December 2018 - Pharo-users

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

by Richard Sargent

On Thu, Dec 13, 2018 at 12:44 PM Roelof Wobben <r.wobben(a)home.nl> wrote: > there is a method but then I ran into another warning > > FindAndDeleteWordsWithForbiddenParts > "deletes all the words that contain forbidden parts" > > | result | > result := words > select: [ :word | > [ (forbiddenWords detect: [ :forbidden | word includesAll: > forbidden ]) = true ] > on: NotFound > do: [ true ] ]. > There are a number of things you can do, such as the variations on detect , for example. But, even those keep you from the answer. The question you are trying to answer is which words satisfy the condition that they don't include any forbidden word as a sub-string. There are other variations than detect for that kind of question. Try examining the API in Collection for any which suggest a correspondence, and if not there, check subclasses in turn down to the classes of the collections you are working with. And don't forget that #select: isn't the only way to extract a subset of a collection. Look around for others. [I'm deliberately not giving you the answer, because I think you will benefit from learning how to work effectively in the environment.] ^ result > > then I see a unnecearry = true but if I deleted that part the code does > not what I was intented to do. > > Roelof > > > > Op 13-12-2018 om 18:15 schreef Richard Sargent: > > It's bad advice in this case, but correctly indicates there is a problem > with your code. See below. > > On Thu, Dec 13, 2018 at 8:13 AM 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 ] >> >> Constructing a pattern and pattern matching is a heavy-handed approach to > find out if a word contains a substring. > I'm not familiar with Pharo's methods, but at the least String inherits > #indexOfSubCollection:startingAt: and probably a number of others. There is > also a good chance String inherits or implements a more specific method > that will test for the presence of a sub-string. > > A good piece of advice I once received is (paraphrased as): Spend more > time reading and less time writing. :-) > The deeper a class hierarchy, the greater the chance that the exact thing > you need has already been implemented in the leaf class or its abstraction. > >> >> 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. 13, 2018

Re: [Pharo-users] Agile Artificial Intelligence book

by mldavis99

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. 13, 2018

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

by Roelof Wobben

Dec. 13, 2018

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

by Richard Sargent

It's bad advice in this case, but correctly indicates there is a problem with your code. See below. On Thu, Dec 13, 2018 at 8:13 AM 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 ] > > Constructing a pattern and pattern matching is a heavy-handed approach to find out if a word contains a substring. I'm not familiar with Pharo's methods, but at the least String inherits #indexOfSubCollection:startingAt: and probably a number of others. There is also a good chance String inherits or implements a more specific method that will test for the presence of a sub-string. A good piece of advice I once received is (paraphrased as): Spend more time reading and less time writing. :-) The deeper a class hierarchy, the greater the chance that the exact thing you need has already been implemented in the leaf class or its abstraction. > > 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. 13, 2018

how to rewrite this to use a stream

by Roelof Wobben

Dec. 13, 2018

Re: [Pharo-users] Pharo 7 under Linux

by Konrad Hinsen

Hi Esteban, > While it should not crash (just disable iceberg), it will be useful to > know which kind of linux are you using. It's Ubuntu 18.10. And it doesn't crash in the sense of crashing the VM, there's just an error message that pops up when I start the image. I didn't try to use that image after that, so perhaps it's just Iceberg that is affected (bad enough, how can you work without Iceberg?). I noticed something weird when inspecting the VM directory more closely: $ cd Pharo/vms/70-x64/lib/pharo/5.0-201806281256/ $ ls -l libgit2* -rw-r--r-- 1 hinsen hinsen 13 dÃ©c. 9 10:54 libgit2.so -rw-r--r-- 1 hinsen hinsen 3019447 dÃ©c. 9 10:54 libgit2.so.0.25.1 -rw-r--r-- 1 hinsen hinsen 17 dÃ©c. 9 10:54 libgit2.so.25 $ cat libgit2.so libgit2.so.25$ $ cat libgit2.so.25 libgit2.so.0.25.1$ What I'd expect to be symbolic links (libgit2.so and libgit2.so.25) are in fact text files containing another filename. I tried replacing them by proper symlinks, but that doesn't change anything in the error message. Konrad.

Dec. 13, 2018

Re: [Pharo-users] [ANN] Pharo Newsletter December 2018

by Marcus Denker

> On 13 Dec 2018, at 01:39, Sean P. DeNigris <sean(a)clipperadams.com> wrote: > > Marcus Denker-4 wrote >> Last we the Pharo Newsletter December 2018 was send. > > Wow, RMapViewer looks really cool :) Is the Units framework mentioned the > same one from SqS? > I guess? It was on smalltalkhub and I moved it to GitHub as we are using it at Zweidenker. Marcus

Dec. 13, 2018

Re: [Pharo-users] [ANN] Pharo Newsletter December 2018

by Alexandre Bergel

Hi Marcus! Maybe a news about the first 6 chapters of Agile Artificial Intelligence. https://agileartificialintelligence.github.io/ Cheers Alexandre http://bergel.eu > Le 12 dÃ©c. 2018 Ã 05:17, Marcus Denker <marcus.denker(a)inria.fr> a Ã©crit : > > > >> On 11 Dec 2018, at 14:08, Marcus Denker <marcus.denker(a)inria.fr> wrote: >> >> Hi, >> >> Last we the Pharo Newsletter December 2018 was send. >> >> You can find the archived version now here: >> >> https://mailchi.mp/pharo/pharo-newsletter-december-2018 >> > > I am looking for content for the January newsletter. > > e.g. if you know a library that has not yet been listed, or just one where some new version has been released. > Or people could write âlargerâ articles (a paragraph or two) about their favourite things. > > Or maybe there are videos that would be nice to be added to the newsletter? > > Marcus

Dec. 13, 2018

Re: [Pharo-users] [ANN] Pharo Newsletter December 2018

by Sean P. DeNigris

Marcus Denker-4 wrote > Last we the Pharo Newsletter December 2018 was send. Wow, RMapViewer looks really cool :) Is the Units framework mentioned the same one from SqS? ----- Cheers, Sean -- Sent from: http://forum.world.st/Pharo-Smalltalk-Users-f1310670.html

Dec. 13, 2018

Re: [Pharo-users] Problem loading seaside with Zinc

by Paul DeBruicker

can also do the code rewriting described here as a stopgap measure: http://forum.world.st/stfx-eu-down-tp5090610p5090624.html Vitor Medina Cruz wrote > Hello, > > Here is my baseline > > BaselineOfEmployees >> baseline: spec [ > > > <baseline> > spec for: #'common' do: [ > spec blessing: #'baseline'; > baseline: 'Seaside3' with: [ spec repository: > 'github://SeasideSt/Seaside:v3.2.2/repository' ]; > baseline: 'Magritte' with: [ spec repository: > 'github://magritte-metamodel/magritte:v3.5.3/source'; > loads: #(Seaside > Core) ]; > > package: 'Employees' with: [ spec requires: #('Seaside3' > 'Magritte'). ]; > > group: 'default' with: #('core'); > group: 'core' with: #('Employees') > ]. > ] > > Then running: > > Metacello new baseline: 'Employees'; > repository: 'tonel:///opt/pharo/employeesSource/pharo/'; > ignoreImage; > onConflict: [ :ex | ex useIncoming ]; > onWarning: [ :ex | Transcript crShow: ex ]; > silently; > load: #(core). > > In the end I got an error: > > Fetched -> Seaside-Tests-Welcome-JohanBrichau.1497083460 --- > https://github.com/SeasideSt/Seaside.git[73adecc] --- > /opt/pharo/pharo-local/iceberg/SeasideSt/Seaside/repository > (Libgit)Project: Zinc Project > stable...RETRY->ConfigurationOfZincHTTPComponents...RETRY->ConfigurationOfZincHTTPComponents...FAILED->ConfigurationOfZincHTTPComponents[31mCould > not resolve: ConfigurationOfZincHTTPComponents > [ConfigurationOfZincHTTPComponents] in > /opt/pharo/pharo-local/package-cache > *http://mc.stfx.eu/ZincHTTPComponents > <http://mc.stfx.eu/ZincHTTPComponents>*[0mMetacelloEnsureFetchingMCSpecLoader(MetacelloCommonMCSpecLoader)>>retryingResolvePackageSpecReferences:gofer:[ > | references nearestReference cachedReference externalReference mcVersion > loadedVersionInfos |cachedReference := nil.packageSpec > searchCacheRepositoryForPackage: [ "check to see if mcz file is already in > cacheRepository" cachedReference := self resolvePackageSpec: packageSpec > cachedGofer: self loaderPolicy cacheGofer. (cachedReference ~~ nil and: [ > packageSpec getFile ~~ nil ]) ifTrue: [ cachedReference name = packageSpec > file ifTrue: [ "exact match between packageSpec file and cache" ^ self > scheduleFetchFor: packageSpec cachedReference: cachedReference ] ] > ].references := self retryingResolvePackageSpecReferences: packageSpec > gofer: gofer. "look up mcz file"nearestReference := references last > > > Note it is trying to reach *http://mc.stfx.eu/ZincHTTPComponents > <http://mc.stfx.eu/ZincHTTPComponents>*, which don't seems to be > hosting > Zinc anymore. Since these are transitive dependencies, I tried to define > Zinc in the root of my baseline, as I did with seaside (magritte baseline > had some problem with it), but the same error happened. Is there any > workaround for this problem? > > Regards, > Vitor -- Sent from: http://forum.world.st/Pharo-Smalltalk-Users-f1310670.html

Dec. 12, 2018