November 2012 - Pharo-dev - lists.pharo.org

Re: [Pharo-project] [ANN] NativeBoost v 1.7

by Igor Stasenko

okay, so it was exactly what i suspected: - MC version history serializer assumes that all strings are bytestring (latin-1) while i put some strange characted in commit log, which turned it into WideString.. and that's why it was unable to read it back. I uploaded a new version of package NativeBoost-Core-IgorStasenko.90 after hot-fixing MC serializer. So, now ConfigurationOfNativeBoost v 1.7 can be loaded. -- Best regards, Igor Stasenko.

Nov. 28, 2012

Re: [Pharo-project] mess in package extensions

by Pavel Krivanek

http://code.google.com/p/pharo/issues/detail?id=7062 -- Pave On Wed, Nov 28, 2012 at 12:01 PM, Camillo Bruni <camillobruni(a)gmail.com> wrote: > we got a method accidentally registered twice with monticello: > - once in the package extension > - once in the real category > > the only solution to get it back to normal, was an intermediate commit > with said method being renamed |-(

Nov. 28, 2012

Re: [Pharo-project] mess in package extensions

by Camillo Bruni

same here, we got a method accidentally registered twice with monticello: - once in the package extension - once in the real category the only solution to get it back to normal, was an intermediate commit with said method being renamed |-( => can you open an issue? On 2012-11-28, at 06:22, Pavel Krivanek <pavel.krivanek(a)gmail.com> wrote: > Hi, > > we have got some problems with package extensions and merging of them > during updates. This problems became evident in this issues: > - http://code.google.com/p/pharo/issues/detail?id=6956 > - http://code.google.com/p/pharo/issues/detail?id=7036 > > In the first one we several times tired to move the methods to the > correct package. At the end we tried to remove them in a fixing slice > but how can you see in the latest image, this methods are still there. > This methods cause 5 of 6 test failures in the Pharo Kernel (the last > one is the Undeclared test with 5 undeclared classes) > > In the second issue the Form class methods are in the correct method > category but as you may see in the Monticello browser, they are not in > the Graphics-Files package. The Pharo Kernel shrinking removes > packages with respect to method category and doesn't take an opinion > of Monticello into account. So this methods are (correctly) removed > during shrinking. But because they are not part of the Graphics-Files > package by Monticello, they are not correctly loaded with the rest of > the Morphic and thus the Jenkins job "Pharo Kernel 2.0 Reload by > Monticello" is failing. > > Cheers, > -- Pavel >

Nov. 28, 2012

[Pharo-project] who are "cipt" and "tbh"?

by Esteban Lorenzano

Hi, Can I ask people to use real names instead name initials? Otherwise I get lost about who pushes changes :) Esteban

Nov. 28, 2012

[Pharo-project] almost ready for 2.0 beta

by Esteban Lorenzano

Hi, I've been working last days to put pharo 2.0 ready to beta. So far we have just two missing integrations: - NativeBoost - Key Chain we deferred Opal to first cycle of next year release (february), because is not ready for inclusion yet. So. we will be on beta as soon as Igor and Ben give me the proper version for their inclusion :) Esteban

Nov. 28, 2012

[Pharo-project] mess in package extensions

by Pavel Krivanek

Hi, we have got some problems with package extensions and merging of them during updates. This problems became evident in this issues: - http://code.google.com/p/pharo/issues/detail?id=6956 - http://code.google.com/p/pharo/issues/detail?id=7036 In the first one we several times tired to move the methods to the correct package. At the end we tried to remove them in a fixing slice but how can you see in the latest image, this methods are still there. This methods cause 5 of 6 test failures in the Pharo Kernel (the last one is the Undeclared test with 5 undeclared classes) In the second issue the Form class methods are in the correct method category but as you may see in the Monticello browser, they are not in the Graphics-Files package. The Pharo Kernel shrinking removes packages with respect to method category and doesn't take an opinion of Monticello into account. So this methods are (correctly) removed during shrinking. But because they are not part of the Graphics-Files package by Monticello, they are not correctly loaded with the rest of the Morphic and thus the Jenkins job "Pharo Kernel 2.0 Reload by Monticello" is failing. Cheers, -- Pavel

Nov. 28, 2012

[Pharo-project] Spec question: menues

by Torsten Bergmann

Can anyone point me to a spec example with a (context) menu with icons and where we can enable/disable specific menue items depending on state?

Nov. 28, 2012

[Pharo-project] [update 2.0] #20421

by Marcus Denker

20421 ----- Issue 7060: New Spec version http://code.google.com/p/pharo/issues/detail?id=7060 Issue 7054: InputEventHandler unregistering should nil the eventFetcher http://code.google.com/p/pharo/issues/detail?id=7054 Issue 7041: remove theme UIThemeStandardSqueak http://code.google.com/p/pharo/issues/detail?id=7041 Issue 7036: move Form class>>#fromFileNamed and fromBinaryStream to Graphic-Files http://code.google.com/p/pharo/issues/detail?id=7036 Issue 7031: Package HistoryCollection should be merged into Nautilus http://code.google.com/p/pharo/issues/detail?id=7031 Diff information: http://ss3.gemstone.com/ss/Pharo20/Spec-Core-MarcusDenker.66.diff http://ss3.gemstone.com/ss/Pharo20/Polymorph-Widgets-MarcusDenker.739.diff http://ss3.gemstone.com/ss/Pharo20/Nautilus-MarcusDenker.380.diff http://ss3.gemstone.com/ss/Pharo20/Kernel-MarcusDenker.1259.diff http://ss3.gemstone.com/ss/Pharo20/HistoryCollection-MarcusDenker.16.diff http://ss3.gemstone.com/ss/Pharo20/Graphics-Files-MarcusDenker.37.diff http://ss3.gemstone.com/ss/Pharo20/Graphics-Display Objects-MarcusDenker.98.diff -- Marcus Denker -- denker(a)acm.org http://www.marcusdenker.de

Nov. 28, 2012

Re: [Pharo-project] Getting started with Athens

by Vanessa PeÃ±a Araya

Yes, the symlink worked in the end. My problem was that I had one called "libcairo.so.2.11..." and I needed one "libcairo.so" :P Thanks for the help! Vanessa. On 11/27/2012 05:29 PM, Esteban Lorenzano wrote: > make a symlink (assuming you are working on linux/mac) where the rest of the plugins are. > > Esteban > > On Nov 27, 2012, at 9:19 PM, Vanessa PeÃ±a Araya <van.c.pena(a)gmail.com> wrote: > >> Hi Sebastian and Esteban, >> >> Thanks for replying. I checked and I already have the libcairo2 installed. I actually can see Roassal working with Cairo in VW so I guess that's ok. I use a 32bits Ubuntu. >> >> But I still can't make it work :( This error in particular is because the connection with cairo? >> >> Thanks, >> Vanessa. >> >> >> On 11/27/2012 01:46 PM, Esteban Lorenzano wrote: >>> be sure to install the 32bits version >>> >>> >>> On Nov 27, 2012, at 5:37 PM, Sebastian Nozzi <sebnozzi(a)gmail.com> wrote: >>> >>>> Hello Araya, >>>> >>>> you might still need to install the library for your host operating system. >>>> >>>> See here: >>>> >>>> http://www.pharocasts.com/2012/06/athens-and-native-boost-at-work.html >>>> >>>> For Debian/Ubuntu systems you would need to do: >>>> >>>> apt-get install libcairo2 # as root in Debian >>>> >>>> Or: >>>> >>>> sudo apt-get install libcairo2 # as non-root in Ubuntu >>>> >>>> Hope this helps! >>>> >>>> Sebastian >>>> >>>> P.S. Future versions of Pharo, if I'm not mistaken, will include this >>>> dependency... ? >>>> >>>> >>>> 2012/11/27 Vanessa PeÃ±a Araya <van.c.pena(a)gmail.com>: >>>>> Hi, >>>>> >>>>> I'm trying to get started with Athens but I'm having the following errors: >>>>> >>>>> Pharo 1.4, loaded last version configurationOfAthens: >>>>> 'failed to get a symbol address: cairo_image_surface_create' >>>>> >>>>> I also tried in this in a 2.0 image and I got the same error. >>>>> >>>>> I'm trying with this VM: >>>>> >>>>> https://ci.lille.inria.fr/pharo/view/Cog/job/NativeBoost-Cog-VM/Architectur… >>>>> >>>>> Any help, please? :) >>>>> >>>>> Thanks, >>>>> Vanessa. >>> >> > >

Nov. 27, 2012

Re: [Pharo-project] a probably stupid question on canvas

by Igor Stasenko

And this is not the only example. same goes for BitBlt GrafPort WarpBlt which are products of adding new features to VM without reconsidering/cleaning up the whole model/interface(s). And of course, as always, keeping backwards compatibility is a good excuse. ;) so we end up with many faces of same beast.. and as original developers gone, nobody dares go inside and clean it up. -- Best regards, Igor Stasenko.

Nov. 27, 2012