November 2013 - Pharo-users

Re: [Pharo-users] NativeBoost Questions while wrapping FMOD

by Igor Stasenko

On 23 November 2013 17:42, Denis Kudriashov <dionisiydk(a)gmail.com> wrote: > Hi > > 2013/11/22 Igor Stasenko <siguctua(a)gmail.com> > >> >> >>> 2. instead of returning useful objects, FMOD returns error codes and you >>> pass a pointer to receive the object. >>> >>> - The first consequence is that I have to wrap all the calls e.g. "self >>> processErrorCode: self primCreate." where >>> processErrorCode: anInteger >>> anInteger = 0 ifFalse: [ self error: 'FMOD returned error code >>> ', anInteger >>> asString ]. >>> Is there a more graceful way to do that? >>> >>> i doubt so.. since it is library API which dictates you to use it in >> certain way. >> The proper error handling never hurts (except from causing extra code >> bloat, of course :) >> > > Is it good idea to override (or create new) #nbCall: method to handle > library common logic? (when any library function returns error code) > of course, the #nbCall: is just a convenience method and meant to be overridden if necessary. nbCall: fnSpec " you can override this method if you need to" ^ (self nbCalloutIn: thisContext sender) convention: self nbCallingConvention; function: fnSpec module: self nbLibraryNameOrHandle And nbCallingConvention, nbLibraryNameOrHandle is there for override as well. A more verbose form is: self nbCallout stdcall; options: #( - optDirectProxyFnAddress optAllowExternalAddressPtr); function: #( NBBootstrapUlong HeapAlloc (ulong heapHandle , 0 , SIZE_T size) ) module: #Kernel32 But be aware though, that this code usually invoked once during code generation, and later no method code is invoked because it just runs a primitive (primitiveNativeCall), and if it succeeds, it just returns with result, without running a method's body. So, nbcall can be changed/extended if you need extra thing at code generation stage, or you want to handle primitive error(s) differently. But not for handling things like function return values (which happens to be a library's error/success code). -- Best regards, Igor Stasenko.

Nov. 24, 2013

Re: [Pharo-users] [Athens] Finding if a line passes through a specific pixel .

by Igor Stasenko

There is some bits which can help you in this regard. The part of it is path tesseleration, which turns any path with complex curves into polygon (which has only straight lines), from there on, i think you can easily calculate the distance between any point and that polygon. Look at AthensCurveFlattener class, and its uses. the entry point is (flattenPath: aPath transform: aTransformation). But be aware that this is not fully integrated yet and subject to change in future. On 23 November 2013 15:20, kilon alios <kilon.alios(a)gmail.com> wrote: > aaaaaahhh come on ... I am not THAT lazy :D > > of course I have searched it before asking. Unless I am blind I see loads > of bezier methods, none that detects whether a point belong to a curve. But > now you mention it I think I remember squeak was able to do add points to a > curve via click or something similar, because thats what I wan to do. So I > will take a look at it , most probably there is a method in there that does > this. > > I want the user to click on a curve or line and there a new control point > will be placed that will allow the user to shape the line or curve to a > finer detail. A standard feature for all vector editors , 3d editors, even > image editors. > > I am sorry if I miss something obvious here, its not laziness , its most > likely stupidity or just the fact I am very new with this. This is the > first time I try to make a vector editor.Again thank you all for trying to > help me, you have done already a lot. > > > On Sat, Nov 23, 2013 at 3:38 PM, StÃ©phane Ducasse < > stephane.ducasse(a)inria.fr> wrote: > > >> >> On Nov 23, 2013, at 11:24 AM, kilon alios <kilon.alios(a)gmail.com> wrote: >> >> would I sound too lazy to ask in which class ? >> >> >> On Sat, Nov 23, 2013 at 11:35 AM, StÃ©phane Ducasse < >> stephane.ducasse(a)inria.fr> wrote: >> >>> there are bezier algo in the system >>> >>> Stef >>> >>> On Nov 20, 2013, at 9:55 PM, kilon alios <kilon.alios(a)gmail.com> wrote: >>> >>> yes I found the formula for straight lines and I was wondering what was >>> the one for bezier curves, so your reply could not have come a better >>> moment. Thanks I will study it and implement it. >>> >>> >>> On Wed, Nov 20, 2013 at 9:21 PM, Hilaire Fernandes < >>> hilaire.fernandes(a)gmail.com> wrote: >>> >>>> Hello, >>>> >>>> You may want to calculate the distance between the mouse position and >>>> your path (I guess a bezier curve). Then given this distance and a >>>> tolerance to zero you decide if the mouse position is more or less on >>>> your path. Calculating the distance you will also get for free the place >>>> where to add a point on your path. >>>> >>>> http://blog.gludion.com/2009/08/distance-to-quadratic-bezier-curve.html >>>> >>>> Hilaire >>>> >>>> >>>> Le 14/11/2013 21:07, kilon alios a Ã©crit : >>>> > In order to do that I will have to check that the line passes through >>>> a >>>> > specific pixel. Or to be more correct that the place that mouse has >>>> > clicked is where my path passes through. How I do that ? Can Athens do >>>> > that ? Can athens tell me which pixels my path passes through ? >>>> > >>>> >>>> >>>> -- >>>> Dr. Geo http://drgeo.eu >>>> >>>> >>>> >>> >>> >> >> >> > -- Best regards, Igor Stasenko.

Nov. 24, 2013

Re: [Pharo-users] spec and double click

by Benjamin

I will have a look tomorrow :) If you are in a hurry, I know such a behaviour is present in EyeInspector Ben On 24 Nov 2013, at 00:07, Lorenz KÃ¶hl <lorenz(a)quub.de> wrote: > > On Nov 23, 2013, at 7:37 PM, Benjamin <Benjamin.VanRyseghem.Pharo(a)gmail.com> wrote: > >> m := NewListModel new >> m doubleClickAction: [ self halt ] >> >> m openWithSpec > > your code and the following (double clicking the list items) doesn't work for me in Pharo3.0 > Latest update: #30591 > > m := NewListModel new items: #(a b c). > m doubleClickAction: [ Transcript traceCr: 'foo' ]. > m openWithSpec

Nov. 24, 2013

Re: [Pharo-users] Nautilus pkg-pane regression in pharo 3.0?

by Benjamin

That a major issue so far. It was not intended, but since the tree widget already handle arrow right by itself, I do not know if Esteban thought about this feature Ben On 23 Nov 2013, at 22:34, Lorenz KÃ¶hl <lorenz(a)quub.de> wrote: > Nautilus has stopped jumping to packages when the focus is in the package pane and keys are pressed (the same behaviour use from eg. osx finder) > > Is that a regression or intended?

Nov. 23, 2013

Re: [Pharo-users] spec and double click

by Lorenz KÃ¶hl

On Nov 23, 2013, at 7:37 PM, Benjamin <Benjamin.VanRyseghem.Pharo(a)gmail.com> wrote: > m := NewListModel new > m doubleClickAction: [ self halt ] > > m openWithSpec your code and the following (double clicking the list items) doesn't work for me in Pharo3.0 Latest update: #30591 m := NewListModel new items: #(a b c). m doubleClickAction: [ Transcript traceCr: 'foo' ]. m openWithSpec

Nov. 23, 2013

Nautilus pkg-pane regression in pharo 3.0?

by Lorenz KÃ¶hl

Nautilus has stopped jumping to packages when the focus is in the package pane and keys are pressed (the same behaviour use from eg. osx finder) Is that a regression or intended?

Nov. 23, 2013

Mugs, mouse mats, stickers

by Jeff Gray

What else would you like in the Pharo store? Personally I'd like to have a t-shirt. -- View this message in context: http://forum.world.st/Mugs-mouse-mats-stickers-tp4724644.html Sent from the Pharo Smalltalk Users mailing list archive at Nabble.com.

Nov. 23, 2013

Re: [Pharo-users] Merchandise in your locale

by Jeff Gray

Nov. 23, 2013

Merchandise in your locale

by Jeff Gray

Just a reminder to have a look in the Pharo store. There are a number of sites around the globe to allow you to shop in your language and to reduce postage charges: USA (English) http://www.zazzle.com USA (EspaÃ±ol) http://www.zazzle.com/?lang=es Canada (English) http://www.zazzle.ca Canada (FranÃ§ais) http://www.zazzle.ca/?lang=fr UK http://www.zazzle.co.uk Deutschland http://www.zazzle.de EspaÃ±a http://www.zazzle.es France http://www.zazzle.fr Portugal http://www.zazzle.pt Sverige http://www.zazzle.se Nederland http://www.zazzle.nl Ãsterreich http://www.zazzle.at Schweiz (Deutsch) http://www.zazzle.ch Suisse (FranÃ§ais) http://www.zazzle.ch/?lang=fr Belgique (FranÃ§ais) http://www.zazzle.be BelgiÃ« (Nederlandse) http://www.zazzle.be/?lang=nl Australia http://www.zazzle.com.au New Zealand http://www.zazzle.co.nz æ¥æ¬ http://www.zazzle.co.jp ëíë¯¼êµ http://www.zazzle.co.kr Brasil http://www.zazzle.com.br -- View this message in context: http://forum.world.st/Merchandise-in-your-locale-tp4724632.html Sent from the Pharo Smalltalk Users mailing list archive at Nabble.com.

Nov. 23, 2013

Re: [Pharo-users] Writing a GUI - Where to start?

by Picci Pharo

Typical Window := Window with buttons and textInput.... > From: stephan(a)stack.nl > Date: Fri, 22 Nov 2013 16:35:09 +0100 > To: pharo-users(a)lists.pharo.org > Subject: Re: [Pharo-users] Writing a GUI - Where to start? > > >About Glamour, isn'it used to create Browser? How can i create a typical window? > > What is a typical window? Spec gives too much freedom and not enough > structure for 'typicalâ applications. > > Almost every window in your smalltalk system is a browser. > Nearly all would be significantly less code when build with Glamour. > > Stephan

Nov. 23, 2013