July 2014 - Pharo-dev - lists.pharo.org

Re: [Pharo-dev] [Article] Elegant Pharo Code

by J.F. Rick

I've often wondered if there might not be a better way to write while loops that output a product, similar to enumeration sum: and count:. | sum count index | sum := 0. count := 0. index := 1. [ index := index + 1. index isPrime ifTrue: [ count := count + 1. sum := sum + index ]. count = 64 ] whileFalse. vs something like | count sum | count := 0. sum := 1 until: [ count = 64 ] sum: [ :index | index isPrime ifTrue: [ count := count + 1. index ] ifFalse: [ 0 ] ]. or | count sum | count := 0. sum := (1 until: [ count = 64 ]) sum: [ :index | index isPrime ifTrue: [ count := count + 1. index ] ifFalse: [ 0 ] ]. On Sun, Jul 13, 2014 at 3:05 PM, Sergi Reyner <sergi.reyner(a)gmail.com> wrote: > It may have been mentioned already, but number 7 is wrong, is states: > > "7. Sum of the first 64 primes" > > when it actually is: > > "7. Sum of the primes not larger than 64" > > > Cheers, > Sergi > > > 2014-07-12 17:02 GMT+01:00 Sven Van Caekenberghe <sven(a)stfx.eu>: > > >> On 08 Jul 2014, at 21:53, kilon alios <kilon.alios(a)gmail.com> wrote: >> >> > Ruby and Pearl look very ugly to me. >> > >> > Programming languages are primarily personal preference. >> > >> > Pharo advantages to me at least are far more than live image, its live >> coding which by the way is different from live image, the IDE tools, the >> factor that is all Pharo objects even when you do assembly coding inside >> Pharo , the libraries of course and last but not least the community. >> > >> > I could name also thousands more advantages that I have found in the >> implementation that I really like as well others I don't. Devil , as they >> say, is in the details. Details are everything , generalisations have >> little meaning. >> > >> > Lovely article Sven , keep them coming. I think Pharo would definitely >> benefit from a cookbook working as a database where pharo coders can use to >> find example code the easy way. >> >> Thanks, Kilon. >> >> > But even articles like this can help a lot, I just wish that there was >> a Pharo wiki to keep these links and see them buried in Pharo news. >> >> Yes, I think we need more/better places to link everything together. A >> curated reading list maybe. >> >> > PS: Sven I am making my own Pharo book called Pharo Universe where I >> want to put things that are not part of PBE or PFTE which can be find here >> https://github.com/kilon/Pharo-Universe and here >> https://ci.inria.fr/pharo-contribution/job/PharoUniverse/ . May I have >> permission to use your articles as chapters for my book ? I will fully >> credit you and link back to your blog posts of course :) >> >> I know you are doing a lot of effort with respect to documentation and >> that is really great and important. I am not sure that copying articles >> that were not meant to be part of a book or larger whole is a good >> solution, for either party, it will give too much repetition and not a lot >> of coherence. >> >> > On Tue, Jul 8, 2014 at 10:33 PM, Paul Davidowitz <pdavidow(a)fastmail.fm> >> wrote: >> > Very nice, except all this could be done in Ruby, also elegantly. >> > The Smalltalk advantage is the live image, and that's where the focus >> > should be. >> > >> > >> >> >> > -- Jochen "Jeff" Rick, Ph.D. http://www.je77.com/ Skype ID: jochenrick

July 14, 2014

Pharo 40 June meeting

by stepharo

Hi guys We run the first checkpoint meeting around meeting with the team at rmod and we analysed why we did not get public sprints and regular hangouts. We were tired I guessed. Now we will have a public calendar with the next 6 months of planified events. The calendar will be (may be it is already) added to the web site. Since the team is reorganizing itself and its priority here is a report of our analysis for the first iteration. If you have resources and you want to get an impact please join and let us know. Pharo 40 features list ====================== - Modularisation is under way - Tools - Morphic - Polymorph package structure is slowly getting cleaner - Infrastructure - Android (Missing management events and graphics) - Raspberry (FFI works, vm works, OS process) - Spur should be finished this week - found a bug in Spur related to weak structure (already fixed by Eliot) - release each image we produce to Spur format - nativeBoost is not working - large integer plugin is not working - Host window integration - OSWindow common interface - Multiple back-ends (VM, SDL2, plugin) - Ready to be integrated. - Need some class comments! - New bytecode set - integrated this morning - needs a new vm for the bytecodeset - Spur 64 bits - starting to work on it in September - Removing old compiler - reloadable - needed for VM - Spur - System Localization - usman should open a bug issue - Leading Char Removal - probably we will not do it. - Guille Core - bootstrap - next fridayâ¦ - Unloading Reloading tools - JB will help - Athens - roadmap - missing txText - TxText - Widget is there and working - Morph is working - Spec integration - Not all tool use Spec - Clean version - Igor MUST release a version and not try to make the complete system working with it. - Remove the old browser - kill the filePackage browser - We should get Nautilus or OndoBrowser browsing Ring - Spec - Should do a pass - Check the design (in particular link between adapter and morph) - Examples-driven are needed - GitSupport - filetree integration is working but not good enough - small progress. - Martin should check with Max the status of libgit2 - Need to build on top of that. - Morph and theme - we will see if we have the time - brainstorming on a solution - Make sure that we do not overuse Ring - Font with FreeType - Glamour - esteban roadmap - first bootstrap - second only a stripped down version - development outside the image - Ombu/Epicea - call for more users - ZnUrl should be used in place of URI URL - Keychain Removal - Morph class side announcer - we should look at the announcer plague - Intersect: - should be finished - Image-level resources - brainstorming prototyping - Opal there is progress! - slots <-> OPAL - subclassing and adding instance compiled - specific slots are on the way - ClassBuilder - should be revisited. Previous wish list ================== - Web site animation - catalog (meta data and ressources) - videos - Infrastructure ppa linux linux [glibc] win32 OSX iOS android (Missing management events and graphics) raspberry (works) solaris CentOS + OSX - Investigate obs openbuild service - extracting VM as dylib + GUI + console

July 14, 2014

Consortium report

by stepharo

Hi pharoers Here is the report of the consortium meeting held in late June. Sorry for the time it took to send it around. It is the same as the one we sent in the consortium mailing-list and is public. There is a draft of a roadmap for Pharo 40. Stef

July 14, 2014

Re: [Pharo-dev] Doc of mustache

by Norbert Hartl

Stef, {{# â¦} opens a section that can be used to do nested operation. If the value of number is a collection the section is evaluated for every element in the value. #(1 2 3) is a collection of size 3 and '1 2 3' is a collection of size 5 hence the output. A common use case for templates would be something like { 'number' -> { 'value' -> 1. 'value' -> 2. 'value' -> 3 } asDictionary } asDictionary Then you can use a template of {{#number}} value is {{value}} {{/number}} Norbert Am 13.07.2014 um 23:33 schrieb stepharo <stepharo(a)free.fr>: > Hi Norbert > > I'm improving the doc. I added some text on section and others.. > And on list with collection I have the following behavior and I do not understand why on string. > > | templateString context | > templateString := 'Shown. > {{#number}} > Shown too! > {{/number}}'. > context := { 'number' -> #(1 2 3) } asDictionary. > (MustacheTemplate on: templateString) value: context 'Shown. > > Shown too! > > Shown too! > > Shown too! > ' > > > > | templateString context | > templateString := 'Shown. > {{#number}} > Shown too! > {{/number}}'. > context := { 'number' -> '1 2 3' } asDictionary. > (MustacheTemplate on: templateString) value: context > 'Shown. > > Shown too! > > Shown too! > > Shown too! > > Shown too! > > Shown too! > ' > > > > > On 11/7/14 10:53, Norbert Hartl wrote: >> >> Am 11.07.2014 um 10:43 schrieb stepharo <stepharo(a)free.fr>: >> >>> Hi guys >>> >>> I would like to experiment with template for a small web project I have (my web page) and >>> I would like to try mustache. Is there any tutorial? doc? >>> I know that "jose maria aguerra" wrote a tutorial but I cannot find it anymore. >> >> http://norbert.hartl.name/blog/2013/10/03/mustache-templates-for-smalltalk/ >> >> Norbert >> >

July 13, 2014

Re: [Pharo-dev] [Pharo4] CachingMorph: anyone ever used it?

by Sean P. DeNigris

Sven Van Caekenberghe-2 wrote > before we say something has no users (in the image), we should also check > (some) external packages. Cool idea! ----- Cheers, Sean -- View this message in context: http://forum.world.st/Pharo4-CachingMorph-anyone-ever-used-it-tp4767095p476… Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com.

July 13, 2014

Re: [Pharo-dev] Doc of mustache

by stepharo

Hi Norbert I'm improving the doc. I added some text on section and others.. And on list with collection I have the following behavior and I do not understand why on string. | templateString context | templateString := 'Shown. {{#number}} Shown too! {{/number}}'. context := { 'number' -> #(1 2 3) } asDictionary. (MustacheTemplate on: templateString) value: context 'Shown. Shown too! Shown too! Shown too! ' | templateString context | templateString := 'Shown. {{#number}} Shown too! {{/number}}'. context := { 'number' -> '1 2 3' } asDictionary. (MustacheTemplate on: templateString) value: context 'Shown. Shown too! Shown too! Shown too! Shown too! Shown too! ' On 11/7/14 10:53, Norbert Hartl wrote: > > Am 11.07.2014 um 10:43 schrieb stepharo <stepharo(a)free.fr > <mailto:stepharo@free.fr>>: > >> Hi guys >> >> I would like to experiment with template for a small web project I >> have (my web page) and >> I would like to try mustache. Is there any tutorial? doc? >> I know that "jose maria aguerra" wrote a tutorial but I cannot find >> it anymore. > > http://norbert.hartl.name/blog/2013/10/03/mustache-templates-for-smalltalk/ > > Norbert >

July 13, 2014

Re: [Pharo-dev] Nautilus - Recover Lost Changes - Class Comments

by stepharo

Hi Ben > > I had SomeClassA that did not have a class comment, and so an alert > icon showed for it in Nautilus. Then I used Recover-Lost-Changes > which added a class comment for SomeClassA, but the icon did not > update. I understood that Pharo 4 will be modifying how the > ".changes" file works. Will this include modifying the mechanism of > Recover-Lost-Changes mechanism? Hence, should create a Case for this > for Pharo 4, or just for Pharo 3 ? Report for 40 and we will see later. Because the community should stress test Ombu before we push it into the arena. > > cheers -ben > >

July 13, 2014

Re: [Pharo-dev] Reporting Segmentation fault

by stepharo

Hi Yuriy Pay attention (It happens to me) that you control which VM is launched. > Iâve encountered a situation where pharo crashes, and when I run it from command line I get: "Segmentation fault: 11". Also before that some trace appears starting with highlighted: "MessageNotUnderstood: ToolDockingBarMorph>>popUpEvent:in:â. Iâve also updated my Pharo launcher and now stack trace disappeared (probably because now I have to specify some flag), but the problem remains. Should I report this to pharo issue tracker, or this is a VM problem and it should be handled in a different way? Can you give us more information: - image - VM version - a reproducible case Stef

July 13, 2014

Reporting Segmentation fault

by Yuriy Tymchuk

Iâve encountered a situation where pharo crashes, and when I run it from command line I get: "Segmentation fault: 11". Also before that some trace appears starting with highlighted: "MessageNotUnderstood: ToolDockingBarMorph>>popUpEvent:in:â. Iâve also updated my Pharo launcher and now stack trace disappeared (probably because now I have to specify some flag), but the problem remains. Should I report this to pharo issue tracker, or this is a VM problem and it should be handled in a different way? Uko

July 13, 2014

Re: [Pharo-dev] handling method/classes updartes

by Yuriy Tymchuk

Itâs fine. I was asking because I am creating entities that should stay in sync with certain methods per entity. So I wandered if an entity could register to that method. Now if I need to register to SystemAnnouncer, I have either to register each entity separately and check if announcement is regarding the method Iâm interested in. Or I can create the manager of entities which will listen to announcer and tell entities what to do. I have to figure out which approach is better. Uko On 13 Jul 2014, at 18:10, stepharo <stepharo(a)free.fr> wrote: > What you should realize is that when we introduce the SystemAnnouncer ancestor with Roel Wuyts, there way no such a thing: a system that to whim you register and notify about changes. So in VW you had to go and find all the places and patch them to get you notified. > So to me this is a huge enhancements to support the creation of great tool. Now working with a real system is complex because > a class can recategorized, a method rename, recatogorized, removed.... and there is magical solution - at least that I know > > Stef > > On 10/7/14 19:38, Yuriy Tymchuk wrote: >> Hi, >> >> is there a way to subscribe directly to a method/class, so I san know when it updates/is removed and so on? Because it looks like a pain to subscribe to system announcer and check if the method changed is the one I need. >> >> Uko >> >> > >

July 13, 2014