January 2016 - Pharo-users - lists.pharo.org

Re: [Pharo-users] [ann] gtdebugger in pharo 5.0

by Andrei Chis

Hi Nicolai, Thanks for reporting these issues. They are indeed bugs and we are working on fixing them. Cheers, Andrei On Tue, Jan 12, 2016 at 11:24 PM, Nicolai Hess <nicolaihess(a)gmail.com> wrote: > > > 2016-01-08 11:24 GMT+01:00 Tudor Girba <tudor(a)tudorgirba.com>: > >> Hi, >> >> We are about to integrate in Pharo a new member of the Glamorous Toolkit: >> the GTDebugger. As this is a significant change that might affect your >> workflow, here is some background information to help you deal with the >> change. >> >> First, you should know that the change is not irreversible and it is >> easily possible to disabled the new debugger through a setting. However, >> please do take the time to provide us feedback if something does not work >> out for you. We want to know what can be improved and we try to react as >> fast as we can. >> >> A practical change comes from the fact that the variables are manipulated >> through a GTInspector, which makes it cheaper to maintain in the longer run. >> > > Accept and Cancel buttons shouldn't be there > or should not act on if the codepane hasn't changed. > (every press on "accept" writes a new method version, although the > contents didn't changed - tested on > Latest update: #50524 ) > > Most (all?) other tools don't have Accept/Cancel buttons. > > - I really miss the "List Methods using 'varname'/List Methods storing > into 'varname' > - is "stackTop" now gone ? I thought you wanted to add it to the stack ? > - thisContext is gone as well ? > - the Bytecode/GT button is badly placed, it looks like the "downarrow" > window menu icon > is a dropdown menu with label "Bytecode" (since when do we put buttons > in the title pane? > - the evaluator pane is shown as "dirty", as it does not make a difference > if we > accept the text in this pane, there shouldn't be a dirty indicator. > - you can not use the inspector pane to change inst var values > - there is no way to refresh the inspector pane > > I don't open bugtracker entries now, I 'll wait maybe this issues aren't > bugs but > features. > > > nicolai > > >

Jan. 14, 2016

Re: [Pharo-users] SSL client certificates

by Sven Van Caekenberghe

> On 14 Jan 2016, at 12:12, Norbert Hartl <norbert(a)hartl.name> wrote: > >> >> Am 12.01.2016 um 16:25 schrieb Sven Van Caekenberghe <sven(a)stfx.eu>: >> >> Given a ZdcSecureSocketStream you can access the #sslSession. In this session object you can use #certificateName: to set the path or name of the certificate (before you #connect !). That is the general idea. >> >> Now, I don't know if this works or not. Be prepared to look in the plugin C code! On Linux this will probably work. >> >> And please let us know how it goes ;-) >> >>> On 12 Jan 2016, at 16:05, Norbert Hartl <norbert(a)hartl.name> wrote: >>> >>> Is there a way to make SSL connections to the outside world using client certificates from pharo? >>> > > There were some issues with certificates. I tested a lot but did an error on the way. Now I did everything again like creating certificates etc. I could establish a connection from a linux machine. At least it doesn't signal anything an tells it is connected. So that *is* good news ! If you could share more details, that might help others as well. > On Mac OS it does not work. Have you seen the C source code of the Mac SSL Plugin ? It is written against an API from Mac OS System 7, from before Mac OS X, that is an OS from last century, ~199X. This is ancient, you won't find many developers who want to work on that, nor does it make sense towards the future. OpenSSL is also available standard on Mac OS X, the Linux plugin could just as well be used there (give or take). > I think it has something to do with the certificate authority. On linux the CA of the apple server is installed system wide. The SSL plugin code only looks for the certificate and the key but not for the CA. So under linux it finds it in the system but on Mac OS I couldn't do the same. I imported the CA in the keychain but openssl does not seem to find it there. It is very simple why it does not work on Mac (like I said), it is just not implemented ! Try finding where certName is used in sqMacSSL.c, there is a setter and getter, but it is simply not used, so it can never work. In sqUnixOpenSSL.c on the other hand, there is /* if a cert is provided, use it */ if(ssl->certName) { if(ssl->loglevel) printf("sqSetupSSL: Using cert file %s\n", ssl->certName); if(SSL_CTX_use_certificate_file(ssl->ctx, ssl->certName, SSL_FILETYPE_PEM)<=0) ERR_print_errors_fp(stderr); if(SSL_CTX_use_PrivateKey_file(ssl->ctx, ssl->certName, SSL_FILETYPE_PEM)<=0) ERR_print_errors_fp(stderr); } Sven > Norbert

Jan. 14, 2016

Re: [Pharo-users] SSL client certificates

by Norbert Hartl

> Am 12.01.2016 um 16:25 schrieb Sven Van Caekenberghe <sven(a)stfx.eu>: > > Given a ZdcSecureSocketStream you can access the #sslSession. In this session object you can use #certificateName: to set the path or name of the certificate (before you #connect !). That is the general idea. > > Now, I don't know if this works or not. Be prepared to look in the plugin C code! On Linux this will probably work. > > And please let us know how it goes ;-) > >> On 12 Jan 2016, at 16:05, Norbert Hartl <norbert(a)hartl.name> wrote: >> >> Is there a way to make SSL connections to the outside world using client certificates from pharo? >> There were some issues with certificates. I tested a lot but did an error on the way. Now I did everything again like creating certificates etc. I could establish a connection from a linux machine. At least it doesn't signal anything an tells it is connected. On Mac OS it does not work. I think it has something to do with the certificate authority. On linux the CA of the apple server is installed system wide. The SSL plugin code only looks for the certificate and the key but not for the CA. So under linux it finds it in the system but on Mac OS I couldn't do the same. I imported the CA in the keychain but openssl does not seem to find it there. Norbert

Jan. 14, 2016

Re: [Pharo-users] I am David Allouche, let me introduce myself.

by Dimitris Chloupis

sure but namespaces do not solve that problem either. They actually make it much worse. You can have a library named Door by X author and then another library named Door by author Y , hello name clash. Even worse those libraries may contain objects than name class, normally that would not be a problem under the expected usage of name space but lazy coders being .... well lazy... they love to collapse the namespace to reduce typing. So House.Door.Handle will become just Handle which may clash with with the name of another collapsed namespace. Pharo is immune to such issues and the enviroment guarantees that you wont have name clashes inside the image. Namespaces , dont. Add to that that we are very much centralised community around STHub and GitHub which makes it dead easy to find any potential name clash, compared to other languages that are spread all over the internet. So I would say its more a luxury solution than a poor man's. What pharo lacks is something that helps us organise our classes even more and even tag them to form relationships between them. But thats easy to implement the questions is who is going to port the thousands of classes to a new system. On Thu, Jan 14, 2016 at 4:50 AM Ben Coman <btc(a)openinworld.com> wrote: > On Thu, Jan 14, 2016 at 2:12 AM, Dimitris Chloupis > <kilon.alios(a)gmail.com> wrote: > > "I have seen someone in another thread expressing the desire for > namespaces. > > As a newcomer, I also feel that is something important and missing. But > > since one of the goals of Pharo is to produce an environment that can be > > understood by a single person, the lack of namespace might actually be a > > feature. Namespace are essential for languages like C++, designed for > large > > projects, produced by large teams, where nobody understands the whole > > system. They embody an industrial approach to programming. In contrast, I > > have the impression that Smalltalk emphasises regularity to minimise > system > > size, rather than modularity to manage system size, and that it embodies > a > > craftsmanship approach." > > > > Pharo contains over 800000 methods spread around several thousands > classes > > and the system has not imploded yet even though we have no namespaces. > The > > reasoning is simple, Pharo does not need namespaces because its not a > > programming language. Namespaces primary role is not organization so > much as > > prevention of name clashes , this is not a problem for Pharo since it > does > > not allow name clashes in the first place. > > Name clashes can still happen between different external libraries. > Effectively we have a poor-mans namespaces with prefixes. But its > harder to port libraries from other Smalltalks that support > namespaces, limiting our ecosystem. Not necessarily the main driver, > but worthy of consideration. Squeak has recently done some namespace > stuff so will be interesting to see their experience. > > cheers -ben > > > > Thats is happening in the > > enviroment and not language level, if you want to organise your classes > you > > use packages . Packages also are not part of the language and are also > > implemented on enviroment level, though one could argue that language is > > also part of the enviroment. > > > > So yes it feelt strange even for not having namespace coming from python > but > > after coding in pharo for over a year its clear to me that pharo does not > > need it and would probably benefit by more sophisticated and specific > > implementation when it comes to organisations. But generally speaking we > > already have plenty of tools that help you locate classes and methods > with > > ease, so that is not much of a problem. > > > > Pharo enviroment is far from simple, quite the contrary Smalltalk > > implementations were always far more complex than you average language > like > > C++, the big diffirence is that the enviroment is constructed in such > way to > > help you deal with complexity. Where modern languages ephasize simplicity > > pharo and smalltalk in general invest on better dealing with complexity , > > hence why Smalltalk has been very popular in specific very complex fields > > like Banking Sector, Shipping etc. C++ is no where as suitable for > complex > > applications , its popularity merely stems from the fact that is OS > language > > with immediate access to OS libraries and massive support. > > > > On Wed, Jan 13, 2016 at 7:30 PM David Allouche <david(a)allouche.net> > wrote: > >> > >> Thank you everyone for the kind responses. I will answer some of the > >> questions and suggestions here. > >> > >> To get faster to the intellectual stimulation, pick a task / project and > >> start asking questions. And have fun :). > >> > >> > >> That is good advice, but right now I think I will just explore the > system, > >> review changes, and wait for the tasks to find me. There are already > more > >> things I want to do than I will probably every be able to accomplish :-) > >> > >> I think when people referring to Smalltalk refer to Smalltalk as the > >> language or OOP itself, but thats only the tip of the iceberg, Pharo > tries > >> to stay true to vision of smalltalk creating the virtual enviroment > where > >> the user can easily create his or her own tools or modifying the > existing > >> ones, something that I feel no other language out there can emulate > because > >> they follow a non monolithic approach where there is a deep dichotomy > >> between language, libraries and the environment itself. You could say > that > >> Smalltalk is the anti-Unix architecture , a great example of monolithic > >> design that works great in practice yet its easy to modify and extend. > >> > >> > >> The fact Smalltalk the language cannot be taken in isolation from > >> Smalltalk the environment became clear rather early on: reading the > UPBE and > >> other materials shows that the language itself does almost nothing: it > >> cannot even define classes or packages. It is all part of the > environment. > >> > >> I have seen someone in another thread expressing the desire for > >> namespaces. As a newcomer, I also feel that is something important and > >> missing. But since one of the goals of Pharo is to produce an > environment > >> that can be understood by a single person, the lack of namespace might > >> actually be a feature. Namespace are essential for languages like C++, > >> designed for large projects, produced by large teams, where nobody > >> understands the whole system. They embody an industrial approach to > >> programming. In contrast, I have the impression that Smalltalk > emphasises > >> regularity to minimise system size, rather than modularity to manage > system > >> size, and that it embodies a craftsmanship approach. > >> > >> Since I am just old enough to remember a time where computer systems > could > >> be understood by a single person, I find it an attractive idea to try > and > >> produce a modern system with that property. > >> > >> I can't remember who famous said something like "The only languages > >> worth learning are the ones that changes the way you think about > >> programming..." and I found that with Smalltalk & Pharo. > >> > >> > >> I am familiar with that idea too. And most good languages provide that. > >> > >> One day I will have learn some language from the ML family. I choose > >> Smalltalk because it clearly had such a strong influence on the > thinking of > >> people who have used it, I thought I could learn > >> > >> A good way to get your hands dirty is to review fixes submitted by > >> others... > >> https://pharo.fogbugz.com/default.asp?pgx=LF&ixFilter=45 > >> There are never enough hands there :) (my own action here fluctuates > >> from time to time) > >> > >> > >> I will try to spend some time reviewing code. That is a good way to get > >> exposed to things one does not even know they exist. > >> > >> Did you look at Seaside? > >> Since you should be able to read french there is a nice tutorial: > >> tinyBlog. In two days you can build a small web app > >> with a mongo back end. > >> > >> > >> I have no looked at Seaside yet, I have not even read the corresponding > >> chapter in UPBE. But I certainly will, eventually. The web is an > essential > >> part of society today. > >> > >> Where are you located ? > >> > >> > >> Paris, France. > >> > >> It did occur to me that the Pharo community does have a unusually large > >> fraction of french people. But that is not a factor for me. > >> > >> Thanks again for all the kind words. I am sure this journey is going to > be > >> a lot of fun. :-) > >> > >> > > > >

Jan. 14, 2016

Re: [Pharo-users] Spec: Populating a playground with a particular content (was Re: Spec: Binding widgets how to)

by Nicolai Hess

2016-01-13 23:07 GMT+01:00 Offray Vladimir Luna CÃ¡rdenas <offray(a)riseup.net> : > Hi, > > I've reading the code at [1] but I'm still having problems understanding > the dynamic spec. I have created a small dictionary and populated with > data, then created the accessors and initializePresenter, initializeWidgets > and rebuildWithdCodeLayout and rebuildWithTextLayout, as you can see on the > following snippets, but nothing happens despite the debugging Transcript is > showing me that the content subwidget has the proper model, but is not > rebuild in the GUI. > > Any help is, as always, very much appreciated. > Hi Offray, I think you have to call self needRebuild: false. in both rebuild... methods. Otherwise the all "inner"-widgets are rebuild: > > [1] > https://github.com/spec-framework/documentation/blob/master/output/book.md#… > > ---------------------- > DynamicDict>>initializeWidgets > > | data | > data := { 'first' -> 'I\"m just text' . 'second' -> 'ProfStef > openPharoZenWorkspace'} asOrderedDictionary. > list := self newList. > list items: data keys. > content := self newText. > self focusOrder > add: list; > add: content. > ---------------------- > DynamicDict>>initializePresenter > list whenSelectedItemChanged: [:item | > item = 'first' > ifTrue: [ self rebuildWithTextLayout ]. > item = 'second' > ifTrue: [ self rebuildWithCodeLayout ]. > ] > ---------------------- > DynamicDict>>rebuildWithCodeLayout > | newLayout | > > content := self instantiate: GlamourPresentationModel. > newLayout := SpecLayout composed > newRow: [:row | > row > add: #list right: 0.7; > add: #content left: 0.3]; > yourself. > self needRebuild: true. > content needRebuild: true. > Transcript show: 'cÃ³digo + ', self content asString, String cr. > self buildWithSpecLayout: newLayout. > ---------------------- > DynamicDict>>rebuildWithTextLayout > | newLayout | > > content := self newText. > newLayout := SpecLayout composed > newRow: [:row | > row > add: #list right: 0.7; > add: #content left: 0.3]; > yourself. > self needRebuild: false. > list needRebuild: false. > content needRebuild: true. > Transcript show: 'texto + ', self content asString, String cr.. > self buildWithSpecLayout: newLayout. > ---------------------- > > Cheers, > > Offray > > On 10/01/16 13:39, Johan Fabry wrote: > > Sorry for the lateness and short reply but Iâm involved in a bunch of > other things that are taking almost all of my time ... > > From what I understand and what I can see in your code, you want to modify > a widget when the UI is open. However you never trigger the rebuilding of > the window. For example, see the Spec docs that I sent, where in 9.1 there > is code that calls buildWithSpecLayout: . > > Actually, I am not sure you need to subclass DynamicComposableModel since > the number of widgets never changes. You can use a normal ComposableModel > subclass and use what is detailed in 9.1 to change the layout between the > two options you want to have (a TextModel or a GlamourPresentationModel). > > On Jan 8, 2016, at 23:07, Offray Vladimir Luna CÃ¡rdenas <offray(a)riseup.net> > wrote: > > Shouldn't be #right dynamically defined according to the values of 'item' > and be replaced on the m2 layout? > > > > > ---> Save our in-boxes! http://emailcharter.org <--- > > Johan Fabry - http://pleiad.cl/~jfabry > PLEIAD and RyCh labs - Computer Science Department (DCC) - University > of Chile > > >

Jan. 14, 2016

Re: [Pharo-users] I am David Allouche, let me introduce myself.

by Ben Coman

On Thu, Jan 14, 2016 at 2:12 AM, Dimitris Chloupis <kilon.alios(a)gmail.com> wrote: > "I have seen someone in another thread expressing the desire for namespaces. > As a newcomer, I also feel that is something important and missing. But > since one of the goals of Pharo is to produce an environment that can be > understood by a single person, the lack of namespace might actually be a > feature. Namespace are essential for languages like C++, designed for large > projects, produced by large teams, where nobody understands the whole > system. They embody an industrial approach to programming. In contrast, I > have the impression that Smalltalk emphasises regularity to minimise system > size, rather than modularity to manage system size, and that it embodies a > craftsmanship approach." > > Pharo contains over 800000 methods spread around several thousands classes > and the system has not imploded yet even though we have no namespaces. The > reasoning is simple, Pharo does not need namespaces because its not a > programming language. Namespaces primary role is not organization so much as > prevention of name clashes , this is not a problem for Pharo since it does > not allow name clashes in the first place. Name clashes can still happen between different external libraries. Effectively we have a poor-mans namespaces with prefixes. But its harder to port libraries from other Smalltalks that support namespaces, limiting our ecosystem. Not necessarily the main driver, but worthy of consideration. Squeak has recently done some namespace stuff so will be interesting to see their experience. cheers -ben > Thats is happening in the > enviroment and not language level, if you want to organise your classes you > use packages . Packages also are not part of the language and are also > implemented on enviroment level, though one could argue that language is > also part of the enviroment. > > So yes it feelt strange even for not having namespace coming from python but > after coding in pharo for over a year its clear to me that pharo does not > need it and would probably benefit by more sophisticated and specific > implementation when it comes to organisations. But generally speaking we > already have plenty of tools that help you locate classes and methods with > ease, so that is not much of a problem. > > Pharo enviroment is far from simple, quite the contrary Smalltalk > implementations were always far more complex than you average language like > C++, the big diffirence is that the enviroment is constructed in such way to > help you deal with complexity. Where modern languages ephasize simplicity > pharo and smalltalk in general invest on better dealing with complexity , > hence why Smalltalk has been very popular in specific very complex fields > like Banking Sector, Shipping etc. C++ is no where as suitable for complex > applications , its popularity merely stems from the fact that is OS language > with immediate access to OS libraries and massive support. > > On Wed, Jan 13, 2016 at 7:30 PM David Allouche <david(a)allouche.net> wrote: >> >> Thank you everyone for the kind responses. I will answer some of the >> questions and suggestions here. >> >> To get faster to the intellectual stimulation, pick a task / project and >> start asking questions. And have fun :). >> >> >> That is good advice, but right now I think I will just explore the system, >> review changes, and wait for the tasks to find me. There are already more >> things I want to do than I will probably every be able to accomplish :-) >> >> I think when people referring to Smalltalk refer to Smalltalk as the >> language or OOP itself, but thats only the tip of the iceberg, Pharo tries >> to stay true to vision of smalltalk creating the virtual enviroment where >> the user can easily create his or her own tools or modifying the existing >> ones, something that I feel no other language out there can emulate because >> they follow a non monolithic approach where there is a deep dichotomy >> between language, libraries and the environment itself. You could say that >> Smalltalk is the anti-Unix architecture , a great example of monolithic >> design that works great in practice yet its easy to modify and extend. >> >> >> The fact Smalltalk the language cannot be taken in isolation from >> Smalltalk the environment became clear rather early on: reading the UPBE and >> other materials shows that the language itself does almost nothing: it >> cannot even define classes or packages. It is all part of the environment. >> >> I have seen someone in another thread expressing the desire for >> namespaces. As a newcomer, I also feel that is something important and >> missing. But since one of the goals of Pharo is to produce an environment >> that can be understood by a single person, the lack of namespace might >> actually be a feature. Namespace are essential for languages like C++, >> designed for large projects, produced by large teams, where nobody >> understands the whole system. They embody an industrial approach to >> programming. In contrast, I have the impression that Smalltalk emphasises >> regularity to minimise system size, rather than modularity to manage system >> size, and that it embodies a craftsmanship approach. >> >> Since I am just old enough to remember a time where computer systems could >> be understood by a single person, I find it an attractive idea to try and >> produce a modern system with that property. >> >> I can't remember who famous said something like "The only languages >> worth learning are the ones that changes the way you think about >> programming..." and I found that with Smalltalk & Pharo. >> >> >> I am familiar with that idea too. And most good languages provide that. >> >> One day I will have learn some language from the ML family. I choose >> Smalltalk because it clearly had such a strong influence on the thinking of >> people who have used it, I thought I could learn >> >> A good way to get your hands dirty is to review fixes submitted by >> others... >> https://pharo.fogbugz.com/default.asp?pgx=LF&ixFilter=45 >> There are never enough hands there :) (my own action here fluctuates >> from time to time) >> >> >> I will try to spend some time reviewing code. That is a good way to get >> exposed to things one does not even know they exist. >> >> Did you look at Seaside? >> Since you should be able to read french there is a nice tutorial: >> tinyBlog. In two days you can build a small web app >> with a mongo back end. >> >> >> I have no looked at Seaside yet, I have not even read the corresponding >> chapter in UPBE. But I certainly will, eventually. The web is an essential >> part of society today. >> >> Where are you located ? >> >> >> Paris, France. >> >> It did occur to me that the Pharo community does have a unusually large >> fraction of french people. But that is not a factor for me. >> >> Thanks again for all the kind words. I am sure this journey is going to be >> a lot of fun. :-) >> >> >

Jan. 14, 2016

Re: [Pharo-users] I am David Allouche, let me introduce myself.

by Serge Stinckwich

On Wed, Jan 13, 2016 at 6:29 PM, David Allouche <david(a)allouche.net> wrote: > Thank you everyone for the kind responses. I will answer some of the > questions and suggestions here. ... > Where are you located ? > > > Paris, France. Great ! I'm also located in Paris. Maybe we could have a Pharo user group in the future. I know 5 Pharoers now in Paris ;-) Regards, -- Serge Stinckwich UCBN & UMI UMMISCO 209 (IRD/UPMC) Every DSL ends up being Smalltalk http://www.doesnotunderstand.org/

Jan. 13, 2016

Re: [Pharo-users] Spec: Populating a playground with a particular content (was Re: Spec: Binding widgets how to)

by Offray Vladimir Luna CÃ¡rdenas

Hi, I've reading the code at [1] but I'm still having problems understanding the dynamic spec. I have created a small dictionary and populated with data, then created the accessors and initializePresenter, initializeWidgets and rebuildWithdCodeLayout and rebuildWithTextLayout, as you can see on the following snippets, but nothing happens despite the debugging Transcript is showing me that the content subwidget has the proper model, but is not rebuild in the GUI. Any help is, as always, very much appreciated. [1] https://github.com/spec-framework/documentation/blob/master/output/book.md#… ---------------------- DynamicDict>>initializeWidgets | data | data := { 'first' -> 'I\"m just text' . 'second' -> 'ProfStef openPharoZenWorkspace'} asOrderedDictionary. list := self newList. list items: data keys. content := self newText. self focusOrder add: list; add: content. ---------------------- DynamicDict>>initializePresenter list whenSelectedItemChanged: [:item | item = 'first' ifTrue: [ self rebuildWithTextLayout ]. item = 'second' ifTrue: [ self rebuildWithCodeLayout ]. ] ---------------------- DynamicDict>>rebuildWithCodeLayout | newLayout | content := self instantiate: GlamourPresentationModel. newLayout := SpecLayout composed newRow: [:row | row add: #list right: 0.7; add: #content left: 0.3]; yourself. self needRebuild: true. content needRebuild: true. Transcript show: 'cÃ³digo + ', self content asString, String cr. self buildWithSpecLayout: newLayout. ---------------------- DynamicDict>>rebuildWithTextLayout | newLayout | content := self newText. newLayout := SpecLayout composed newRow: [:row | row add: #list right: 0.7; add: #content left: 0.3]; yourself. self needRebuild: false. list needRebuild: false. content needRebuild: true. Transcript show: 'texto + ', self content asString, String cr.. self buildWithSpecLayout: newLayout. ---------------------- Cheers, Offray On 10/01/16 13:39, Johan Fabry wrote: > Sorry for the lateness and short reply but Iâm involved in a bunch of > other things that are taking almost all of my time ... > > From what I understand and what I can see in your code, you want to > modify a widget when the UI is open. However you never trigger the > rebuilding of the window. For example, see the Spec docs that I sent, > where in 9.1 there is code that calls buildWithSpecLayout: . > > Actually, I am not sure you need to subclass DynamicComposableModel > since the number of widgets never changes. You can use a normal > ComposableModel subclass and use what is detailed in 9.1 to change the > layout between the two options you want to have (a TextModel or a > GlamourPresentationModel). > >> On Jan 8, 2016, at 23:07, Offray Vladimir Luna CÃ¡rdenas >> <offray(a)riseup.net <mailto:offray@riseup.net>> wrote: >> >> Shouldn't be #right dynamically defined according to the values of >> 'item' and be replaced on the m2 layout? > > > > ---> Save our in-boxes! http://emailcharter.org <--- > > Johan Fabry - http://pleiad.cl/~jfabry <http://pleiad.cl/%7Ejfabry> > PLEIAD and RyCh labs - Computer Science Department (DCC) - > University of Chile >

Jan. 13, 2016

Re: [Pharo-users] I am David Allouche, let me introduce myself.

by Offray Vladimir Luna CÃ¡rdenas

Hi, On 10/01/16 17:16, Dimitris Chloupis wrote: [...] > > I think when people referring to Smalltalk refer to Smalltalk as the > language or OOP itself, but thats only the tip of the iceberg, Pharo > tries to stay true to vision of smalltalk creating the virtual > enviroment where the user can easily create his or her own tools or > modifying the existing ones, something that I feel no other language > out there can emulate because they follow a non monolithic approach > where there is a deep dichotomy between language, libraries and the > environment itself. You could say that Smalltalk is the anti-Unix > architecture , a great example of monolithic design that works great > in practice yet its easy to modify and extend. > [...] A place where I finally got this clear was in the book Tracing the Dynabook by John W. Maxwell[1], where its author says that a fair comparison wouldn't be between Smalltalk and other programming languages, but between Smalltalk and the operative systems paradigm. I had found Squeak in 2005 and even reading something about the Dynabook, but was until reading this text on 2007 when I finally get it. It has a different approach that the classical programming book, but was one of the most enlightening reading for me about Smalltalk, the dynabook and the popular computer paradigm we have today. [1] http://tkbr.ccsp.sfu.ca/dynabook/ Cheers, Offray

Jan. 13, 2016

Re: [Pharo-users] Understanding the role of the sources file

by Dimitris Chloupis

"I am getting off-topic, but since I have already written what follows, and I think it is informative, I am leaving it there :-)" Thats ok my question has been answered by Sven's first reply anyway, some offtopic wont hurt "Python is a rather large language, with a large number of quirks that need to be preserved to provide full compatibility. It is also a language with advanced introspection, and very dynamic semantics, even more than Smalltalk in some aspects. And its main implementation (CPython) has very highly optimised data structures and memory management. That makes it really difficult to implement correctly and and to improve performance without slowing down other parts. Before PyPy, Google funded the Unladen Swallow project, that tried to produce a Python JIT using LLVM, this project was abandoned. Even PyPy needs to make frequent and expensive guard tests to ensure that invariants required for JIT are preserved. It does bring a significant speed boost (2x-10x), but at the expense of slow startup and increased memory footprint (~2x). However, it seems clear that the long term future of Python lies in PyPy. In particular, they have high hopes to solve the lack of multicore support by implementing software transactional memory in the medium term, until hardware transactional memory is advanced enough to be usable there." Well I have watched a couple of talks of Guido (creator of cpython for those who dont know) on the subject, and I think his overall argument "why should I give a damn ?" His line of argument is that Python offers such a great support for wrapping C libraries that makes a JIT complier not that important. Python initial missions was to be a scripting language for C/C++ , the fact it grew to an entity by itself is the side effect of its popularity. However Guido even lately stresses the idea what Python should remain one of the best if not the best way to wrap C code. So if you start using and relying on C code so much a JIT makes less and less sense. For example even cpython itself over 50% of its code base is coded in C. Cpython developers are essentially C developers. Pharo is not on the same boat, Pharo is a self hosting enviroment, that means that a great deal of code is written in pharo , some of it Slang that is basically smalltalk that compiles to C and some of the really performance dependent parts in pure C. Also Pharo libraries are not C libraries wrapped for Pharo, they are just Pharo libraries. So it is a very diffirent mentality . So the reason why CPython has not a JIT VM is that it does really need it, python coders already rely on performance orientated python libraries( see numpy) that are written in C to do the heavy lifiting, I dont think a JIT VM would make much of a difference anyway. On the other hand you rarely see someone here asking for the Pharo FFI, or how to make a VM plugin , in 99% of the cases people work on pure pharo code. On Wed, Jan 13, 2016 at 8:04 PM David Allouche <david(a)allouche.net> wrote: > On 13 Jan 2016, at 18:14, Dimitris Chloupis <kilon.alios(a)gmail.com> wrote: > > Why the Cog VM is mentioned in my documentation is no accident. There is > another VM that is called Stackless VM or old VM that was the VM before Cog > that had no JIT and its a VM we still use for platforms that Cog VM does > not currently run like Raspberry Pi, Android and I think iOS too. > > > "But nowadays, JIT compiling byte-code to machine language are the norm > rather than the exception, so I do not think it is worth mentioning." > > Definitely not the norm Python and Ruby VMs are not JIT, generally > speaking JIT compilers are rare for Dynamic languages like Pharo. Python > has PyPy which is a JIT VM but is nowhere as popular as Cpython. > > > You are right. The JIT is worth mentioning there. > > > CPython people try to avoid JIT because it makes VM architecture much more > complex. At least thats their argument. > > > I am getting off-topic, but since I have already written what follows, and > I think it is informative, I am leaving it there :-) > > Python is a rather large language, with a large number of quirks that need > to be preserved to provide full compatibility. It is also a language with > advanced introspection, and very dynamic semantics, even more than > Smalltalk in some aspects. And its main implementation (CPython) has very > highly optimised data structures and memory management. > > That makes it really difficult to implement correctly and and to improve > performance without slowing down other parts. Before PyPy, Google funded > the Unladen Swallow project, that tried to produce a Python JIT using LLVM, > this project was abandoned. Even PyPy needs to make frequent and expensive > guard tests to ensure that invariants required for JIT are preserved. It > does bring a significant speed boost (2x-10x), but at the expense of slow > startup and increased memory footprint (~2x). > > However, it seems clear that the long term future of Python lies in PyPy. > In particular, they have high hopes to solve the lack of multicore support > by implementing software transactional memory in the medium term, until > hardware transactional memory is advanced enough to be usable there. > > Another big advantage of Pharo is that it has luxury to come with a JIT > included and well tested. > > > I guess the simplicity of the language makes it relatively easy to produce > a JIT. This is indeed a very good thing. > > > The only popular JITs out there are Java's and .NET, dont know if > Javascript V8 use JIT as well, but thats pretty much it. But bytecode is > very popular indeed. > > > V8 does use JIT. I believe it was the first widely deployed language > implementation to use a tracing interpreter to JIT a language with latent > typing. I believe all the other major Javascript implementations have > followed suite. > > This is also the approach used by PyPy. Java and .Net have static typing, > so they do not require a tracing interpreter to compile to machine code. > > On Wed, Jan 13, 2016 at 6:43 PM David Allouche <david(a)allouche.net> wrote: > >> On 13 Jan 2016, at 16:27, Dimitris Chloupis <kilon.alios(a)gmail.com> >> wrote: >> >> "The *virtual machine* (VM) provides the environment where the Pharo >> system lives. It is different for each operating system and hardware >> architecture, and runs as a machine language executable in the operating >> system. It implements the details of managing memory, executing Pharo >> byte-code, and communicating with the world outside of the Pharo system: >> files, other operating system process, and the network." >> >> No the environment is the image, the VM is basically what its names says, >> a machine emulated by software. The vast majority of tools, even the >> language itself reside on the image. VM is there in order for the code to >> be able to execute and to interface with the underlying Operating System. >> You could completely modify the VM , for example move it to the JVM and >> still the pharo enviroment would be intact. >> >> >> That is indeed the idea I tried to convey by "where the Pharo system >> lives", but I see how that can be misunderstood. However, I try to avoid >> using "big words" like "abstraction". >> >> How about this? >> >> "The virtual machine (VM) provides the portable environment to execute >> Pharo images. Its implementation needs to be different for each operating >> system and hardware architecture, as it runs as a machine language >> executable in the operating system. It implements the details of managing >> memory, executing Pharo byte-code, and communicating with the world outside >> of the Pharo system: files, other operating system process, and the >> network. >> >> >> "Hi Dimitris, >> your formulation "...Pharo bytcode...and convert it to machine code..." >> is insofar irritating to me as "convert it to machine code" would >> suggest to me that a compiler is at work here. Davids "executing Pharo >> byte-code" seems more understandable to me here." >> >> Thats correct its a compiler, a byte compiler, it compiles bytecode to >> machine code and it does it while the code executes, this is why its called >> JIT , which has the meaning of Just In Time compilation, meaning that >> machine code is compiled just before the code is executed so several >> optimizations can be applied that would not be known before the execution >> of the code. Similar to JAVA's JIT compiler. >> >> Note here that a compiler is not just something that produces machine >> code, a compiler for example can take one language and compile it to >> another language. >> >> >> That's technically true. But most readers will probably unconsciously >> assume that a compiler is something that produces machine language. The >> document should be careful to avoid misunderstandings caused by such common >> assumptions. >> >> As for the JIT, that is totally an implementation detail, and I believe >> it is only worth mentioning if you want to prevent the reader from assuming >> that Pharo is slow because its VM executes byte-code. But nowadays, JIT >> compiling byte-code to machine language are the norm rather than the >> exception, so I do not think it is worth mentioning. >> >> >> >> On Wed, Jan 13, 2016 at 4:58 PM Werner Kassens <wkassens(a)libello.com> >> wrote: >> >>> Hi Dimitris, >>> your formulation "...Pharo bytcode...and convert it to machine code..." >>> is insofar irritating to me as "convert it to machine code" would >>> suggest to me that a compiler is at work here. Davids "executing Pharo >>> byte-code" seems more understandable to me here. >>> werner >>> >>> On 01/13/2016 02:22 PM, Dimitris Chloupis wrote: >>> > I assume you have never read a an introduction to C++ then :D >>> > >>> > here is the final addition for the vm >>> > >>> > (Vm) is the only component that is different for each operating system. >>> > The main purpose of the VM is to take Pharo bytcode that is generated >>> > each time user accepts a piece of code and convert it to machine code >>> in >>> > order to be executed, but also to generally handle low level >>> > functionality like interpreting code, handling OS events (mouse and >>> > keyboard), calling C libraries etc. Pharo 4 comes with the Cog VM a >>> very >>> > fast JIT VM. >>> > >>> > I think its clear, precise and does not leave much room for confusion. >>> > Personally I think its very important for the absolute begineer to have >>> > strong foundations of understanding the fundamental of Pharo and not >>> for >>> > things to appear magical and "dont touch this". >>> > >>> > On Wed, Jan 13, 2016 at 2:54 PM Sven Van Caekenberghe <sven(a)stfx.eu >>> > <mailto:sven@stfx.eu>> wrote: >>> > >>> > >>> > > On 13 Jan 2016, at 13:42, Dimitris Chloupis >>> > <kilon.alios(a)gmail.com <mailto:kilon.alios@gmail.com>> wrote: >>> > > >>> > > I mentioned bytecode because I dont want the user to see at some >>> > point bytecode and say "What the hell is that" I want the reader to >>> > feel confident that at least understands the basic in Pharo. Also >>> > very brief explanations about bytecode I have seen in similar >>> python >>> > tutorials. Obviously I dont want to go any deeper than that because >>> > the user wont have to worry about the technical details on a daily >>> > basis anyway. >>> > > >>> > > I agree that I could add a bit more on the VM description >>> similar >>> > to what you posted. I am curious though, wont even the interpreter >>> > generate machine code in order to execute the code or does it use >>> > existing machine code inside the VM binary ? >>> > >>> > No, a classic interpreter does not 'generate' machine code, it is >>> > just a program that reads and executes bytes codes in a loop, the >>> > interpreter 'is' machine code. >>> > >>> > No offence, but you see why I think it is important to not try to >>> > use or explain too much complex concepts in the 1st chapter. >>> > >>> > Learning to program is hard. It should first be done abstractly. >>> > Think about Scratch. The whole idea of Smalltalk is to create a >>> > world of interacting objects. (Even byte code is not a necessary >>> > concept at all, for example, in Pharo, you can compile (translate) >>> > to AST and execute that, I believe. There are Smalltalk >>> > implementations that compile directly to C or JavaScript). Hell, >>> > even 'compile' is not necessary, just 'accept'. See ? >>> > >>> > > On Wed, Jan 13, 2016 at 2:25 PM Sven Van Caekenberghe >>> > <sven(a)stfx.eu <mailto:sven@stfx.eu>> wrote: >>> > > Sounds about right. >>> > > >>> > > Now, I would swap 1 and 4, as the image is the most important >>> > abstraction. >>> > > >>> > > There is also a bit too much emphasis on (byte|source)code. This >>> > is already pretty technical (it assume you know what compilation is >>> > and so on). But I understand it must be explained here, and you did >>> > it well. >>> > > >>> > > However, I would start by saying that the image is a snapshot of >>> > the object world in memory that is effectively a live Pharo system. >>> > It contains everything that is available and that exists in Pharo. >>> > This includes any objects that you created yourself, windows, >>> > browsers, open debuggers, executing processes, all meta objects as >>> > well as all representations of code. >>> > > >>> > > <sidenote> >>> > > The fact that there is a sources and changes file is an >>> > implementation artefact, not something fundamental. There are ideas >>> > to change this in the future (but you do not have to mention that). >>> > > </sidenote> >>> > > >>> > > Also, the VM not only executes code, it maintains the object >>> > world, which includes the ability to load and save it from and to >>> an >>> > image. It creates a portable (cross platform) abstraction that >>> > isolates the image from the particular details of the underlying >>> > hardware and OS. In that role it implements the interface with the >>> > outside world. I would mention that second part before mentioning >>> > the code execution. >>> > > >>> > > The sentence "The purpose of the VM is to take Pharo bytcode >>> that >>> > is generated each time user accepts a piece of code and convert it >>> > to machine code in order to be executed." is not 100% correct. It >>> is >>> > possible to execute the byte code without converting it. This is >>> > called interpretation. JIT is a faster technique that includes >>> > converting (some often used) byte code to machine code and caching >>> that. >>> > > >>> > > I hope this helps (it is hard to write a 'definitive >>> explanation' >>> > as there are some many aspects to this and it depends on the >>> > context/audience). >>> > > >>> > > > On 13 Jan 2016, at 12:58, Dimitris Chloupis >>> > <kilon.alios(a)gmail.com <mailto:kilon.alios@gmail.com>> wrote: >>> > > > >>> > > > So I am correct that the image does not store the source code, >>> > and that the source code is stored in sources and changes. The only >>> > diffirence is that the objects have a source variable that points >>> to >>> > the right place for finding the source code. >>> > > > >>> > > > This is the final text if you find anything incorrect please >>> > correct me >>> > > > >>> > > > --------------- >>> > > > >>> > > > 1. The virtual machine (VM) is the only component that is >>> > different for each operating system. The purpose of the VM is to >>> > take Pharo bytcode that is generated each time user accepts a piece >>> > of code and convert it to machine code in order to be executed. >>> > Pharo 4 comes with the Cog VM a very fast JIT VM. The VM executable >>> > is named: >>> > > > >>> > > > â¢ Pharo.exe for Windows; â¢ pharo for Linux ; and >>> > > > >>> > > > â¢ Pharo for OSX (inside a package also named Pharo.app). >>> > > > The other components below are portable across operating >>> > systems, and >>> > > > >>> > > > can be copied and run on any appropriate virtual machine. >>> > > > >>> > > > 2. The sources file contains source code for parts of Pharo >>> > that donât change frequently. Sources file is important because the >>> > image file format stores only the bytecode of live objects and not >>> > their source code. Typically a new sources file is generated once >>> > per major release of Pharo. For Pharo 4.0, this file is named >>> > PharoV40.sources. >>> > > > >>> > > > 3. The changes file logs of all source code modifications >>> since >>> > the .sources file was generated. This facilitates a per method >>> > history for diffs or re- verting.That means that even if you dont >>> > manage to save the image file on a crash or you just forgot you can >>> > recover your changes from this file. Each release provides a near >>> > empty file named for the release, for example Pharo4.0.changes. >>> > > > >>> > > > 4. The image file provides a frozen in time snapshot of a >>> > running Pharo system. This is the file where the Pharo bytecode is >>> > stored and as such its a cross platform format. This is the heart >>> of >>> > Pharo, containing the live state of all objects in the system >>> > (including classes and methods, since they are objects too). The >>> > file is named for the release (like Pharo4.0.image). >>> > > > >>> > > > The .image and .changes files provided by a Pharo release are >>> > the starting point for a live environment that you adapt to your >>> > needs. Essentially the image file containes the compiler of the >>> > language (not the VM) , the language parser, the IDE tools, many >>> > libraries and acts a bit like a virtual Operation System that runs >>> > on top of a Virtual Machine (VM), similarly to ISO files. >>> > > > >>> > > > As you work in Pharo, these files are modified, so you need to >>> > make sure that they are writable. The .image and .changes files are >>> > intimately linked and should always be kept together, with matching >>> > base filenames. Never edit them directly with a text editor, as >>> > .images holds your live object runtime memory, which indexes into >>> > the .changes files for the source. It is a good idea to keep a >>> > backup copy of the downloaded .image and .changes files so you can >>> > always start from a fresh image and reload your code. However the >>> > most efficient way for backing up code is to use a version control >>> > system that will provide an easier and powerful way to back up and >>> > track your changes. >>> > > > >>> > > > The four main component files above can be placed in the same >>> > directory, although itâs also possible to put the Virtual Machine >>> > and sources file in a separate directory where everyone has >>> > read-only access to them. >>> > > > >>> > > > If more than one image file is present in the same directory >>> > pharo will prompt you to choose an image file you want to load. >>> > > > >>> > > > Do whatever works best for your style of working and your >>> > operating system. >>> > > > >>> > > > >>> > > > >>> > > > >>> > > > >>> > > > On Wed, Jan 13, 2016 at 12:13 PM Sven Van Caekenberghe >>> > <sven(a)stfx.eu <mailto:sven@stfx.eu>> wrote: >>> > > > >>> > > > > On 13 Jan 2016, at 10:57, Dimitris Chloupis >>> > <kilon.alios(a)gmail.com <mailto:kilon.alios@gmail.com>> wrote: >>> > > > > >>> > > > > I was adding a short description to the UPBE about sources >>> > file , I always thought that the sources file is the file that >>> > contains the source code of the image because the image file itself >>> > stores only the bytecode. >>> > > > > >>> > > > > However its just came to my attention that the sources file >>> > does not contain code that is recently installed in the image. >>> > > > > >>> > > > > So how exactly the sources file works and what it is ? >>> > > > >>> > > > The main perspective is from the object point of view: methods >>> > are just objects like everything else. In order to be executable >>> > they know their byte codes (which might be JIT compiled on >>> > execution, but that is an implementation detail) and they know >>> their >>> > source code. >>> > > > >>> > > > Today we would probably just store the source code strings in >>> > the image (maybe compressed) as memory is pretty cheap. But way >>> back >>> > when Smalltalk started, that was not the case. So they decided to >>> > map the source code out to files. >>> > > > >>> > > > So method source code is a magic string (RemoteString) that >>> > points to some position in a file. There are 2 files in use: the >>> > sources file and the changes file. >>> > > > >>> > > > The sources file is a kind of snapshot of the source code of >>> > all methods at the point of release of a major new version. That is >>> > why there is a Vxy in their name. The source file never changes >>> once >>> > created or renewed (a process called generating the sources, see >>> > PharoSourcesCondenser). >>> > > > >>> > > > While developing and creating new versions of methods, the new >>> > source code is appended to another file called the changes file, >>> > much like a transaction log. This is also a safety mechanism to >>> > recover 'lost' changes. >>> > > > >>> > > > The changes file can contain multiple versions of a method. >>> > This can be reduced in size using a process called condensing the >>> > changes, see PharoChangesCondenser. >>> > > > >>> > > > On a new release, the changes file will be (almost) empty. >>> > > > >>> > > > HTH, >>> > > > >>> > > > Sven >>> > > > >>> > > > >>> > > > >>> > > >>> > > >>> > >>> > >>> >>>

Jan. 13, 2016