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

[Pharo-project] MC and shared pools

by Igor Stasenko

I just stumbled upon this.. I was assuming that it is perfectly fine to: - define a shared pool in package - make a class which using given pool in same package but when MC loads it , it looks like it doesn't care about loading order and if it loads the class which using pool first, it says that pool not exists and asks me to create it. if you create it, then when it loads the pool , it barks that given class already in system... i know that solution is to have 2 packages, 1 with pool, and another with classes which using them, but somehow i was assumed that it should work fine in single package.. So, here 2 questions: - does anybody knows if this worked ever? - do we need that? -- Best regards, Igor Stasenko.

Feb. 29, 2012

[Pharo-project] [update 1.4] #14369

by Marcus Denker

14369 ----- Issue 5331: use the glamour style for pretty print http://code.google.com/p/pharo/issues/detail?id=5331 Issue 5396: Deprecate TTMorph classes http://code.google.com/p/pharo/issues/detail?id=5396 -- Marcus Denker -- http://marcusdenker.de

Feb. 29, 2012

Re: [Pharo-project] [Bug Tracker] Issues tagged 1.3

by Igor Stasenko

On 29 February 2012 14:57, Schwab,Wilhelm K <bschwab(a)anest.ufl.edu> wrote: > Sig, > > I appear to a 1.3 image in preliminary state. Â I'm sure stuff will be broken, and Cog is rearing its head by looking for external libraries in the wrong places; symlinks should clear that up, but it's ugly. > Bill, it's not about 1.3 or 1.4. it is how VM works (symlinks/lib search). In this regard, any pharo version is not anything better than other. Yes, we need to change that at some day. > I used the 1.3 seaside image as a base, and will have Seaside 2-3 problems to fix. Â Alien was the problem, probably breaking the compiler, as you suggested. > To correct you: it was not a suggestion. When Alien was released there was need to change compiler/parser in order to make it work. Eliot made a separate package for that. Now, after those changes were incorporated into Pharo itself, they are no longer needed. and moreover they might override existing methods with older/incompatible versions of it coming from Alien. Overrides are evil. Overrides in kernel are evil^2 :) > Bill > > > > ________________________________________ > From: pharo-project-bounces(a)lists.gforge.inria.fr [pharo-project-bounces(a)lists.gforge.inria.fr] on behalf of Igor Stasenko [siguctua(a)gmail.com] > Sent: Sunday, February 26, 2012 9:50 AM > To: Pharo-project(a)lists.gforge.inria.fr > Subject: Re: [Pharo-project] [Bug Tracker] Issues tagged 1.3 > > On 25 February 2012 11:15, Schwab,Wilhelm K <bschwab(a)anest.ufl.edu> wrote: >> Sig, >> >> This is helpful, as you have told me Â what is killing my images. Â Then you say to load then you say to load the latest config. Â Happy to comply, but how do I do that? >> > > (ConfigurationOfWhatever project version: 'x.y.z') load > > or better learn more about metacello, there's a well-written chapter > of it from the guys who care > https://gforge.inria.fr/scm/viewvc.php/*checkout*/PharoByExampleTwo-Eng/Met… > >> Bill >> >> ________________________________________ >> From: pharo-project-bounces(a)lists.gforge.inria.fr [pharo-project-bounces(a)lists.gforge.inria.fr] on behalf of Igor Stasenko [siguctua(a)gmail.com] >> Sent: Saturday, February 25, 2012 12:39 AM >> To: Pharo-project(a)lists.gforge.inria.fr >> Subject: Re: [Pharo-project] [Bug Tracker] Issues tagged 1.3 >> >> On 25 February 2012 06:52, Igor Stasenko <siguctua(a)gmail.com> wrote: >>> On 25 February 2012 06:35, Schwab,Wilhelm K <bschwab(a)anest.ufl.edu> wrote: >>>> Sig, >>>> >>>> Sorry about the empty reply. Â Migrate is fair game, and the image was healthy enough to file out the base (attached). Â Please look over how I install FFI to see if there are any mistakes. Â You might need to stub out a basic subclass (e.g. MigrateSig or anything else you want to call it) with a few methods (#me, #homeGrownClasses, etc.) to make it happy. Â Content of them can be simple to start. Â The "private stuff" has yet to enter the picture so shouldn't matter to us. >>>> >>>> THANKS!!! >>>> >>> Good. I will take a look. >>> >>>> Bill >> >> Ok, things for starting: >> >> >> Â Â Â Â Gofer new >> Â Â Â Â Â Â Â Â url:self mirror, 'MetacelloRepository'; >> Â Â Â Â Â Â Â Â package: 'ConfigurationOfAlien'; >> Â Â Â Â Â Â Â Â load. >> Â Â Â Â (Smalltalk globals classNamed: 'ConfigurationOfAlien') >> perform:#loadLatestVersion. >> >> >> Never ever use #loadLatestVersion, always use concrete version >> numbers, if you want to have stability. >> >> In our case, #loadLatestVersion is not quite latest, you can check it >> by yourself: >> >> >> ConfigurationOfAlien project updateProject. >> ConfigurationOfAlien project latestVersion Â "inspect it" >> >> 0.1 [ConfigurationOfAlien] >> so it actually loads not the latest available version, but the most >> old one. Because it is the only version with #release blessing. >> >> And i can tell you without even proceeding further, that it won't work >> in 1.3 neither in 1.4, because it is too old and not takes into >> account >> certain changes made in pharo. >> >> For instance: >> >> package: 'Alien-Prereqs' >> >> is no longer needed, as i told here: >> >> http://forum.world.st/Alien-Prerequisites-is-now-not-required-td3258432.html >> >> but its even worse, this version of config will be loading >> Alien-Prereqs-mha.3 which having even more overrides, >> and will definitely put image to its knees because compiler will be broken. >> >> So, try to load latest config first, if it would not help we will continue. >> >> P.S. to all people involved in using Alien.. some love needed to >> update configs for pharo 1.3 and 1.4 to finally get rid of >> Â 'Alien-Prereqs' package loading. >> >> -- >> Best regards, >> Igor Stasenko. >> >> > > > > -- > Best regards, > Igor Stasenko. > -- Best regards, Igor Stasenko.

Feb. 29, 2012

Re: [Pharo-project] [Bug Tracker] Issues tagged 1.3

by Schwab,Wilhelm K

Sig, I appear to a 1.3 image in preliminary state. I'm sure stuff will be broken, and Cog is rearing its head by looking for external libraries in the wrong places; symlinks should clear that up, but it's ugly. I used the 1.3 seaside image as a base, and will have Seaside 2-3 problems to fix. Alien was the problem, probably breaking the compiler, as you suggested. Bill ________________________________________ From: pharo-project-bounces(a)lists.gforge.inria.fr [pharo-project-bounces(a)lists.gforge.inria.fr] on behalf of Igor Stasenko [siguctua(a)gmail.com] Sent: Sunday, February 26, 2012 9:50 AM To: Pharo-project(a)lists.gforge.inria.fr Subject: Re: [Pharo-project] [Bug Tracker] Issues tagged 1.3 On 25 February 2012 11:15, Schwab,Wilhelm K <bschwab(a)anest.ufl.edu> wrote: > Sig, > > This is helpful, as you have told me what is killing my images. Then you say to load then you say to load the latest config. Happy to comply, but how do I do that? > (ConfigurationOfWhatever project version: 'x.y.z') load or better learn more about metacello, there's a well-written chapter of it from the guys who care https://gforge.inria.fr/scm/viewvc.php/*checkout*/PharoByExampleTwo-Eng/Met… > Bill > > ________________________________________ > From: pharo-project-bounces(a)lists.gforge.inria.fr [pharo-project-bounces(a)lists.gforge.inria.fr] on behalf of Igor Stasenko [siguctua(a)gmail.com] > Sent: Saturday, February 25, 2012 12:39 AM > To: Pharo-project(a)lists.gforge.inria.fr > Subject: Re: [Pharo-project] [Bug Tracker] Issues tagged 1.3 > > On 25 February 2012 06:52, Igor Stasenko <siguctua(a)gmail.com> wrote: >> On 25 February 2012 06:35, Schwab,Wilhelm K <bschwab(a)anest.ufl.edu> wrote: >>> Sig, >>> >>> Sorry about the empty reply. Migrate is fair game, and the image was healthy enough to file out the base (attached). Please look over how I install FFI to see if there are any mistakes. You might need to stub out a basic subclass (e.g. MigrateSig or anything else you want to call it) with a few methods (#me, #homeGrownClasses, etc.) to make it happy. Content of them can be simple to start. The "private stuff" has yet to enter the picture so shouldn't matter to us. >>> >>> THANKS!!! >>> >> Good. I will take a look. >> >>> Bill > > Ok, things for starting: > > > Gofer new > url:self mirror, 'MetacelloRepository'; > package: 'ConfigurationOfAlien'; > load. > (Smalltalk globals classNamed: 'ConfigurationOfAlien') > perform:#loadLatestVersion. > > > Never ever use #loadLatestVersion, always use concrete version > numbers, if you want to have stability. > > In our case, #loadLatestVersion is not quite latest, you can check it > by yourself: > > > ConfigurationOfAlien project updateProject. > ConfigurationOfAlien project latestVersion "inspect it" > > 0.1 [ConfigurationOfAlien] > so it actually loads not the latest available version, but the most > old one. Because it is the only version with #release blessing. > > And i can tell you without even proceeding further, that it won't work > in 1.3 neither in 1.4, because it is too old and not takes into > account > certain changes made in pharo. > > For instance: > > package: 'Alien-Prereqs' > > is no longer needed, as i told here: > > http://forum.world.st/Alien-Prerequisites-is-now-not-required-td3258432.html > > but its even worse, this version of config will be loading > Alien-Prereqs-mha.3 which having even more overrides, > and will definitely put image to its knees because compiler will be broken. > > So, try to load latest config first, if it would not help we will continue. > > P.S. to all people involved in using Alien.. some love needed to > update configs for pharo 1.3 and 1.4 to finally get rid of > 'Alien-Prereqs' package loading. > > -- > Best regards, > Igor Stasenko. > > -- Best regards, Igor Stasenko.

Feb. 29, 2012

[Pharo-project] Presenter/Acceptor

by Friedrich Dominicus

I don't know if it's good idea but what about adding it to aidaDescriptor? e.g we would have: (descriopton textFieldOn: #someAspect) Presenter: StandardTextPresenter on: Accepter: StandardAccepter for: or the like In CLIM the presenter depends on the stream for output see clim specification http://bauhh.dyndns.org:8000/clim-spec/index.html There we may have a different presentatoin depending on the outputstream it looks like this (in pseudo Lisp defmethod present ((anEmployee emp) (view +text-view+)) presents an emp on a text-view we may also have defmethod present ((anEmploeyee emp) (view +overview-view+) and this may e.g have name, firstname etc. and it will have different kinds of accepters also. But that's just an consideration. Regards Friedrich -- Q-Software Solutions GmbH; Sitz: Bruchsal; Registergericht: Mannheim Registriernummer: HRB232138; Geschaeftsfuehrer: Friedrich Dominicus

Feb. 29, 2012

Re: [Pharo-project] Nearly limitless Image: revisited.

by Janko MivÅ¡ek

S, Marcus Denker piÅ¡e: > On Feb 29, 2012, at 1:35 PM, Sven Van Caekenberghe wrote: >> On 29 Feb 2012, at 12:58, Janko MivÅ¡ek wrote: >> >>> For Pharo there seems everyone is avoiding this idea, mostly because of >>> image corruption fear. Well, we need to improve the robustness of Pharo >>> image and VM then this fear will vanish. It is obviously doable, if VW >>> is reliable enough, why not once Pharo? >> I am trying to be polite: remarks like this are unfair and not based on the experiences of most of the people in the Pharo community or of the companies using Pharo. It is perfectly possible to have stable long running servers based on Pharo. Sneaking in these phrases in simply spreading FUD. > It's not fair to expect, for nothing, the same as you get from a very expensive product. Especially if even just > entering a bug report is seemingly already too much to ask (!), while for VisualWorks, you have to pay real money. > > "It is obviously doable" has a differnt meaning with some Millions of EUR per year that one can spend for engineering > than without. Without, it's *far* from obvious! Guys, calm down a bit please. There will be a Pharo Consortium, we will actually pay something, so it is a time to start improving the support infrastructure as well. With users vs internal developer distinction included. I'm using both VW and Pharo now, so why won't you listen my observations and experiences a bit? They are for good, not for bad! And one of attributes of a professional is to stand the constructive critics and not feel offended immediately. Best regards Janko -- Janko MivÅ¡ek Aida/Web Smalltalk Web Application Server http://www.aidaweb.si

Feb. 29, 2012

Re: [Pharo-project] VNC client

by Jan van de Sandt

Hi Norbert, I have the same setup. But I must confess that I use the VNC client as little as possible, only for troubleshooting. If I have to do real work I copy the image to my local development laptop. So I don't really know if the key mapping works. I will test this tonight. Jan. On Wed, Feb 29, 2012 at 11:44 AM, Norbert Hartl <norbert(a)hartl.name> wrote: > Jan, > > I tried chicken of the vnc and basically it works. The "chicken" version > (is newer) that is based on chicken of the vnc does not work. I use Mac as > client to connect to a linux server machine (well it's pharo). Do you have > a different setup? However how do you map the keys? I cannot use alt, cmd, > etc. > > Norbert > > Am 28.02.2012 um 14:49 schrieb Jan van de Sandt: > > I use Chicken of the VNC > > > > On Tue, Feb 28, 2012 at 2:43 PM, Milan Mimica <milan.mimica(a)gmail.com>wrote: > >> I use remmina. >> >> >> On 28 February 2012 14:06, Norbert Hartl <norbert(a)hartl.name> wrote: >> >>> I wonder what vnc client do you use if you use pharo + RFB? I had a >>> running via macports. Now that I'm about to switch to homebrew I have >>> problems finding a working one. Be it in brew or app store, etc. >>> >>> thanks, >>> >>> Norbert >>> >> >> >> >> -- >> Milan Mimica >> http://sparklet.sf.net >> > > >

Feb. 29, 2012

Re: [Pharo-project] Pharo user vs. Pharo developer (was Nearly limitless Image)

by Frank Shearar

2012/2/29 Janko MivÅ¡ek <janko.mivsek(a)eranova.si>: > S, Marcus Denker piÅ¡e: >> Janko MivÅ¡ek wrote: > >>> I also support the image as a sole persistence idea and I'm actually >>> using it for years on VisualWorks. >>> >>> For Pharo there seems everyone is avoiding this idea, mostly because of >>> image corruption fear. Well, we need to improve the robustness of Pharo >>> image and VM then this fear will vanish. It is obviously doable, if VW >>> is reliable enough, why not once Pharo? > >> Someone needs to start fixing and improving. Why not, for example, you? > > I'm a Pharo user, not Pharo developer. Developer of Pharo itself, that > is. I'm just using it as a tool to contribute on other fields like on > the web. > > Therefore it is not to be expected to know Pharo internals enough to > contribute directly, but I can report wishes, errors and help to someone > more knowledgeable to fix them. And that's what I'm actually doing for > improving Pharo reliability, ask Igor and Stef. > > About issue tracker: it is not a duty of us Pharo users to open and > track issues there, IMHO. This is a tool for Pharo developers, so I > propose that one of you "harvest" them from mailing lists and open > issues. You know better how to open an issue than me. I can be then > invited to describe the problem better and help with you solving it. Realise, of course, that this puts the burden on the Pharo developers to trawl through mail. That means they have less time to work on fixing bugs. The more work you do - reporting the issue in the designated tracker, describing the problem precisely, etc. - that YOU do means the faster THEY can solve your problem. I fully expect with _any_ software that I "use" to be told "please use the official bugtracker". frank > And yes, this distinction between Pharo user and developer is necessary > part of Pharo becoming more professional and business friendly. > > Best regards > Janko > >> >> Here is a link: >> >> Â Â Â http://code.google.com/p/pharo/issues/list >> >> It's much better than talking. >> >> Â Â Â Marcus >> >> >> -- >> Marcus Denker -- http://marcusdenker.de >> >> >> > > -- > Janko MivÅ¡ek > Aida/Web > Smalltalk Web Application Server > http://www.aidaweb.si >

Feb. 29, 2012

Re: [Pharo-project] Nearly limitless Image: revisited.

by Marcus Denker

On Feb 29, 2012, at 1:35 PM, Sven Van Caekenberghe wrote: > > On 29 Feb 2012, at 12:58, Janko MivÅ¡ek wrote: > >> For Pharo there seems everyone is avoiding this idea, mostly because of >> image corruption fear. Well, we need to improve the robustness of Pharo >> image and VM then this fear will vanish. It is obviously doable, if VW >> is reliable enough, why not once Pharo? > > I am trying to be polite: remarks like this are unfair and not based on the experiences of most of the people in the Pharo community or of the companies using Pharo. It is perfectly possible to have stable long running servers based on Pharo. Sneaking in these phrases in simply spreading FUD. It's not fair to expect, for nothing, the same as you get from a very expensive product. Especially if even just entering a bug report is seemingly already too much to ask (!), while for VisualWorks, you have to pay real money. "It is obviously doable" has a differnt meaning with some Millions of EUR per year that one can spend for engineering than without. Without, it's *far* from obvious! Marcus -- Marcus Denker -- http://marcusdenker.de

Feb. 29, 2012

Re: [Pharo-project] Nearly limitless Image: revisited.

by Sven Van Caekenberghe

On 29 Feb 2012, at 12:58, Janko MivÅ¡ek wrote: > For Pharo there seems everyone is avoiding this idea, mostly because of > image corruption fear. Well, we need to improve the robustness of Pharo > image and VM then this fear will vanish. It is obviously doable, if VW > is reliable enough, why not once Pharo? I am trying to be polite: remarks like this are unfair and not based on the experiences of most of the people in the Pharo community or of the companies using Pharo. It is perfectly possible to have stable long running servers based on Pharo. Sneaking in these phrases in simply spreading FUD. Sven -- Sven Van Caekenberghe http://stfx.eu Smalltalk is the Red Pill

Feb. 29, 2012