December 2014 - Pharo-dev - lists.pharo.org

Re: [Pharo-dev] Duration year

by Chris Muller

> I was working on a GenericYear that was intended to be used for DateAndTime > calculus - to be able to add and subtract an arbitrary number of years. The > class by itself would not know the days - it is an arbitrary (generic) Year. > > It would only know the number of days when it is added (or subtracted) from > an actual date (or timespan). The idea was that it would have a very > specific and limited scope of responsibility. I liked it. Hopefully a few loose ends can be fixed up so it can be put into trunk.

Dec. 9, 2014

Re: [Pharo-dev] [squeak-dev] Does Monticello try to install shared pools before methods using them?

by Eliot Miranda

Hi All, Hi Bert, On Mon, Dec 8, 2014 at 12:02 PM, Chris Muller <asqueaker(a)gmail.com> wrote: > I'm pretty sure the responsibility falls to the loader. I'm not too > familiar with it but if I wanted to be I would probably start tracing > from McVersionLoader. > +1. Craig, I thought that Monticello had been modified to load additions first, then changes, then removals. But when I read the code I can't see this sorting in trunk. Bert, what's your recollection? And if you agree that Monticello does this, where's the code? I see that MCPatch>>applyTo: doesn't do anything special and when it appears to collect its definitions in MCPatch>>initializeWithBase:target: there is no sorting going on. What am I missing? > > Do your pool references end up being valid? That is the most > important but I agree it would be an annoyance to see those messages > in the Transcript. > Hi Craig, > > On Mon, Dec 8, 2014 at 7:24 AM, Craig Latta <craig(a)netjam.org> wrote: > > > > Hoi-- > > > > I'm attempting to create a Monticello package which provides a > > shared pool (as a subclass of SharedPool) and methods which use its > > variables. On installation, does Monticello try to ensure that a pool is > > installed before methods which use it? When I install the package in a > > fresh system, I get messages on the Transcript complaining that shared > > variables are undefined. > As I said above I thought Monticello implemented that sensible load order of add, change, remove, in which case poll definitions should be included in additions, and before any methods. But I can't see where in current trunk this is happening. AFAIC, not loading in this order is a bug. We need as predictable aand useful a general load order as possible, and it should be straight-forward to arrange that pools are added early. > > > Is there some way one fiddles with the writing process, or with the > > package after it is written, to avoid ordering problems? > > > > > > thanks! > > > > -C > > > > -- > > Craig Latta > > netjam.org > > +31 6 2757 7177 (SMS ok) > > + 1 415 287 3547 (no SMS) > > > > > > -- best, Eliot

Dec. 8, 2014

Re: [Pharo-dev] [Esug-list] [Pharo-users] [ann] gtspotter: a moldable interface for spotting objects

by Tudor Girba

Thanks for the suggestions, Johan. I will change the text and we will definitely look into making the shortcuts more apparent. Cheers, Doru On Mon, Dec 8, 2014 at 5:17 PM, Johan Fabry <jfabry(a)dcc.uchile.cl> wrote: > > OK now I get it â¦ thanks for clarifying! > I think my confusion stems from the use of âcategory' and then talking > about 'the collection objectâ, which made me think about collections and > source code and protocols and packages and I got lost. > > I think that the sentence of the blog would be clearer like this > (uppercase to show changes): > > To do this, GTSpotter offers an extra action: diving in a RESULTS > category. Pressing Cmd+Shift+ArrowRight dives in the collection OF RESULTS > OF that category. Thus, we can continue refining the search inside the > category. > > So then, my suggestion for a legend at the bottom of the results list > would be: > Cmd+Shift+ArrowUp/ArrowDown = Next/Prev category, Cmd+RightArrow = Dive > into Result, Cmd+Shift+RightArrow = Dive into Category > > > On Dec 8, 2014, at 12:48, Tudor Girba <tudor(a)tudorgirba.com> wrote: > > > > Hi, > > > > Ok. Let's take it step by step and see if we cannot find a better way of > explaining. > > > > Take a look at the first picture in the blog post. Entering GTSpo shows > results both for Classes and for Packages. These "Classes" and "Packages" > are what we call search categories, and they have associated a query > processor that can populate them with results (see the "Spotting your > objects" section from the bottom of the post). > > > > In our case, we get 39 classes (of which only 5 are shown) and 1 package > that match the query. If you want to look at all those 39 classes, you can > dive in the whole collection behind the category in a separate step. This > is achieved through Cmd+Shift+ArrowRight. Does it make more sense now? > > > > I did not consider the category to be confusing. Would you propose > another name? > > > > Cheers, > > Doru > > > > > > On Mon, Dec 8, 2014 at 4:02 PM, Johan Fabry <jfabry(a)dcc.uchile.cl> > wrote: > > Sorry, but no :-( > > > > I am always confused when people say âcategoryâ because the word has so > many overloaded meanings. The same happens in the blog post, it is not > clear to me what category means here, and what does it have to do with the > collection object? > > > > > On Dec 7, 2014, at 11:16, Tudor Girba <tudor(a)tudorgirba.com> wrote: > > > > > > Hi, > > > > > > Yes, we will still evolve the UI. At the very least you will get the > shortkeys directly on the actions. > > > > > > The answer to your question is in the blog post: > > > GTSpotter offers an extra action: diving in a category. Pressing > Cmd+Shift+ArrowRight dives in the collection object containing only the > items from that category. Thus, we can continue refining the search inside > the category. > > > > > > So, you will open the collection of that sub-category and you will see > more items at once (not just 5). Is it clearer now? > > > > > > Cheers, > > > Doru > > > > > > > > ---> Save our in-boxes! http://emailcharter.org <--- > > > > Johan Fabry - http://pleiad.cl/~jfabry > > PLEIAD lab - Computer Science Department (DCC) - University of Chile > > > > > > > > > > -- > > www.tudorgirba.com > > > > "Every thing has its own flow" > > > > ---> Save our in-boxes! http://emailcharter.org <--- > > Johan Fabry - http://pleiad.cl/~jfabry > PLEIAD lab - Computer Science Department (DCC) - University of Chile > > -- www.tudorgirba.com "Every thing has its own flow"

Dec. 8, 2014

Re: [Pharo-dev] Investigating MultiByteFileStream for adding tests

by Sven Van Caekenberghe

> On 08 Dec 2014, at 22:27, sheridan(a)sheridan-mahoney.com wrote: > > Hi, > I've been looking into MultiByteFileStream for adding tests in this area, and have come up against what appears to me to be an anomaly. I ran the following 4 tests: > > > FileStream fileNamed: 'test.txt' > do: [ :str | str > nextPutAll: '123'; > crlf ]. > > MultiByteFileStream fileNamed: 'test2.txt' > do: [ :str | str > nextPutAll: '123'; > crlf ]. > > FileStream fileNamed: 'test3.txt' > do: [ :str | str > nextPutAll: 'Ã©lÃ¨ve en FranÃ§ais'; > crlf ]. > > MultiByteFileStream fileNamed: 'test4.txt' > do: [ :str | str > nextPutAll: 'Ã©lÃ¨ve en FranÃ§ais'; > crlf ]. > > > The results are: > > Input string Result Unix File > Class Class Type > --------------------------------------------------------------- > FileStream '123' MultiByte- ASCII text > FileStream > > MultiByte- '123' MultiByte- ASCII text > FileStream FileStream > > FileStream 'Ã©lÃ¨ve en MultiByte- UTF-8 Unicode > FranÃ§ais' FileStream > > MultiByte- 'Ã©lÃ¨ve en MultiByte- UTF-8 Unicode > FileStream FranÃ§ais' FileStream > > > Do operations involving FileStream as input always return a MultiByteFileStream? Yes. They also use UTF-8 by default (ASCII is a proper subset of UTF-8). > Thanks, > Sheridan >

Dec. 8, 2014

Re: [Pharo-dev] playground vs workspace

by Tudor Girba

Hi everyone, My intention was to have a discussion based on arguments and to ask for permission to change something I believe in. If I give the impression that I want to name for the sake of naming, I am sorry. And I am sorry I had to disturb so many people with my infatuation. At the same time, I am also quite abated by the fact that too many people look only at what exists and not invest trust in the commitment behind the effort. I will stop responding to this thread now. Thanks for the feedback. Cheers, Doru On Mon, Dec 8, 2014 at 9:30 PM, HernÃ¡n Morales Durand < hernan.morales(a)gmail.com> wrote: > Hello Doru, > > I think you have a new different implementation of the classic Workspace, > for better or worst. > > The problem is everybody wants their tool be named as the highest > taxonomic rank. Compare it with naming your cat as "Cat" and not as Siamese > Felis Catus which is a specific breed. > > On the other side, what prevents for anyone else doing a "better"/new > implementation and reclaiming the Workspace name? And this could happen > every year? I don't know you but I get bored of seeing "I want that name" > discussions. > > Workspace is a type of tool, it happens that there is an implementation > which owned such name. It should be renamed as OldWorkspace, > DeveloperWorkspace, ClassicWorkspace, or whatever. > > For me, Workspace in the World menu should display a sub-menu with both > implementations. And community gains a place to display/register more > implementations discriminated by its type. > > HernÃ¡n > > > > > > > 2014-12-08 3:18 GMT-03:00 Tudor Girba <tudor(a)tudorgirba.com>: > > Hi, >> >> As you might see, the GTPlayground is called playground not workspace. >> The main reason for this is that workspace implies the place where work is >> done, and work is typically associated with creating code. >> >> This is not the intention of the workspace. >> >> Workspace also comes with a history that is associated with actual >> implementations. While the current GTPlayground might appear to be similar >> to the workspace, I think the way you can use the latter produces a >> significant shift. >> >> Furthermore, I think the term playground is more inline with the idea of >> "playing with live objects". And, in the future, the implementation will >> likely evolve towards even more liveliness. >> >> It is for this reason that I would prefer that the World menu item gets >> renamed to Playground. >> >> What do you think? >> >> Cheers, >> Doru >> >> -- >> www.tudorgirba.com >> >> "Every thing has its own flow" >> > > -- www.tudorgirba.com "Every thing has its own flow"

Dec. 8, 2014

Re: [Pharo-dev] Investigating MultiByteFileStream for adding tests

by Stephan Eggermont

Sheridan wrote: >Do operations involving FileStream as input always return a >MultiByteFileStream? That is a well-known pattern, class cluster. A base class defining an interface and acting as a factory hiding the specific class returned. Cocoa uses that a lot, it results in a much cleaner architecture than that of .net and java. https://developer.apple.com/library/ios/documentation/General/Conceptual/Co… Stephan

Dec. 8, 2014

Investigating MultiByteFileStream for adding tests

by sheridan＠sheridan-mahoney.com

Hi, I've been looking into MultiByteFileStream for adding tests in this area, and have come up against what appears to me to be an anomaly. I ran the following 4 tests: FileStream fileNamed: 'test.txt' do: [ :str | str nextPutAll: '123'; crlf ]. MultiByteFileStream fileNamed: 'test2.txt' do: [ :str | str nextPutAll: '123'; crlf ]. FileStream fileNamed: 'test3.txt' do: [ :str | str nextPutAll: 'Ã©lÃ¨ve en FranÃ§ais'; crlf ]. MultiByteFileStream fileNamed: 'test4.txt' do: [ :str | str nextPutAll: 'Ã©lÃ¨ve en FranÃ§ais'; crlf ]. The results are: Input string Result Unix File Class Class Type --------------------------------------------------------------- FileStream '123' MultiByte- ASCII text FileStream MultiByte- '123' MultiByte- ASCII text FileStream FileStream FileStream 'Ã©lÃ¨ve en MultiByte- UTF-8 Unicode FranÃ§ais' FileStream MultiByte- 'Ã©lÃ¨ve en MultiByte- UTF-8 Unicode FileStream FranÃ§ais' FileStream Do operations involving FileStream as input always return a MultiByteFileStream? Thanks, Sheridan

Dec. 8, 2014

Re: [Pharo-dev] playground vs workspace

by Tudor Girba

Hi, We do not yet have an IDE (where I stands for integrated). We have a set of tools and some of them are quite strong. But, we do not have an overall experience yet. Yes, the underlying language model does offer a beautiful experience, but the experience of the tools is far from that. We need tools that are not just powerful in the hands of few specialists, but tools that are beautiful and pleasurable to use in the hands of many. This is the goal. When we will get that, I believe we will change the game. Cheers, Doru On Mon, Dec 8, 2014 at 9:36 PM, kilon alios <kilon.alios(a)gmail.com> wrote: > "Our "battle" is with the non-Smalltalk world. This is what we need to > convince. And that world does not care about the history of Smalltalk. It > cares about how we can solve their problem. That's all." > > But you already offer an impressive arsenal of tools. The world can > complain for several things but what it cannot complain is that Pharo as an > IDE is not really powerful as a whole. Your competition would be I assume > specialised tools that come as third party. There is no IDE I know out > there that offers the level of refactoring of Pharo, in terms of tools and > libraries. > > Also we should not forget that those tools are not just powerful as is the > example of GTPlayground but are engineered in such way to cooperate which > each other with ease. While on the other hand , tools for other programming > languages are by default separate and independent and not necessarily play > nice with each other which the price one pays to keep things modular and > unattached to a specific language or IDE. So the way I see it Pharo already > is beyond competition. > > Unless of course you refer to a comparison tool per tool, which mean you > take this pharo tool and you compare with another tool for another language > out there. That case would very difficult for you because there are tons of > development tools out there. > > But the overall idea that is "hey there I give you a language, take this > enviroment too, live coding, how about an IDE and also get these coding > tools too " is really a dreamy situation already. > > The question is how badly coders want a very powerful IDE at least the > level of Pharo. > -- www.tudorgirba.com "Every thing has its own flow"

Dec. 8, 2014

Re: [Pharo-dev] playground vs workspace

by kilon alios

"Our "battle" is with the non-Smalltalk world. This is what we need to convince. And that world does not care about the history of Smalltalk. It cares about how we can solve their problem. That's all." But you already offer an impressive arsenal of tools. The world can complain for several things but what it cannot complain is that Pharo as an IDE is not really powerful as a whole. Your competition would be I assume specialised tools that come as third party. There is no IDE I know out there that offers the level of refactoring of Pharo, in terms of tools and libraries. Also we should not forget that those tools are not just powerful as is the example of GTPlayground but are engineered in such way to cooperate which each other with ease. While on the other hand , tools for other programming languages are by default separate and independent and not necessarily play nice with each other which the price one pays to keep things modular and unattached to a specific language or IDE. So the way I see it Pharo already is beyond competition. Unless of course you refer to a comparison tool per tool, which mean you take this pharo tool and you compare with another tool for another language out there. That case would very difficult for you because there are tons of development tools out there. But the overall idea that is "hey there I give you a language, take this enviroment too, live coding, how about an IDE and also get these coding tools too " is really a dreamy situation already. The question is how badly coders want a very powerful IDE at least the level of Pharo.

Dec. 8, 2014

Re: [Pharo-dev] playground vs workspace

by HernÃ¡n Morales Durand

Hello Doru, I think you have a new different implementation of the classic Workspace, for better or worst. The problem is everybody wants their tool be named as the highest taxonomic rank. Compare it with naming your cat as "Cat" and not as Siamese Felis Catus which is a specific breed. On the other side, what prevents for anyone else doing a "better"/new implementation and reclaiming the Workspace name? And this could happen every year? I don't know you but I get bored of seeing "I want that name" discussions. Workspace is a type of tool, it happens that there is an implementation which owned such name. It should be renamed as OldWorkspace, DeveloperWorkspace, ClassicWorkspace, or whatever. For me, Workspace in the World menu should display a sub-menu with both implementations. And community gains a place to display/register more implementations discriminated by its type. HernÃ¡n 2014-12-08 3:18 GMT-03:00 Tudor Girba <tudor(a)tudorgirba.com>: > Hi, > > As you might see, the GTPlayground is called playground not workspace. The > main reason for this is that workspace implies the place where work is > done, and work is typically associated with creating code. > > This is not the intention of the workspace. > > Workspace also comes with a history that is associated with actual > implementations. While the current GTPlayground might appear to be similar > to the workspace, I think the way you can use the latter produces a > significant shift. > > Furthermore, I think the term playground is more inline with the idea of > "playing with live objects". And, in the future, the implementation will > likely evolve towards even more liveliness. > > It is for this reason that I would prefer that the World menu item gets > renamed to Playground. > > What do you think? > > Cheers, > Doru > > -- > www.tudorgirba.com > > "Every thing has its own flow" >

Dec. 8, 2014