April 2010 - Pharo-dev - lists.pharo.org

Re: [Pharo-project] FFI in 1.1

by StÃ©phane Ducasse

> > Frankly, its a dev tool. What else it is if not a dev tool? > > Just a library, but not a dev tool from my point of view. I understand from "dev tool" a tool aimed for make the developrs life better: good browsers, autocompletion, color hightitlying, refactoring tools, etc. yes communication with C library. so this is just a tool too.

April 25, 2010

Re: [Pharo-project] [Metacello] Problem with ConfigurationOfGofer in Pharo1.0 [WAS] Re: ConfigurationOfGofer class

by Mariano Martinez Peck

Thanks for the investigation Dale...now...suppose I want to fix it and I want to force to load gofer again. How can I do that ? because if I try to install it normally, it won't do it as it is already installed. Thanks Mariano On Tue, Apr 20, 2010 at 6:21 PM, Dale Henrichs <dale.henrichs(a)gemstone.com>wrote: > Mariano, > > The Pharo-1.0.changes file looks to be corrupted. If you edit the the > changes file and search for ConfigurationOfGofer here's what you see: > > !ConfigurationOfGofer methodsFor: 'accessing' stamp: 'nk 3/5/2005 15:28'! > sNil! ! > > !ConfigurationOfGofer methodsFor: 'baselines'! > 1 into: [:sum :each | sum * (self foo: each)]! ! > > !ConfigurationOfGofer methodsFor: 'versions'! > tring]. > ^string! ! > > !ConfigurationOfGofer methodsFor: 'versions'! > RefactoryTestDataApp methodsFor: 'inline'! ! > > !ConfigurationOfGofer methodsFor: 'versions'! > inlineComponent > | a | > a := 5. > ^a class > superclass; > hasImmediateInstances; > yourself! ! > > !ConfigurationOfGofer class methodsFor: 'accessing'! > h | > each printString. > ^anObject]! ! > > !ConfigurationOfGofer class methodsFor: 'loading'! > ! ! > > !ConfigurationOfGofer class methodsFor: 'metacello tool support' stamp: 'md > 3/14/2006 16:47'! > malltalk at: #MyTest > ifAbsent: > [| collection | > collection := #(). > Smalltalk at: #MyTest put: collection]! ! > > !ConfigurationOfGofer class methodsFor: 'metacello tool support'! > ng new! ! > > !ConfigurationOfGofer class methodsFor: 'metacello tool support'! > or: 'inline'! ! > > !ConfigurationOfGofer class methodsFor: 'private' stamp: 'DaleHenrichs > 1/25/2010 13:14'! > bootstrapPackage: aString from: aPath > | reposito > ! ! > > !ConfigurationOfGofer class methodsFor: 'private'! > 4:51'! ! > > So somehow the source was corrupted ... the compiled methods are correct > since: > > ConfigurationOfGofer project versions > > gives the correct answer. The source following ConfigurationOfGofer looks > fine so I can't imagine how the source would have been corrupted especially > since the methods were compiled correctly... > > Dale > > ----- "Mariano Martinez Peck" <marianopeck(a)gmail.com> wrote: > > | wow...you are right....but I have NO idea how that could have been > | done. > | > | Adrian: look at the instance methods of the class ConfigurationOfGofer > | and > | you will notice what Erwann says. > | > | Dale, do you have an idea how this can be done ? As I remember > | correctly, I > | just evaluated: > | > | > | Gofer new > | squeaksource: 'MetacelloRepository'; > | package: 'ConfigurationOfPharo'; > | load. > | > | ((Smalltalk at: #ConfigurationOfPharo) project version: '1.0-10517') > | perform: #silently: with: true; perform: #load. > | > | > | to generate the Pharo1.0 dev image. This code will bootstrap Metacello > | and > | thus, install Gofer. Do you think there could be a problem there ? > | > | Thanks Erwann for reporting! Open a bug ticket please. > | > | Cheers > | > | Mariano > | > | On Tue, Apr 20, 2010 at 4:52 PM, Adrian Lienhard > | <alienhard(a)netstyle.ch>wrote: > | > | > Hi Erwann > | > > | > What does scrambled mean? > | > > | > I cc-ed to Mariano, he may more quickly see whether there is > | something > | > wrong. > | > > | > Cheers, > | > Adrian > | > > | > ___________________ > | > http://www.adrian-lienhard.ch/ > | > > | > On Apr 20, 2010, at 16:34 , Erwann Wernli wrote: > | > > | > > "ConfigurationOfGofer class" seem scrambled in the download of > | Pharo 1.0 > | > form the website -- don't know what happened or if it's only on my > | image. > | > > > | > > Erwann > | > > > | > > > | > > > | > > | > > | > | > | -- > | Subscription settings: > | http://groups.google.com/group/metacello/subscribe?hl=en >

April 25, 2010

Re: [Pharo-project] Fwd: Easy SqueakVM build

by Mariano Martinez Peck

Excellent. There is something I dont understand.... (probably because my shell scripts are very limited) When you do: echo "Download PharoCore 1.0" if [ ! -e $pharo_archive ]; then wget $download_url fi but... download_url is download_url="https://gforge.inria.fr/frs/download.php/26775/$pharo_archive" and $pharo_archive is pharo_archive=$pharo_dir.zip and $pharo_dir is pharo_dir=$working_dir/$pharo_version and finally, $working_dir is working_dir=`pwd`/out So...I don't understand how $download_url can be https://gforge.inria.fr/frs/download.php/26775/PharoCore-1.0-10517.zip Cheers Mariano 2010/4/20 laurent laffont <laurent.laffont(a)gmail.com> > If some people are interested ... > > Cheers, > > Laurent Laffont > > > ---------- Forwarded message ---------- > From: laurent laffont <laurent.laffont(a)gmail.com> > Date: Tue, Apr 20, 2010 at 10:16 PM > Subject: Easy SqueakVM build > To: Squeak Virtual Machine Development Discussion < > vm-dev(a)lists.squeakfoundation.org> > > > Hi, > > I've written a script to build the squeak vm from scratch on Linux in one > command. It's here http://github.com/lolgzs/easy_squeakvm > To build the VM, just get the the script and > ./easy_squeakvm.sh > > This will > - checkout squeakvm sources from svn repository > - download PharoCore image > - load VMMaker into PharoCore > - generate the interpreter > - build the VM > - put binaries in out/squeakvm, ready to use > > I was tired of typing the same commands over and over :) > > Cheers, > > Laurent Laffont > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >

April 25, 2010

Re: [Pharo-project] FFI in 1.1

by Mariano Martinez Peck

On Wed, Apr 21, 2010 at 5:54 AM, Igor Stasenko <siguctua(a)gmail.com> wrote: > 2010/4/20 Mariano Martinez Peck <marianopeck(a)gmail.com>: > > > > > > On Thu, Apr 15, 2010 at 9:31 PM, StÃ©phane Ducasse > > <stephane.ducasse(a)inria.fr> wrote: > >> > >> Hi all > >> > >> Do we add FFI by default in 1.1? > > > > Noooooooooo!!! > > > > I wouldn't include neither FFI or Alien FFI in neither PharoCore or > PharoDev > > image. > > > > FFI is an EXTERNAL package, NOT core, neither a dev tool. > > > > Frankly, its a dev tool. What else it is if not a dev tool? > Just a library, but not a dev tool from my point of view. I understand from "dev tool" a tool aimed for make the developrs life better: good browsers, autocompletion, color hightitlying, refactoring tools, etc. > You have to be extremely careful, as with anything which using C or > binary interfaces, > but apart from it, i see nothing wrong with it. > How else you can easily bind an external API to image? Writing a plugin? > You could, but what the point in spending a lot of hours writing a > primitives for each > API function, when in the end, you still will be doing pretty same > thing: using external api? > > > That's only my opinion. > > > > Cheers > > > > Mariano > > > >> > >> This was the plan so. > >> > >> Stef > >> > >> _______________________________________________ > >> Pharo-project mailing list > >> Pharo-project(a)lists.gforge.inria.fr > >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > > > > > _______________________________________________ > > Pharo-project mailing list > > Pharo-project(a)lists.gforge.inria.fr > > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > > > > > -- > Best regards, > Igor Stasenko AKA sig. > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >

April 25, 2010

Re: [Pharo-project] Serialization of font object

by Mariano Martinez Peck

On Fri, Apr 23, 2010 at 9:22 PM, Henrik Johansen < henrik.s.johansen(a)veloxit.no> wrote: > > On Apr 23, 2010, at 7:37 33PM, StÃ©phane Ducasse wrote: > > > may be using the storeOn: aStream and readFrom: mechanism? > > > I doubt it would work for FreeType fonts, which hold external handles. > mmmmm but do you want also to serialize external handles ? what are you going to do with them when you load the objects again to memory ? > For StrikeFonts, it may, but even if it did, you'd probably end up > importing a new Font instance (complete with an XX KB bitmap), for each > Morph... > > Cheers, > Henry > > > > > On Apr 23, 2010, at 4:28 PM, Henrik Johansen wrote: > > > >> On Apr 23, 2010, at 4:01 48PM, nullPointer wrote: > >> > >>> > >>> > >>> Regards > >>> > >>> I see in Pharo a few ways for create a font, StrikeFont, LogicalFont... > I > >>> need generate dinamically code in Smalltalk for build UIs, and my > problem > >>> now is when I have serialize the font of a widget. Which could be the > best > >>> way for do that? In other objects like arrays, points, associations is > >>> possible get that calling #printString method. Exists some equivalence > for > >>> that in fonts, or is needed implement that mechanism of zero? > >>> > >>> Thanks for answers > >> > >> I would store the the name/boldness/size etc., then use (LogicalFont > familyName: pointSize: stretchValue:weightValue: slantValue: ) realFont > >> to discover the closest font available when loading the morph into an > image again. > >> > >> Cheers, > >> Henry > >> _______________________________________________ > >> Pharo-project mailing list > >> Pharo-project(a)lists.gforge.inria.fr > >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > > > > > _______________________________________________ > > Pharo-project mailing list > > Pharo-project(a)lists.gforge.inria.fr > > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >

April 25, 2010

[Pharo-project] #findString:startingAt:caseSensitive: broken

by Lukas Renggli

The method #findString:startingAt:caseSensitive: seems to be quite broken. This is sort of a critical bug, because the widely used #includesSubstring:caseSensitive: (in Pier and Seaside) depends on it. http://code.google.com/p/pharo/issues/detail?id=2353 First, the method #findString:startingAt:caseSensitive: has all kinds of strange comments and warning from a unfinished implementation. Furthermore -- and this is the bug I am encountering -- it replaces the string to be searched with an empty WideString if either the search or the searched string are WideString instances. Thus in the case any of the participating strings is a WideString the code never finds a match. WideString new findSubstring: key in: self startingAt: start matchTable: CaseInsensitiveOrder In addition, when used in combination with #includesSubstring:caseSensitive: the code raises a "subscript is out of bounds" exception, because the start index points to an different string than the empty one the search is actually performed on. Lukas -- Lukas Renggli www.lukas-renggli.ch

April 25, 2010

Re: [Pharo-project] making O/Ecompletion even better

by StÃ©phane Ducasse

On Apr 25, 2010, at 3:55 PM, Tudor Girba wrote: > Hi Stef, > > No you are not :). > > I also mentioned once two more issues: > - Enter should not trigger completion because it interferes with code formatting. I would only keep tab as a means to trigger completion and to select something yes you are probably right. I was in favor of this inverse... but I learned :) > - right now, when pressing the down key because you saw a good completion item in the second position there is a visible delay because the ECompletion is triggered. ECompletion should either be integrated in the OCompletion directly, or should be triggered only when we hit the bottom of the three items. In the current form, having the completion item in the second position is almost as bad as not having it at all. > > Unfortunately, I do not have time to look into these issues, but I can offer testing :) romain :) please....... with the eye of the cat in shrek > > Cheers, > Doru > > > On 25 Apr 2010, at 15:45, StÃ©phane Ducasse wrote: > >> Hi romain >> >> Often when I edit code, the completion works well but I end up been forced to delete part of the text manually. >> Here is the scenario >> >> >> classDefinedSelector foo... >> ^ >> cursor >> >> Then I type meta >> >> the system proposes me metaclassDefinedSelector >> I type enter >> >> and I get >> >> metaclassDefinedSelector classDefinedSelector foo >> ^ >> >> so I have to remove by hand classDefinedSelector. >> >> So it would be good if we could eat the current text when doing the substitution. >> >> Am I the only one getting such behavior? >> >> Stef >> _______________________________________________ >> Pharo-project mailing list >> Pharo-project(a)lists.gforge.inria.fr >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > -- > www.tudorgirba.com > > "From an abstract enough point of view, any two things are similar." > > > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

April 25, 2010

Re: [Pharo-project] making O/Ecompletion even better

by Tudor Girba

Hi Stef, No you are not :). I also mentioned once two more issues: - Enter should not trigger completion because it interferes with code formatting. I would only keep tab as a means to trigger completion and to select something - right now, when pressing the down key because you saw a good completion item in the second position there is a visible delay because the ECompletion is triggered. ECompletion should either be integrated in the OCompletion directly, or should be triggered only when we hit the bottom of the three items. In the current form, having the completion item in the second position is almost as bad as not having it at all. Unfortunately, I do not have time to look into these issues, but I can offer testing :) Cheers, Doru On 25 Apr 2010, at 15:45, StÃ©phane Ducasse wrote: > Hi romain > > Often when I edit code, the completion works well but I end up been > forced to delete part of the text manually. > Here is the scenario > > > classDefinedSelector foo... > ^ > cursor > > Then I type meta > > the system proposes me metaclassDefinedSelector > I type enter > > and I get > > metaclassDefinedSelector classDefinedSelector foo > ^ > > so I have to remove by hand classDefinedSelector. > > So it would be good if we could eat the current text when doing the > substitution. > > Am I the only one getting such behavior? > > Stef > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project -- www.tudorgirba.com "From an abstract enough point of view, any two things are similar."

April 25, 2010

Re: [Pharo-project] making O/Ecompletion even better

by Hilaire Fernandes

StÃ©phane Ducasse a Ã©crit : > > so I have to remove by hand classDefinedSelector. > > So it would be good if we could eat the current text when doing the substitution. > > Am I the only one getting such behavior? Yes, it makes sense. Otherwise I have to admit I found the completion pretty neat and smart in some part. I guess it does some statistick on the search terms because I get back quite quickly what I am looking for. Hilaire

April 25, 2010

[Pharo-project] making O/Ecompletion even better

by StÃ©phane Ducasse

Hi romain Often when I edit code, the completion works well but I end up been forced to delete part of the text manually. Here is the scenario classDefinedSelector foo... ^ cursor Then I type meta the system proposes me metaclassDefinedSelector I type enter and I get metaclassDefinedSelector classDefinedSelector foo ^ so I have to remove by hand classDefinedSelector. So it would be good if we could eat the current text when doing the substitution. Am I the only one getting such behavior? Stef

April 25, 2010