April 2015 - Pharo-dev - lists.pharo.org

Re: [Pharo-dev] in Pharo4.0, Latest update: #40581: push down -> DNU

by stepharo

Yes this is really strange because I tried to reproduce it in 4.0 and 4.0 plus pillar and I could not. However I was really not doing anything special. So we can probably close this issue but it was annoying. Stef Le 30/3/15 15:48, Ben Coman a Ã©crit : > > I also cannot reproduce this. In 40586... > 1. To AnnouncementsHelp I added method #y > 2. Right-clicking on #y > Refactoring > Push down > 3. Clicked <Ok> in the Changes Browser > 4. Observed #y had moved from AnnouncementsHelp to both subclasses. > > Repeated for existing AnnouncementsHelp-class>>introduction. > > cheers -ben > > > On Mon, Mar 30, 2015 at 3:03 PM, Nicolai Hess <nicolaihess(a)web.de > <mailto:nicolaihess@web.de>> wrote: > > I can not reproduce this error, do you use > refactoring from the Nautilus menu? > > 2015-03-30 8:47 GMT+02:00 stepharo <stepharo(a)free.fr > <mailto:stepharo@free.fr>>: > > This bug is really severe > Once we have it it breaks loading MC files too. > > Le 29/3/15 21:09, stepharo a Ã©crit : > > Hi > > > When I select an instance method and use push down in a > subclass without obvious conflicts. > I get two DNUs. > > realClass > > ^baseClass realClass theMetaClass > > dnu: theMetaClass > > while baseClass is a class > realClass -> nil > > > the environement does not contain the class. > Now I did a class rename before and I wonder if it is not > the one that is broken :( > > Stef > > > > > > >

April 1, 2015

Re: [Pharo-dev] Bloc Contributions

by stepharo

cool I cannot because only the owner can update such features such alain will certainly do it. I started to write tests to understand some aspects of block. Not much but I found bugs that way. Stef Le 30/3/15 14:46, Sean P. DeNigris a Ã©crit : > Alain, maybe we should move Bloc to a repo owned by the Pharo team? In the > mean time, will you add me as a developer? Thanks. > > Small refactoring uploaded to > http://smalltalkhub.com/#!/~SeanDeNigris/SeansOutbox > Bloc-Core-SeanDeNigris.410 > Cleanup BlAlarmQueue > - #queue protocol -> private > - Extract all "BlUniverse default" -> factory method > - Rename #(un)subscribesToTicks -> #(un)subscribeToTicks (proper English) > > > > ----- > Cheers, > Sean > -- > View this message in context: http://forum.world.st/Bloc-Contributions-tp4816082.html > Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com. > >

April 1, 2015

Re: [Pharo-dev] Bloc Roadmap

by Sean P. DeNigris

> yes, we know, we are preparing a documentation. Great! Is the WIP publicly available (e.g. in one of our Pillar/GitHub repos)? Btw. the class comments are reeeally great! I'm only talking about documenting design rationale. > Weâve implemented a lot of examples. please have a look. Thanks - I have and they are also great. But AFAICT it's impossible to tell from the examples /why/ e.g. we went from Morphs owning their step actions to block-based steps registered with a scheduler that is universal to all worlds. I have a doubt about that because one use case I've had in Morphic is to speed up time in a world to see what happens e.g. in 10 minutes from now, but #schedule hard-codes the reference to the BlUniverse singleton. ----- Cheers, Sean -- View this message in context: http://forum.world.st/Bloc-Roadmap-tp4816171p4816769.html Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com.

April 1, 2015

PharoV30.sources in Pharo 4.0 download

by Dmitri Zagidulin

I noticed that the Pharo 4.0 full distribution zip file off of Pharo.org/download contains PharoV30.sources (and not, say, PharoV40.sources). This is true for all 3 platforms: - http://files.pharo.org/platform/Pharo4.0-mac.zip - http://files.pharo.org/platform/Pharo4.0-win.zip - http://files.pharo.org/platform/Pharo4.0-linux.zip Is there a typo in a CI job somewhere?

April 1, 2015

Re: [Pharo-dev] Bloc Roadmap

by Alain Plantec

Hello Sean, > On 01 Apr 2015, at 20:39, Sean P. DeNigris <sean(a)clipperadams.com> wrote: > >> did you read the class comments? >> or Morph and Views > Yes. I see that the Morph is delegating it's drawing to a View, but what I'm interested in is the rationale. I believe that documenting the design at a higher level, so everyone can understand the rationale and features, is the only way to keep Bloc clean and lean and prevent another "Morphic is so cool but impossible to understandâ. yes, we know, we are preparing a documentation. Weâve implemented a lot of examples. please have a look. Cheers Alain

April 1, 2015

Re: [Pharo-dev] Bloc Roadmap

by Sean P. DeNigris

> did you read the class comments? > or Morph and Views Yes. I see that the Morph is delegating it's drawing to a View, but what I'm interested in is the rationale. I believe that documenting the design at a higher level, so everyone can understand the rationale and features, is the only way to keep Bloc clean and lean and prevent another "Morphic is so cool but impossible to understand". Let's take the current example of the introduction of View classes Scenario #1: Design choice because of higher-level vision for Bloc? Right now views are hot-swappable: 1. Make a clock with "clock :=BlPolygonViewExamples exampleSimpleClock." 2. Now turn it into a vintage Bulova with "clock withRectangleView" That is awesome! But is it on-purpose? a documented feature? Those are important so a well-meaning refactorer doesn't come in and smash the design by bloating or removing important capabilities Scenario #2: Views extracted simply because Morphic Morphs were so bloated Morph methodDict size "962" BlMorph methodDict size "473" If #1 is the case, it may be important to preserve the feature, but if it's #2 we may not be as attached to that particular choice. > Decorators are not modularly extensible Bloc use a chain of command. I'm not sure what you mean. Right now, borders are an intrinsic part of BlEnclosedView. If Decorators were used (and I'm not saying they're appropriate, just asking), one could e.g. add multiple borders, or a border and scroll bars, etc. How are Decorators not modularly extensible when used that way? ----- Cheers, Sean -- View this message in context: http://forum.world.st/Bloc-Roadmap-tp4816171p4816747.html Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com.

April 1, 2015

Re: [Pharo-dev] Brand new modern Spotter design!

by kilon alios

If I understand what you are trying to say, then I don't agree, I prefer tools that work as part of the whole and that includes theming too. But the good news is that out of coincidence it fits my blue theme and I also love GTSpotter :) â On Wed, Apr 1, 2015 at 2:56 PM, Tudor Girba <tudor(a)tudorgirba.com> wrote: > I am saying that the current look is universally applicable and should be > used as is without alteration for all eternity :). > > Doru > > -- > www.tudorgirba.com > > "Every thing has its own flow" > > On 01 Apr 2015, at 13:22, kilon alios <kilon.alios(a)gmail.com> wrote: > > can you explain a bit more what you mean ? are you referring to the pharo > central theme or GTSpotter theme ? > > It will be a huge pain for users to set a separate theme for each tool > inside pharo. > > On Wed, Apr 1, 2015 at 2:19 PM, Tudor Girba <tudor(a)tudorgirba.com> wrote: > >> But we think that complementary universal colors work best in this >> context. >> >> Doru >> >> -- >> www.tudorgirba.com >> >> "Every thing has its own flow" >> >> > On 01 Apr 2015, at 13:07, kilon alios <kilon.alios(a)gmail.com> wrote: >> > >> > >> > My opinion : >> > >> > Even when a tool has its own theme support should still support and >> respect the central theme. >> >> >

April 1, 2015

Re: [Pharo-dev] New Pharo-Launcher package for Mac and Windows

by Ben Coman

On Wed, Apr 1, 2015 at 12:34 AM, Damien Cassou <damien.cassou(a)gmail.com> wrote: > > Cyril Ferlicot <cyril.ferlicot(a)gmail.com> writes: > > > On 31 March 2015 at 17:22, Damien Cassou <damien.cassou(a)gmail.com> > wrote: > > > >> Ben Coman <btc(a)openinworld.com> writes: > >> > >> > Regarding Windows... Last that I looked, the PharoLauncher image is > >> > installed into the "Program Files" System Folder. This causes a > problem > >> > since when a PharoLauncher Image is saved by a Standard User, they > cannot > >> > write to the System Folder so Windows *fakes* it and actually writes > to a > >> > User Fodler, but makes it look like the file is in the System Folder. > >> That > >> > is, the Standard User is looking at a Fake System Folder!! > >> > >> that's very strange. Can someone please evaluate this expression on your > >> Windows box and give us the result? > >> > >> PlatformResolver forCurrentPlatform cache > > > File @ C:\Users\Cyril\AddData\Roaming > > > looks good to me. So, I don't understand Ben's problem. > > > > >> > This confused me for quite a while - even as an IT Administrator of 15 > >> > years. Its likely to cause chaos and pain for others down the track. > So > >> I > >> > recommend that the installer be changed so that the PharoLauncher > Image > >> is > >> > installed into a User Folder - at least by default, and maybe as the > only > >> > option. > >> > >> > >> I disagree. The launcher should be installed with other applications. > >> But the data is per-user. On Ubuntu, the launcher is happily installed > >> under /usr/share/pharo-launcher and images are download to > >> ~/.cache/Pharo/images. > The location of the downloaded images is fine. It is the location of PharoLauncher's own image that is the problem on windows if the user happens to save the Image. Try the following on Windows 7+... 1. Install pharo_installer-0.x.x.exe as administrator. 2. Run Pharolauncher as a standard user. 3. Change to development mode, open Playground and type a note to yourself like "This should not show after a reinstall". Quit Pharolauncher and restart to check it still appears, and quit again. 4. Re-install pharo_installer-0.x.x.exe as administrator (like you might when a newer version is available) 5. Run Pharolauncher as a standard user. You would expect a fresh image - one that doesn't show the note to yourself, but I had found (a while ago) that the note still shows, indicating the upgrade effectively failed. That is a "typical" workflow in a corporate environment, which I expect Pharo would like to target. So when you: * run PharoLauncher as administrator and get the new Pharolauncher, but * run PharoLauncher as a standard user and get the old Pharolauncher you may rightfully ask wtf ?!? What do you observe? cheers -ben >> > >> > I had meant to fix this in PharoLauncher myself, and also would offer > to > >> > test it - but unfortunately I currently don't have access to a Windows > >> > machine. > >> > >> I hope someone else will try > >> > >> -- > >> Damien Cassou > >> http://damiencassou.seasidehosting.st > >> > >> "Success is the ability to go from one failure to another without > >> losing enthusiasm." --Winston Churchill > >> > >> > > -- > Damien Cassou > http://damiencassou.seasidehosting.st > > "Success is the ability to go from one failure to another without > losing enthusiasm." --Winston Churchill > >

April 1, 2015

Re: [Pharo-dev] Bloc Roadmap

by stepharo

did you read the class comments? or Morph and Views Decorators are not modularly extensible Bloc use a chain of command. Much better Le 30/3/15 19:41, Sean P. DeNigris a Ã©crit : > I created a page at > https://github.com/pharo-project/pharo-workingRoadmaps/blob/master/BlocRoad… > > I started with some questions and my basic understanding after a few > experiments > > > > ----- > Cheers, > Sean > -- > View this message in context: http://forum.world.st/Bloc-Roadmap-tp4816171.html > Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com. > >

April 1, 2015

Re: [Pharo-dev] 14606 Really Important bug - Can't see some variables values while debugging closure

by stepharo

Thanks ben Several rmoder are busy with paper deadline for this evening. ... Stef Le 1/4/15 18:56, Ben Coman a Ã©crit : > Issue 14606 [1] is tagged "Really Important" where evaluating.... > > | t1 t2 t3 t4 t5 t6 | > t1 := 1. > t2 := 2. > t3 := 3. > t4 := 4. > t5 := 5 . > t6 := 6. > [ t5 := 50. > t6 := 60. > t3 + t4. > 1 halt. > ] value > > => can't see the value of t4, t5, t6. > > I've just submitted a possible solution for this, but its a bit late > before release for such a change - so I'm asking for "all hands in" to > help stress test it. You don't need to review code, just please merge > latest slice 14606 into your normal working image and report your > experience. > > cheers -ben > > [1] https://pharo.fogbugz.com/default.asp?14606#124466 >

April 1, 2015