- Pharo-dev - lists.pharo.org

Red in Dark theme
by stepharo March 5, 2016

March 5, 2016

Just that you know (because other people may have the same problem than my view :) I cannot read error in red in the dark theme. Probably because the hue of the red is not strong enough. Stef

1 0

feedback about 50628
by stepharo March 5, 2016

March 5, 2016

Hi I tried to restart shooting videos for the mooc. - When I use Spotter I get a rubric error (subscript ....) Doru did you tell me that it was fixed? or did I got it wrong? - then DisplayScreen hostWindowSize: 1920@1080 is crashing the VM (and I'm not resizing the window during package load) from time to time stef

2 1

menu items shifted around
by stepharo March 5, 2016

March 5, 2016

Hi I think that we have a bug in menu with rubric. I spotted this with Frank last summer and it is visible in profstef. When you bring the profstef menu the do it menu item is moved out of the list and put at the end of the items. In the past we got the same at the beginning of the list. Stef

1 0

CollectionValueHolder ValueAdded ValueRemoved
by Peter UhnÃ¡k March 5, 2016

March 5, 2016

This is something that was discussed some time ago but never resolved. To summarize: When working with CollectionValueHolder one can only see whether the collection as a whole changed and not how, e.g. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ col := OrderedCollection new asValueHolder. col add: 1. col add: 2. col add: 3. col add: 4. col whenChangedDo: [ :c :changedValue | 1halt. ]. "col = #(1 2 3 4)" col add: 4. "whenChangedDo: c = col, changedValue = 4" col remove: 4. "whenChangedDo: c = col, changedValue = 4" ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ In the last two cases the changedValue is the same so I can't tell from the block if the value was added or removed. My proposition: * introduce (at least) two more Announcements: ValueAdded and ValueRemoved * add messages #whenAddedDo:/#whenChangedDo: * maybe add #when:do: so it can be used as an announcable object (valueHolder when: ValueSomething do: [ :ann | ... ]) The ValueAdded/ValueRemoved would be fired when an item is added, resp. removed. Thierry also proposed to add an index (I'd make it optional). ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ col whenAddedDo: [ :col :addedValue :optionalIndex | "here we know for sure that value was added" ]. col whenRemovedDo: [ :col :addedValue :optionalIndex | "here we know for sure that value was removed" ] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Finally Thierry mentioned a way to handle situations where values are replaced (either single one or multiple ones). As this is edge use-case and the current behavior is pretty much random I wouldn't address the replacement unless a defined use case emerges. Any thoughts, Peter

2 1

Fwd: Re: Call about Numerical Methods in Pharo :)
by stepharo March 5, 2016

March 5, 2016

-------- Message transfÃ©rÃ© -------- Sujet : Re: Call about Numerical Methods in Pharo :) Date : Wed, 02 Mar 2016 22:58:37 +0100 De : Souissi Sami <Sami.Souissi(a)univ-lille1.fr> Pour : stepharo <stepharo(a)free.fr> Copie Ã : Pharo Development List <pharo-dev(a)lists.pharo.org>, Any question about pharo is welcome <pharo-users(a)lists.pharo.org>, Didier Besset <didier(a)ieee.org> Dear StÃ©phane, dear all Many thanks StÃ©phane for this excellent initiative, Although I am a 'new' memer in pharo 'world' but I will be interested to see how can we enrich numerically the tool and hope that we can be able to have more generators for probability distributions (random numbers, Gamma, Log-Normal, etc) as well as some algorithms used to generale different random walks that can be used in several agent-based models (fractal; multifractal, LÃ©vy etc) I have also good experience with time-series analysis and other approaches (Bayesian), Happy to see that Serge started such development, Looking to have more interactions on these numerical issues and looking to meet you in the near future Best wishes Sami Sami Souissi, Professor UniversitÃ© Lille Sciences & Technologies UMR CNR Le Mercredi 2 Mars 2016 14:33 CET, stepharo <stepharo(a)free.fr> a Ã©crit: > Hi guys > > I met Didier Besset and we had a great hacking session and discussions > with Serge Stinckwich. > Didier would like to help Pharo and the numerical part of it. ***Big > thanks*** Didier. > > We would like to do several things: > > - Work on "Hows to" > The numerical methods in Pharo is good but the gap between us and > the math is too large :) > so the idea is to have a series of "how to ..." > - histomgram (simple, based on distribution) > > - Improve the SciPharo/NumPha (previously SciSmalltalk) library > This morning we started to implement a > ComponentPrincipalDecomposition by combining two of the > objects available in SciPharo. > Then we started to enhance the distributions to make sure that > we can plug other distribution for having > controlled random number. > > - Do a public call to know what is missing for you: this is this > mail :) > Didier would like to work on concrete cases. I love that attitude > So tell us :) > > Hernan ?? > Alex: ?? pvalue? better distribution? > Vincent: covariance? CPA? > Philippe: times series > Serge R frames? > Sami: Better random number and various distributions? > > - Organise a two day lectures with practices on concrete case in > September with a recording session. > Either at IRD Bondy or Lille. > > - Make sure that the Numerical Method book will get on lulu :) with > a better cover and title :) > > > Stef

1 0

[Pharo-users] Numerical Methods in Pharo without FFI ?
by Nicolas Cellier March 5, 2016

March 5, 2016

> Can you add enhancement issues in the SciSmalltalk issue tracker : > https://github.com/SergeStinckwich/SciSmalltalk/issues > in order to keep them in mind ? > > We are trying to create a pure Smalltalk library, so we try to avoid > FFI as much as possible :-) > > snip... > -- > Serge Stinckwich > UCBN & UMI UMMISCO 209 (IRD/UPMC) > Every DSL ends up being Smalltalk > http://www.doesnotunderstand.org/ > > I agree with using Smalltalk when we can, and avoid FFI, except maybe one thing... In Smalltalk, we could completely eliminate the Float primitives and emulate them with Integer arithmetic. But we don't. IMO we should adopt same pragmatic approach with basic operations on vectors/matrices, and maybe also for more involved algebra. The guys of LAPACK decided to use a "primitive" layer BLAS because the FORTRAN compiler was too naive to compete with specially crafted optimizations like that of MKL. It does not make sense to compete in this area, our language is not taylored for number crunching whatever the JIT efforts we can make, So let's use some kind of numeric primitives. That's how Matlab, NumPy, R or evry other interpreter would do it, don't they? It may be interesting to have some fallback code written in Smalltalk, but it ain't gonna scale. So don't forget that Smallapack could leverage number crunching, that's what it's made for...

2 1

Re: [Pharo-dev] Call about Numerical Methods in Pharo :)
by Eliot Miranda March 5, 2016

March 5, 2016

On Thu, Mar 3, 2016 at 11:29 PM, Benoit St-Jean via Pharo-dev < pharo-dev(a)lists.pharo.org> wrote: > > > ---------- Forwarded message ---------- > From: Benoit St-Jean <bstjean(a)yahoo.com> > To: Pharo Development List <pharo-dev(a)lists.pharo.org> > Cc: > Date: Fri, 4 Mar 2016 07:27:45 +0000 (UTC) > Subject: Re: [Pharo-dev] Call about Numerical Methods in Pharo :) > Why just don't we honor Didier H. Besset and leave it as it was, Numerical > Methods in Smalltalk? > > That way, regardless of your favorite Smalltalk implementation, there will > always be a "Numerical Methods in Smalltalk" package somewhere! > > The same way there's no GlorpPharo : it's Glorp everywhere! > +1 > > ----------------- > BenoÃ®t St-Jean > Yahoo! Messenger: bstjean > Twitter: @BenLeChialeux > Pinterest: benoitstjean > IRC: lamneth > Blogue: endormitoire.wordpress.com > "A standpoint is an intellectual horizon of radius zero". (A. Einstein) > > > ------------------------------ > *From:* Serge Stinckwich <serge.stinckwich(a)gmail.com> > *To:* Pharo Development List <pharo-dev(a)lists.pharo.org> > *Sent:* Wednesday, March 2, 2016 1:43 PM > *Subject:* Re: [Pharo-dev] Call about Numerical Methods in Pharo :) > > On Wed, Mar 2, 2016 at 7:11 PM, Nicolas Cellier > <nicolas.cellier.aka.nice(a)gmail.com> wrote: > > SciPharo? Not so great news from my POV. > > What is so much pharo specific in this library? > > Is Smalltalk scientific community large enough for yet another split? > > We can think of a more neutral name. > Any idea for a new name ? > > Regards, > > -- > Serge Stinckwich > UCBN & UMI UMMISCO 209 (IRD/UPMC) > Every DSL ends up being Smalltalk > http://www.doesnotunderstand.org/ > > > > > -- _,,,^..^,,,_ best, Eliot

1 0

Re: [Pharo-dev] Call about Numerical Methods in Pharo :)
by stepharo March 4, 2016

March 4, 2016

We are changing the code and removing mistakes (except if you prefer to have double initialization for example). We are maintaining the book and we will produce a new. So focus on the good energy and our world will get much better. Le 4/3/16 08:32, Benoit St-Jean via Pharo-dev a Ã©crit :

1 0

Point#>>scaleTo:
by Aliaksei Syrel March 4, 2016

March 4, 2016

> > *scaleTo: anExtent* > "Return a Point scalefactor for shrinking a thumbnail of the receiver's > extent to fit within anExtent." 5@5 scaleTo: 10@10 *=> 2.0@2.0* 10@10 scaleTo: 5@5 *=> 0.5@0.5* 10@10 scaleTo: 5@10 *=> 1.0@1.0 (wat)* 5@10 scaleTo: 10@0 *=> **0.6666666666666666@0.0 (wat?)* 5@0 scaleTo: 10@0 *=> ZeroDivide (still, wat?)* The idea of the method is nice, especially result as float point, so it can be nicely used for UI. I just don't understand why by contract it does not work with zeros. Imagine we have two UI elements of arbitrary extent, meaning that 5@0 is completely valid extent. Then imagine we have another UI element with extent 10@0. I would expect to have at least any valid scaling factor that would give correct result if applied. For example for corner cases result could be: 5@10 scaleTo: 10@0 => 2.0@0.0 (to scale 10 to 0 we need to multiply by 0) 5@0 scaleTo: 10@10 => ZeroDivide 5@0 scaleTo: 10@0 => 2.0@1.0 (when scaling factor is 1.0 it means that scaled arguments are equal, so the same rule can be applied when scaling 0 to 0. Please realize, scaling is not a division!) Otherwise this method is completely useless (no senders, btw) *So, let's discuss. :)* Cheers, Alex

2 2

Space low on CI?
by Max Leske March 4, 2016

March 4, 2016

I minimized the storage size required for the Fuel development builds because they kept erroring out with the message "pharo-vm/PharoV40.sources: write error (disk full?). Continue? (y/n/^C) â Now that Iâve freed up some space the builds work again, the question is for how long. Is the space that low on the CI? Cheers, Max

3 3