- Pharo-dev - lists.pharo.org

Re: [Pharo-project] Destiny of changed:/update:

by Schwab,Wilhelm K

The short answer is "look at other dialects." Pharo's weak collections replace objects with nil. Dolphin's completely "forget" the object. It also does a robust job of handling the trickier side of finalization, such as the fact that an object is allowed to rescue itself at the last minute. All of this put together is what I mean by "thread safe and self-repairing." See it done well, and Pharo looks bad by comparison. I have put wrappers around the weak collections. ________________________________________ From: pharo-project-bounces(a)lists.gforge.inria.fr [pharo-project-bounces(a)lists.gforge.inria.fr] on behalf of Igor Stasenko [siguctua(a)gmail.com] Sent: Tuesday, December 27, 2011 5:39 AM To: Pharo-project(a)lists.gforge.inria.fr Subject: Re: [Pharo-project] Destiny of changed:/update: i cannot get it, how weak type containers are related to thread safety? A garbage collection happens instantly for language side , so there is no way how a weak collection can be in inconsistent state when it is not known yet, if object is dead or alive. By accessing an object in weak container, you put it's reference on stack, which means that if GC will happen at that stage, a corresponding object will not be collected until at least next GC phase. Now what is "thread-safe weak collections" and what they should serve for? WeakIdentityKeyDictionary? It is not thread-safe. And why it should, when you can easily make it thread-safe by wrapping it with appropriate container object. What else? WeakArray? it is already thread safe: reads and writes are primitives and size is immutable. What else? weak set? again, put a wrapper around it and you got thread-safety. And what is 'self-repairing'? Can you explain what is it? -- Best regards, Igor Stasenko.

Dec. 27, 2011

Re: [Pharo-project] Programmatic creation of classes Re: [ANN][Beta Release] DBXTools - Magritte Scaffolding - Phoseydon

by Eliot Miranda

On Tue, Dec 27, 2011 at 7:53 AM, Ben Coman <btc(a)openinworld.com> wrote: > Guillermo Polito wrote: > >> *What does Phoseydon beta includes?* >> By now, it reads a database schema, builds a model description from it and >> then allows us to: >> - write classes from that description. >> >> > I am very interested in the programmatic creation of classes. Although my > application is not quite related to database schema. Instead I have a UML > model stored in a database basically defined by two tables t_objects and > t_connector from which I need to create Smalltalk classes, instance > variables, and association & aggregate relationships. I tracked down a lot > of what I want to the methods... > > DBXMagritteClassBuilderVisitor**>>visitDBXEntity: aDBXEntity > for which the answer was obvious in hindsight - just replicating the call > to Object>>subclass::: > > DBXMagritteClassBuilderVisitor**>>buildAccessorForAttribute: attribute > inClass: aClass > could you give me a pointer to read up on what the curly-brace-numbers are? > > btw, Can you advise how is the compiling activated? If you had about 800 > classes that need to be created, would temporarily turning off the compile > assist the program execution time significantly? > regards, Ben > profile it using spyOn: to see where the time goes, and then you'll know. -- best, Eliot

Dec. 27, 2011

Re: [Pharo-project] can't infer base LD_LIBRARY_PATH. Aborting.

by Eliot Miranda

Hi Phillipe, what does ldd answer for the squeak vm? The script is trying to work out what default LD_LIBRARY_PATH to supply. The choice is between /lib/tls:/lib:/usr/lib/tls:/usr/lib and /lib:/usr/lib. Older linuxes had two versions of the pthreads library, an experimental one (now the default) using thread-local storage (hence tls) and an older one, I think using the stack pointer to identify the current thread. The script uses ldd to find out what the VM is linked against. Here's the code: # On some old linuxes there is a /lib/tls thread-local-storage version of the C # library which the VM may use and if so should take precedence over /lib libc. case `/usr/bin/ldd "$BIN/squeak" | /bin/fgrep /libc. | sed 's/^.*=> //'` in /lib/tls/libc*) SVMLLP="/lib/tls:/lib:/usr/lib/tls:/usr/lib";; /lib/libc*) SVMLLP="/lib:/usr/lib";; *) echo "can't infer base LD_LIBRARY_PATH. Aborting." 1>&2; exit 1 esac On Tue, Dec 27, 2011 at 9:51 AM, Philippe Marschall <kustos(a)gmx.net> wrote: > Hi > > With the latest cog VMs I get: > > can't infer base LD_LIBRARY_PATH. Aborting. > > Gentoo 64bit, Linux 3.1 > > LD_LIBRARY_PATH is not set but libtool is present. > > Cheers > Philippe > > > -- best, Eliot

Dec. 27, 2011

[Pharo-project] can't infer base LD_LIBRARY_PATH. Aborting.

by Philippe Marschall

Hi With the latest cog VMs I get: can't infer base LD_LIBRARY_PATH. Aborting. Gentoo 64bit, Linux 3.1 LD_LIBRARY_PATH is not set but libtool is present. Cheers Philippe

Dec. 27, 2011

Re: [Pharo-project] Programmatic creation of classes Re: [ANN][Beta Release] DBXTools - Magritte Scaffolding - Phoseydon

by Guillermo Polito

DBXMagritteClassBuilderVisitor>>buildAccessorForAttribute: attribute inClass: aClass > could you give me a pointer to read up on what the curly-brace-numbers are? > The curly braces in the string are just for string formatting, the second one is a dynamic array: 'hello {1}! How's your {2}?' format: { 'Ben' 'metaprogramming' }. What that method does is creating the accessors, ending up in a final string like this for the mutator: 'someAccessor: aParam instVar := aParam' > > btw, Can you advise how is the compiling activated? If you had about 800 > classes that need to be created, would temporarily turning off the compile > assist the program execution time significantly? > I cannot give a good answer here. Maybe someone has better knowledge on the compiler and it's settings. Maybe creating a giant script with all the class creations is faster than 800 scripts. Or Maybe not significantly :P. Maybe disabling the UI notification enhances it a little bit. I've never worried about it a lot :P. Cheers, Guille

Dec. 27, 2011

Re: [Pharo-project] System Browser - shifting class comments into pane beneath class definition

by Ben Coman

Mariano Martinez Peck wrote: > Hi Ben. For me it would make a lot of sense. I have already argued several > times that "?" is amazingly hidden. NOBODY can see it. You must have a 5 > years Pharo degree to find it ;) > > The problem is OB, I am not sure if it is supported by OB and if it is, I > don't know how to enable it. Moreover, if you take a look to Pharo 1.4 > (where OB is not present), you will find that the System Browser in there, > has what you want. > Good to know. thanks Mariano. Pharo 1.4 is just what I was thinking, except that when the window is resized, rather than holding the split at 50%, I'd prefer the class definition pane size held fixed and let more of the class comment show. > On Tue, Dec 27, 2011 at 4:04 PM, Ben Coman <btc(a)openinworld.com> wrote: > >> I could not find any previous discussion of this so I'll just throw this >> out there... >> >> I find that I forget to click the "?" button to read the class comments as >> I browse around the system. They are "kind of" hidden. Perhaps also I stop >> looking because they are often empty. If others do the same then it is a >> downward spiral... if it seems no one looks at them then no one values >> writing them. (Except for the good work of the COTDC) >> >> However System Browser opens by default with 16 lines. >> Most class definitions (eg [1]) use 6 lines, which leaves a lot of unused >> whitespace. >> I wonder if the class comment was to appear (in a separate pane) beneath >> the class definition then the class definition would be filled out more >> often. Bring the class comments to the fore and perhaps they'll be valued >> more. At least for me, it will ensure that I don't miss where there are >> good comments I can learn from. Also more attention may focus on the poor >> comments due to being more exposed. >> just a random thought... >> cheers, Be

Dec. 27, 2011

[Pharo-project] Programmatic creation of classes Re: [ANN][Beta Release] DBXTools - Magritte Scaffolding - Phoseydon

by Ben Coman

Guillermo Polito wrote: > *What does Phoseydon beta includes?* > By now, it reads a database schema, builds a model description from it and > then allows us to: > - write classes from that description. > I am very interested in the programmatic creation of classes. Although my application is not quite related to database schema. Instead I have a UML model stored in a database basically defined by two tables t_objects and t_connector from which I need to create Smalltalk classes, instance variables, and association & aggregate relationships. I tracked down a lot of what I want to the methods... DBXMagritteClassBuilderVisitor>>visitDBXEntity: aDBXEntity for which the answer was obvious in hindsight - just replicating the call to Object>>subclass::: DBXMagritteClassBuilderVisitor>>buildAccessorForAttribute: attribute inClass: aClass could you give me a pointer to read up on what the curly-brace-numbers are? btw, Can you advise how is the compiling activated? If you had about 800 classes that need to be created, would temporarily turning off the compile assist the program execution time significantly? regards, Ben

Dec. 27, 2011

Re: [Pharo-project] System Browser - shifting class comments into pane beneath class definition

by StÃ©phane Ducasse

This is good point. we should give more space to class comments. On Dec 27, 2011, at 4:04 PM, Ben Coman wrote: > > I could not find any previous discussion of this so I'll just throw this out there... > > I find that I forget to click the "?" button to read the class comments as I browse around the system. They are "kind of" hidden. Perhaps also I stop looking because they are often empty. If others do the same then it is a downward spiral... if it seems no one looks at them then no one values writing them. (Except for the good work of the COTDC) > > However System Browser opens by default with 16 lines. > Most class definitions (eg [1]) use 6 lines, which leaves a lot of unused whitespace. > I wonder if the class comment was to appear (in a separate pane) beneath the class definition then the class definition would be filled out more often. Bring the class comments to the fore and perhaps they'll be valued more. At least for me, it will ensure that I don't miss where there are good comments I can learn from. Also more attention may focus on the poor comments due to being more exposed. > just a random thought... > cheers, Ben > > [1] Object subclass: #Behavior > uses: TPureBehavior > instanceVariableNames: 'superclass methodDict format' > classVariableNames: 'ObsoleteSubclasses' > poolDictionaries: '' > category: 'Kernel-Classes' > > >

Dec. 27, 2011

Re: [Pharo-project] System Browser - shifting class comments into pane beneath class definition

by Mariano Martinez Peck

Hi Ben. For me it would make a lot of sense. I have already argued several times that "?" is amazingly hidden. NOBODY can see it. You must have a 5 years Pharo degree to find it ;) The problem is OB, I am not sure if it is supported by OB and if it is, I don't know how to enable it. Moreover, if you take a look to Pharo 1.4 (where OB is not present), you will find that the System Browser in there, has what you want. Cheers On Tue, Dec 27, 2011 at 4:04 PM, Ben Coman <btc(a)openinworld.com> wrote: > > I could not find any previous discussion of this so I'll just throw this > out there... > > I find that I forget to click the "?" button to read the class comments as > I browse around the system. They are "kind of" hidden. Perhaps also I stop > looking because they are often empty. If others do the same then it is a > downward spiral... if it seems no one looks at them then no one values > writing them. (Except for the good work of the COTDC) > > However System Browser opens by default with 16 lines. > Most class definitions (eg [1]) use 6 lines, which leaves a lot of unused > whitespace. > I wonder if the class comment was to appear (in a separate pane) beneath > the class definition then the class definition would be filled out more > often. Bring the class comments to the fore and perhaps they'll be valued > more. At least for me, it will ensure that I don't miss where there are > good comments I can learn from. Also more attention may focus on the poor > comments due to being more exposed. > just a random thought... > cheers, Ben > > [1] Object subclass: #Behavior > uses: TPureBehavior > instanceVariableNames: 'superclass methodDict format' > classVariableNames: 'ObsoleteSubclasses' > poolDictionaries: '' > category: 'Kernel-Classes' > > > > -- Mariano http://marianopeck.wordpress.com

Dec. 27, 2011

[Pharo-project] System Browser - shifting class comments into pane beneath class definition

by Ben Coman

I could not find any previous discussion of this so I'll just throw this out there... I find that I forget to click the "?" button to read the class comments as I browse around the system. They are "kind of" hidden. Perhaps also I stop looking because they are often empty. If others do the same then it is a downward spiral... if it seems no one looks at them then no one values writing them. (Except for the good work of the COTDC) However System Browser opens by default with 16 lines. Most class definitions (eg [1]) use 6 lines, which leaves a lot of unused whitespace. I wonder if the class comment was to appear (in a separate pane) beneath the class definition then the class definition would be filled out more often. Bring the class comments to the fore and perhaps they'll be valued more. At least for me, it will ensure that I don't miss where there are good comments I can learn from. Also more attention may focus on the poor comments due to being more exposed. just a random thought... cheers, Ben [1] Object subclass: #Behavior uses: TPureBehavior instanceVariableNames: 'superclass methodDict format' classVariableNames: 'ObsoleteSubclasses' poolDictionaries: '' category: 'Kernel-Classes'

Dec. 27, 2011