October 2013 - Pharo-dev - lists.pharo.org

Re: [Pharo-dev] SmalltalkHub red

by Camillo Bruni

so is seaside: https://ci.inria.fr/pharo-contribution/job/Seaside/ On 2013-10-09, at 17:48, Damien Cassou <damien.cassou(a)gmail.com> wrote: > On Wed, Oct 9, 2013 at 4:47 PM, Yuriy Tymchuk <yuriy.tymchuk(a)me.com> wrote: >> Do you know who is working with grease? It doesn't sound too hard to solve this. > > > and Grease seems to be ok: https://ci.inria.fr/pharo-contribution/job/Grease/

Oct. 9, 2013

Re: [Pharo-dev] Strane error while loading code > Syntax Error : Message pattern expected

by Camillo Bruni

can you get the method source? try inspecting the MCMethodAddition further down the stack... On 2013-10-09, at 19:19, "phil(a)highoctane.be" <phil(a)highoctane.be> wrote: > > I am encountering a weird error (in 2.0) while loading code from a monticello package. > > <09-10-13 19-10-36.png> > > This comes after a "compiling" progress box, then I get "Reloading..." and that one. > > A right click on the top entry gives me the following: > > <09-10-13 19-14-36.png> > > The method is empty and its protocol is #'as yet unclassified' > > Really annoying > > The package was saved from a working image w/ the app working well. > > Ideas on how to fix this? > Phil > >

Oct. 9, 2013

Re: [Pharo-dev] Strane error while loading code > Syntax Error : Message pattern expected

by ClÃ©ment Bera

This kind of errors are triggered by the parser. Something is probably miswritten in the method. Are you loading something with the old syntax ? (I mean <- instead of :=). Try in 3.0 to see if Opal can handle this case. 2013/10/9 phil(a)highoctane.be <phil(a)highoctane.be> > > I am encountering a weird error (in 2.0) while loading code from a > monticello package. > > [image: Inline image 1] > > This comes after a "compiling" progress box, then I get "Reloading..." and > that one. > > A right click on the top entry gives me the following: > > [image: Inline image 2] > > The method is empty and its protocol is #'as yet unclassified' > > Really annoying > > The package was saved from a working image w/ the app working well. > > Ideas on how to fix this? > Phil > > >

Oct. 9, 2013

Re: [Pharo-dev] ShortRunArray not in Pharo 3.0 anymore, leading to problems with FFI-Kernel-tbn.25

by phil＠highoctane.be

Well, no issues. It is too complex for me to get things loaded in 3.0 and be delivering what we need to, so I am back to 2.0 now. Spent the day on dealing with configs and packages. I am more educated about all that now anyway. Phil On Wed, Oct 9, 2013 at 7:43 PM, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr>wrote: > Phil > > we are ready to help but then give the expression that shows the problem. > May be this is just that snapshotcello was simply never loaded in 30 > because moose is in 20. > and that there is just a platform definition missing. > > Stef > > > Snashotcello not loading in 3.0 as well and the 2.x trick. > > Also, Gofer gives me some headaches when loading from directories. > > Phil > > - > On Wed, Oct 9, 2013 at 4:05 PM, Esteban Lorenzano <estebanlm(a)gmail.com>wrote: > >> >> On Oct 9, 2013, at 3:30 PM, Marcus Denker <marcus.denker(a)inria.fr> wrote: >> >> > >> > On Oct 9, 2013, at 3:22 PM, phil(a)highoctane.be wrote: >> > >> >> There is an "ExternalObject class >>initialize" that attemps to place >> itself after ShortRunArray, which isn't there, leasgin to "Error: nil >> cannot be found in Smalltalk dictionary". >> >> >> >> Just to let you know, >> >> >> > >> > Do we support non-nativeBoost FFI in Pharo3? >> >> not yet >> >> > >> > >> >> (Was testing the loading of my confs in 3.0 with the 2.0 monticello >> trick". >> >> >> >> Phil >> > >> >> >> > >

Oct. 9, 2013

Re: [Pharo-dev] ShortRunArray not in Pharo 3.0 anymore, leading to problems with FFI-Kernel-tbn.25

by StÃ©phane Ducasse

Phil we are ready to help but then give the expression that shows the problem. May be this is just that snapshotcello was simply never loaded in 30 because moose is in 20. and that there is just a platform definition missing. Stef > Snashotcello not loading in 3.0 as well and the 2.x trick. > > Also, Gofer gives me some headaches when loading from directories. > > Phil > > - > On Wed, Oct 9, 2013 at 4:05 PM, Esteban Lorenzano <estebanlm(a)gmail.com> wrote: > > On Oct 9, 2013, at 3:30 PM, Marcus Denker <marcus.denker(a)inria.fr> wrote: > > > > > On Oct 9, 2013, at 3:22 PM, phil(a)highoctane.be wrote: > > > >> There is an "ExternalObject class >>initialize" that attemps to place itself after ShortRunArray, which isn't there, leasgin to "Error: nil cannot be found in Smalltalk dictionary". > >> > >> Just to let you know, > >> > > > > Do we support non-nativeBoost FFI in Pharo3? > > not yet > > > > > > >> (Was testing the loading of my confs in 3.0 with the 2.0 monticello trick". > >> > >> Phil > > > > >

Oct. 9, 2013

Weird errors while loading code... reason why

by phil＠highoctane.be

There is stuff like this in the fileouts and in the MCZ source.st !XXXrView methodsFor: 'as yet unclassified' stamp: 'PhilippeBack 9/28/2013 00:25'! ! ! So, nothing in the source of the method. Don't know how this occured, maybe refactorings Removing these things allow me to filein the source.st Phil

Oct. 9, 2013

Strane error while loading code > Syntax Error : Message pattern expected

by phil＠highoctane.be

I am encountering a weird error (in 2.0) while loading code from a monticello package. [image: Inline image 1] This comes after a "compiling" progress box, then I get "Reloading..." and that one. A right click on the top entry gives me the following: [image: Inline image 2] The method is empty and its protocol is #'as yet unclassified' Really annoying The package was saved from a working image w/ the app working well. Ideas on how to fix this? Phil

Oct. 9, 2013

Re: [Pharo-dev] ShortRunArray not in Pharo 3.0 anymore, leading to problems with FFI-Kernel-tbn.25

by ClÃ©ment Bera

I guess we can put back ShortRunArray, RunArray and the 3rd one. It was in Ballon, but there were no users in the image, so I removed them. We need to support FFI without NativeBoost I guess, or we would need to reimplement all what was done with FFI in NativeBoost. 2013/10/9 phil(a)highoctane.be <phil(a)highoctane.be> > Not directly related, I am loading my full stack and it includes both > projets. > > Speaking of which, ConfigurationOfSeaside3 understands Pharo3.x things I > think but loaded with a 2.x compatibility trick, fails (like in loading > ConfigurationOfGrease which loads ConfigurationOfRefactoringBrowser, which > fails miserably. > > So, if a config is available in for 3.x, metacello platforms should be 3.x > and 2.x if not there yet. > > Well, kind of difficult to use Pharo3 with its fixes when one wants to use > the other projects. > > Loads to Gofer and manual loads to do to get out of this. > > Phil >

Oct. 9, 2013

Re: [Pharo-dev] Please contributors fill up this form...

by Sean P. DeNigris

ClÃ©ment Bera-4 wrote > Is the mail display on the website... on purpose ? Ha ha, I sent out the same response the first time we set that page up... ----- Cheers, Sean -- View this message in context: http://forum.world.st/Please-contributors-fill-up-this-form-tp4713432p47134… Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com.

Oct. 9, 2013

Re: [Pharo-dev] SlotVirus

by Martin Dias

I just created the issue: https://pharo.fogbugz.com/f/cases/11832/Better-name-for-SlotVirus On Wed, Oct 2, 2013 at 12:41 PM, Marcus Denker <marcus.denker(a)inria.fr>wrote: > Nearly all are already on the bug tracker. I added a "cleanup" tag: > > https://pharo.fogbugz.com/f/search/?sSearchFor=tag%3A%22cleanup%22 > > > On Tue, Oct 1, 2013 at 5:53 PM, StÃ©phane Ducasse < > stephane.ducasse(a)inria.fr> wrote: > >> marcus >> >> I love what you did for the codeHolder removal (please do the same for >> these items). >> Yes I want to remove the duplicates. >> >> Stef >> >> On Oct 1, 2013, at 9:31 AM, Marcus Denker <marcus.denker(a)inria.fr> wrote: >> >> > >> > On Oct 1, 2013, at 9:23 AM, Marcus Denker <marcus.denker(a)inria.fr> >> wrote: >> > >> >> >> >> On Oct 1, 2013, at 9:07 AM, Marcus Denker <marcus.denker(a)inria.fr> >> wrote: >> >>> >> >>> And we should not forget to do these cleanups: we now have nearly >> everything double in the system: >> >>> -> Compiler >> >>> -> Inspector >> >>> -> Debugger >> >>> -> ClassBuilder >> >>> -> ClassOrganizer >> >>> -> Browserâ¦. >> >> >> >> -> Rpackage vs PackaInfo vs. SystemCategory >> >> -> Ring vs. PseudeClass&co vs. RBClass >> >> -> AST vs RB AST >> > >> > -> Url vs. ZnUrl >> > -> ZnMimeType vs MimeType >> > -> NewList vs whichever-of-the 30 classes with "List" in its name >> > -> FreeType vs. Bitmap fonts >> > -> TreeInspector vs. ObjectExplorer >> > >> > We should really clean up. >> > >> > Marcus >> > >> > >> >> > > > -- > -- > Marcus Denker -- denker(a)acm.org > http://www.marcusdenker.de >

Oct. 9, 2013