- Pharo-users - lists.pharo.org

SmaCC 2.0.5 released
by Thierry Goubier Aug. 20, 2015

Aug. 20, 2015

Hi All, I have released SmaCC 2.0.5, aligned on the main SmaCC[1] with a small fix on the Java parser, a few additional parsers (Cucumber), the addition of the Javascript parser, and continuous improvements on the performance and source code coverage of SmaCC generated parsers (with many thanks to John Brant and Don Roberts). The project is available on github[2] [1] http://www.refactoryworkers.com/SmaCC.html [2] https://github.com/ThierryGoubier/SmaCC This release is targeted at Pharo4. All parsers should work on Pharo 5, but the GUI doesn't work. Most parsers come with full AST generation and a dedicated visitor for each, and support for native language refactoring. To retrieve a specific parser, please use Metacello with the following script: Metacello new baseline: 'SmaCC'; repository: 'github://ThierryGoubier/SmaCC:v2.0.5'; load: 'SmaCC-Javascript' (SmaCC-Python, SmaCC-Java, SmaCC-Cucumber, etc...) On Pharo4 only, to retrieve the full set (gui and tests), please use Metacello with the following expressions: Metacello new baseline: 'SmaCC'; repository: 'github://ThierryGoubier/SmaCC:v2.0.5'; load Regards, Thierry

4 4

Is RBClassNotReferencedRule ~200x slower in Pharo 4 vs Pharo 3 ?
by Paul DeBruicker Aug. 19, 2015

Aug. 19, 2015

I think I'm setting this up wrong in Pharo 4 but am not sure. I use the following code for the test: rule:=RBClassNotReferencedRule new. environment:=RBBrowserEnvironment new forPackageNames: #('MyPackage'). AndreasSystemProfiler spyOn:[RBSmalllintChecker runRule: rule onEnvironment: environment] In Pharo 4 it spends most of the time in Behavior>>#includesDefinedSelector:ofClassName: and takes 300s. In Pharo 3 its only 1.5 s Is there something that can be done to get Pharo 4 up to the speed of Pharo 3? Thanks Paul

4 11

Making a Rubric text morph movable
by Stephan Eggermont Aug. 19, 2015

Aug. 19, 2015

We have a Morph containing a RubTextFieldMorph that we want to be able to drag around. At the same time, we want selection extension with the mouse to work. That feels ok when the dragging can only be initiated when the text field has no keyboard focus. The way we've made that work feels somewhat clumsy. What would be a better way? The containing morph has a default mouseDownPriority of 10 MindmapNode>>initialize super initialize. self addMorph: ( (MindmapTextField on: self) withGhostText: self defaultTitle; autoAccept: true; bounds: (self bounds); yourself). prio := 10. When it is clicked, the prio is set to 0 MindmapNode>>click: evt submorphs first takeKeyboardFocus. prio := 0 The text field is subclassed RubTextFieldMorph subclass: #MindmapTextField and overrides whenFocusLost: MindmapTextFieldwhenLostForcus: anAnnouncement owner resetPrio. super whenLostForcus: anAnnouncement MindmapNode>>resetPrio prio := 10 Noah & Stephan

1 0

Nautilus questions
by Peter UhnÃ¡k Aug. 19, 2015

Aug. 19, 2015

Hi, maybe I'm too picky, but I have couple of questions/remarks regarding current Nautilus UI. Now, I'm happy that new features were added/reorganized, but it does feel it is a little bit chaotic right now. I would expect the left/right history buttons being right next to the History Navigator selector. That way history-related controls are together and the purpose of the buttons is obvious "Hier.", "Com."... abbreviating button labels? Really? âAnd finally, why was part of the buttons moved up? And part stayed down? I know that you wanted to distinguish "Scoped", because it opens a new browser, and Variables, because they have intermediate menu (and later open another menu). But I think the fact that Variables are tied to a particular selected class would make much more sense to place the button beside hierachy/class button. And opening Scoped can be denoted with an appropriate icon ( https://www.google.com/search?site=&tbm=isch&source=hp&biw=&bih=&q=new+wind… ) Placing them all on a single line would also save up vertical space, which considering we've added QualityAssistant and Rubric adds another line is quite valuable. (And finally the width of the buttons should perhaps match more closely the actual needs of the label, because now it's bit cramped.) â And finally, code-wise I've never really interacted with Morphic and stayed in Spec, so when I was digging through code it seems that the way Morphs are composed are through these "addMorph:fullFrame: (LayoutFrame identity rightFraction: 0.10; yourself);" constructs sprinkled haphazardly throughout the code. Is there nothing like SpecLayout? Maybe it wasn't so hard once you get used to Morphic, because I spent probably more time doing elementary changes to layout than learning half of Spec... (Also since Spec can interoperate with Morphic, maybe the layout could also be adapted? Or slowly migrate Nautilus to Spec (if there are even plans like that)?) Now as I've mentioned I'm not an UI/UX expert, but you don't need to be a chef to tell when food isn't great (which is of course amplified by personal tastes). So unless I am the only person bothered by this, I would like to raise a discussion (or send a slice, but the way I've done the layout weirds me out, so I still want to look into that regardless). Cheers, Peter

5 4

Chinese Character Data
by Sean P. DeNigris Aug. 18, 2015

Aug. 18, 2015

Can I e.g. paste Chinese characters from a web page and save as data? I don't need to be able to use Chinese characters for code. I see pasting creates a wide string but the characters do not display. I assume this is because my font does not support them, but in the Settings Browser I only see 3 choices on Mac, none of which seem to display them. ----- Cheers, Sean -- View this message in context: http://forum.world.st/Chinese-Character-Data-tp4843635.html Sent from the Pharo Smalltalk Users mailing list archive at Nabble.com.

4 7

Re: [Pharo-users] ***UNCHECKED*** Re: [ANN] Magma 1.5
by Pablo R. Digonzelli Aug. 18, 2015

Aug. 18, 2015

i was using voyage for these tests. Pablo Ing. Pablo Digonzelli Software Solutions IP-Solutiones SRL Metrotec SRL 25 de Mayo 521 San Miguel de TucumÃ¡n Email: pdigonzelli(a)softsargentina.com pdigonzelli(a)gmail.com Cel: 5493815982714 ----- Mensaje original ----- De: "Esteban Lorenzano" <estebanlm(a)gmail.com> Para: "Any question about pharo is welcome" <pharo-users(a)lists.pharo.org> CC: "stephan" <stephan(a)stack.nl>, "Torsten Bergmann" <astares(a)gmx.de>, hilaire(a)drgeo.eu Enviados: Martes, 18 de Agosto 2015 13:55:10 Asunto: Re: [Pharo-users] ***UNCHECKED*** Re: [ANN] Magma 1.5 > On 18 Aug 2015, at 18:35, Pablo R. Digonzelli <pdigonzelli(a)gmail.com> wrote: > > Yes I checkesd. It is by far better approach to pharo and I like it very much. > The problem i found was when i stress mongo, it lost some transactions (i.e 10000 rewrites lost about 80 or more don't remember exactly) . > In bussiness apps i cannot take that risk. targeting plain mongo or using voyage? I find hard to believe a production database as is mongo can have those problems so I would think more on voyage (and in that case, we can find a solution because this is obviously a bug :P) Esteban > > > > > > Ing. Pablo Digonzelli > Software Solutions > IP-Solutiones SRL > Metrotec SRL > 25 de Mayo 521 > San Miguel de TucumÃ¡n > Email: pdigonzelli(a)softsargentina.com > pdigonzelli(a)gmail.com > Cel: 5493815982714 > > ----- Mensaje original ----- > De: "Esteban Lorenzano" <estebanlm(a)gmail.com> > Para: "Any question about pharo is welcome" <pharo-users(a)lists.pharo.org> > CC: stephan(a)stack.nl, "Torsten Bergmann" <astares(a)gmx.de>, hilaire(a)drgeo.eu > Enviados: Martes, 18 de Agosto 2015 12:45:05 > Asunto: ***UNCHECKED*** Re: [Pharo-users] [ANN] Magma 1.5 > >> On 18 Aug 2015, at 17:20, Pablo R. Digonzelli <pdigonzelli(a)gmail.com> wrote: >> >> I can colaborate. I do not know if i have the time and technical skills to do it alone, but for shure i can colaborate. >> I think magma will be a very interesting option for object persistence in pharo. >> I am currently working with garage and glorp and i suffers the innecesary complexity and other bad things thet persistence layer introduce in developing pharo apps. > > have you checked voyage? (to work with a MongoDB)â¦ it will avoid all the pain of using a relational database. > > Esteban > > ps: But yes, I do believe having magma would be cool too. > >> >> Pharo >> >> >> Ing. Pablo Digonzelli >> Software Solutions >> IP-Solutiones SRL >> Metrotec SRL >> 25 de Mayo 521 >> San Miguel de TucumÃ¡n >> Email: pdigonzelli(a)softsargentina.com >> pdigonzelli(a)gmail.com >> Cel: 5493815982714 >> >> ----- Mensaje original ----- >> De: "Torsten Bergmann" <astares(a)gmx.de> >> Para: "Any question about pharo is welcome" <pharo-users(a)lists.pharo.org> >> CC: stephan(a)stack.nl, stepharo(a)free.fr, hilaire(a)drgeo.eu, "Pablo Digonzelli" <pdigonzelli(a)gmail.com> >> Enviados: Martes, 18 de Agosto 2015 10:26:44 >> Asunto: Fwd: [ANN] Magma 1.5 >> >> Stephan Eggermont wrote: >>> It is possible, and it needs someone who can deal with low-level >>> details. Magma has its own mailing list. Chris has repeatedly said he'd >>> be willing to help out, but is not willing to be the Pharo port maintainer. >> >> I asked Chris directly and you can find his answer here on the magma list: >> >> http://lists.squeakfoundation.org/pipermail/magma/2015-August/001955.html >> >> As ever it requires someone to move this forward to make it happen. >> But as there was an initiative with Magma 1.3 for Pharo 1.3 already it does not >> seem impossible. >> >> Also - as Stef already pointed out - community services like CI could be used >> to maintain it once an initial up to date port was done. >> >> Bye >> T. >> >

1 0

Re: [Pharo-users] ***UNCHECKED*** Re: [ANN] Magma 1.5
by Pablo R. Digonzelli Aug. 18, 2015

Aug. 18, 2015

Yes I checkesd. It is by far better approach to pharo and I like it very much. The problem i found was when i stress mongo, it lost some transactions (i.e 10000 rewrites lost about 80 or more don't remember exactly) . In bussiness apps i cannot take that risk. Ing. Pablo Digonzelli Software Solutions IP-Solutiones SRL Metrotec SRL 25 de Mayo 521 San Miguel de TucumÃ¡n Email: pdigonzelli(a)softsargentina.com pdigonzelli(a)gmail.com Cel: 5493815982714 ----- Mensaje original ----- De: "Esteban Lorenzano" <estebanlm(a)gmail.com> Para: "Any question about pharo is welcome" <pharo-users(a)lists.pharo.org> CC: stephan(a)stack.nl, "Torsten Bergmann" <astares(a)gmx.de>, hilaire(a)drgeo.eu Enviados: Martes, 18 de Agosto 2015 12:45:05 Asunto: ***UNCHECKED*** Re: [Pharo-users] [ANN] Magma 1.5 > On 18 Aug 2015, at 17:20, Pablo R. Digonzelli <pdigonzelli(a)gmail.com> wrote: > > I can colaborate. I do not know if i have the time and technical skills to do it alone, but for shure i can colaborate. > I think magma will be a very interesting option for object persistence in pharo. > I am currently working with garage and glorp and i suffers the innecesary complexity and other bad things thet persistence layer introduce in developing pharo apps. have you checked voyage? (to work with a MongoDB)â¦ it will avoid all the pain of using a relational database. Esteban ps: But yes, I do believe having magma would be cool too. > > Pharo > > > Ing. Pablo Digonzelli > Software Solutions > IP-Solutiones SRL > Metrotec SRL > 25 de Mayo 521 > San Miguel de TucumÃ¡n > Email: pdigonzelli(a)softsargentina.com > pdigonzelli(a)gmail.com > Cel: 5493815982714 > > ----- Mensaje original ----- > De: "Torsten Bergmann" <astares(a)gmx.de> > Para: "Any question about pharo is welcome" <pharo-users(a)lists.pharo.org> > CC: stephan(a)stack.nl, stepharo(a)free.fr, hilaire(a)drgeo.eu, "Pablo Digonzelli" <pdigonzelli(a)gmail.com> > Enviados: Martes, 18 de Agosto 2015 10:26:44 > Asunto: Fwd: [ANN] Magma 1.5 > > Stephan Eggermont wrote: >> It is possible, and it needs someone who can deal with low-level >> details. Magma has its own mailing list. Chris has repeatedly said he'd >> be willing to help out, but is not willing to be the Pharo port maintainer. > > I asked Chris directly and you can find his answer here on the magma list: > > http://lists.squeakfoundation.org/pipermail/magma/2015-August/001955.html > > As ever it requires someone to move this forward to make it happen. > But as there was an initiative with Magma 1.3 for Pharo 1.3 already it does not > seem impossible. > > Also - as Stef already pointed out - community services like CI could be used > to maintain it once an initial up to date port was done. > > Bye > T. >

3 2

Re: [Pharo-users] [ANN] Magma 1.5
by Pablo R. Digonzelli Aug. 18, 2015

Aug. 18, 2015

I can colaborate. I do not know if i have the time and technical skills to do it alone, but for shure i can colaborate. I think magma will be a very interesting option for object persistence in pharo. I am currently working with garage and glorp and i suffers the innecesary complexity and other bad things thet persistence layer introduce in developing pharo apps. Pharo Ing. Pablo Digonzelli Software Solutions IP-Solutiones SRL Metrotec SRL 25 de Mayo 521 San Miguel de TucumÃ¡n Email: pdigonzelli(a)softsargentina.com pdigonzelli(a)gmail.com Cel: 5493815982714 ----- Mensaje original ----- De: "Torsten Bergmann" <astares(a)gmx.de> Para: "Any question about pharo is welcome" <pharo-users(a)lists.pharo.org> CC: stephan(a)stack.nl, stepharo(a)free.fr, hilaire(a)drgeo.eu, "Pablo Digonzelli" <pdigonzelli(a)gmail.com> Enviados: Martes, 18 de Agosto 2015 10:26:44 Asunto: Fwd: [ANN] Magma 1.5 Stephan Eggermont wrote: >It is possible, and it needs someone who can deal with low-level >details. Magma has its own mailing list. Chris has repeatedly said he'd >be willing to help out, but is not willing to be the Pharo port maintainer. I asked Chris directly and you can find his answer here on the magma list: http://lists.squeakfoundation.org/pipermail/magma/2015-August/001955.html As ever it requires someone to move this forward to make it happen. But as there was an initiative with Magma 1.3 for Pharo 1.3 already it does not seem impossible. Also - as Stef already pointed out - community services like CI could be used to maintain it once an initial up to date port was done. Bye T.

2 1

ClassCategoryReader and ClassCommentReader
by Guillermo Polito Aug. 18, 2015

Aug. 18, 2015

These two classes seem to be unused. They implement the #scanFor: method but the only senders of #scanFor: send it explicitly to RunArray, like: RunArray scanFor: ... Does anybody use it? Because it looks like an old class that could be removed. Guille

2 2

Fwd: [ANN] Magma 1.5
by Torsten Bergmann Aug. 18, 2015

Aug. 18, 2015

Stephan Eggermont wrote: >It is possible, and it needs someone who can deal with low-level >details. Magma has its own mailing list. Chris has repeatedly said he'd >be willing to help out, but is not willing to be the Pharo port maintainer. I asked Chris directly and you can find his answer here on the magma list: http://lists.squeakfoundation.org/pipermail/magma/2015-August/001955.html As ever it requires someone to move this forward to make it happen. But as there was an initiative with Magma 1.3 for Pharo 1.3 already it does not seem impossible. Also - as Stef already pointed out - community services like CI could be used to maintain it once an initial up to date port was done. Bye T.

1 0