July 2010 - Pharo-dev - lists.pharo.org

Re: [Pharo-project] Symbol >> #asClass ?

by StÃ©phane Ducasse

>> >>>> 2. We are trying to remove all direct references to the Smalltalk >>>> namespace, this however hardcodes the Smalltalk namespace again. >>> >>> self environment can solve this. >> >> No, what will #asClass return if you have different instances of Smalltalk? > > A Class or nil. what lukas wants to say is that using asClass you cannot write code that can used on different instances of systemDictionary and this was one of the goal of a lot of cleaning. Stef

July 22, 2010

Re: [Pharo-project] Symbol >> #asClass ?

by StÃ©phane Ducasse

indeed this is a nice point. >>> 1. #World asClass does not return a class. >> >> #classNamed: can solve this. > > No, it still returns an instance of PasteUpMorph. > >>> 2. We are trying to remove all direct references to the Smalltalk >>> namespace, this however hardcodes the Smalltalk namespace again. >> >> self environment can solve this. > > No, what will #asClass return if you have different instances of Smalltalk? > > Lukas

July 22, 2010

Re: [Pharo-project] How disable debugger?

by StÃ©phane Ducasse

specializing UImanager is not enough? On Jul 22, 2010, at 4:31 PM, Mariano Martinez Peck wrote: > > > 2010/7/22 Denis Kudriashov <dionisiydk(a)gmail.com> > Yes, I can do this. > > But I hope there is existed solution. > > I was think scriptLoader>>cleanForProduction do this (disable debugger). > > > I think I don't agree. That is not cleaning, but a matter of bussines. I know a lot of seaside apps running headless with RFB...in that case I do want a debugger. > > But I agree, it would be good to have a Strategy and 2 implementation. DebuggerUnhandledError and TranscriptUhandledError. You put the first as default. > Then, you put the straegy as an (class?) instVar of UnhandledError. you define an api method and implement it in both. > > Then, at runtime, in your app you can set that insVar and you are done. In addition, it can be added to Core and you create a setting that let you choose the strategy. > > Cheers > > Mariano > > 2010/7/22 StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> > > have a look at the handler of > > UnhandledError>>defaultAction > > ^ UIManager default unhandledErrorDefaultAction: self exception > > so you can define a different UIManager for example > > Stef > On Jul 22, 2010, at 1:18 PM, Denis Kudriashov wrote: > > > Hello > > > > Is it possible write all exceptions to log and dont show debugger? > > How I can do this? > > _______________________________________________ > > Pharo-project mailing list > > Pharo-project(a)lists.gforge.inria.fr > > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

July 22, 2010

Re: [Pharo-project] State of Sound in Morph

by Mariano Martinez Peck

2010/7/22 Gary Chambers <gazzaguru2(a)btinternet.com> > Do we have anyone responsibly for these kind of "near core" packages? > > The idea behind ConfigurationOfPharoNonCore is that WE (pharo developers) are the maintainers, and not "external maintainers". So, this is the difference between this PharoNonCore and PharoDev. In PharoDev we have a lot of development tools, and mostly, maintained by external people. PharoNonCore should be part as Pharo, but that shouldn't be part of the core. However, it is maitained by the Pharo developers. Cheers Mariano > Regards, Gary > > ----- Original Message ----- > *From:* Mariano Martinez Peck <marianopeck(a)gmail.com> > *To:* Pharo-project(a)lists.gforge.inria.fr > *Sent:* Thursday, July 22, 2010 3:40 PM > *Subject:* Re: [Pharo-project] State of Sound in Morph > > > > On Thu, Jul 22, 2010 at 3:45 PM, Adrian Lienhard <adi(a)netstyle.ch> wrote: > >> >> On Jul 22, 2010, at 14:42 , StÃ©phane Ducasse wrote: >> >> > probably because we were stupid to remove it from the image. >> >> I don't think so. >> >> Rather because we have: >> - nobody who takes responsibility to maintain this package >> - no tests for this package >> - no build and test server to detect problems automatically >> >> > I think it is a combination of the 2. First, we are not stupid to remove it > from core. They HAVE to be removed and put it as separate package. > > Then, I agree with ALL adrian comments. But I also agree that this problem > can be minimized. When you remove methods/classes in PharoCore, it would be > good to search them in "Pharo". That's why I did > ConfigurationOfPharoNonCore. > > Cheers > > Mariano > > >> Cheers, >> Adrian >> >> BTW, the two expressions posted below work for me on a recent Mac VM in a >> Pharo 1.0 image. >> >> >> > If you can check there are two fixes on the bug archives and it would be >> great that somebody compares with squeak >> > version. I did not see that much changes from this part of the system on >> their side. >> > >> > Stef >> > On Jul 22, 2010, at 2:36 PM, Hilaire Fernandes wrote: >> > >> >> I tried the Pharo sound package, so far it looks sound is not working. >> >> Is it true or is it a mistake of myself. >> >> >> >> For example: >> >> >> >> No sound output when executing: >> >> FMSound bass1 play >> >> >> >> DNU FMSound copy when executing: >> >> (FMSound lowMajorScaleOn: FMSound bass1) play >> >> >> >> Hilaire >> >> >> >> >> >> _______________________________________________ >> >> Pharo-project mailing list >> >> Pharo-project(a)lists.gforge.inria.fr >> >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >> > >> > >> > _______________________________________________ >> > Pharo-project mailing list >> > Pharo-project(a)lists.gforge.inria.fr >> > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >> >> >> _______________________________________________ >> Pharo-project mailing list >> Pharo-project(a)lists.gforge.inria.fr >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >> > > ------------------------------ > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >

July 22, 2010

Re: [Pharo-project] Literal array with evaluated value of stmts, vars etc..

by Eliot Miranda

2010/7/22 Sudhakar Krishnamachari <skrishnamachari(a)gmail.com> > And of course the extensibility it gives: > > [ 1->'asd'. 2->'asd'. 3->'dsd' ] asDictionary > I think Travis Griggs' once is much much nicer: [{ 1->'asd'. 2->'asd'. 3->'dsd' } asDictionary] once when once is sent to a BlockClosure it becomes itself into a CachedBlockClosure that remembers and answers the evaluated value. You could add selectors to this such as onceEachSession - destroy the cached value on image startup It doesn't require any special compiler machinery (unlike ##(...)) and so there are no problems with browsing references or senders within compile-time evaluated forms. It is easily extensible to different life-times. It is portable. my 2Â¢ Eliot > > ... > > > On Thu, Jul 22, 2010 at 7:48 PM, Sudhakar Krishnamachari < > skrishnamachari(a)gmail.com> wrote: > >> * >> *>* > It's different because evaluated only once, either at method compilation >> *>* or at first execution (lazy initialization). >> * >> >> > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >

July 22, 2010

Re: [Pharo-project] about hex

by Eliot Miranda

On Wed, Jul 21, 2010 at 9:59 AM, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr > wrote: > > On Jul 21, 2010, at 6:49 PM, Nicolas Cellier wrote: > > > I don't think Eliot changed that, he only changed Integer>>hex... > > No he did > http://code.google.com/p/pharo/issues/detail?id=2705 > > Item was changed: > ----- Method: Character>>hex (in category 'printing') ----- > hex > + ^value printStringBase: 16! > - ^value hex! > and that is a mistake. It should stay the way it was and then Character hex would also have the 16r prefix. But I'm caught between two stools, whether to move forward and make Character hex answer 16rXX or to move back and make Integer hex answer XX, and e.g. use phex for 16r... Our Teleplace images make heavy use of hex and assume it answers 16rXX so if we don't keep the 16r prefix I'll have to maintain a fork of the Cog VMMaker and that would make life even more difficult (I already have three forks, open-source Cog, production Teleplace and development Teleplace (threaded & separate ObjectMemory and Interpreter) but they all agree w.r.t. hex. If you think adding the prefix to Character won't do too much harm then revert the change above so Character hex reads (as it does in our Teleplace images) Character hex ^value hex > > It seems that every implementation of hex did avoid to print the radix in > Squeak 3.10. Was it intentional ? > > > > Anyway, it will be hard to reach homogeneity because: > > Integer>>hex should print the radix for VMMaker compatibility > > ByteArray>>hex shouldn't for cypto compatibility > > but crypto could use printStringHex so yes we can reach it. > > > > > Nicolas > > > > 2010/7/21 StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> > > Levente /Eliot > > > > I noticed that > > > > $A hex on squeak > > -> 41 > > and in pharo > > -> '16r41' > > > > I would like to have both consistent what do you think to have > > > > $A hex > > -> '16r41' > > > > $A printStringHex > > -> '41' > > > > like that this is consistent with the behavior on integer. > > Eliot I saw that to change the behavior on characters and I would like to > know why. > > I thought that it was for debugging purpose. > > > > Stef > > > > > > > > _______________________________________________ > > Pharo-project mailing list > > Pharo-project(a)lists.gforge.inria.fr > > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > > > _______________________________________________ > > Pharo-project mailing list > > Pharo-project(a)lists.gforge.inria.fr > > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >

July 22, 2010

Re: [Pharo-project] Mac like editing?

by Gary Chambers

I believe some of us are working on soting out the mess... Perhaps a thread on "standardised keyboard control" would be in order? Regards, Gary ----- Original Message ----- From: Max Leske To: pharo-project(a)lists.gforge.inria.fr Sent: Thursday, July 22, 2010 4:58 PM Subject: [Pharo-project] Mac like editing? Hi everyone. I have no idea what the status is on the editor in pharo and if there are any plans to do something to / with it. If there's some resource, could you point me to it? In the meantime I thought I'd share my impressions with you; doesn't hurt at least. Since I started developing in Pharo I've struggled with the editor, in particular with the fact that I need to use the mouse a lot to position the cursor which, as all of you know, makes development quite a bit slower. Before I came to Pharo I used to develop in Eclipse and Eclipse supports all shortcuts used inside OS X (and if not you can change the shortcut settings). My favorite shortcuts: a.. cmd + -> : go to end of line b.. cmd + <- : go to beginning of line c.. alt + -> : go to end of the current word d.. alt + -> : go to beginning of current word e.. alt + arrow up : switch the line above with the current line (current line has line number -1, line above has line number +1) f.. alt + arrow down : switch the line below with the current line g.. cmd + l : delete line (I know this shortcut is used to cancel changes at the moment. Still, I haven't found a shortcut yet that deletes a single line) I'd LOVE to see this in Pharo, or at least be able to configure the shortcuts myself. I think, being able to change the shortcuts is important anyway because each developer has an own style. I'd try to come up with a solution myself but I just don't have the time at the moment. Thanks for listening to my yapping :) Max ------------------------------------------------------------------------------ _______________________________________________ Pharo-project mailing list Pharo-project(a)lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

July 22, 2010

Re: [Pharo-project] State of Sound in Morph

by Gary Chambers

Do we have anyone responsibly for these kind of "near core" packages? Regards, Gary ----- Original Message ----- From: Mariano Martinez Peck To: Pharo-project(a)lists.gforge.inria.fr Sent: Thursday, July 22, 2010 3:40 PM Subject: Re: [Pharo-project] State of Sound in Morph On Thu, Jul 22, 2010 at 3:45 PM, Adrian Lienhard <adi(a)netstyle.ch> wrote: On Jul 22, 2010, at 14:42 , StÃ©phane Ducasse wrote: > probably because we were stupid to remove it from the image. I don't think so. Rather because we have: - nobody who takes responsibility to maintain this package - no tests for this package - no build and test server to detect problems automatically I think it is a combination of the 2. First, we are not stupid to remove it from core. They HAVE to be removed and put it as separate package. Then, I agree with ALL adrian comments. But I also agree that this problem can be minimized. When you remove methods/classes in PharoCore, it would be good to search them in "Pharo". That's why I did ConfigurationOfPharoNonCore. Cheers Mariano Cheers, Adrian BTW, the two expressions posted below work for me on a recent Mac VM in a Pharo 1.0 image. > If you can check there are two fixes on the bug archives and it would be great that somebody compares with squeak > version. I did not see that much changes from this part of the system on their side. > > Stef > On Jul 22, 2010, at 2:36 PM, Hilaire Fernandes wrote: > >> I tried the Pharo sound package, so far it looks sound is not working. >> Is it true or is it a mistake of myself. >> >> For example: >> >> No sound output when executing: >> FMSound bass1 play >> >> DNU FMSound copy when executing: >> (FMSound lowMajorScaleOn: FMSound bass1) play >> >> Hilaire >> >> >> _______________________________________________ >> Pharo-project mailing list >> Pharo-project(a)lists.gforge.inria.fr >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project _______________________________________________ Pharo-project mailing list Pharo-project(a)lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project ------------------------------------------------------------------------------ _______________________________________________ Pharo-project mailing list Pharo-project(a)lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

July 22, 2010

[Pharo-project] Mac like editing?

by Max Leske

Hi everyone. I have no idea what the status is on the editor in pharo and if there are any plans to do something to / with it. If there's some resource, could you point me to it? In the meantime I thought I'd share my impressions with you; doesn't hurt at least. Since I started developing in Pharo I've struggled with the editor, in particular with the fact that I need to use the mouse a lot to position the cursor which, as all of you know, makes development quite a bit slower. Before I came to Pharo I used to develop in Eclipse and Eclipse supports all shortcuts used inside OS X (and if not you can change the shortcut settings). My favorite shortcuts: cmd + -> : go to end of line cmd + <- : go to beginning of line alt + -> : go to end of the current word alt + -> : go to beginning of current word alt + arrow up : switch the line above with the current line (current line has line number -1, line above has line number +1) alt + arrow down : switch the line below with the current line cmd + l : delete line (I know this shortcut is used to cancel changes at the moment. Still, I haven't found a shortcut yet that deletes a single line) I'd LOVE to see this in Pharo, or at least be able to configure the shortcuts myself. I think, being able to change the shortcuts is important anyway because each developer has an own style. I'd try to come up with a solution myself but I just don't have the time at the moment. Thanks for listening to my yapping :) Max

July 22, 2010

Re: [Pharo-project] Pharo-project Digest, Vol 27, Issue 118

by Norbert Hartl

On 20.07.2010, at 19:15, Larry Gadallah wrote: > On 19 July 2010 12:16, <pharo-project-request(a)lists.gforge.inria.fr> wrote: >> Message: 6 >> Date: Mon, 19 Jul 2010 21:15:54 +0200 >> From: Norbert Hartl <norbert(a)hartl.name> >> Subject: Re: [Pharo-project] XML Demarshalling >> To: Pharo-project(a)lists.gforge.inria.fr >> Message-ID: <6BCBFCA9-367D-4848-8D4C-9EA724017B99(a)hartl.name> >> Content-Type: text/plain; charset=us-ascii >> >> >> On 16.07.2010, at 18:46, Larry Gadallah wrote: >> >>> Hi all: >>> >>> I am a noob searching for an example of how to demarshall a set of >>> simple and complex XML nodes into an object using Pastell or something >>> similar. I have found many examples of how to pick out all of the >>> elements of a certain type into a collection, but I'm trying to grab >>> the data from a mixed collection of (~20) elements that are children >>> of a particular element type and unpack them into an object. >>> >>> Does anyone know of some good examples of how to do this using Pastell >>> or something similar? >>> >> Can you give more detailled information about how the XML maps onto the data model? I mean if you have childr elements will that map to objects that are referenced similar? Do attributes map to variables and such? Or do you want to freely define how the structure of the XML resembles into your objects? >> >> Norbert > > Hi Norbert: > > Alas, I am laboring under the burden of being a newbie to both Pharo > and XML. It seems to me that it would be better to be able to > dynamically pick up all of the child elements of an element of > interest, however, I can't imagine how to define an object that > doesn't know it's own contents until run-time. In addition, in my > case, some of the child elements are complex (i.e. they also contain > children) which complicates the traversal a little bit. I'm assuming > for now that I will have to read the source XML, define the instance > variables of my object based on them, and simply populate them at run > time by traversing the XML tree. If the XML model changes, the object > definition will have to change too. > Hi again, just yesterday I discovered that I have a similar problem to solve. Usually I would try to solve it with magritte but this time I need some quick and dirty hack to achieve the goal. And somehow I like Pastell. Let's take an XML Snippet <Root> <Name>Mister RootNode</Name> <SubNode> <Name>Mister SubNode</Name> </SubNode> </Root> And now let's assume you want everything under Root in an object of class RootClass. In the case of SubNode you want a new object from class SubNodeClass that is referenced from RootClass object. In this example both classes RootClass and SubNodeClass are subclasses of NodeBaseClass NodeBaseClass class>>node: aNode ^ self new node: aNode NodeBaseClass>>node ^ node NodeBaseClass>>node: aNode node := aNode NodeBaseClass>>at: aSymbol ^ (self class xmodel at: aSymbol) value: self node Here is the RootClass: RootClass class>>xmodel ^ Dictionary new; at: #name put: [:node| (node / #Name ) contentString ]; at: #subObject put: [:node| SubNodeClass node: (node / #SubNode) first ]; yourself and another class that is referenced from RootClass. Btw. "(node / #Name)" is using Pastell SubNodeClass class>>xmodel ^ Dictionary new; at: #name put: [:node| (node / #Name ) contentString ]; yourself Then you could do: document := XMLDOMParser parse.... rootObject := RootClass node: document root. Transcript show: ((rootObject at: #subObject) at: #name and you should get "Mister SubNode" . I hope there aren't too much errors in there. The basic idea is that every object gets an XML node and a description how to retrieve the values relative to this node. In the case of #name just a string is received. In the case of #subObject you can see how this gets recursive. #subObject returns itself an object that is just initialized with an XML node. Just like the rootObject This is just a quick way to get some objects from an XML file. There is much room for improvement. I just wanted to give you an idea how you could solve it. Hope this helps, Norbert

July 22, 2010