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

Direct messaging using Slack

by Alexandre Bergel

Hi! In case you are not aware of it, a great number of the Pharo, Moose, and Roassal community members are on Slack. Slack is a very cool messaging platform. If you want to chat with developers and get instantaneous help on a topic, then Slack simply rocks. More info on: http://pharo.org/community Cheers, Alexandre -- _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: Alexandre Bergel http://www.bergel.eu ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.

Nov. 19, 2015

Re: [Pharo-dev] Preparing for the (already here) future world of multi-core PCs

by Dimitris Chloupis

Now days the real CPU is the GPU. Most of the acceleration and parrarel computing happens in there. Eventhough my fellow pharo collogues are mostly purists that prefer everything implemented in Pharo I am a parasitist preffering to leach off any technology I can find, my favourite choice being python. Its possible to use my socket bridge to bridge pharo with python in a few dozen lines of code to access to 2 python libraries, multiprocessing and foremost PyCuda so you can both take advantage the multiple cores of a CPU and an GPU though GPU should be your prime focus if you want considerably speed ups to your code. So none stops a pharoers making code that runs on multiple CPU and GPUs , all it takes is a bit of imagination and an open mind. Sometimes the quickest way to reach a goal is not a straight line . The truth however is even I that works with 3d graphics I find myself barely needing 1 CPU core, so you need to first make sure that you need such a feature before implementing it. On Thu, Nov 19, 2015 at 4:19 AM EuanM <euanmee(a)gmail.com> wrote: > I agree that parallelism and concurrency are different things. > > For easy parallelism : > - a multi-VM version of something like PharoLauncher. > - tools for tailor each VM + Image > - tools to manage and track each running VM + Image. > - tolls to try to mak sure that different running VMs and images have > their own core, and that VMs can be closed down so that VMs never are > sharing (unless you want them to) > > All we need to do is make sure that the VM and image don't have > baked-in assumptions about being the only VM on the physical machine > baked in to each VM and the Smalltalk above it.) > > The task for the end-user app developers then becomes seeing what > bundles of functionality are sufficiently decoupled from the others > that they can be set adrift in their own VM. > > (I'm even starting to think of reinventing the multi-user server. A > single 16 core PC, with 16 4-core Raspberry Pis connected, and each > RPi's user taking control of a single app on a single VM single core > on the PC). > > (Then you could have a quite a decent network for $140 per workstation > (PiCEED + keyboard + mouse) , and one fast laptop as the central > server Cheap, and easy to transport to remote locations. (But I > digress) > > > Concurrency is a deeper and harder problem for a Smalltalk, I agree. > > > > On 18 November 2015 at 23:36, Michael J. Forster <mike(a)sharedlogic.ca> > wrote: > > This is no small issue, nor a single one. > > > > Be careful to distinguish concurrency from parallelism. See Rob Pike's > > talk/slides on this. > > > > Note that many systems can benefit from a concurrent design--even on a > > single core, where no parallel execution is possible--because the > problem is > > inherently concurrent; note how effectively shared-nothing approaches > with > > Erlang, Go, or Scala/Akka have been used in such domains. > > > > In other domains, parallelism--especially nested data parallelism--is > more > > important than concurrency (see > > > http://yosefk.com/blog/parallelism-and-concurrency-need-different-tools.html > ). > > The Haskell and other communities have been exploring this with shared > > memory approaches like STM. > > > > So, to better use the slowing but increasing number of cores in today's > > hardware, we need to understand the many very different use cases and > > approaches. The lowest-hanging fruit in the Pharo/Squeak world might be > to > > rework the vm and process scheduler to support massive numbers of lighter > > weight processes similar to Erlang. Check out Andrew Blacks "Erlang in > > Pharo", http://www.squeaksource.com/Erlang.html. > > > > Best, > > > > Mike > > > > > > > > > > -- > > View this message in context: > http://forum.world.st/Preparing-for-the-already-here-future-world-of-multi-… > > Sent from the Pharo Smalltalk Developers mailing list archive at > Nabble.com. > > > >

Nov. 19, 2015

Re: [Pharo-dev] News about FastTable

by Alexandre Bergel

Hi Cyril! Excellent work! I have trouble executing: exampleTreeWithAllItemsFilter exampleTreeWithAllItemsSearch The filter with FTTableMorph cannot have a condition? Something like each > 50 and: [ each < 100 ] ? Alexandre -- _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: Alexandre Bergel http://www.bergel.eu ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. > On Nov 18, 2015, at 6:53 AM, Ferlicot D. Cyril <cyril.ferlicot(a)gmail.com> wrote: > > Hi everyone! > > I just wanted to say that I added a first version of Tree in FastTable > and I corrected some bugs. > > If someone want to try it and give some feedback you can see some > examples on the latest Pharo 5 in FTExamples class. > > With Usman we also worked on the glamour version of FastTable begun by > Esteban. > You can see some examples of FastList, FastTable, FastOutlineList and > FastTree in the GLMExamplesBrowser after executing this: > > (ConfigurationOfFastTable project version: #stable) load: #glamour > > Every feedback is welcome :) > > -- > Cyril Ferlicot (cyril.ferlicot(a)synectique.eu) > > http://www.synectique.eu > > 165 Avenue Bretagne > Lille 59000 France >

Nov. 19, 2015

Re: [Pharo-dev] Hapao2>>addMethodEdges:scope:view MNU

by Alexandre Bergel

[ For people not aware of Hapao: Hapao is a visual test coverage tool for Pharo. More info on: http://bergel.eu/download/papers/Berg12c-HapaoSCP.pdf ] Hi Sheridan, How did you load Hapao? Roassal seems to be missing. The following should work: Gofer it smalltalkhubUser: 'ObjectProfile' project: 'Spy2'; configurationOf: 'Spy2'; loadStable. I can get the coverage of the class String. Here is the result by the way: https://dl.dropboxusercontent.com/u/31543901/TMP/StringCoverage/index.html Interesting to see that, despite all the unit test that uses String, String is not even 50% covered. I have attached the html coverage report to this email. Cheers, Alexandre -- _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: Alexandre Bergel http://www.bergel.eu ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. > On Nov 18, 2015, at 9:39 PM, sheridan(a)sheridan-mahoney.com wrote: > > Hi, > > I'm trying to run Hapao2 on the String class and its subclasses, and encountered an error. > > I downloaded the latest image (50452) for Windows, loaded Roassal2 and S2py, and tried to run the following code: > > (Hapao2 runTestsForClasses: String withAllSubclasses , StringTest withAllSubclasses) visualize. > > This worked in previous versions. This time I got an MNU error in > > Hapao2 >> addMethodEdges:scope:view: > > It looks like RTEdgeBuilder isn't loaded. > > Can someone tell me if I am missing a step? > > Many thanks, > > Sheridan >

Nov. 19, 2015

Re: [Pharo-dev] News about FastTable

by Thierry Goubier

2015-11-19 12:41 GMT+01:00 Stephan Eggermont <stephan(a)stack.nl>: > On 18-11-15 22:01, Thierry Goubier wrote: > >> Normal GUI behavior is that selection switch to collapsed item if >> selection was previously among that item children. >> > > Using a different icon to show partial selection, and perhaps a > contextmenu to do 'select all subnodes, deselect all subnodes'? Maybe. I haven't checked if the rule above holds for multi-selection. I checked: In multi-selection, if main/first selection is among the item children, that selection switch to parent item upon collapse all all secondary selections among the children are deselected. (Reference: Thunderbird 38.3.0) I had difficulties finding the right decomposition to express that between the model and the view, and I'm not entirely satisfied with the solution. Maybe it could be left to the application using the widget, but, even at that level, writing it properly is hard (and it would be good that the toolkit provides for proper behavior to make the life of the application designer easier). Thierry > > > Stephan > > > >

Nov. 19, 2015

Re: [Pharo-dev] News about FastTable

by Nicolai Hess

2015-11-18 10:53 GMT+01:00 Ferlicot D. Cyril <cyril.ferlicot(a)gmail.com>: > Hi everyone! > > I just wanted to say that I added a first version of Tree in FastTable > and I corrected some bugs. > > If someone want to try it and give some feedback you can see some > examples on the latest Pharo 5 in FTExamples class. > > With Usman we also worked on the glamour version of FastTable begun by > Esteban. > You can see some examples of FastList, FastTable, FastOutlineList and > FastTree in the GLMExamplesBrowser after executing this: > > (ConfigurationOfFastTable project version: #stable) load: #glamour > > Every feedback is welcome :) > Looks great. It is really fast. And it comes with class comments ( <- really important ) > > -- > Cyril Ferlicot (cyril.ferlicot(a)synectique.eu) > > http://www.synectique.eu > > 165 Avenue Bretagne > Lille 59000 France > >

Nov. 19, 2015

Re: [Pharo-dev] News about FastTable

by Stephan Eggermont

On 18-11-15 22:01, Thierry Goubier wrote: > Normal GUI behavior is that selection switch to collapsed item if > selection was previously among that item children. Using a different icon to show partial selection, and perhaps a contextmenu to do 'select all subnodes, deselect all subnodes'? Stephan

Nov. 19, 2015

Re: [Pharo-dev] A Question about Metacello's #repository:

by Sven Van Caekenberghe

> On 18 Nov 2015, at 00:30, Dale Henrichs <dale.henrichs(a)gemtalksystems.com> wrote: > > > > On 11/17/15 1:34 PM, Sven Van Caekenberghe wrote: >> Hi Dale, >> >> Thanks for the elaborate answer. >> >> Yes, I think that BaselineOf is what I was looking for, I will experiment with it. >> >> I do suspect that a Baseline can only be loaded using the Metacello API, not using the Pharo config handler which seems to fall back to Gofer. >> > > The Metacello Scripting API should be used for all loads ... I know that does not always happen in Pharo, but that doesn't change the fact that the Metacello API _should_ be used:) Yes, we should have a look at that. > Also make sure that you've loaded the latest Metacello into Pharo[1], because there are definitely bugs that have been fixed In Metacello that haven't made into Pharo ... I just don't know when the last sync to place, the baseline stuff seemed to work OK. > Dale > > [1] https://github.com/dalehenrich/metacello-work#pharo30-pharo40-and-pharo50 >

Nov. 19, 2015

Re: [Pharo-dev] A Question about Metacello's #repository:

by Sven Van Caekenberghe

> On 17 Nov 2015, at 22:34, Sven Van Caekenberghe <sven(a)stfx.eu> wrote: > > Hi Dale, > > Thanks for the elaborate answer. > > Yes, I think that BaselineOf is what I was looking for, I will experiment with it. I tried the following and it worked perfectly: $ git clone https://github.com/dalehenrich/sample.git And the in a Pharo 4 image: | repo | repo := (FileLocator imageDirectory / 'sample' / 'repository') resolve asUrl scheme: #filetree; printString. Metacello new baseline: 'Sample'; repository: repo; load. Like you said this is limited to the top level project that you load, but that would seem like the most common use case. > I do suspect that a Baseline can only be loaded using the Metacello API, not using the Pharo config handler which seems to fall back to Gofer. > > Sven > >> On 17 Nov 2015, at 20:08, Dale Henrichs <dale.henrichs(a)gemtalksystems.com> wrote: >> >> Sven, >> >> I've read through the follow-on messages to this post and I'm not quite sure I know what you are asking, so I will just reply here ... >> >> On 11/17/2015 06:12 AM, Sven Van Caekenberghe wrote: >>> Hi, >>> >>> I have a question about the way repositories are specified in Metacello (i.e. by means on #repository:). >>> >>> Say I use FileTree where the files come from a classic VCS or just from some archive. Using ZeroConf I can get an image+vm, etc. I can use the config command line handler to load the configuration from a local file directory (I haven't tried this yet, but I assume no scheme results in a MCDirectoryRepository). So far so good. >> Okay I'll put in some examples here to make sure that we're on the same page. First here's a linek[1] to the documentation for the various schemes that can be used as Metacello repository descriptions (the arg to the #repository: message). >> >> If you want to use FileTree for your local repository then the repository description has the form: >> >> filtree://<full path to directory> >> >> For an MCDirectoryRepository, the repository description has the form: >> >> client://<full path to directory> >> >> or >> >> <full path to directory> >> >> [1] https://github.com/dalehenrich/metacello-work/blob/master/docs/MetacelloScr… >>> >>> The question is, what do I put in as argument to #repository: in my baseline ? >>> >>> baseline1: spec >>> <version: '1-baseline'> >>> >>> spec for: #common do: [ >>> spec >>> blessing: #baseline; >>> repository: 'http://mc.stfx.eu/Neo'; >>> package: 'Neo-Console-Core'; >>> group: 'default' with: #('Neo-Console-Core'); >>> group: 'Core' with: #('Neo-Console-Core') ] >> Okay, the baseline1: above is from your ConfigurationOfNeo configuration, correct? >> >> Since you are referencing `http://mc.stfx.eu/Neo`, I assume that the primary repository is on http://mc.stfx.eu, correct? >> >> When you ask "what do I put in as argument to #repository: in my baseline?", I am confused ... >> >> Perhaps my assumptions are incorrect? I assume that the above came from a ConfigurationOfNeo and that the primary location is `http://mc.stfx.eu/Neo`. >>> Obviously not a remote http repository. But also not an absolute path, since I do not know where the user placed the files. >> >> If the Neo project's primary repository is `http://mc.stfx.eu/Neo`, then how does the user get a local copy of the repository - filetree:// or client://? >> >>> Ideally, it should be possible to leave out repository and let it resolve to wherever the configuration was located. Can that be done ? Or is there another way to do this ? >>> >> >> The BaselineOf was invented to solve this particular problem and I will talk about github/git scenario, since the question of "how does the user get a local copy of the repository?" has a straightforward answer.... >> >> To start with let's use the Sample project[2] as an example and lets look how the project is specified in the BaselineOfSample>>baseline:[3]: >> >> baseline: spec >> <baseline> >> spec >> for: #'common' >> do: [ >> spec package: 'Sample-Core'. >> spec package: 'Sample-Tests' with: [ spec requires: 'Sample-Core' ]. >> spec >> group: 'default' with: #('Core'); >> group: 'Core' with: #('Sample-Core'); >> group: 'Tests' with: #('Sample-Tests') ] >> >> Note that there is no repository: specification .... which makes me think that this is the answer to the question that you are asking: >>> It should be possible to leave out repository and let it resolve to wherever the configuration was located. >> >> A BaselineOf sets the internal repository: to the location where the BaselineOfSample spec was loaded from. So you use the following expression to load a BaselineOf into your image: >> >> Metacello new >> baseline: 'Sample'; >> repository: 'filetree://<fulll-path-to-directory>; >> load. >> >> Given that the user knows where she's cloned the https://github.com/dalehenrich/sample.git repository, then it is reasonable for the developer to know how to load the project from that location ... >> >> However, this is not a complete answer to the the overall problem ... >> >> For example a corollary to your question is: >> >>> How do I ensure that other projects that reference the Sample project load from my local clone? >> >> So If you have a reference to the Sample project in another project, the reference is likely to look like the following: >> >> spec >> baseline: 'Sample'; >> repository: 'github://dalehenrich/sample:master/repository'; >> loads: 'Core'. >> >> Here we've got a reference to the github repo, but we'd like to use the local filetree:// clone. There is a Metacello Scripting API command that you can use to inform Metacello that the local filetree:// clone should be used: >> >> Metacello new >> baseline: 'Sample'; >> repository: 'filetree://<fulll-path-to-directory>; >> lock. >> >> Again we're using the full-path, but the developer is the one that knows a) the path to the local clone and b) whether or not the local clone should be used ... >> >> Now I will admit that having to remember and keep track of making sure that the `lock` expression get's run in each image, can be a bit difficult .... >> >> In tODE I've addressed this issue by creating a project entry object that is shared by "all images" and specifies not only the particulars of project locking but includes additional information about what how you want a particlar project loaded in "all images" (i.e., a specification for the `loads` expression as well) ... >> >> So I think I've addressed your question, but maybe I've completely missed your point:) >> >> Dale >> >> [2] https://github.com/dalehenrich/sample >> [3] https://github.com/dalehenrich/sample/blob/master/repository/BaselineOfSamp… >> >

Nov. 19, 2015

Re: [Pharo-dev] EyeInspector with subclasses of ProtoObject

by Stephan Eggermont

On 19-11-15 10:33, Guillermo Polito wrote: > I know we can :) and I did some prototypes already as part of my thesis. > > The question is if that is a design we *want* or not. 1 Is it an improvement on the current situation? 2 Is it blocking an improvement direction we want to take? 3 Can it be tested quickly? Stephan

Nov. 19, 2015