October 2013 - Pharo-dev - lists.pharo.org

References to Obsolete classes

by Noury Bouraqadi

Hi, Consider the following scenario: 1- create class A 2- create class B with a method m that references class A 3- delete class A . As a result B>>#m holds a reference to ObsoleteA 4- create a new class A. B>>#m continues to hold a reference to ObsoleteA. This is even more disturbing, since the source of B>>#m displays A. Is this behavior desired? If yes, in which situations. because I'd rather prefer to have B>>#m updated. This can be done, by moving the association of a removed class (or global in the general case) from the system dictionary to the Undeclared. Side note: just noticed that Undeclared is a dictionary. I was expecting it to be a WeakSet. Noury Ecole des Mines de Douai http://car.mines-douai.fr/noury -- Afin de contribuer au respect de l'environnement, merci de n'imprimer ce courriel qu'en cas de necessite Please consider the environment before you print

Oct. 4, 2013

RPackage-SystemIntegration problem

by Pavel Krivanek

Hi, When I look at the package RPackage-SystemIntegration (v. 173) and I do browse it in Monticello GUI, it doesn't contain the extension of the removed class ClassOrganizer. But when I do the same and I browse the this package in the same version from the Pharo repository, it does contain extension of the ClassOrganizer. As the result the job that tries to load Gofer to PharoKernel fails because the exported .st file contains extension of the nonexisting class. Can you, please, try to save the new version of the package or something like that to remove this extension from the package content? Cheers, -- Pavel

Oct. 4, 2013

Re: [Pharo-dev] Pharo (cog) Vm on Windows with 2GB virtual mem

by Tudor Girba

Hi, As far as I can tell, you should set the limit in MBytes. So, for 2G, you should have: AddressSpaceLimit=2048 I am using 1GB for some Moose analyses. I benchmarked and got no sign of slowdown. Cheers, Doru On Fri, Oct 4, 2013 at 6:10 AM, Oscar EA CallaÃº <oalvarez(a)dcc.uchile.cl>wrote: > Thanks a lot, I added**** > > ** ** > > AddressSpaceLimit = 2147483648**** > > ** ** > > To the .ini file and all is working smoothly. **** > > ** ** > > FYI, I donât have speed problems with such memory. Of course, If Iâll > process some big data, Iâll wait a while to get the answer, but the same > happens in any other lang/platt. Certainty, I can do fast experiments in C, > but I donât get all Pharo benefits, like pausing my analysis and resuming > it later, such thing in C (or may be any other language) would require to > store the intermediate data into files or a database. With Pharo, I donât > care about that.**** > > ** ** > > Cheers.**** > > ** ** > > *From:* Pharo-dev [mailto:pharo-dev-bounces@lists.pharo.org] *On Behalf > Of *Guillermo Polito > *Sent:* Thursday, October 3, 2013 2:37 AM > > *To:* Pharo Development List > *Subject:* Re: [Pharo-dev] Pharo (cog) Vm on Windows with 2GB virtual mem* > *** > > ** ** > > I can only guess by the comment in the commit:**** > > ** ** > > > https://gitorious.org/cogvm/blessed/commit/341e6c2e150bcde80dfea6c890bab774… > **** > > ** ** > > Can you check that?**** > > ** ** > > Guille**** > > ** ** > > On Thu, Oct 3, 2013 at 6:36 AM, Oscar EA CallaÃº <oalvarez(a)dcc.uchile.cl> > wrote:**** > > Thanks, but what would be the parameter?**** > > **** > > This is the default .ini**** > > **** > > [Global]**** > > DeferUpdate=1**** > > ShowConsole=0**** > > DynamicConsole=1**** > > ReduceCPUUsage=1**** > > ReduceCPUInBackground=0**** > > 3ButtonMouse=0**** > > 1ButtonMouse=0**** > > UseDirectSound=1**** > > PriorityBoost=1**** > > B3DXUsesOpenGL=0**** > > CaseSensitiveFileMode=0**** > > **** > > **** > > So there is no mention the paratemer to change or add.**** > > **** > > Cheers.**** > > **** > > *From:* Pharo-dev [mailto:pharo-dev-bounces@lists.pharo.org] *On Behalf > Of *Guillermo Polito > *Sent:* Friday, September 27, 2013 5:18 AM > *To:* Pharo Development List > *Subject:* Re: [Pharo-dev] Pharo (cog) Vm on Windows with 2GB virtual mem* > *** > > **** > > I think latest Pharo VM supports to change the memory limit by only > changing an ini file. Igor did that not long ago.**** > > **** > > Cheers,**** > > Guille**** > > **** > > On Fri, Sep 27, 2013 at 2:09 PM, Oscar E. A. CallaÃº < > oalvarez(a)dcc.uchile.cl> wrote:**** > > Hi Guys,**** > > **** > > I read in this thread**** > > **** > > http://forum.world.st/out-of-memory-cog-on-windows-td3465621.html**** > > **** > > that 512 mb limit can be avoided in windows, but a re-compilation is > needed. I was wondering if the situation changed, and now we can pass a > parameter. Otherwise, I would be interested if any of you has a > pre-compiled vm with more mem.**** > > **** > > I need to run some experiments and I need 2GB (or 1.5 GB at least ) on a > cog VM for Windows. **** > > **** > > Thanks in advance.**** > > **** > > Oscar**** > > **** > > ** ** > -- www.tudorgirba.com "Every thing has its own flow"

Oct. 4, 2013

[regression reporter]regression occurred

by no-reply＠ci.inria.fr

https://ci.inria.fr/pharo/job/Pharo-3.0-Update-Step-2.1-Validation/label=ma… 1 regressions found. ToolsTest.Debugger.DebuggerTest.testBasic

Oct. 4, 2013

[regression reporter]regression occurred

by no-reply＠ci.inria.fr

https://ci.inria.fr/pharo/job/Pharo-3.0-Update-Step-2.1-Validation/label=wi… 2 regressions found. ToolsTest.Debugger.DebuggerTest.testBasic Zinc.Tests.ZnServerTests.testEntityTooLarge

Oct. 4, 2013

[regression reporter]regression occurred

by no-reply＠ci.inria.fr

https://ci.inria.fr/pharo/job/Pharo-3.0-Update-Step-2.1-Validation/label=li… 1 regressions found. ToolsTest.Debugger.DebuggerTest.testBasic

Oct. 4, 2013

[update 3.0] #30446

by Marcus Denker

30446 ----- 11719 integrate all the changes from the repository of the new debugger https://pharo.fogbugz.com/f/cases/11719 This commit contains the following main changes: - debugging actions are first class entities; they are defined in separate classes. All actions from the toolbar are added dynamically based on pragmas. - separated the spec interface into multiple widgets: stack, toolbar, buttons - have dedicated filters to hide contexts from the stack Thanks to Clara Allende and Andrei Chis! Diff information: http://smalltalkhub.com/mc/Pharo/Pharo30/main/Spec-Debugger-MarcusDenker.16… http://smalltalkhub.com/mc/Pharo/Pharo30/main/Settings-Tools-MarcusDenker.4… http://smalltalkhub.com/mc/Pharo/Pharo30/main/DebuggerModel-MarcusDenker.76… http://smalltalkhub.com/mc/Pharo/Pharo30/main/DebuggerActions-MarcusDenker.… http://smalltalkhub.com/mc/Pharo/Pharo30/main/DebuggerFilters-MarcusDenker.…

Oct. 4, 2013

Re: [Pharo-dev] Pharo (cog) Vm on Windows with 2GB virtual mem

by btc＠openinworld.com

That's a great quote for marketing Pharo. Thanks Oscar. Oscar EA CallaÃº wrote: > > FYI, I donât have speed problems with such memory. Of course, If Iâll > process some big data, Iâll wait a while to get the answer, but the same > happens in any other lang/platt. Certainty, I can do fast experiments in C, > but I donât get all Pharo benefits, like pausing my analysis and resuming it > later, such thing in C (or may be any other language) _would require to store > the intermediate data into files or a database. With Pharo, I donât care > about that._ > >

Oct. 4, 2013

Re: [Pharo-dev] "as yet unclassified" protocol remains when empty

by btc＠openinworld.com

Thanks Thierry. Searching didn't turn up anything, so I logged it https://pharo.fogbugz.com/f/cases/11772/as-yet-unclassified-protocol-remain… Goubier Thierry wrote: > Hi Ben, > > I think it has already been noticed as a bug, so it will be corrected. > Not sure there is already an issue about it, however. > > Regards, > > Thierry > > Le 01/10/2013 15:43, btc(a)openinworld.com a Ã©crit : >> I wanted to check this was not a feature before reporting an issue. >> Previously in Pharo 2.0, when the protocol "as yet unclassified" became >> empty, it disappeared. >> In Pharo 30434, it does not disappear when empty. >> Really a small thing, but I prefer the previous behaviour. >> >> cheers -ben >> >> >> >

Oct. 4, 2013

Re: [Pharo-dev] [ANN] Mustache templates

by Norbert Hartl

Am 04.10.2013 um 08:56 schrieb Torsten Bergmann <astares(a)gmx.de>: >> I like to announce my smalltalk port of mustache templates. > > Cool! > >> To load it do > > Consider making a copy into MetaRepoForPharo20 or MetaReporForPharo30 > if it is working in these versions. > > That way it is easily accessible via Config browser. > Good idea. Done! thanks, Norbert

Oct. 4, 2013