October 2015 - Pharo-dev - lists.pharo.org

Re: [Pharo-dev] [Vm-dev] Re: Random forest in Pharo

by Eliot Miranda

Hi Ken, I have this very paper on my desk, printed out from (alas, much) earlier in the year. Anyone who has energy to address this would be very very welcome. Writing a concurrent incremental mark-sweep that handles ephemeron is the task. Making it function both in small bursts, e.g. after scavenging, and in its own thread is I think important; a non-threaded solution can be extremely efficient in certain contexts. On Mon, Oct 19, 2015 at 12:04 PM, KenD <Ken.Dickey(a)whidbey.com> wrote: > On Mon, 19 Oct 2015 07:10:31 -0700 > Eliot Miranda <eliot.miranda(a)gmail.com> wrote: > > > - an incremental global mark-sweep GC for Spur > > One experiment with multi-cores might be interesting -- concurrent GC. > > The attached paper indicates an interesting strategy which worked OK for > SML in terms of short pause times. Multiple cores could really reduce the > total run (gc+mutate) time. > > The basic strategy seems simple enough that it would not be tons of work > to try out. > > Not that I have much time either.. 8^( > > Cheers, > KenD <Ken.Dickey(a)Whidbey.com> > -- _,,,^..^,,,_ best, Eliot

Oct. 23, 2015

Re: [Pharo-dev] powerSet

by Nicolas Cellier

2015-10-23 18:02 GMT+02:00 monty <monty2(a)programmer.net>: > The wiki entry says powersets have 2^n elements where n is the number of > elements in the original set. 2^n grows exponentially fast: > > 2^5 = 35 > 2^10 = 1024 > 2^50 = 1125899906842624 > 2^100 = 1267650600228229401496703205376 > > So if it's added, bounding it to only support small collections is not a > bad idea. 2^30 = 1,073,741,824, so 30 element collections are too big (with > 1GB address space). But <= 25 elements would be OK: 2^25 = 33,554,432 > > I actually needed something like this a few months ago, so there are > usecases. > > But as said by Stephan Eggermont, do you really need to store all the subsets together in memory, or would you just need to iterate them sequentially? If you create the subsets, then it should better be kind of lazy. > > Sent: Thursday, October 22, 2015 at 4:58 PM > > From: stepharo <stepharo(a)free.fr> > > To: "Pharo Development List" <pharo-dev(a)lists.pharo.org> > > Subject: [Pharo-dev] powerSet > > > > Hi > > > > I was programming an exercise with one of my son (well in Python.... > > arghhhhhh) > > and I end it up doing it in Pharo (I'm save now). > > > > The idea was to write one function that computes the powerset > > > > powerset(4) > > = a Set(a Set(1) a Set(1 2) a Set(3) a Set(2) a Set(1 3) a Set(2 3) > > a Set(1 2 3) a Set(4) a Set(1 4) a Set(2 4) a Set(1 2 4) a Set(3 4) a > > Set(1 3 4) a Set(2 3 4) a Set(1 2 3 4)) > > > > I did it without thinking too much in fact > > > > | s n ps | > > ps := Set new. > > > > 1 to: ((2 raisedTo: 4) -1) > > do: [ :i | > > s := Set new. > > n := 0. > > 1 to: 4 do: [ :b | > > n := n + 1. > > ((i bitAt: b) = 1 ) > > ifTrue: [ s add: n]. > > ps add: s ]]. > > ps > > > > but I wonder if we want to add it to our lib. > > > > Stef > > > > > >

Oct. 23, 2015

issue #16806: basic inspector font

by Miguel Campusano

As said in: https://pharo.fogbugz.com/f/cases/16806/Wrong-front-in-basic-inspector There is an issue about using a wrong font in basic inspector. I was looking at this and it is easily fixed if we modify the method RubScrolledTextMorph>>beForSmalltalkScripting to change the font of the text area (as made in RubScrolledTextMorph>>beForSmalltalkCode). I think the font for Smalltalk scripting should be the same as the font for a Smalltalk code. Is there a reason why is not like that in Rubric?

Oct. 23, 2015

Re: [Pharo-dev] powerSet

by monty

The wiki entry says powersets have 2^n elements where n is the number of elements in the original set. 2^n grows exponentially fast: 2^5 = 35 2^10 = 1024 2^50 = 1125899906842624 2^100 = 1267650600228229401496703205376 So if it's added, bounding it to only support small collections is not a bad idea. 2^30 = 1,073,741,824, so 30 element collections are too big (with 1GB address space). But <= 25 elements would be OK: 2^25 = 33,554,432 I actually needed something like this a few months ago, so there are usecases. > Sent: Thursday, October 22, 2015 at 4:58 PM > From: stepharo <stepharo(a)free.fr> > To: "Pharo Development List" <pharo-dev(a)lists.pharo.org> > Subject: [Pharo-dev] powerSet > > Hi > > I was programming an exercise with one of my son (well in Python.... > arghhhhhh) > and I end it up doing it in Pharo (I'm save now). > > The idea was to write one function that computes the powerset > > powerset(4) > = a Set(a Set(1) a Set(1 2) a Set(3) a Set(2) a Set(1 3) a Set(2 3) > a Set(1 2 3) a Set(4) a Set(1 4) a Set(2 4) a Set(1 2 4) a Set(3 4) a > Set(1 3 4) a Set(2 3 4) a Set(1 2 3 4)) > > I did it without thinking too much in fact > > | s n ps | > ps := Set new. > > 1 to: ((2 raisedTo: 4) -1) > do: [ :i | > s := Set new. > n := 0. > 1 to: 4 do: [ :b | > n := n + 1. > ((i bitAt: b) = 1 ) > ifTrue: [ s add: n]. > ps add: s ]]. > ps > > but I wonder if we want to add it to our lib. > > Stef > >

Oct. 23, 2015

[pharo-project/pharo-core] 2f8ec5: 50403

by GitHub

Branch: refs/heads/5.0 Home: https://github.com/pharo-project/pharo-core Commit: 2f8ec5a53adcf29c1926148aadd4075201ee0e45 https://github.com/pharo-project/pharo-core/commit/2f8ec5a53adcf29c1926148a… Author: Jenkins Build Server <board(a)pharo-project.org> Date: 2015-10-23 (Fri, 23 Oct 2015) Changed paths: M Nautilus.package/MethodWidget.class/instance/protocol/selectNextMethod.st R ScriptLoader50.package/ScriptLoader.class/instance/pharo - scripts/script50402.st A ScriptLoader50.package/ScriptLoader.class/instance/pharo - scripts/script50403.st R ScriptLoader50.package/ScriptLoader.class/instance/pharo - updates/update50402.st A ScriptLoader50.package/ScriptLoader.class/instance/pharo - updates/update50403.st M ScriptLoader50.package/ScriptLoader.class/instance/public/commentForCurrentUpdate.st M Tool-Base.package/AbstractTool.class/class/private/defineClassTemplate.st M Tool-Base.package/MethodClassifier.class/definition.st A Tool-Base.package/MethodClassifier.class/instance/classification-rules/classifyByKnownPragma_.st M Tool-Base.package/MethodClassifier.class/instance/classification/classify_.st A Tool-Base.package/MethodClassifier.class/instance/initialization/buildPragmaDictionary.st M Tool-Base.package/MethodClassifier.class/instance/initialization/initialize.st Log Message: ----------- 50403 16835 Error when removing a trait in a class being browsed https://pharo.fogbugz.com/f/cases/16835 16841 MethodClassifier should classify on pragma https://pharo.fogbugz.com/f/cases/16841 16855 Class template mentions poolDictionary https://pharo.fogbugz.com/f/cases/16855 http://files.pharo.org/image/50/50403.zip

Oct. 23, 2015

[pharo-project/pharo-core]

by GitHub

Branch: refs/tags/50403 Home: https://github.com/pharo-project/pharo-core

Oct. 23, 2015

Re: [Pharo-dev] Reading validation reports

by Stephan Eggermont

On 23/10/15 16:17, Ben Coman wrote: > On Fri, Oct 23, 2015 at 2:32 PM, Marcus Denker <marcus.denker(a)inria.fr> wrote: >> It runs twice the tests (first to get false positives) and some Lint rules. > > btw, I always wondered, whether the first run of tests to record > existing errors could be done as the last step of creating the > validation image, then it would not need to be run each time. Maybe > that might cause problems running the validation image on different > platforms? Running it once on all platforms would solve that (and show platform differences) Stephan

Oct. 23, 2015

Re: [Pharo-dev] Reading validation reports

by Ben Coman

On Fri, Oct 23, 2015 at 2:32 PM, Marcus Denker <marcus.denker(a)inria.fr> wrote: > >> On 22 Oct 2015, at 19:45, Stephan Eggermont <stephan(a)stack.nl> wrote: >> >> At the moment, most of the isues I provide a slice for time out in validation. What is exactly tested in the validation process? > > It runs twice the tests (first to get false positives) and some Lint rules. btw, I always wondered, whether the first run of tests to record existing errors could be done as the last step of creating the validation image, then it would not need to be run each time. Maybe that might cause problems running the validation image on different platforms? cheers -ben > >> Sometimes nothing is visible in the report, sometimes there are lots of deprecation warnings and other problems that look unrelated to the slice. When should they be ignored? >> > > Yes, this started last week and honestly I have not even a clue what is happening. To me it looks that it might even be caused by the underlying virtual linux machines? > > The deprecated we should fixâ¦ > > Marcus > >

Oct. 23, 2015

Re: [Pharo-dev] R: dear contributor

by Cyril Ferlicot

For Pharo contributor there is a licence agreement on the Pharo website: http://pharo.org/contribute. On Fri, Oct 23, 2015 at 3:34 PM, Lorenzo Schiavina <lorenzo(a)edor.it> wrote: > Hi Steff, > > I have not understood what I have to do. > > Ciao > > Lorenzo > > -----Messaggio originale----- > Da: Pharo-dev [mailto:pharo-dev-bounces@lists.pharo.org] Per conto di stepharo > Inviato: giovedÃ¬ 22 ottobre 2015 17:18 > A: Pharo Development List > Oggetto: [Pharo-dev] dear contributor > > please sign the license agreeement and send us a little picture for the web site ... > > Stef > > -- Cheers Cyril Ferlicot

Oct. 23, 2015

R: dear contributor

by Lorenzo Schiavina

Hi Steff, I have not understood what I have to do. Ciao Lorenzo -----Messaggio originale----- Da: Pharo-dev [mailto:pharo-dev-bounces@lists.pharo.org] Per conto di stepharo Inviato: giovedÃ¬ 22 ottobre 2015 17:18 A: Pharo Development List Oggetto: [Pharo-dev] dear contributor please sign the license agreeement and send us a little picture for the web site ... Stef

Oct. 23, 2015