbiddenPatterns collection. (4) You are saying "(coll reject: [..test..]) notEmpty", building an entire collection just to ask whether it has any members. What you want to know is "are there any elements of coll that do not satisfy the test", which can be expressed as (coll allSatisfy: [..test..]) not. (Sadly there is no #notAllSatisfy:, although #notAnySatisfy: *does* exist under the name #noneSatisfy:. (5) #match: is a highly risky way to look for a substring. What if you want to look for a literal # or * ? What if you need to look for 'Boring' but not 'boring'? ("Boring" is a perfectly good surname.) (6) You really don't need the "result" variable. allowedWords "Answer those words which do not contain any forbiddenWord as a substring." ^words select: [:each | forbiddenWords noneSatisfy: [:forbidden | <<each includes forbidden as a substring>>]] (7) You can find out how to do the <<part>> by looking at the methods for String. You will find several candidate methods. (8) Of course, if there are W words and F forbidden words, in the worst case you'll do W*F substring inclusion tests. This is not good. "Multiple pattern matching" is a much-studied problem, see for example Wu and Manber's paper http://web.archive.org/web/20130203031412/http://www.inf.fu-berlin.de/inst/… There is more recent work also. Does anyone know whether there's a multiple pattern algorithm for Pharo? One could always fall back on ternary search trees... On Fri, 14 Dec 2018 at 05:13, Roelof Wobben <r.wobben(a)home.nl> wrote: > Hello, > > I have this code : > > FindAndDeleteWordsWithForbiddenParts > "deletes all the words that contain forbidden parts" > > | result | > result := words > select: [ :word | > [ (forbiddenWords reject: [ :forbidden | '*' , forbidden , '*' match: word ]) isNotEmpty ] > on: NotFound > do: [ false ] ]. > ^ result > > > but I see warnings that I have to use a stream instead of string concentation. > > Anyone hints how to do so ? > > Roelof > > >

Dec. 15, 2018

Re: Pharo 7 and Zero conf problem

by John Kimber

Hi, I recently installed Pharo-Launcher under Mint 19. Version 70 build 1384 seems to be running fine. HTH, john -- Sent from: http://forum.world.st/Pharo-Smalltalk-Users-f1310670.html

Dec. 15, 2018

Pharo 7 and Zero conf problem

by Vitor Medina Cruz

Hello, Under debian, loading curl https://get.pharo.org/64/70+vm | bash And then executing: ./pharo Pharo.image eval "Class allInstances" I got: self recordInitializationSuccess ] in LGitLibrary>>initializeLibGit2 in Block: [ self libgit2_init.... BlockClosure>>on:do: LGitLibrary>>initializeLibGit2 LGitLibrary class>>startUp: ClassSessionHandler>>startup: [ :each | each startup: isImageStarting ] in WorkingSession>>runStartup: in Block: [ :each | each startup: isImageStarting ] [ aBlock value: each ] in [ :each | [ aBlock value: each ] on: Exception do: [ :error | self errorHandler handleError: error ] ] in WorkingSession>>runList:do: in Block: [ aBlock value: each ] BlockClosure>>on:do: [ :each | [ aBlock value: each ] on: Exception do: [ :error | self errorHandler handleError: error ] ] in WorkingSession>>runList:do: in Block: [ :each | ... Array(SequenceableCollection)>>do: WorkingSession>>runList:do: WorkingSession>>runStartup: WorkingSession>>start: SessionManager>>launchSnapshot:andQuit: [ isImageStarting := self launchSnapshot: save andQuit: quit. wait signal ] in SessionManager>>snapshot:andQuit: in Block: [ isImageStarting := self launchSnapshot: save and...etc... [ self value. Processor terminateActive ] in BlockClosure>>newProcess in Block: [ self value....*#()* If I use pharo 6 this error don't happened and I got *#()* Can I ignore this error? Should I use Pharo 6? Something missed? Regards, Vitor <http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_cam…> Livre de vÃrus. www.avg.com <http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_cam…>. <#m_5683544748734822669_DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>

Dec. 14, 2018

Re: [Pharo-users] Agile Artificial Intelligence book

by Alexandre Bergel

Hi! Thanks for having spotted it. Code for the first chapters can be downloaded using: Metacello new baseline: 'NeuralNetwork'; repository: 'github://AgileArtificialIntelligence/PharoSourceCode/src' ; load. But note that the code is not officially released. We will package it betterâ¦ Cheers, Alexandre -- _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: Alexandre Bergel http://www.bergel.eu ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. > On Dec 13, 2018, at 3:35 PM, mldavis99 <mldavis(a)chaparralwest.com> wrote: > > Hi, the Book so far is great, I am looking forward to more. However, when I > try to load the code with: > Metacello new > package: 'NeuralNetwork'; > repository: 'github://AgileArtificialIntelligence/PharoSourceCode/src' ; > load. > Both Pharo 6.1 stable and 7 Development Version do not understand: package: > When I dig into it a bit this message is implemented in an abstract method > of Metacello which does not appear to be in Metacello hierarchy?? What did > I miss? > mike > > > > -- > Sent from: http://forum.world.st/Pharo-Smalltalk-Users-f1310670.html >

Dec. 14, 2018