- Pharo-dev - lists.pharo.org

Re: [Pharo-project] Pharo VMs should be more Pharo-like

by Esteban Lorenzano

Hi, So, I took "the bull by the horns" and hacked a jenkins job to do a Pharo app. I also created a new squeaksource project (PharoVM) to keep the specific configurations for Pharo. Right now, I made it just for Macs, but if everything is ok, I'll add a job for Linux and when windows slave is working, also for windows. Take a look: https://ci.lille.inria.fr/pharo/view/VM/job/Pharo-Mac-Cocoa/ Cheers, Esteban El 29/11/2011, a las 9:56p.m., Serge Stinckwich escribiÃ³: > On Tue, Nov 29, 2011 at 9:46 PM, Esteban Lorenzano <estebanlm(a)gmail.com> wrote: >> Hi, >> >> Currently, we are linking official VMs for pharo taking directly from Jenkins a version and packaging for download (or for Pharo One-Click). >> I think this is a temporal solution, we need something better: >> >> 1) Pharo VMs should have Pharo Icons. >> 2) Pharo VMs should be named "pharo" or "pharovm" (better the fist), not "CogVM", which is an internal name. >> 3) We should include a PLUGINS.txt file with information about present plugins (internal and external) >> 4) We should include a LICENCE.txt file with all licence information. >> >> ... I think that's all :) >> >> >> >> I know... this is very dumb and clearly not mandatory, but it says a lot about the maturity of our Pharo :) >> >> All agree? Should I create a ticket? > > Yes, nobody could disagree with that ;-) > >> Cheers, >> Esteban >> >> pd: I'm the only one who does not likes One-Click versions? I would prefer a different package for each platform (Is ugly to have a directory named Pharo.app in linux or windows. Also, no need to have all those duplicated-but-for-different-platform files) > > Most of the time, i'm using One-click images. This is really > convenient to move images between machines and give them to friends. > If the process of creating releases is automated, both can be provided. > Regards, > -- > Serge Stinckwich > UMI UMMISCO 209 (IRD/UPMC), Hanoi, Vietnam > Matsuno Laboratory, Kyoto University, Japan (until 12/2011) > http://www.mechatronics.me.kyoto-u.ac.jp/ > Every DSL ends up being Smalltalk > http://doesnotunderstand.org/ >

Dec. 3, 2011

Re: [Pharo-project] VM freezes sending #basicNew to Stream subclass

by Igor Stasenko

On 3 December 2011 01:30, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> wrote: > We will have to track when this error was introduced may be this is the runtimeError fix that we pushed. > We should check on images to spot this regression. > > Now printing the contents of a stream by default looks to me a bad idea. > May be when the stream is positionable it makes sense but why would I force to read the complete contents of a stream > when I open an inspector on it. > > I think that 'contents' concept in streams are controversial. Just consider: - a stream which generates random numbers. every time you send #next, you get another random number. what is a contents of such stream? all random numbers it can generate till universe die? - a real-time stream broadcasting video/audio. what is a contents of such stream? you cannot capture whole contents of it, because these kinds of streams has no 'end'. It is because streams are for processing data in portions, and when you requesting contents of it, you turning it into collection. Here the Collection class comment: " I am the abstract superclass of all classes that represent a group of elements. " And Stream: " I am an abstract class that represents an accessor for a sequence of objects. This sequence is referred to as my "contents". " Yes. Of course, abstractly speaking, all streams containing something , but you can access it only by portions. Otherwise they will be no different to collections and therefore we won't need them. -- Best regards, Igor Stasenko.

Dec. 3, 2011

Re: [Pharo-project] VM freezes sending #basicNew to Stream subclass

by StÃ©phane Ducasse

Thanks I entered this description http://code.google.com/p/pharo/issues/detail?id=5054 And I will start to see if I can identify what changed using dichotomyâ¦. On Dec 2, 2011, at 7:37 PM, Nicolas Cellier wrote: > Yes. With user interaction... > > In a workspace, type this and debug it: > Stream basicNew printString. > > Step over basicNBew. > Step into printString. > Step into printStringLimitedTo: 50000 > Step into streamContents: [:s | self printOn: s] limitedTo: limit > Step over 100 min: sizeLimit > Step over self new: (100 min: sizeLimit) > --> error: the debugger tries to execute (String new: String) > > I also note that all class methods in stack window left column are > miss-printed - like 'Block in class)' new:

Dec. 3, 2011

Re: [Pharo-project] [Pharo-users] [COTDC] 89 - NullStream

by StÃ©phane Ducasse

I'm a stream doing nothing (NullObject pattern). I can be handy when a stream is needed and that we need to do nothing (for example a logging in off mode). Stef On Dec 3, 2011, at 11:29 AM, sean(a)clipperadams.com wrote: > Continuing in Collections... > > Today: NullStream > > > Comment Of The Day Contest - One Day One Comment > Rules: > #1: Each day a not commented class is elected. Each day the best comment will be integrated with name of the author(s). > #2: If you cannot comment it, deprecate it. > Results: http://code.google.com/p/pharo/wiki/CommentOfTheDayContest > > p.s. this message was sent from a pharo image. Did you know you could easily do that?! >

Dec. 3, 2011

Re: [Pharo-project] GUI Testing Library

by Sean P. DeNigris

Tudor Girba-2 wrote > > It's just cool. > Excellent! The latest package in the repo adds right clicking, choosing from menus, and a bug fix for Morphs that use local coordinates. -- View this message in context: http://forum.world.st/GUI-Testing-Library-tp4097866p4153835.html Sent from the Pharo Smalltalk mailing list archive at Nabble.com.

Dec. 3, 2011

Re: [Pharo-project] Monticello enhancing

by Tobias Pape

Am 2011-12-03 um 15:52 schrieb Mariano Martinez Peck: > > > On Sat, Dec 3, 2011 at 3:46 PM, Tobias Pape <Das.Linux(a)gmx.de> wrote: > > Am 2011-12-03 um 14:34 schrieb Mariano Martinez Peck: > > > > > > >> On Sat, Dec 3, 2011 at 12:23 AM, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> wrote: > >> Thanks tobias > >> This is cool to have that! > >> With your experiment are loading MC definitions using fuel? > >> > > yes, loading and writing. Basically, for writing, for example, it is to change #writeSnapshot: or #serializeInBinary: to use Fuel instead of DataStream. > > For loading, #loadDefinitions > > No. that is the point. > you add a loderFuel to the class side of MCMc{d,z}Reader that > returns an association of priority -> oneargblock. > > My proposal is about experimenting with the MC stuff _without_ changing > the present logic, but augmenting them. > > See the FuelMonticello-topa.1.mcz for an example. > (note that this mcz already contains a snapshot.fuel ;) ) > > > Yes, yes, yes, I saw that :) > I just wanted to point to Stef where MC was doing such magic. I know you didn't modify that directly with instead add different loaders. I saw your code :) Ah, ok. nevermind. Best -Tobias

Dec. 3, 2011

Re: [Pharo-project] Monticello enhancing

by Mariano Martinez Peck

On Sat, Dec 3, 2011 at 3:46 PM, Tobias Pape <Das.Linux(a)gmx.de> wrote: > > Am 2011-12-03 um 14:34 schrieb Mariano Martinez Peck: > > > > > > >> On Sat, Dec 3, 2011 at 12:23 AM, StÃ©phane Ducasse < > stephane.ducasse(a)inria.fr> wrote: > >> Thanks tobias > >> This is cool to have that! > >> With your experiment are loading MC definitions using fuel? > >> > > yes, loading and writing. Basically, for writing, for example, it is to > change #writeSnapshot: or #serializeInBinary: to use Fuel instead of > DataStream. > > For loading, #loadDefinitions > > No. that is the point. > you add a loderFuel to the class side of MCMc{d,z}Reader that > returns an association of priority -> oneargblock. > > My proposal is about experimenting with the MC stuff _without_ changing > the present logic, but augmenting them. > > See the FuelMonticello-topa.1.mcz for an example. > (note that this mcz already contains a snapshot.fuel ;) ) > > Yes, yes, yes, I saw that :) I just wanted to point to Stef where MC was doing such magic. I know you didn't modify that directly with instead add different loaders. I saw your code :) > best > -Tobias > > > > -- Mariano http://marianopeck.wordpress.com

Dec. 3, 2011

Re: [Pharo-project] Monticello enhancing

by Tobias Pape

Am 2011-12-03 um 14:34 schrieb Mariano Martinez Peck: > > >> On Sat, Dec 3, 2011 at 12:23 AM, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> wrote: >> Thanks tobias >> This is cool to have that! >> With your experiment are loading MC definitions using fuel? >> > yes, loading and writing. Basically, for writing, for example, it is to change #writeSnapshot: or #serializeInBinary: to use Fuel instead of DataStream. > For loading, #loadDefinitions No. that is the point. you add a loderFuel to the class side of MCMc{d,z}Reader that returns an association of priority -> oneargblock. My proposal is about experimenting with the MC stuff _without_ changing the present logic, but augmenting them. See the FuelMonticello-topa.1.mcz for an example. (note that this mcz already contains a snapshot.fuel ;) ) best -Tobias

Dec. 3, 2011

Re: [Pharo-project] A commercial question

by Larry White

Must. Have. It! On Sat, Dec 3, 2011 at 8:40 AM, Esteban Lorenzano <estebanlm(a)gmail.com>wrote: > Hi, > > http://smallworks.com.ar/products/mars/ > > (but it is sleeping "the dream of the just", right now :) > > cheers, > Esteban > > El 03/12/2011, a las 10:33a.m., Larry White escribiÃ³: > > Esteban, > What is Mars? > > > On Sat, Dec 3, 2011 at 8:07 AM, Esteban Lorenzano <estebanlm(a)gmail.com>wrote: > >> Hi GermÃ¡n, >> >> As others said, mac users will not be willing to accept pharo look&feel >> except for rare situations (thats one of the reasons I started Mars in >> first instance...). Also, Mars is still far for being really usable >> (sadly), so I don't think you have a real native option right now. >> I'm commited to finish Mars "some time"... but time is being a big >> bottleneck this year. >> I hope in the future I can, at least, advance the project enough to make >> other people contributions possible, so Carla, German, and any other can >> help me to finish :) >> >> cheers, >> Esteban >> >> >> El 02/12/2011, a las 11:04p.m., Sean P. DeNigris escribiÃ³: >> >> > >> > fstephany wrote >> >> >> >> you can probably have a look at MacRuby >> >> >> > >> > MacRuby is pretty cool, and if you go this route, maybe you can have a >> very >> > thin MacRuby GUI layer on top of Pharo (I would start with REST as the >> IPC, >> > but you could go with sockets if it wasn't fast enough). That would be >> my >> > first experiment if I wanted to make a Mac desktop app while waiting for >> > Mars. >> > >> > >> > HTH, >> > Sean >> > >> > >> > -- >> > View this message in context: >> http://forum.world.st/A-commercial-question-tp4150440p4151542.html >> > Sent from the Pharo Smalltalk mailing list archive at Nabble.com. >> > >> >> >> > >

Dec. 3, 2011

Re: [Pharo-project] A commercial question

by Esteban Lorenzano

Hi, http://smallworks.com.ar/products/mars/ (but it is sleeping "the dream of the just", right now :) cheers, Esteban El 03/12/2011, a las 10:33a.m., Larry White escribiÃ³: > Esteban, > What is Mars? > > > On Sat, Dec 3, 2011 at 8:07 AM, Esteban Lorenzano <estebanlm(a)gmail.com> wrote: > Hi GermÃ¡n, > > As others said, mac users will not be willing to accept pharo look&feel except for rare situations (thats one of the reasons I started Mars in first instance...). Also, Mars is still far for being really usable (sadly), so I don't think you have a real native option right now. > I'm commited to finish Mars "some time"... but time is being a big bottleneck this year. > I hope in the future I can, at least, advance the project enough to make other people contributions possible, so Carla, German, and any other can help me to finish :) > > cheers, > Esteban > > > El 02/12/2011, a las 11:04p.m., Sean P. DeNigris escribiÃ³: > > > > > fstephany wrote > >> > >> you can probably have a look at MacRuby > >> > > > > MacRuby is pretty cool, and if you go this route, maybe you can have a very > > thin MacRuby GUI layer on top of Pharo (I would start with REST as the IPC, > > but you could go with sockets if it wasn't fast enough). That would be my > > first experiment if I wanted to make a Mac desktop app while waiting for > > Mars. > > > > > > HTH, > > Sean > > > > > > -- > > View this message in context: http://forum.world.st/A-commercial-question-tp4150440p4151542.html > > Sent from the Pharo Smalltalk mailing list archive at Nabble.com. > > > > >

Dec. 3, 2011