- Pharo-users - lists.pharo.org

Re: [Pharo-users] Speeding-up >>#instVarNamed: in Pharo-5.0 and beyond?

by stepharo

> Hi, > > Magritte and my TagLengthValue (TLV) both use >>#instVarNamed:/>>#instVarNamed:put: to read and write from an object. I was just running >>#bench on my SMPP library and noticed that besides Spur Pharo5 is slower than Pharo3. > > > I added this to PointerLayout: > >>> # instVarIndexFor: aString ifAbsent: aBlockClosure > | idx | > idx := 1. > slotScope do: [:each | > each isVisible ifTrue: [ > each name = aString ifTrue: [^idx]. > idx := idx +1]]. > ^aBlockClosure value > > and modified ClassDescription/TClassDescription to use it: > > "protocol: instance variables" > instVarIndexFor: instVarName ifAbsent: aBlock > "Answer the index of the named instance variable." > > | index | > index := self classLayout instVarIndexFor: instVarName ifAbsent: [0]. > index = 0 ifTrue: > [^self superclass == nil > ifTrue: [aBlock value] > ifFalse: [self superclass instVarIndexFor: instVarName ifAbsent: aBlock]]. > ^self superclass == nil > ifTrue: [index] > ifFalse: [index + self superclass instSize] > > > The speed-up comes from filtering allSlots to allVisibleSlots (and creating an Array), then collecting the slot names and finally searching the name. Does it make sense to integrate such speed-ups? Yes!!!!!!!!!!!!!!!!!!!!!! if you can just improve a bit the method comments it would be really great! > > > cheers > holger > > >

Sept. 29, 2016

turning a collection (interval for example) into a stream

by stepharo

Hi I was looking at how we can turn a collection into a stream. I played with g := Generator on: [ :gen | | current | current := 0. [ gen yield: (current := current + 1) ] repeat ]. g next. and now imagine that we have (1 to: 1000) collect: [:each | {each -1 . each . each + 1}]. I was wondering how we could turn it into a stream (1 to: 1000) asStream collect: [:each | {each -1 . each . each + 1}]. and what would be the asStream method. for index based I can see that next next is something like next ^ self at: (current := current + 1) and collect: could be rewritten to use a generator. Now does anybody play with this already? Does xtream provide a way to do something in the same vein? We can have a stream of blocks. Stef

Sept. 29, 2016

trying to run Pier3 in Pharo6 with latest Pillar

by Peter H. Meadows

Am I right that PRDocumentDescription>>parserClass should be PRPillarParserMain instead of PRDocumentParser ?

Sept. 28, 2016

Re: [Pharo-users] Amazon drive API's for Pharo ?

by Paul DeBruicker

There is this https://github.com/newapplesho/aws-sdk-smalltalk I've been (slowly) working on adding the streaming-encrypted-upload-to-S3 code which I could publish soonish (this weekend?) in a fork if you want to take a look at it. CÃ©drick BÃ©ler wrote > Hi all, > > Iâve just seen that the unlimited offer that Amazon proposes for its drive > is now available in France. !! > Itâs 70â¬/year so quite cheapâ¦ > > Do people know if there are pharo packages to uses its RESTful API ? > https://developer.amazon.com/public/apis/experience/cloud-drive/content/res… > > TIA (Iâll investigate deeper shortly), > > CÃ©drik -- View this message in context: http://forum.world.st/Amazon-drive-API-s-for-Pharo-tp4917106p4917317.html Sent from the Pharo Smalltalk Users mailing list archive at Nabble.com.

Sept. 28, 2016

Speeding-up >>#instVarNamed: in Pharo-5.0 and beyond?

by Holger Freyther

Hi, Magritte and my TagLengthValue (TLV) both use >>#instVarNamed:/>>#instVarNamed:put: to read and write from an object. I was just running >>#bench on my SMPP library and noticed that besides Spur Pharo5 is slower than Pharo3. I added this to PointerLayout: >># instVarIndexFor: aString ifAbsent: aBlockClosure | idx | idx := 1. slotScope do: [:each | each isVisible ifTrue: [ each name = aString ifTrue: [^idx]. idx := idx +1]]. ^aBlockClosure value and modified ClassDescription/TClassDescription to use it: "protocol: instance variables" instVarIndexFor: instVarName ifAbsent: aBlock "Answer the index of the named instance variable." | index | index := self classLayout instVarIndexFor: instVarName ifAbsent: [0]. index = 0 ifTrue: [^self superclass == nil ifTrue: [aBlock value] ifFalse: [self superclass instVarIndexFor: instVarName ifAbsent: aBlock]]. ^self superclass == nil ifTrue: [index] ifFalse: [index + self superclass instSize] The speed-up comes from filtering allSlots to allVisibleSlots (and creating an Array), then collecting the slot names and finally searching the name. Does it make sense to integrate such speed-ups? cheers holger

Sept. 28, 2016

SPLASH'16 Amsterdam CFP: early registration ends Sept 30

by Tijs van der Storm

################################################# ACM Conference on Systems, Programming, Languages, and Applications: Software for Humanity (SPLASH'16) ################################################# Amsterdam, The Netherlands Sun 30th October - Fri 4th November , 2016 http://2016.splashcon.org https://twitter.com/splashcon https://www.facebook.com/SPLASHCon/ ** REGISTRATION ** 30 September 2016 (Early Deadline) Contact: info(a)splashcon.org http://2016.splashcon.org/attending/registration # What's happening at SPLASH? ## Keynotes - Benjamin Pierce (SPLASH) The Science of Deep Specification - Andy Ko (SPLASH) A Human View of Programming Languages - Martin Odersky (SPLASH) - Guy Steele Jr. (SPLASH-I) - Robby Findler (SLE) Redex: Lightweight Semantics Engineering - Tiark Rompf (GPCE) Lightweight Modular Staging: Generate all the things! - Simon Peyton Jones (SPLASH-I/E) The dream of a lifetime: shaping how our children learn computing - Laurence Tratt (Scala) Fine-grained language composition without a common VM - Jan Vitek (Scala) This is not a Type: Gradual typing in practice ## Workshop Keynotes - Andrew Black (NOOL) The Essence of Inheritance - Alan Blackwell (PLATEAU) How to Design a Programming Language - Felienne Hermans (DSLDI) Small, simple and smelly: What we can learn from examining end-user artifacts? - Ivano Malavolta (Mobile!) Beyond native apps: Web technologies to the rescue! - Betsy Pepels (ITSLE) Model Driven Software Engineering (MDSE) in the large - Markus Voelter (ITSLE) Lessons Learned about Language Engineering from the Development of mbeddr - Beverly Sanders (SEPS) Patterns for Parallel Programming: New and Improved! ** Conference Program ** http://2016.splashcon.org/program/program-splash-2016 ** SPLASH-I Track ** SPLASH-I is a series of invited and solicited talks that address topics relevant to the SPLASH community. Speakers are world-class experts in their field, selected and invited by the organizers. The SPLASH-I talks series is held in parallel with the rest of SPLASH during the week days. Talks are open to all attendees. A selection of confirmed talks: - Edwin Brady Type-driven Development in Idris - JÃ¼rgen Cito Using Docker Containers to Improve Reproducibility in PL/SE Research - Yvonne Coady Exploratory Analysis in Virtual Reality: The New Frontier - Adam Chlipala Rapid Development of Web Applications with Typed Metaprogramming in Ur/Web - Tudo Girba Software Environmentalism - Robert Grimm Adventures in Software Evolution - Brian Harvey Snap! Scheme Disguised as Scratch - Lennart Kats Responsive Language Tooling For Cloud-based IDEs - Ralf Laemmel The basic skill set of software language engineering - Crista Lopes Simulating Cities: The Spacetime Framework - Heather Miller Language Support for Distributed Systems - Mark Miller & Bill Tulloh The elements of decision alignment: Large programs as complex organizations - Boaz Rosenan & David Lorenz Define Your App, Donât Implement It: Building a Scalable Social Network in 45 minutes - Emmanuel Schanzer Bootstrap - Chris Seaton Truffle and Graal: Fast Programming Languages With Modest Effort - Emma SÃ¶derbergh From Tricorder to Tricium: Useful Static Analysis and the Importance of Workflow Integration - Emma Tosch Designing and Debugging Surveys with SurveyMan - Todd Veldhuizen Fast Datalog - Markus VÃ¶lter How Domain Requirements Shape Languages - Jos Warmer Making Mendix Meta Model Driven - Andy Zaidman Fact or fiction? What software analytics can do for us (developers and researchers) More information here: http://2016.splashcon.org/track/splash-2016-splash-i#program ** Research tracks - OOPSLA http://2016.splashcon.org/track/splash-2016-oopsla#event-overview - Onward! http://2016.onward-conference.org/track/onward-2016-papers#event-overview - Onward! Essays http://2016.onward-conference.org/track/onward2016-essays#program - Software Language Engineering (SLE) http://conf.researchr.org/track/sle-2016/sle-2016-papers#event-overview - Generative Programming: Concepts & Experiences (GPCE) http://conf.researchr.org/track/gpce-2016/gpce-2016-papers#event-overview - Dynamic Languages Symposium (DLS) http://conf.researchr.org/track/dls-2016/dls-2016-papers#event-overview - Scala Symposium http://conf.researchr.org/track/scala-2016/scala-2016#event-overview ** Other Events - Doctoral Symposium http://2016.splashcon.org/track/splash-2016-ds#event-overview - Programming Language Mentoring Workshop (PLMW) http://2016.splashcon.org/track/splash-2016-plmw - Student Research Competition (SRC) http://2016.splashcon.org/track/splash-2016-src - Posters http://2016.splashcon.org/track/splash-2016-posters#event-overview ** Workshops SPLASH'16 is hosting a record number of 15 workshops: - AGERE! Programming based on Actors, Agents, and Decentralized Control http://2016.splashcon.org/track/agere2016 - DSLDI: Domain-Specific Language Design and Implementation http://2016.splashcon.org/track/dsldi2016 - DSM: Domain-Specific Modeling http://2016.splashcon.org/track/dsm2016 - FOSD: Feature-oriented Software Development http://www.fosd.net/workshop2016 - ITSLE: Industry Track Software Language Engineering http://2016.splashcon.org/track/itsle2016 - LWC@SLE: Language Workbench Challenge http://2016.splashcon.org/track/lwc2016 - META http://2016.splashcon.org/track/meta2016 - Mobile! http://2016.splashcon.org/track/mobile2016 - NOOL: New Object-Oriented Languages http://2016.splashcon.org/track/nool2016 - PLATEAU: Evaluation and Usability of Programming Languages and Tools http://2016.splashcon.org/track/plateau2016 - Parsing@SLE http://2016.splashcon.org/track/parsing2016 - REBLS: Reactive and Event-based Languages & Systems http://2016.splashcon.org/track/rebls2016 - SA-MDE: Tutorial on MDD with Model Catalogue and Semantic Booster http://2016.splashcon.org/track/samde2016 - SEPS: Software Engineering for Parallel Systems http://2016.splashcon.org/track/seps2016 - VMIL: Virtual Machines and Intermediate Languages http://2016.splashcon.org/track/vmil2016 - WODA: Workshop on Dynamic Analysis http://2016.splashcon.org/track/woda2016 ## SPLASH'16 is kindly supported by the following organizations: - ACM: http://www.acm.org/ - SIGPLAN: http://www.sigplan.org/ - LogicBlox (Gold): http://www.logicblox.com/ - Universal Robots (PLMW, Gold): http://www.universal-robots.com/ - Oracle (Silver): http://www.oracle.com/index.html - TU Delft (Silver): http://tudelft.nl/ - Huawei (Bronze): http://www.huawei.com/en/ - Facebook (Bronze): https://research.facebook.com/ - IBM Research (Bronze): http://www.research.ibm.com/ - Google (Bronze): https://www.google.com - Itemis (Bronze): https://www.itemis.com/en/ - ING (Bronze): https://www.ing.nl Interested in supporting SPLASH'16? See our options here: http://2016.splashcon.org/attending/support-program.

Sept. 28, 2016

Re: [Pharo-users] bug in debugger?

by Nicolai Hess

2016-09-28 14:15 GMT+02:00 Usman Bhatti <usman.bhatti(a)gmail.com>: > Hello, > > I have a strange error in my image that results in image start taking up > system memory and crashes when it attains the max. the VM can provide on my > machine (2Gb). I am working on Pharo 4.0 + Spec debugger. > > After some digging in the image, I know that the error comes from an > assignment operation (below in bold). And the bug only occurs when the > method initialize is executed as inherited method. > The strange part is that that it does not crash when the value assigned is > slightly different e.g. "nosignature1" instead of "nosignature". > > FAMIXInvocation>>initialize > super initialize. > sender := nil. > receiver := nil. > receiverSourceCode := nil. > *signature := 'nosignature'.* > candidates := FMMultiMultivalueLink on: self opposite: > #incomingInvocations. > > The crash doesn't happen when I override initialize in the subclass > without sending initialize message to the superclass. So this report is > about a very very specific case, but it is there. I tried to analyze > further in the debugger but with announcements, etc. I wasn't able to > achieve anything. > > So, if someone's interested to have a look, I can provide the image. > I am interested. > > regards. > > Usman >

Sept. 28, 2016

Re: [Pharo-users] Essential Documentation

by stepharo

Le 27/9/16 Ã 17:17, julius a Ã©crit : > Thanks for all replies and the good advice. Grafoscopio looks impressive and > very interesting! After working more and more with Pharo I see and > understand the approach to documentation better I think. So my next steps > will be finishing the application I'm writing and documenting that, also if > I see class comments that are hard to understand I will report them, maybe > even try to improve them myself. super. I want to have a look at your application. I can do a code review if you like. > > I'm thinking a lot about how to improve the programming experience and one > thing I came up with was, that maybe showing the comment of a message while > having the autocompletion dialog open or when selecting it in the code > editor would help to grasp the frameworks faster. Since I still don't know > much about the whole system: would that be hard to implement? If would be good. The autocompletion is a bit rusty code but getting the comments of a method is easy check compiledMethod I think that there is one comment or getComment or firstcomment method. > > Best, > Julius > > > > -- > View this message in context: http://forum.world.st/Essential-Documentation-tp4916861p4917224.html > Sent from the Pharo Smalltalk Users mailing list archive at Nabble.com. > >

Sept. 28, 2016

Re: [Pharo-users] Subscript Out Of Bounds: 1 when commiting from Versionner

by stepharo

so do you think that we should back port these two big fixes? at least 18927 I think Le 28/9/16 Ã 13:48, HernÃ¡n Morales Durand a Ã©crit : > May be this issue : > https://pharo.fogbugz.com/f/cases/18684/Versionner-Error-while-commiting-a-… > > And here for the fix : > https://pharo.fogbugz.com/f/cases/18927/Komitter-throws-up-if-it-has-commit… > > HernÃ¡n > > > 2016-09-28 1:29 GMT-03:00 stepharo <stepharo(a)free.fr > <mailto:stepharo@free.fr>>: > > Yes it was fix in pharo 60 and we should port it back but I forgot > the issue number > > > > Le 27/9/16 Ã 21:00, HernÃ¡n Morales Durand a Ã©crit : > > When commiting with Versionner in Pharo 5.0 I always have to > manually override this method > > KomitStagingArea>>remotes > | result | > > result := self packages first remotes. > self packages allButFirst do: [ :each | > result := result intersection: each remotes ]. > > ^ result collect: [ :each | each koRemote ] > > to > > KomitStagingArea>>remotes > > self packages isEmpty > ifFalse: [ > | result | > result := self packages first remotes. > self packages allButFirst do: [ :each | > result := result intersection: each remotes ]. > ^ result collect: [ :each | each koRemote ] ]. > ^ Array empty. > > Screenshot attached. > Is this known issue or was this reported before? > > Hernan > > > >

Sept. 28, 2016

Re: [Pharo-users] Slots in Pharo 5

by Denis Kudriashov

2016-09-28 17:06 GMT+02:00 Jan Cada <jan(a)cada.cz>: > Hi, > > I have opened the issue #19153 > <https://pharo.fogbugz.com/f/cases/19153/Change-of-class-structure-with-Rela…> > . > Thank's

Sept. 28, 2016