- Pharo-dev - lists.pharo.org

[pharo-project/pharo-core]

by GitHub

Branch: refs/tags/30495 Home: https://github.com/pharo-project/pharo-core

Oct. 17, 2013

IMPORTANT: RPackage behavior changed and Nautilus updated

by Esteban Lorenzano

Hi, I just important an important change in Pharo. I made a new step over the integration of RPackage into the system, and as a result there is a different behavior. Although 99% of users shouldn't notice it, that can drive to headaches to tool developers, like Thierry and his AltBrowser (but I'm not sure about that) :) I'm sorry, but this is inevitable... and part of a long path discussed and adopted since almost 1 1/2 years. So, what's the change? Now 1 RPackage 0 1 MCPackage. As a result, package granularity is coarse, and then categories for class are stored as package tags. Other consequence is that RPackageSet is now more or less useless, but I'm keeping it because it provides caching services for monticello, and the speed up resulting is good :) This change opens the path to remove old PackageInfo, finally :) To take the new structure into account (and to add a long time awaited feature), I modified nautilus to use a tree instead a simple list (finally!) And since I was in the mood, I moved groups to same list, taking the idea of the screenshot Stef sent the other day (lots of good ideas that we should adopt there), and then I removed the "groups" button... in functionality is good, but visually not so much... I accept ideas about :) THERE WILL BE BUGS! This is a really deep and important change, I'm pushing it because we need to test and fix. So, do not hesitate to send your reports! Also, there are features in modified-nautilus that are missing. It would be good to have a list of them so I can implement it :) Please, report any issue around. Enjoy, Esteban

Oct. 17, 2013

[regression reporter]regression occurred

by no-reply＠ci.inria.fr

https://ci.inria.fr/pharo/job/Pharo-3.0-Update-Step-2.1-Validation-M-Z/labe… 3 regressions found. RPackage.Tests.RPackageReadOnlyCompleteSetupTest.testAddTag RPackage.Tests.RPackageReadOnlyCompleteSetupTest.testAddTagNames RPackage.Tests.RPackageSetTest.testPolymorphicWithPackageInfo

Oct. 17, 2013

[regression reporter]regression occurred

by no-reply＠ci.inria.fr

https://ci.inria.fr/pharo/job/Pharo-3.0-Update-Step-2.1-Validation-M-Z/labe… 3 regressions found. RPackage.Tests.RPackageReadOnlyCompleteSetupTest.testAddTag RPackage.Tests.RPackageReadOnlyCompleteSetupTest.testAddTagNames RPackage.Tests.RPackageSetTest.testPolymorphicWithPackageInfo

Oct. 17, 2013

Re: [Pharo-dev] [PROVENANCE INTERNET] Re: default monospaced code font

by Igor Stasenko

On 17 October 2013 12:02, Henrik Johansen <henrik.s.johansen(a)veloxit.no>wrote: > > On Oct 17, 2013, at 11:40 , Esteban Lorenzano <estebanlm(a)gmail.com> wrote: > > > > > On Oct 17, 2013, at 11:35 AM, Goubier Thierry <thierry.goubier(a)cea.fr> > wrote: > > > >> Once the correct fonts are installed, the pdf looks very different :) > >> > >> Revised judgement: > >> > >> Source code pro is not bad for a monospaced font. > >> Source sans pro is impressive of density (9), and has this additional > readability on DejaVu sans for the $l. > >> > >> I'll take Source Sans Pro, then. > >> > >> But Pharo is not seeing it in the dialog. Is that because I did a user > specific install of the fonts (and not system wide) and Pharo is not > checking the user-specific fonts? > > yes, that's exactly what happens :) > > but also pharo should contain those fonts (it should not need an > installation)... currently I'm working on figure out why that does not work. > > > > Esteban > > I noticed display of fallback-fonts is broken for StrikeFonts now, (select > Bitmap Deja Vu and any non-western language (or French) to have the > palindrome preview X out) I guess that's not a big issue if the plan for > 3.0 is embedding true type/freetype fonts in the image and removing > StrikeFonts altogether though. > > there will be a single fallback raster font (single resolution, to not take much space), for the cases when freetype unavail. > Cheers, > Henry > -- Best regards, Igor Stasenko.

Oct. 17, 2013

[regression reporter]regression occurred

by no-reply＠ci.inria.fr

https://ci.inria.fr/pharo/job/Pharo-3.0-Update-Step-2.1-Validation-M-Z/labe… 3 regressions found. RPackage.Tests.RPackageReadOnlyCompleteSetupTest.testAddTag RPackage.Tests.RPackageReadOnlyCompleteSetupTest.testAddTagNames RPackage.Tests.RPackageSetTest.testPolymorphicWithPackageInfo

Oct. 17, 2013

Re: [Pharo-dev] [PROVENANCE INTERNET] Re: default monospaced code font

by Henrik Johansen

On Oct 17, 2013, at 11:40 , Esteban Lorenzano <estebanlm(a)gmail.com> wrote: > > On Oct 17, 2013, at 11:35 AM, Goubier Thierry <thierry.goubier(a)cea.fr> wrote: > >> Once the correct fonts are installed, the pdf looks very different :) >> >> Revised judgement: >> >> Source code pro is not bad for a monospaced font. >> Source sans pro is impressive of density (9), and has this additional readability on DejaVu sans for the $l. >> >> I'll take Source Sans Pro, then. >> >> But Pharo is not seeing it in the dialog. Is that because I did a user specific install of the fonts (and not system wide) and Pharo is not checking the user-specific fonts? > yes, that's exactly what happens :) > but also pharo should contain those fonts (it should not need an installation)... currently I'm working on figure out why that does not work. > > Esteban I noticed display of fallback-fonts is broken for StrikeFonts now, (select Bitmap Deja Vu and any non-western language (or French) to have the palindrome preview X out) I guess that's not a big issue if the plan for 3.0 is embedding true type/freetype fonts in the image and removing StrikeFonts altogether though. Cheers, Henry

Oct. 17, 2013

Re: [Pharo-dev] Spec error

by Benjamin

Fixed: https://pharo.fogbugz.com/default.asp?11916 Ben On Oct 16, 2013, at 9:50 PM, Sean P. DeNigris <sean(a)clipperadams.com> wrote: > I manually fed in an instance of by changing the code to: > actionToPerformWithSelector: selector arguments: args > | instance | > instance := spec instance. > instance = #DialogWindowSpec ifTrue: [ instance := SpecDialogWindow new ]. > ^ instance perform: selector withArguments: args > > But then I got: MessageNotUnderstood: ByteSymbol>>changeTableLayout (where the symbol was #Panel) > > It seems that #actionToPerformWithSelector:arguments: expects "spec instance" to return an actual instance, but is getting Symbols instead. It seems to be a bug in Spec because none of this has anything to do with my code, and if I change "self new openDialogWithSpec" to "self new openWithSpec" it opens correctly. > > On Oct 16, 2013, at 1:01 PM, "Benjamin Van Ryseghem-2 [via Smalltalk]" <[hidden email]> wrote: > >> Try replacing it with DlalogWindowModel >> >> Ben >> >> On Oct 16, 2013, at 5:27 PM, Sean P. DeNigris <<a href="x-msg://206/user/SendEmail.jtp?type=node&node=4715064&i=0" target="_top" rel="nofollow" link="external">[hidden email]> wrote: >> >>> Benjamin Van Ryseghem-2 wrote >>>> Without further information, I can not do much more >>> >>> I sent #openDialogWithSpec, and the offending symbol is #DialogWindowSpec >>> >>> >>> >>> ----- >>> Cheers, >>> Sean >>> -- >>> View this message in context: http://forum.world.st/Spec-error-tp4714069p4715018.html >>> Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com. >>> >> >> >> >> If you reply to this email, your message will be added to the discussion below: >> http://forum.world.st/Spec-error-tp4714069p4715064.html >> To unsubscribe from Spec error, click here. >> NAML > > Cheers, > Sean > > View this message in context: Re: Spec error > Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com.

Oct. 17, 2013

Re: [Pharo-dev] Spec error

by Benjamin

It is indeed some left over from the Spec refactoring :) I thought the code you were talking about was your code where actually it's part of Spec :) I am fixing it right now :P Ben On Oct 16, 2013, at 9:50 PM, Sean P. DeNigris <sean(a)clipperadams.com> wrote: > I manually fed in an instance of by changing the code to: > actionToPerformWithSelector: selector arguments: args > | instance | > instance := spec instance. > instance = #DialogWindowSpec ifTrue: [ instance := SpecDialogWindow new ]. > ^ instance perform: selector withArguments: args > > But then I got: MessageNotUnderstood: ByteSymbol>>changeTableLayout (where the symbol was #Panel) > > It seems that #actionToPerformWithSelector:arguments: expects "spec instance" to return an actual instance, but is getting Symbols instead. It seems to be a bug in Spec because none of this has anything to do with my code, and if I change "self new openDialogWithSpec" to "self new openWithSpec" it opens correctly. > > On Oct 16, 2013, at 1:01 PM, "Benjamin Van Ryseghem-2 [via Smalltalk]" <[hidden email]> wrote: > >> Try replacing it with DlalogWindowModel >> >> Ben >> >> On Oct 16, 2013, at 5:27 PM, Sean P. DeNigris <<a href="x-msg://206/user/SendEmail.jtp?type=node&node=4715064&i=0" target="_top" rel="nofollow" link="external">[hidden email]> wrote: >> >>> Benjamin Van Ryseghem-2 wrote >>>> Without further information, I can not do much more >>> >>> I sent #openDialogWithSpec, and the offending symbol is #DialogWindowSpec >>> >>> >>> >>> ----- >>> Cheers, >>> Sean >>> -- >>> View this message in context: http://forum.world.st/Spec-error-tp4714069p4715018.html >>> Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com. >>> >> >> >> >> If you reply to this email, your message will be added to the discussion below: >> http://forum.world.st/Spec-error-tp4714069p4715064.html >> To unsubscribe from Spec error, click here. >> NAML > > Cheers, > Sean > > View this message in context: Re: Spec error > Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com.

Oct. 17, 2013

Re: [Pharo-dev] Smalltalkhub is exposing private information

by Yuriy Tymchuk

Yes, and also post-commit url's should not be visible to everyone, as this can provide someone the data which can be used to spam ci server with jobs for example. I don't know how important is smalltalkhub anyway. uko On Oct 17, 2013, at 11:40 AM, Norbert Hartl <norbert(a)hartl.name> wrote: > When you look at someones page you can read commit messages of packages that belong to a private project. You can't see the project nor the package but it is listed in recent activities. I filed a bug > > https://code.google.com/p/smalltalk-hub/issues/detail?id=32 > > Norbert

Oct. 17, 2013