Squeak-dev July 2007

squeak-dev@lists.squeakfoundation.org

140 participants
214 discussions

RE: Win32Window how to get Squeak's window Handle? VM?
by J J 23 Jul '07

23 Jul '07

This message sounds like Ffenestri is the planned future GUI interface of Squeak. Is that so, or I read it wrong? I was under the impression that the BltBlt method was faster then using native windows (though it could be that just the wxWidget libraries are themselves slow). Is this true? Sig has mentioned he can get a 3 to 5 times speed up from BltBlt using OpenGL.Note, I'm not expressing an opinion here, just trying to understand.Thanks,Jason> From: tim(a)rowledge.org> Date: Thu, 12 Jul 2007 09:19:35 -0700> To: squeak-dev(a)lists.squeakfoundation.org> Subject: Re: Win32Window how to get Squeak's window Handle? VM?> > > On 12-Jul-07, at 2:14 AM, Herbert König wrote:> > > > >> > the VM does what I need actually and I already thought of just making> > a new VM which gives access to the C-variable stWindow to Squeak.> But that variable is slated for retirement; it is irrelevant once > Ffenestri is actually used.> > tim> --> tim Rowledge; tim(a)rowledge.org; http://www.rowledge.org/tim> "Bother!" said Pooh, searching for the $10m winning lottery ticket.> > > _________________________________________________________________ PC Magazine’s 2007 editors’ choice for best web mail—award-winning Windows Live Hotmail. http://imagine-windowslive.com/hotmail/?locale=en-us&ocid=TXT_TAGHM_migrati…

3 2

[Monticello2] Saving slices to a directory
by Damien Cassou 23 Jul '07

23 Jul '07

Hi Colin and Avi, it's again about MC2. Avi seems to think that I should forget dealing with slices that are not based on PackageInfo. However, the possibility to save any kind of slice is one of the big features (maybe THE feature) of MC2. What would happen if I try to mimic the behavior of MC1: - have a directory - when the user saves a slice, a MDFileRepository is created in the directory and the data is put into. So, each FileRepository contains only 1 memo. Wouldn't that work? -- Damien Cassou

1 1

[Monticello2] Loading from a repository
by Damien Cassou 23 Jul '07

23 Jul '07

Hi Colin and Avi, I'm not sure how to load data from a repository into the image. Currently, I have something like: loadFromMemo | snapshot versions | snapshot := selectedMemo getSnapshotFrom: repository. versions := snapshot getVersionsFrom: repository. versions do: [:version | MDWorkingCopy default loadVersion: version] Is that the right way of loading? The problem with this approach is when a class A contain a method m1 in the image, but you are loading a version which does not contain this method because it has been removed. Then, the method is not removed from the image. -- Damien Cassou

3 8

[Monticello2] Time is important for #=
by Damien Cassou 23 Jul '07

23 Jul '07

Hi Colin and Avi, I do not understand why the second assertion fails: slice1 := MDPackageInfoSlice named: 'MyCategory'. slice2 := MDPackageInfoSlice named: 'MyCategory'. self assert: slice1 = slice2. (Delay forSeconds: 1) wait. slice3 := MDPackageInfoSlice named: 'MyCategory'. self assert: slice1 = slice3. "<=== fails" Is it important that the time is taken into account while generating the id of slices? -- Damien Cassou

1 1

Re: Linux install - dumb questions
by Bill Schwab 23 Jul '07

23 Jul '07

Subbu, If that means what I think it means<g>, it will be a huge help. Thanks!!!! Bill =========================== subbukk <subbukk@...> subbukk@... wrote: On Friday 20 July 2007 7:06 pm, Bill Schwab wrote: > Subbu mentioned packages. I am not above that, but want to be able to > do a good job of installing the latest. If I can turn the archives into > packages and load them, that would be fine by me. Any pointers? If you wish to build your own deb from a tar.gz, see: http://ubuntuforums.org/showthread.php?t=51003 Jens maintains the latest stable debs at: http://www.impara.de/~jens/ubuntu Apart from installing them into standard locations (which require root permission), you could also extract the files from the deb archive into your own home locations (see -x option of dpkg(1)) to check them out. Hope this helps .. Subbu Wilhelm K. Schwab, Ph.D. University of Florida Department of Anesthesiology PO Box 100254 Gainesville, FL 32610-0254 Email: bschwab(a)anest.ufl.edu Tel: (352) 846-1285 FAX: (352) 392-7029

1 0

Morphic!!!
by Jerome Peace 23 Jul '07

23 Jul '07

Morphic! (from - methods for license conversion - thread) *** > >tim Rowledge tim at rowledge.org >Thu Jul 5 20:52:26 UTC 2007 > >On 4-Jul-07, at 12:40 PM, Pavel Krivanek wrote: > >> Hi all, > >> >> I have prepared an image with small code base that is ready for >> license conversion and is able to load a sequence of packages up to >> working Morphic system. > >I'm rather inclined to the idea that "Morphic Must Die!" > It hurts me to hear you say that, Tim. You have a responsible position on the board of squeakfoundation and what you say carries weight. And can cause harm. It might also bring about the very thing you wish for. And I believe if that happened you would want to unwish it. I have been making efforts and progress (with self funded resources) to repair squeakdev morphic. Most of what Ive done Ive done on spec and the toughest part of my task has been to sell the repairs to the various and ever changing harvesting/releasing teams. Smalltalk/Squeak is a very rich language base on a half page response to a challenge to do something interesting and powerful in that half page. Morphic is an interation or two beyond that and to me represents a challenge to let even children do interesting things graphically just by assembling pre-existing morphs. Morphic also represents a first pass at doing things. Its been so popular lots of people have tried their hand at it. Some with less understanding than others. The harvesting teams have never before had the ability to do quality control. So poor smelling code has accumulated in squeak in general and it will look like it is mostly in morphic because people mostly see morphic. As Ive tracked down these bugs I have found the most profound ones to pre-date morphic. Now it would be nice if someone gave you a gazillion dollars and stood back and watched what you produced with it. It would be nice if people of the quality who could create a Cairo Canvas or whatever did so with their own self funded time or some of your gazillions. But we both are probably aware that that is NGH (Not-Gonna_Happen). You got what you got. You CAN start from where you are. You can use the wonderful tools that exist in squeak to produce, patiently, slowly, and incrementally, a morphic and squeak that is much better than it is today. My question for you is stated in POSITIVE terms, what did you really mean to say? What as a user story would you like to see come about? What (other than the NGH gazillions) is keeping it from happening? What is keeping you from making progress towards the goal? Or stated in postive terms what would get you(us) started along the path to that goal? Your in curiosity and service, --Jerome Peace ____________________________________________________________________________________ Shape Yahoo! in your own image. Join our Network Research Panel today! http://surveylink.yahoo.com/gmrs/yahoo_panel_invite.asp?a=7

2 1

RE: Ergonomics - Rant
by J J 22 Jul '07

22 Jul '07

I had this problem too all the time, until I discovered that if instead of putting the paren at the end, you just highlight the whole code that goes in parens and press the left one. It puts both on around the highlighted code, so now it's gone from an irritation to something I miss when it's not there. :)> From: tblanchard(a)mac.com> Date: Tue, 17 Jul 2007 12:32:32 -0700> To: squeak-dev(a)lists.squeakfoundation.org> Subject: Re: Ergonomics - Rant> > Oh, I'll chime in - it doesn't work the way I do and it drives me nuts.> > I often type most of a line, realize I need parens around part of it > at the end of the line, put in the closing paren, then go back to put > in the opening one - where it then insists on providing a matching > closing one. In general, I often end up typing the closing quote/ > paren first, then go back to put in the opening ones.> > In other words, I never code left to right - I code like an artist > builds a picture - starting with a sketch and gradually filling in > detail jumping around. eCompletion is a huge impediment when you are > doing this as you spend more time deleting spurious matching quotes > and parens than you do just typing stuff.> > I'm quite willing to agree that I might not be typical - but that's > how I work - successive approximations of proper code rather than > left to right perfection.> > -Todd Blanchard> > On Jul 16, 2007, at 10:58 AM, Damien Cassou wrote:> > > 2007/7/16, Gary Chambers <gazzaguru2(a)btinternet.com>:> >> Since we do business devlopment we like to keep things a bit more > >> stable,> >> though try to keep up with the main release.> >> Besides, I find the eCompletion stuff a bit problematic (some of > >> us use it> >> though).> >> > What is problematic for you might be problematic for others too. > > What is it?> >> > -- > > Damien Cassou> >> > _________________________________________________________________ Don't get caught with egg on your face. Play Chicktionary! http://club.live.com/chicktionary.aspx?icid=chick_wlmailtextlink

1 0

RE: Ergonomics - Rant
by Bill Schwab 22 Jul '07

22 Jul '07

Jason, There is room for debate on what it should do, but this one sounds like a preference to me. Options would be to drop, keep, or launch a new browser on the method. You want ergonomics? We should move the Browse-it command to just under inspect; I have to go to the second menu quite often. With my current project, I found a modal way to select a folder, but not files, other than a chain of menus that (pro) show all hidden files, (con) cannot hide them, force the user to resort to a more... option, and do not allow sorting. It screams for (sorry) modal dialog. I created a really cheesy looking "tool bar" composed to ordinary buttons because that was pretty much the only option. I have to "balance" the mouse to keep the input focus in the method name field of the method finder; great tool, pathetic interface. I could go on. The IDE, and sadly the basic pieces available to give to a would-be user, are 25 years out of date. How's that for rant? :) Bill =========================== J J <azreal1977@...> azreal1977@... wrote: > Date: Mon, 16 Jul 2007 08:47:32 -0500 > From: david.mitchell@... > To: squeak-dev@... > Subject: Re: Ergonomics - Rant > > Damien, I agree with Todd on this one. > > While it is certainly a defensible position to say that the method > should disappear because after editing, it doesn't match the browser's > criteria, but I think it is less useful. > > That is, I think of the senders browser as being a filter on open not > a continuously updating filter. Once open, I want to edit my code. I > don't want it to keep on filtering. This is part of the whole MVC paradigm, no? When you change a method some updates will get generated and the window will know to check this method that changed, it does fit anymore so it's dropped. I personally have never had this problem, because knowing this was MVC I always open a class browser on that method (it's either double click it or a right click menu, don't remember) if I want to change the code not to send the browsed message anymore *but* I still want to look at it after. Wilhelm K. Schwab, Ph.D. University of Florida Department of Anesthesiology PO Box 100254 Gainesville, FL 32610-0254 Email: bschwab(a)anest.ufl.edu Tel: (352) 846-1285 FAX: (352) 392-7029

1 0

RE: Ergonomics - Rant
by J J 22 Jul '07

22 Jul '07

Now I'm confused. This has been the state of affairs since ST-80 or earlier no? This isn't a morphic thing, the event stuff is in Object itself.> From: gazzaguru2(a)btinternet.com> To: squeak-dev(a)lists.squeakfoundation.org> Date: Mon, 16 Jul 2007 15:55:44 +0100> Subject: RE: Ergonomics - Rant> > Is that new for 3.10, doesn't seem to happened in my 3.9...> > -----Original Message-----> From: squeak-dev-bounces(a)lists.squeakfoundation.org> [mailto:squeak-dev-bounces@lists.squeakfoundation.org] On Behalf Of Damien> Cassou> Sent: 16 July 2007 3:47 pm> To: The general-purpose Squeak developers list> Subject: Re: Ergonomics - Rant> > > 2007/7/16, Gary Chambers <gazzaguru2(a)btinternet.com>:> > Indeed, would be nice if it monitored new senders of the method > > selector too!> > > This is already done: the sender browser is informed when a new sender> appears.> > > > Though, as I know, that would be rather cpu heavy.> > This does not consume more. The notification system informs the browser when> a method is changed or when a new method is created.> > > Best to treat it as a snapshot at the time it was asked for.> > -- > Damien Cassou> > _________________________________________________________________ Don't get caught with egg on your face. Play Chicktionary! http://club.live.com/chicktionary.aspx?icid=chick_wlmailtextlink

1 0

RE: Ergonomics - Rant
by J J 22 Jul '07

22 Jul '07

> Date: Mon, 16 Jul 2007 08:47:32 -0500> From: david.mitchell(a)gmail.com> To: squeak-dev(a)lists.squeakfoundation.org> Subject: Re: Ergonomics - Rant> > Damien, I agree with Todd on this one.> > While it is certainly a defensible position to say that the method> should disappear because after editing, it doesn't match the browser's> criteria, but I think it is less useful.> > That is, I think of the senders browser as being a filter on open not> a continuously updating filter. Once open, I want to edit my code. I> don't want it to keep on filtering.This is part of the whole MVC paradigm, no? When you change a method some updates will get generated and the window will know to check this method that changed, it does fit anymore so it's dropped. I personally have never had this problem, because knowing this was MVC I always open a class browser on that method (it's either double click it or a right click menu, don't remember) if I want to change the code not to send the browsed message anymore *but* I still want to look at it after. _________________________________________________________________ Missed the show? Watch videos of the Live Earth Concert on MSN. http://liveearth.msn.com

1 0

← Newer
1
...
6
7
8
9
10
11
12
...
22
Older →

Jump to page: