March 2012 - Pharo-dev - lists.pharo.org

[Pharo-project] Object>>deprecated:on:in: doesn't respect state of Deprecation class>>showWarning:

by Martin Sandiford

Hi all, Kind of new to Pharo, so not sure if this is intended behaviour or not, but "Object>>deprecated:on:in:" doesn't respect the current state of the "Deprecation class>>showWarning" setting. This was unexpected for me --- in a headless setting the warning dialog causes the VM to exit. To reproduce (in 1.4 14399): Deprecation showWarning: false. true and: [ true ] and: [ true ]. "Boolean>>and:and: sends deprecated:on:in:" Fix seems straightforward: wrap the current contents of "deprecated:on:in:" in "Deprecation showWarning ifTrue: [...]". If there are no objections, I'll raise a bug in the tracker. Cheers, Martin

March 19, 2012

[Pharo-project] Pharo Kernel Reload Tests-Unix - Build # 21 - Still Unstable!

by jenkins-pharo.ci.inria.fr＠jenkins-server.lille.inria.fr

March 19, 2012

Re: [Pharo-project] [update 1.4] #14403

by StÃ©phane Ducasse

Thanks! Stef On Mar 19, 2012, at 10:08 PM, Marcus Denker wrote: > 14403 > ----- > > Issue 5497: SystemNavigation depending on Cursor > http://code.google.com/p/pharo/issues/detail?id=5497 > > Issue 5516: supporting code to rename package > http://code.google.com/p/pharo/issues/detail?id=5516 > > Issue 5518: deprecated #dejaVuAreInstalled > http://code.google.com/p/pharo/issues/detail?id=5518 > > Issue 5519: String has categories *morphic and *Morphic > http://code.google.com/p/pharo/issues/detail?id=5519 > > Issue 5517: #subPixelRenderColorFonts and #subPixelRenderFonts just return true > http://code.google.com/p/pharo/issues/detail?id=5517 > > -- > Marcus Denker -- http://marcusdenker.de > >

March 19, 2012

[Pharo-project] Pharo 1.4 Tests-Unix - Build # 297 - Still Unstable!

by jenkins-pharo.ci.inria.fr＠jenkins-server.lille.inria.fr

March 19, 2012

[Pharo-project] [update 1.4] #14403

by Marcus Denker

14403 ----- Issue 5497: SystemNavigation depending on Cursor http://code.google.com/p/pharo/issues/detail?id=5497 Issue 5516: supporting code to rename package http://code.google.com/p/pharo/issues/detail?id=5516 Issue 5518: deprecated #dejaVuAreInstalled http://code.google.com/p/pharo/issues/detail?id=5518 Issue 5519: String has categories *morphic and *Morphic http://code.google.com/p/pharo/issues/detail?id=5519 Issue 5517: #subPixelRenderColorFonts and #subPixelRenderFonts just return true http://code.google.com/p/pharo/issues/detail?id=5517 -- Marcus Denker -- http://marcusdenker.de

March 19, 2012

[Pharo-project] Polymorph layout: images in a column

by Schwab,Wilhelm K

Gary, If I create a column and place an alpha image in it, is there a way to get the image to scale, ideally to the width of the column? As it is, I'm setting fixed extents, but somewhat works against what I would like to leave to the column? Bill

March 19, 2012

Re: [Pharo-project] non-interactive compiling with feedback

by Milan Mimica

On 19 March 2012 21:04, Eliot Miranda <eliot.miranda(a)gmail.com> wrote: > > a) compiling source containing an undeclared reference merits only a > warning > > IMO, it's not an error, and a) is a good option. Note that if you did > make it an error in all circumstances you'd badly break the system. > Hi Eliot, a) works for me. I just need a string to display to the user, on compile time. So how do I get a warning? -- Milan Mimica http://sparklet.sf.net

March 19, 2012

Re: [Pharo-project] non-interactive compiling with feedback

by Eliot Miranda

Hi Milan, On Mon, Mar 19, 2012 at 11:22 AM, Milan Mimica <milan.mimica(a)gmail.com>wrote: > On 18 March 2012 23:20, Guillermo Polito <guillermopolito(a)gmail.com>wrote: > >> Hmm, you can compile a method with Behavior>>#compile:notifying: >> >> and send an observer implementing: >> >> #notify:at:in: >> >> and some other stuff like #selection #deselect... >> >> That way you can capture the errors... >> > > This almost works. > It doesn't raise errors for undeclared variables, instead it prompts for > variable replacement. > > | method requestor | > requestor := Mock new. > (requestor stub: #selectionInterval) returns: (1 to: 2). > requestor stub: #selectFrom:to:. > (requestor stub: #bindingOf:) returns: nil. > (requestor stub: #notify:at:in:) will: [ self halt ]. > method := Compiler new compiledMethodFor: 'a:=1' in: nil to: nil > notifying: requestor proxy ifFail: nil logged: false. > method inspect > > I would like to catch the error. It is an error, right? > the answer is "not necessarily". Smalltalk deals with circular references between globals in source by using Undeclared. So if the source of class A references class B and vice verse, and they are filed-in one after another, when class A's source is compiled, a reference to B will be created in undeclared. When B's source is compiled, that Undeclared binding will be moved to Smalltalk and updated to reference class B. So one could say in non-interactive use a) compiling source containing an undeclared reference merits only a warning b) compiling source containiing an undeclared reference in the form of a variable beginning with lower-case is an error, since globals should be capitalized c) force classes to be defined before their source is compiled IMO, it's not an error, and a) is a good option. Note that if you did make it an error in all circumstances you'd badly break the system. -- > Milan Mimica -- HTH, Eliot

March 19, 2012

Re: [Pharo-project] Any theme or settings that will speed up a Pharo 1.3 image?

by Marcus Denker

> > > > You used your own images for Pinocchio, or was Pinocchio just fast enough for the standard images? > Pinnocio can't run the image... it's ("just ;-)") a compiler from smalltalk to binary code... the .exe then boostraps an image, but there is no support to persist that (nor support for all the primitives.. and no GC other than Boehm). Marcus -- Marcus Denker -- http://marcusdenker.de

March 19, 2012

Re: [Pharo-project] Any theme or settings that will speed up a Pharo 1.3 image?

by Marcus Denker

>> >> I noticed some absurd preference checking in standard images a while back, but turning this off made no noticeable different in optimized Cog. Look for uses of Preferences subPixelRenderFonts, e.g. installStrikeFont:foregroundColor:backgroundColor:. There's a non-local return in the default path on preference access. So not having subPixelRenderFonts a defined preference method creates lots on noise when doing VM debugging. >> > > in Pharo Preferences are not anymore... #subPixelRenderFonts is send to StrikeFont and just > returns true (the default of the Preference). I think the idea was to add a Setting later if someone > requests it... and nobody ever did. > > So maybe we can simplify that method a bit. > http://code.google.com/p/pharo/issues/detail?id=5517 -- Marcus Denker -- http://marcusdenker.de

March 19, 2012