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

Re: [Pharo-project] About versions on PharoInbox

by StÃ©phane Ducasse

no this is working and simple On Jan 17, 2010, at 6:44 PM, Marcus Denker wrote: > > On Jan 17, 2010, at 6:22 PM, Miguel Enrique CobÃ¡ Martinez wrote: > >> El dom, 17-01-2010 a las 09:51 +0100, Levente Uzonyi escribiÃ³: >> >>> Sure, 1.0 and 1.1 shares the same inbox. >>> >>> >> >> Oh, :) that explains it. > > Maybe we should change that at some point... > > Marcus > > > -- > Marcus Denker -- http://www.marcusdenker.de > INRIA Lille -- Nord Europe. Team RMoD. > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

Jan. 17, 2010

Re: [Pharo-project] NewCompiler load on PharoCore

by StÃ©phane Ducasse

jean-baptite should help after the tools paper. Stef On Jan 17, 2010, at 6:31 PM, Marcus Denker wrote: > > On Jan 14, 2010, at 7:56 PM, Eliot Miranda wrote: > >> >> >> On Thu, Jan 14, 2010 at 12:45 AM, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> wrote: >> Elliot >> >> it would be realllllllly cool to have this in the squeak-vm what should be done to get it? >> >> I'll upload the Newspeak VMMaker package to a suitable repository. I could also try and do an extract and merge. >> Alternatively someone who is interested could do the same starting from the Newspeak release. >> >> > > I would be interested to have a look at some point... maybe not the next weeks, but it would really be nice to have > that. So I am sure I will find time at some point. > > Marcus > > -- > Marcus Denker -- http://www.marcusdenker.de > INRIA Lille -- Nord Europe. Team RMoD. > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

Jan. 17, 2010

Re: [Pharo-project] Smalltalk (Pharo) and String equality

by Nicolas Cellier

2010/1/17 Levente Uzonyi <leves(a)elte.hu>: > On Sun, 17 Jan 2010, Martin McClure wrote: > >> Levente Uzonyi wrote: >>> On Sun, 17 Jan 2010, Nicolas Cellier wrote: >>> >>>> One of my first bug in st80 was >>>> >>>> strange >>>> Â Â ^('foo' writeStream) nextPutAll: '-bar'; contents. >>>> >>>> At that time, writing pastEnd did use become: >>>> >>>> But if you really have time to pick an immutability bit in Newspeak, >>>> we don't have to care anymore. >>> >>> If I had a free bit in the object header, I'd use it for extending the >>> identity hash instead of useless things like immutability. IMO it was a >>> mistake to add immutability for literals in other smalltalks only to >>> avoid possible errors generated by the broken stream/collection semantics. >> >> Literal immutability and other object immutability is only one use of >> the so-called "immutability" bit. In reality, this bit is a "track > > http://www.cincomsmalltalk.com/blog/blogView?entry=3232889180 > Sure, object databases are more interesting applications, since you generally make such mistake only once. As a beginner, I put a halt in the method once identified as the source of the problem, and did not understand my mistake at first shot since it started working again (being recompiled). The good thing is that it helped me learning some Smalltalk superpowers faster :). Nicolas >> modifications" bit, and I find the other uses much more valuable. > > I wonder how that works. > > > Levente > >> >> Regards, >> >> -Martin >> >> _______________________________________________ >> 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 >

Jan. 17, 2010

[Pharo-project] About metacello meta repository and persistency of code

by StÃ©phane Ducasse

Hi guys something I would really like to be able to have is that projects described using metacello and explicitly targetting for pharo are also saved in a specific repository in addition to their own projects. This way I would like to have a possible archive of all the projects described because I saw some sources like the one of salty pickles disappeared and you cannot load or browse old code. Right now I imagine that most of the projects are either on squeak source or ob or lukas one. Now with Gofer it should be easy to copy files around. Let me know what you think but I really hate to get a deadlink when I'm looking for something. This way it also ensures that the metacatalog is always complete. Stef

Jan. 17, 2010

Re: [Pharo-project] Smalltalk (Pharo) and String equality

by Levente Uzonyi

On Sun, 17 Jan 2010, Martin McClure wrote: > Levente Uzonyi wrote: >> On Sun, 17 Jan 2010, Nicolas Cellier wrote: >> >>> One of my first bug in st80 was >>> >>> strange >>> ^('foo' writeStream) nextPutAll: '-bar'; contents. >>> >>> At that time, writing pastEnd did use become: >>> >>> But if you really have time to pick an immutability bit in Newspeak, >>> we don't have to care anymore. >> >> If I had a free bit in the object header, I'd use it for extending the >> identity hash instead of useless things like immutability. IMO it was a >> mistake to add immutability for literals in other smalltalks only to >> avoid possible errors generated by the broken stream/collection semantics. > > Literal immutability and other object immutability is only one use of > the so-called "immutability" bit. In reality, this bit is a "track http://www.cincomsmalltalk.com/blog/blogView?entry=3232889180 > modifications" bit, and I find the other uses much more valuable. I wonder how that works. Levente > > Regards, > > -Martin > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >

Jan. 17, 2010

Re: [Pharo-project] Smalltalk (Pharo) and String equality

by Martin McClure

Levente Uzonyi wrote: > On Sun, 17 Jan 2010, Nicolas Cellier wrote: > >> One of my first bug in st80 was >> >> strange >> ^('foo' writeStream) nextPutAll: '-bar'; contents. >> >> At that time, writing pastEnd did use become: >> >> But if you really have time to pick an immutability bit in Newspeak, >> we don't have to care anymore. > > If I had a free bit in the object header, I'd use it for extending the > identity hash instead of useless things like immutability. IMO it was a > mistake to add immutability for literals in other smalltalks only to > avoid possible errors generated by the broken stream/collection semantics. Literal immutability and other object immutability is only one use of the so-called "immutability" bit. In reality, this bit is a "track modifications" bit, and I find the other uses much more valuable. Regards, -Martin

Jan. 17, 2010

Re: [Pharo-project] Smalltalk (Pharo) and String equality

by Levente Uzonyi

On Sun, 17 Jan 2010, Nicolas Cellier wrote: > One of my first bug in st80 was > > strange > ^('foo' writeStream) nextPutAll: '-bar'; contents. > > At that time, writing pastEnd did use become: > > But if you really have time to pick an immutability bit in Newspeak, > we don't have to care anymore. If I had a free bit in the object header, I'd use it for extending the identity hash instead of useless things like immutability. IMO it was a mistake to add immutability for literals in other smalltalks only to avoid possible errors generated by the broken stream/collection semantics. Levente > > Nicolas > > 2010/1/17 Marcus Denker <marcus.denker(a)inria.fr>: >> >> On Jan 17, 2010, at 6:30 PM, laurent laffont wrote: >> >>> Hi, >>> >>> Something strange to me: >>> >>> '2' = '2'. -> true Â (OK) >>> (2 asString) = (2 asString). -> true Â (OK) >>> >>> (2 asString) == (2 asString). -> false Â (OK) >>> '2' == '2'. -> true Â ????? >>> >> >> Yes, the same string in the same method points to the same entry in the literalframe. >> >> Which explains why this method works: >> >> strange >> Â Â Â Â 'hello world' isString ifTrue: ['hello world' become: {0}]. >> Â Â Â Â 'hello world' at: 1 put: ('hello world' at:1) + 1 . >> Â Â Â Â ^'hello world' Â at: 1. >> >> Put this method in Object, than print multiple times: >> >> Â Â Â Â Object new strange >> >> :-) >> >> -- >> Marcus Denker Â -- http://www.marcusdenker.de >> INRIA Lille -- Nord Europe. Team RMoD. >> >> >> _______________________________________________ >> 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 >

Jan. 17, 2010

Re: [Pharo-project] us the vm -memory in an image ?

by John M McIntosh

On 2010-01-17, at 9:08 AM, HernÃ¡n Morales Durand wrote: > Hi John, > > 2010/1/16 John M McIntosh <johnmci(a)smalltalkconsulting.com>: >> >> On 2010-01-16, at 4:08 PM, csrabak(a)bol.com.br wrote: >> >>> >>> Is there a way to download a file of the presentation (be it powerpoint or PDF)? >> >> A subset at >> http://www.esug.org/data/ESUG2006/pres/gctalkjohnmacintosh.pdf >> >> Generally I give a talk on the current paper to clients who pay for GC consulting. It takes about 3-4 hours to give at the moment. >> > > Have you started experiments or do you plan to include information > about (dead) timestamped garbage collection mechanisms? > > I don't understand some of the statistical graphics (pages 80, 82-86), There aren't any graphs on pages 80,83, do yo have the page numbers correct? > have you used MMU or BMU curves to display the GC responsiveness? > (just thinking, it would be cool to show application-centric GC > metrics) do you have some links, google is not helpful this morning > > Best regards, > > HernÃ¡n -- =========================================================================== John M. McIntosh <johnmci(a)smalltalkconsulting.com> Twitter: squeaker68882 Corporate Smalltalk Consulting Ltd. http://www.smalltalkconsulting.com ===========================================================================

Jan. 17, 2010

Re: [Pharo-project] Smalltalk (Pharo) and String equality

by Nicolas Cellier

2010/1/17 Marcus Denker <marcus.denker(a)inria.fr>: > > On Jan 17, 2010, at 6:57 PM, Nicolas Cellier wrote: >> But if you really have time to pick an immutability bit in Newspeak, >> we don't have to care anymore. >> > Hmm... the code below should even work with immutable strings. I *never* touch > the string! I do not even manipulate the literal array via #at:put:... > > I just become the pointer to an Array generated with {}, which would not > be immutable... > > So I wonder if the immutability solved the problem. (immutable literals > do solve many other bugs when the object itself gets modified...). > I extrapolated a bit... In VW, this leads to an immutability error: ('abc' become: #(1 2 3)) at: 1 Nicolas >>> strange >>> Â Â Â Â 'hello world' isString ifTrue: ['hello world' become: {0}]. >>> Â Â Â Â 'hello world' at: 1 put: ('hello world' at:1) + 1 . >>> Â Â Â Â ^'hello world' Â at: 1. > > > > Â Â Â Â Marcus > > -- > Marcus Denker Â -- http://www.marcusdenker.de > INRIA Lille -- Nord Europe. Team RMoD. > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >

Jan. 17, 2010

Re: [Pharo-project] Smalltalk (Pharo) and String equality

by Marcus Denker

On Jan 17, 2010, at 6:57 PM, Nicolas Cellier wrote: > But if you really have time to pick an immutability bit in Newspeak, > we don't have to care anymore. > Hmm... the code below should even work with immutable strings. I *never* touch the string! I do not even manipulate the literal array via #at:put:... I just become the pointer to an Array generated with {}, which would not be immutable... So I wonder if the immutability solved the problem. (immutable literals do solve many other bugs when the object itself gets modified...). >> strange >> 'hello world' isString ifTrue: ['hello world' become: {0}]. >> 'hello world' at: 1 put: ('hello world' at:1) + 1 . >> ^'hello world' at: 1. Marcus -- Marcus Denker -- http://www.marcusdenker.de INRIA Lille -- Nord Europe. Team RMoD.

Jan. 17, 2010