- Pharo-dev - lists.pharo.org

Re: [Pharo-project] Helping getting pharo license-clean

by StÃ©phane Ducasse

Hi rob you are in the list of ok commiters now when I run the script I did not add the pharoers to the list of authors. I should do that. Now are you telling to me that you did the following steps and run it without getting a debugger? On which system are you? Stef On May 15, 2009, at 11:02 PM, Rob Rothwell wrote: > I tried to see if my signature is filed; I was quite excited to send > postage all the way to France, but do not see my name in the > signature list, or anywhere for that matter...but (unfortunately), > "RobRothwell" shows up, for example, in Class > DisplayScreen>>fullScreenOn... > > Rob > > On Fri, May 15, 2009 at 1:55 PM, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr > > wrote: > Hi all (and matthew we need your guidance/help) > > We really want to get pharo 1.0 license clean. It would be bad that > 1.0 is not. > But 1.0 should get out before summer. > > Here is a collection of all the information I collected on the topic: > > Here what matthew wrote > > I've done the first step and done a full audit, using Yoshiki's > tools [1], of all the code > in Squeak 3.10.2, with the exception of four packages: > - Monticello > - SUnit, TestRunner, and SUnitImproved > - Universes > - Traits > > I also applied all of yoshiki's etoys changes to 3.10.2. The > algorithm that Yoshiki seemed to follow when cleaning etoys is: > - If the method can be easily deleted, delete it > - Otherwise, use the historical image [1] to revert to the > latest version before the problematic author > - Fix any bugs this causes > > I think we can use the same procedure. Perhaps Yoshiki will > comment. > > So, to help with the relicense, grab the historical image[1], > check out the license audit [2], leave a note on mantis about > what you think you'll work on, and start rolling out change > sets. For your convenience, all of the license audits are > included both in the relicensing tools download and on mantis. > I'll try to keep both up to date as we progress with the > relicense > > [1] The tools I've been using to do the relicense: > http://ftp.squeak.org/4.0alpha/squeak4.0- > relicensingTools.zip > see also: > http://lists.squeakfoundation.org/pipermail/squeak-dev/2008-September/13158… > http://www.squeaksource.com/MethodAuthorship.html > > [2] The 4.0 relicense mantis issue: > http://bugs.squeak.org/view.php?id=6989 > > > The most up-to-date list of ok contributors is at > http://www.squeaksource.com/311/MethodAuthorship-kph.17.mcz > See the class-side methods in Authorship, category data > > > here is another email from matthew > > On Wed, Apr 22, 2009 at 11:10:12AM +0200, stephane ducasse wrote: > > I need your guidance for the MIT conversion effort of Pharo > > Do you have some cycles to help me and telling me how to proceed? > > I'm on holidays (so working remotely so I may allocate some time on > > that). > > I can help you after school gets out in a week. I need to > do a quick count of how many lines of code need to be fixed in > squeak for submission to SFLC. I can do the same for Pharo. I > don't know how strenuous you want to be in your license > conversion. > > The next 3 steps in the license conversion process are, in no > particular order: > > - get a document from yoshiki documenting how etoys was > relicensed (and, by extension, cuis). The board will overview > it with the help of SFLC and determine if it is good enough. > - The board will publish a statement about what is and is not a > trivial-enough change to ignore. > - I will count up the lines of code to be converted, and > determine how many of them are trivial and non-trivial > > You should chat with Randal Schwartz about what you need to > decide before you can start relicensing, and what you need to > document in case your relicense needs to be reviewed in the > future. Yoshiki and Juan did not do this and thus we have no > evidence yet that Yoshiki's relicense is strong enough. > > I documented the priorities for squeak 4.0 in this email: > http://lists.squeakfoundation.org/pipermail/squeak-dev/2009-February/134083… > > You should probably figure out your priorities for the relicense > as well, so you know whose standards you need to meet. > > Since that email was written, Randal has indeed heard back from > SFLC, but I haven't heard what they had to say yet. > > -- > Matthew Fulmer -- http://mtfulmer.wordpress.com/ > > > > > > _______________________________________________ > 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

May 16, 2009

Re: [Pharo-project] [ANN] OCompletion improves the accuracy of eCompletion

by Robbes Romain

Hi Hernan, OCompletion is actually an add-on for ECompletion, so it does not disable it. As for the class name issue, and the case where the receiver is self, I will look into it. Thanks for the feedback, Romain ________________________________________ From: pharo-project-bounces(a)lists.gforge.inria.fr [pharo-project-bounces(a)lists.gforge.inria.fr] On Behalf Of Hernan Wilkinson [hernan.wilkinson(a)gmail.com] Sent: Friday, May 15, 2009 3:32 PM To: Pharo-project(a)lists.gforge.inria.fr Subject: Re: [Pharo-project] [ANN] OCompletion improves the accuracy of eCompletion Yes, I tried it on a pharo-dev image. I thought that there could be a conflict between ecomp and ocomp but I also thought that ocomp would disable ecomp when loading... anyway, I did not have the time to check that so I went to the previous image. (BTW I also tried to load ocomp on a pharo-core image and I could not because there is no Installer class. I could not either with the monticello brower... ) So, the performance issue was when typing class names. For example, I typed 'Ar' press tab and it took a lot to show the valid names. Selected Array and the started to delete it and it also took a lot of time to delete each character. I tried the same thing on an pharo-dev image without ocomp and worked fine... so there must be something going on. Another thing that I found rare was this: When loading Ocomp I selected the packages Kernel, Collection and the one I'm working on. In a test class, when writing a test I wrote: "self as" and "assert:" was not in the options. I think it is because I did not select SUnit as a package, but I found this annoying... If I sending a message to self I think it should show all the messages that object could understood (that is, the ones defined in the class I working with and it superclasses. Ok, object has a lot of messages that could bother here, but at least it show show the messages up to object minus one I think) Hope this help. Hernan. On Fri, May 15, 2009 at 3:36 AM, Damien Cassou <damien.cassou(a)gmail.com<mailto:damien.cassou@gmail.com>> wrote: 2009/5/15 Romain Robbes <romain.robbes(a)lu.unisi.ch<mailto:romain.robbes@lu.unisi.ch>>: > Could you describe the symptoms more precisely? I was especially careful so > that the completion should be actually faster > than the default one (ECompletion looks at all the symbols in the image, > whereas OCompletion looks in a much more > restricted set of symbols), so I'm especially surprised of this. > Have you installed other packages in the same image that could cause this? Since he uses pharo-dev, I guess Hernan has eCompletion too on his image. Is that a problem? -- Damien Cassou http://damiencassou.seasidehosting.st "Lambdas are relegated to relative obscurity until Java makes them popular by not having them." James Iry _______________________________________________ Pharo-project mailing list Pharo-project(a)lists.gforge.inria.fr<mailto:Pharo-project@lists.gforge.inria.fr> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

May 16, 2009

Re: [Pharo-project] How to log in console in Windows

by Schwab,Wilhelm K

Mariano, Hmmmm. Something not working in Windows - imaging that :) Sorry, I feel your pain and am trying to run for the Penguin-filled hills as we speak. Dumb question: have you tried the OS Process plugin. I've only skimmed it in Squeak/Pharo, but there is a similar goodie for Dolphin, and it goes to great pains to put streams around the OS streams. No surprise, Windows manages to make a mess of it. Sorry that I can't be of any real help. Bill ________________________________ From: pharo-project-bounces(a)lists.gforge.inria.fr [mailto:pharo-project-bounces@lists.gforge.inria.fr] On Behalf Of Mariano Martinez Peck Sent: Friday, May 15, 2009 9:37 AM To: The general-purpose Squeak developers list; Pharo Development Subject: [Pharo-project] How to log in console in Windows Hi folks! I am calling a external shared library with FFI. This library does some puts(""); or printf(); In Linux, I open Squeak from command line and I can see the outputs of puts() and printf() in the console where I started Squeak. The problem is in Windows. I tried opening Squeak from cmd but didn't work. Nothing is shown in the console. Is there a way of doing this ? Thanks a lot in advance, Mariano

May 16, 2009

Re: [Pharo-project] [ANN] OCompletion improves the accuracy of eCompletion

by Hernan Wilkinson

Thanks! I understand now. On Fri, May 15, 2009 at 6:12 PM, Damien Cassou <damien.cassou(a)gmail.com>wrote: > 2009/5/15 Hernan Wilkinson <hernan.wilkinson(a)gmail.com>: > > ups... I'm confused... in previous mails I read that one image is > > Pharo-core, another pharo-dev and another pharo-web... > > Why should we not make that distinction? sorry but I do not > understand.... > > This is not your fault, it's a bit complex :-). The ultimate truth is: > > - Pharo-Core: a small core which only interests the Pharo-Core > developers and for deployment > - Pharo: based on Pharo-Core, you have a complete Smalltalk system for > developers > > pharo-dev is the old name of Pharo and we should try to avoid using > it. Pharo-web is just a Pharo with Seaside, Magritte and Pier. > > Bye > > -- > Damien Cassou > http://damiencassou.seasidehosting.st > > "Lambdas are relegated to relative obscurity until Java makes them > popular by not having them." James Iry > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >

May 15, 2009

Re: [Pharo-project] [ANN] OCompletion improves the accuracy of eCompletion

by Damien Cassou

2009/5/15 Hernan Wilkinson <hernan.wilkinson(a)gmail.com>: > ups... I'm confused... in previous mails I read that one image is > Pharo-core, another pharo-dev and another pharo-web... > Why should we not make thatÂ distinction? sorry but I do not understand.... This is not your fault, it's a bit complex :-). The ultimate truth is: - Pharo-Core: a small core which only interests the Pharo-Core developers and for deployment - Pharo: based on Pharo-Core, you have a complete Smalltalk system for developers pharo-dev is the old name of Pharo and we should try to avoid using it. Pharo-web is just a Pharo with Seaside, Magritte and Pier. Bye -- Damien Cassou http://damiencassou.seasidehosting.st "Lambdas are relegated to relative obscurity until Java makes them popular by not having them." James Iry

May 15, 2009

Re: [Pharo-project] Helping getting pharo license-clean

by Rob Rothwell

I tried to see if my signature is filed; I was quite excited to send postage all the way to France, but do not see my name in the signature list, or anywhere for that matter...but (unfortunately), "RobRothwell" shows up, for example, in Class DisplayScreen>>fullScreenOn... Rob On Fri, May 15, 2009 at 1:55 PM, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr > wrote: > Hi all (and matthew we need your guidance/help) > > We really want to get pharo 1.0 license clean. It would be bad that > 1.0 is not. > But 1.0 should get out before summer. > > Here is a collection of all the information I collected on the topic: > > Here what matthew wrote > > I've done the first step and done a full audit, using Yoshiki's > tools [1], of all the code > in Squeak 3.10.2, with the exception of four packages: > - Monticello > - SUnit, TestRunner, and SUnitImproved > - Universes > - Traits > > I also applied all of yoshiki's etoys changes to 3.10.2. The > algorithm that Yoshiki seemed to follow when cleaning etoys is: > - If the method can be easily deleted, delete it > - Otherwise, use the historical image [1] to revert to the > latest version before the problematic author > - Fix any bugs this causes > > I think we can use the same procedure. Perhaps Yoshiki will > comment. > > So, to help with the relicense, grab the historical image[1], > check out the license audit [2], leave a note on mantis about > what you think you'll work on, and start rolling out change > sets. For your convenience, all of the license audits are > included both in the relicensing tools download and on mantis. > I'll try to keep both up to date as we progress with the > relicense > > [1] The tools I've been using to do the relicense: > http://ftp.squeak.org/4.0alpha/squeak4.0-relicensingTools.zip > see also: > > http://lists.squeakfoundation.org/pipermail/squeak-dev/2008-September/13158… > http://www.squeaksource.com/MethodAuthorship.html > > [2] The 4.0 relicense mantis issue: > http://bugs.squeak.org/view.php?id=6989 > > > The most up-to-date list of ok contributors is at > http://www.squeaksource.com/311/MethodAuthorship-kph.17.mcz > See the class-side methods in Authorship, category data > > > here is another email from matthew > > On Wed, Apr 22, 2009 at 11:10:12AM +0200, stephane ducasse wrote: > > I need your guidance for the MIT conversion effort of Pharo > > Do you have some cycles to help me and telling me how to proceed? > > I'm on holidays (so working remotely so I may allocate some time on > > that). > > I can help you after school gets out in a week. I need to > do a quick count of how many lines of code need to be fixed in > squeak for submission to SFLC. I can do the same for Pharo. I > don't know how strenuous you want to be in your license > conversion. > > The next 3 steps in the license conversion process are, in no > particular order: > > - get a document from yoshiki documenting how etoys was > relicensed (and, by extension, cuis). The board will overview > it with the help of SFLC and determine if it is good enough. > - The board will publish a statement about what is and is not a > trivial-enough change to ignore. > - I will count up the lines of code to be converted, and > determine how many of them are trivial and non-trivial > > You should chat with Randal Schwartz about what you need to > decide before you can start relicensing, and what you need to > document in case your relicense needs to be reviewed in the > future. Yoshiki and Juan did not do this and thus we have no > evidence yet that Yoshiki's relicense is strong enough. > > I documented the priorities for squeak 4.0 in this email: > > http://lists.squeakfoundation.org/pipermail/squeak-dev/2009-February/134083… > > You should probably figure out your priorities for the relicense > as well, so you know whose standards you need to meet. > > Since that email was written, Randal has indeed heard back from > SFLC, but I haven't heard what they had to say yet. > > -- > Matthew Fulmer -- http://mtfulmer.wordpress.com/ > > > > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >

May 15, 2009

Re: [Pharo-project] [ANN] OCompletion improves the accuracy of eCompletion

by Hernan Wilkinson

>Try the very latest Pharo (please stop using the name pharo-dev (I >know I'm still doing it but that's a mistake :-)) ups... I'm confused... in previous mails I read that one image is Pharo-core, another pharo-dev and another pharo-web... Why should we not make that distinction? sorry but I do not understand.... On Fri, May 15, 2009 at 12:06 PM, Damien Cassou <damien.cassou(a)gmail.com>wrote: > 2009/5/15 Hernan Wilkinson <hernan.wilkinson(a)gmail.com>: > > Yes, I tried it on a pharo-dev image. I thought that there could be a > > conflict between ecomp and ocomp but I also thought that ocomp would > disable > > ecomp when loading... anyway, I did not have the time to check that so I > > went to the previous image. > > Try the very latest Pharo (please stop using the name pharo-dev (I > know I'm still doing it but that's a mistake :-))(this looks like > Lisp)) which now has OCompletion installed: > http://pharo-project.org/download. > > > (BTW I also tried to load ocomp on a pharo-core image and I could not > > because there is no Installer class. I could not either with the > monticello > > brower... ) > > Installer is not installed by default. ScriptLoader new > installingInstaller. > > > So, the performance issue was when typing class names. For example, I > typed > > 'Ar' press tab and it took a lot to show the valid names. Selected Array > and > > the started to delete it and it also took a lot of time to delete each > > character. I tried the same thing on an pharo-dev image without ocomp and > > worked fine... so there must be something going on. > > Can't reproduce on latest Pharo. > > > -- > Damien Cassou > http://damiencassou.seasidehosting.st > > "Lambdas are relegated to relative obscurity until Java makes them > popular by not having them." James Iry > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >

May 15, 2009

Re: [Pharo-project] vw inspired gui tool

by Dale Henrichs

----- "Steve Wirts" <stevewirts(a)gmail.com> wrote: | Stef, | | Only submorphs, the gui hierarchy are shown, this is a really useful | capability when trying to understand deeply nested complex gui | structures. | I suppose we could put a toggle button on a menu option on explorer to | turn | on differernt tree-paths to show as the sub-children. | | I will try to do that. | | thanks | steve Steve, Not sure if this applies or not. But the OT inspectors (regular inspector and chasing inspector) allow for one to provide custom inspectors on a class by class basis (even the ability to provide multiple views)... If that's what you're getting at, I can give you more information about what it takes to extend the inspector. Dale

May 15, 2009

Re: [Pharo-project] vw inspired gui tool

by StÃ©phane Ducasse

Yes this would be great. I think that this is important to get better tools not more tools :) Stef On May 15, 2009, at 10:17 PM, Steve Wirts wrote: > Stef, > > Only submorphs, the gui hierarchy are shown, this is a really useful > capability when trying to understand deeply nested complex gui > structures. I suppose we could put a toggle button on a menu option > on explorer to turn on differernt tree-paths to show as the sub- > children. > > I will try to do that. > > thanks > steve > > > On Fri, May 15, 2009 at 3:33 PM, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr > > wrote: > Hi steve > > what is the difference with the explorer. > If you want to have flash just send us a new menu for explorer? > > Stef > > On May 15, 2009, at 8:46 PM, Steve Wirts wrote: > > > Folks, > > > > Attached is a very simple gui tool inspired from a tool that used to > > exist in VW smalltalk years ago. > > I suspect it could be rewritten as a single function. Please > > incorporate into Pharo if you think > > it's useful. > > > > Steve > > <ViewExplorer.st>_______________________________________________ > > 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

May 15, 2009

Re: [Pharo-project] vw inspired gui tool

by Steve Wirts

Stef, Only submorphs, the gui hierarchy are shown, this is a really useful capability when trying to understand deeply nested complex gui structures. I suppose we could put a toggle button on a menu option on explorer to turn on differernt tree-paths to show as the sub-children. I will try to do that. thanks steve On Fri, May 15, 2009 at 3:33 PM, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr > wrote: > Hi steve > > what is the difference with the explorer. > If you want to have flash just send us a new menu for explorer? > > Stef > > On May 15, 2009, at 8:46 PM, Steve Wirts wrote: > > > Folks, > > > > Attached is a very simple gui tool inspired from a tool that used to > > exist in VW smalltalk years ago. > > I suspect it could be rewritten as a single function. Please > > incorporate into Pharo if you think > > it's useful. > > > > Steve > > <ViewExplorer.st>_______________________________________________ > > 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 >

May 15, 2009