- Pharo-dev - lists.pharo.org

[Pharo-project] Bug with valueWithReceiver:arguments:

by Frederic Pluquet

Hello, I found a fondamental bug in Squeak and Pharo. The next code 11 / 2 gives the fraction (11/2). It's correct. But the next code (SmallInteger>>#/) valueWithReceiver: 11 arguments: {2} gives 1 ! The problem is the method valueWithReceiver:arguements: is used hugely with method wrappers... After long time of debugging, I found a point to debug: this method don't have the good behavior with compiled methods having a primitive that fails and executes some code after (as in SmallInteger>>#/ method when the division don't give a whole integer). In fact, when I send this message, the vm executes normally the compiled method but, in place of returns simply the good result, seems to rerun the the compiled method with other arguments (completly wrong) and returns so a wrong result. For example, (SmallInteger>>#/) valueWithReceiver: 11 arguments: {2} has the following execution trace : 2 isZero | 2 = 0 | returns: false returns: false 2 isMemberOf: SmallInteger | 2 class | returns: SmallInteger | SmallInteger == SmallInteger | returns: true returns: true Fraction numerator: 101 denominator: 2 | Fraction new | | Fraction basicNew | | returns: a Fraction instance | | (a Fraction instance) initialize | | returns: a Fraction instance | returns: a Fraction instance | a Fraction instance setNumerator: 101 denominator: 2 | | 2 = 0 | | returns: false | | 101 asInteger | | returns: 101 | | 2 asInteger | | returns: 2 | | 2 abs | | | 2 < 0 | | | returns: false | | returns: 2 | | 2 < 0 | | returns: false | returns: (101/2) returns: (101/2) (101/2) reduced | 101 = 0 | returns: false | 101 gcd: 2 | | 101 = 0 | | returns: false | | 2 \\ 101 | | returns: 2 | | 2 = 0 | | returns: false | | 101 \\ 2 | | returns: 1 | | 1 = 0 | | returns: false | | 2 \\ 1 | | returns: 0 | | 0 = 0 | | returns: true | | 1 abs | | | 1 < 0 | | | returns: false | | returns: 1 | returns: 1 | 101 // 1 | returns: 101 | 2 // 1 | returns: 2 | 2 = 1 | returns: false | Fraction numerator: 101 denominator: 2 | | Fraction new | | | Fraction basicNew | | | returns: a Fraction instance | | | (a Fraction instance) initialize | | | returns: a Fraction instance | | returns: a Fraction instance | | (a Fraction instance) setNumerator: 101 denominator: 2 | | | 2 = 0 | | | returns: false | | | 101 asInteger | | | returns: 101 | | | 2 asInteger | | | returns: 2 | | | 2 abs | | | | 2 < 0 | | | | returns: false | | | returns: 2 | | | 2 < 0 | | | returns: false | | returns: (101/2) | returns: (101/2) returns: (101/2) 2 isZero | 2 = 0 | returns: false returns: false false isMemberOf: SmallInteger | false class | returns: False | False == SmallInteger | returns: false returns: false Please help me to fix this bug. I really need it works fine ! FrÃ©d -- FrÃ©dÃ©ric Pluquet UniversitÃ© Libre de Bruxelles (ULB) Assistant http://www.ulb.ac.be/di/fpluquet

Oct. 26, 2008

Re: [Pharo-project] about MC browser changes

by Damien Pollet

This update breaks if you have MC windows open, because MCWorkingCopyBrowser doesn't initialize its order instvar lazily. On Sat, Oct 25, 2008 at 8:46 PM, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> wrote: > so I get it when I do updates -> 10106 > I retried from a 10104 I got your fixes. > > Stef > > On Oct 25, 2008, at 9:24 PM, Lukas Renggli wrote: > >> I am also talking about the sorting fixes. >> >> Lukas >> >> On 10/25/08, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> wrote: >>> >>> wait I was talking about your sorting menu. >>> Were you talking about the jf fixes? >>> >>> Stef >>> >>> On Oct 25, 2008, at 8:59 PM, Lukas Renggli wrote: >>> >>>> On Sat, Oct 25, 2008 at 8:41 PM, StÃ©phane Ducasse >>>> <stephane.ducasse(a)inria.fr> wrote: >>>>> >>>>> 10106 contained your fix. >>>>> Ok excellent! This was a really cool addition. >>>> >>>> Wired. I loaded all updates but it was not there? So I loaded it >>>> manually. >>>> >>>> Anyway, if it is in now this is good. People should just make sure to >>>> close their Monticello Browser before pdating, otherwise there is a >>>> debugger coming up. >>>> >>>> Cheers, >>>> Lukas >>>> >>>> -- >>>> Lukas Renggli >>>> http://www.lukas-renggli.ch >>>> >>>> _______________________________________________ >>>> 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 >>> >> >> >> -- >> Lukas Renggli >> http://www.lukas-renggli.ch >> >> _______________________________________________ >> 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 > -- Damien Pollet type less, do more [ | ] http://people.untyped.org/damien.pollet

Oct. 26, 2008

[Pharo-project] [ann] new core image for #10107 online

by Marcus Denker

https://gforge.inria.fr/frs/download.php/10317/Pharo0.1Core-10107.zip -- Marcus Denker -- denker(a)iam.unibe.ch http://www.iam.unibe.ch/~denker

Oct. 26, 2008

Re: [Pharo-project] changes file

by Marcus Denker

On 22.10.2008, at 22:01, John M McIntosh wrote: > I note I downloaded the current pharo image/changes for pushing out > on the iPhone and > noticed the changes was 10.5MB, but if I condense it becomes 4.2MB > which is much easier to > deal with if I include it in an iPhone.app > Yes, the changes file starts to bug me, too. Especially as we will hit the 32MB wall at some point. > I'm wondering here if you need to push out the changes with all the > method versions in it, or > should it be condenced first and rely on changes from MC to show the > history of how the methods > evolved? mcz seems to be a bit too cumbersom... they get very large very fast, and then it's very slow to unpack all of them. I think we should make a file-system based solution: the history browser should just look into a directort ClassName/methodName/ and then show all the files there (named by timestamp). Later one can even simply make a web-server based solution: http-get to a server for all the history. I love .changes as a changelog, this we should keep. But any release done (even the intermediate ones) should come with an empty log. Marcus -- Marcus Denker -- denker(a)iam.unibe.ch http://www.iam.unibe.ch/~denker

Oct. 26, 2008

Re: [Pharo-project] OldSocket removal

by StÃ©phane Ducasse

On Oct 25, 2008, at 10:24 PM, Dan Corneanu wrote: > On Sat, Oct 25, 2008 at 10:45 PM, StÃ©phane Ducasse > <stephane.ducasse(a)inria.fr> wrote: >> I would really that other get a chances to review what you are doing. >> So I will harvest up to dc 49 and after wait that you get some >> other eyes to >> check what you are doing. >> Ok? > > Sure. > Had I been directed sooner to PharoTaskForces, I would have not messed > up the PharoInbox and OB repositories. You did not mess anything. Don't worry! > Regarding the license agreement: > I'll send a signed copy. > > Regarding the socket tests: > Maybe it's better to wait for John to upload the code to a monticello > repository. > After all I have just filled the changesets in my image > and fixed some constants. No do not wait for other. if you have the code publish it in MC. This is good. I'm not saying that you are doing a bad job. I'm just bad in network so I would like have somebody else looking at what you did.

Oct. 26, 2008

Re: [Pharo-project] OldSocket removal

by Dan Corneanu

On Sat, Oct 25, 2008 at 10:45 PM, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> wrote: > I would really that other get a chances to review what you are doing. > So I will harvest up to dc 49 and after wait that you get some other eyes to > check what you are doing. > Ok? Sure. Had I been directed sooner to PharoTaskForces, I would have not messed up the PharoInbox and OB repositories. Regarding the license agreement: I'll send a signed copy. Regarding the socket tests: Maybe it's better to wait for John to upload the code to a monticello repository. After all I have just filled the changesets in my image and fixed some constants.

Oct. 25, 2008

[Pharo-project] [ANN] 10107

by StÃ©phane Ducasse

This provides a Fix and Test for String numArgs http://code.google.com/p/pharo/issues/detail?id=237 - Nicolas Cellier: Just a small improvment required to speed up a little number reading: rewrite a condition (a & b) as (a and: [b]). - Issue 275: TODO: remove OldSocket. Change HTTPSocket superclass from OldSimpleClientSocket to Socket. Reimplementing OldSocket protocol into Socket (some methods were missing). Changing CrLf variable access to (String crlf).

Oct. 25, 2008

Re: [Pharo-project] about MC browser changes

by StÃ©phane Ducasse

so I get it when I do updates -> 10106 I retried from a 10104 I got your fixes. Stef On Oct 25, 2008, at 9:24 PM, Lukas Renggli wrote: > I am also talking about the sorting fixes. > > Lukas > > On 10/25/08, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> wrote: >> wait I was talking about your sorting menu. >> Were you talking about the jf fixes? >> >> Stef >> >> On Oct 25, 2008, at 8:59 PM, Lukas Renggli wrote: >> >>> On Sat, Oct 25, 2008 at 8:41 PM, StÃ©phane Ducasse >>> <stephane.ducasse(a)inria.fr> wrote: >>>> 10106 contained your fix. >>>> Ok excellent! This was a really cool addition. >>> >>> Wired. I loaded all updates but it was not there? So I loaded it >>> manually. >>> >>> Anyway, if it is in now this is good. People should just make sure >>> to >>> close their Monticello Browser before pdating, otherwise there is a >>> debugger coming up. >>> >>> Cheers, >>> Lukas >>> >>> -- >>> Lukas Renggli >>> http://www.lukas-renggli.ch >>> >>> _______________________________________________ >>> 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 >> > > > -- > Lukas Renggli > http://www.lukas-renggli.ch > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >

Oct. 25, 2008

Re: [Pharo-project] OldSocket removal

by StÃ©phane Ducasse

I would really that other get a chances to review what you are doing. So I will harvest up to dc 49 and after wait that you get some other eyes to check what you are doing. Ok? Stef On Oct 25, 2008, at 9:36 PM, Dan Corneanu wrote: > On Thu, Oct 23, 2008 at 12:24 AM, John M McIntosh > <johnmci(a)smalltalkconsulting.com> wrote: >> Ok, Michael R suggested I bundle these up, recheck them, so who >> would I send >> them to? >> > I've played with the change sets from > http://www.smalltalkconsulting.com/html/OTNotes4.html. > I've made some changes to the tests, to fit the current Socket and > NetNameResolver implementations. If it is of any help, I can put up a > mcz so others can have a look to the code. > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >

Oct. 25, 2008

Re: [Pharo-project] OldSocket removal

by StÃ©phane Ducasse

Yes you can also use PharoTaskForces for that. Now dan could you sign the license agreement and send it to me. Stef On Oct 25, 2008, at 9:36 PM, Dan Corneanu wrote: > On Thu, Oct 23, 2008 at 12:24 AM, John M McIntosh > <johnmci(a)smalltalkconsulting.com> wrote: >> Ok, Michael R suggested I bundle these up, recheck them, so who >> would I send >> them to? >> > I've played with the change sets from > http://www.smalltalkconsulting.com/html/OTNotes4.html. > I've made some changes to the tests, to fit the current Socket and > NetNameResolver implementations. If it is of any help, I can put up a > mcz so others can have a look to the code. > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >

Oct. 25, 2008