April 2014 - Pharo-dev - lists.pharo.org

Bug in Spec layout?
by Martin Dias April 9, 2014

April 9, 2014

Hello folks, When I evaluate in latest Pharo 3: DynamicComposableModel new instantiateModels: #( a TreeModel b TreeModel c TreeModel); openWithSpecLayout: (SpecLayout composed newColumn: [ :column | column newRow: [ :row | row newColumn: #a. row newColumn: #b. row newColumn: #c width: 80 ] ]; yourself) ===> [image: Inline image 2] Maybe I'm wrong: I expected that #a and #b have the same width. But it looks like the layout algorithm first splits the space half/half between #a and #b, and later it only shrinks #b to give the fixed space to #c. Should I report an issue? or specify layout in other way? Thanks, MartÃn

2 3

[pharo-project/pharo-core] aecb6d: 30817
by GitHub April 9, 2014

April 9, 2014

Branch: refs/heads/3.0 Home: https://github.com/pharo-project/pharo-core Commit: aecb6d98f78d3658e4dda35fdea84971892c63a2 https://github.com/pharo-project/pharo-core/commit/aecb6d98f78d3658e4dda35f… Author: Jenkins Build Server <board(a)pharo-project.org> Date: 2014-04-09 (Wed, 09 Apr 2014) Changed paths: A Komitter.package/KomitClass.class/class/instance creation/trackedClass_forExtension_.st M Komitter.package/extension/MCAddition/instance/koClass.st A Komitter.package/extension/MCClassDefinition/instance/koClass.st A Komitter.package/extension/MCMethodDefinition/instance/koClass.st M Komitter.package/extension/MCModification/instance/koClass.st A Komitter.package/extension/MCOrganizationDefinition/instance/koClass.st M Komitter.package/extension/MCRemoval/instance/koClass.st M Nautilus.package/AbstractNautilusUI.class/instance/menus behavior/restrictedBrowseClasses_.st M NautilusRefactoring.package/ChangesBrowser.class/class/specs/defaultSpec.st A ScriptLoader30.package/ScriptLoader.class/instance/pharo - scripts/script470.st A ScriptLoader30.package/ScriptLoader.class/instance/pharo - updates/update30817.st M ScriptLoader30.package/ScriptLoader.class/instance/public/commentForCurrentUpdate.st Log Message: ----------- 30817 13014 Hide Refactoring Scope chooser for 3.0 release https://pharo.fogbugz.com/f/cases/13014 12738 Nautilus: Scoped on Class-side = Empty Class Pane https://pharo.fogbugz.com/f/cases/12738 13103 Komitter mixes up when some methods of extension are modified https://pharo.fogbugz.com/f/cases/13103 http://files.pharo.org/image/30/30817.zip

1 0

[pharo-project/pharo-core]
by GitHub April 9, 2014

April 9, 2014

Branch: refs/tags/30817 Home: https://github.com/pharo-project/pharo-core

1 0

Pharo Consortium: New Bronze Member Ta MÃ¨re
by Marcus Denker April 9, 2014

April 9, 2014

The Pharo Consortium is very happy to announce that Ta MÃ¨re has joined the Consortium as a Bronze Industrial Member. More about - Ta MÃ¨re: http://tamere.eu - Pharo Consortium: http://consortium.pharo.org The goal of the Pharo Consortium is to allow companies to support the ongoing development and future of Pharo. Individuals can support Pharo via the Pharo Association: http://association.pharo.org

6 6

What about a Pharo day in May-June?
by Pharo4Stef April 9, 2014

April 9, 2014

Hi guys I think that it would be great to organize a gathering/talks/show me your stuff/networking day before spring. What do you think that this idea? We could host it at Lille. Stef

3 2

Mailinglist Archive not working anymore - 403 Forbidden
by Torsten Bergmann April 9, 2014

April 9, 2014

The mailinglist archives at lists.pharo.org do not work anymore for all three lists (dev, user, business). I think it worked yesterday. Try http://lists.pharo.org/mailman/listinfo/pharo-users_lists.pharo.org and click on the archive. You get a 403 Forbidden Same on http://lists.pharo.org/mailman/listinfo/pharo-dev_lists.pharo.org

1 1

[OT] Finding your workbench muse by David of Basecamp
by Sven Van Caekenberghe April 8, 2014

April 8, 2014

[OT] Finding your workbench muse by David of Basecamp I love Pharo as my tool or workbench, and indeed it is not a 100% rational and certainly a difficult to explain decision, but that's OK, as more eloquently written down here: http://signalvnoise.com/posts/3727-finding-your-workbench-muse And it is also OK for others to have other preferences, there is no need to convert those that have found their own thing. Sven -- Sven Van Caekenberghe Proudly supporting Pharo http://pharo.org http://association.pharo.org http://consortium.pharo.org

1 0

Case 13007 Cannot-integrate-a-slice-with-monticello-red-square-in-the-UI
by Ben Coman April 8, 2014

April 8, 2014

Priority 2 fix ready for review. Please come pick it apart. https://pharo.fogbugz.com/f/cases/13007/Cannot-integrate-a-slice-with-monti… cheers -ben

2 3

Suggested Syntax Highlighting Features for Beginners
by J.F. Rick April 8, 2014

April 8, 2014

Smalltalk is a pretty great language, but its syntax is unusual. I've taught introductory programming in Pharo three times and, in my experience, there are two places where novices get repeatedly confused with syntax. First, there is the differences between assignment := and comparison = . Obviously, Smalltalk is not the only language where that is a problem. Java is even more confusing, with assignment = and comparison == . In Squeak, there was the attempt to address this problem head on with using _ instead of := and having it show up as an arrow pointing left; while that nicely addressed that problem, it proved awkward for many reasons (e.g., it only appeared as an arrow in Squeak, it wasn't compatible with other Smalltalks). In addition, comparison is a simple message send, where assignment is the most critical language element that is not an object or a message. In other words, there is a huge difference in meaning between the two and a really minor syntactical one. That's confusing to novices. Even after covering it repeatedly in class I still find novices making that mistake. As their debugging skills are minimal, this can be a real obstacle. Second, there is the problem of keyword messages with more than one argument. In Java, "this.someMessage( arg1, arg2)" is pretty comprehensible to a novice. In Smalltalk, "self someMessage: arg1 and: arg2" seems like two message sends (i.e., "this.someMessage(arg1).and(arg2)"). While novices won't see multi-argument messages very often, they do come up (e.g., ifTrue:ifFalse:). I have my class do regular quizzes on identifying messages and this still tricks them up even if they've already seen in 10 times. Now, I'm not advocating changing the Smalltalk syntax; that's just unlikely to happen. But, perhaps syntax highlighting could be employed to good use. For instance, it is extremely helpful that blocks and parentheses show by color which ( [ go with which ] ). Perhaps something similar could be done for messages with more than 1 argument. For instance, the message could be underlined in different colors. That way I could easily see which ifFalse: went with which ifTrue:. It would also give a visual indicator: (for novices) Be careful, this message is one of those weird ones that's longer than it first appears. For assignment vs comparison, you similarly need a powerful visual cue that assignment is an usual thing, where comparison is a pretty ordinary thing. Perhaps an arrow decoration could appear behind the :=. These may seem like trivial issues. For experts, they are. I don't need any of it. I don't find syntax highlighting and command completion that useful in my programming. But, I'm proficient with a debugger. Should I occasionally make such a trivial mistake (e.g., forgetting to type : in :=), I immediately find it while debugging and easily fix it. For novices, using a debugger is not an option. Remember, they don't understand what a stack is. Syntax highlighting is the best debugging feature that they understand. And these are two real common mistakes that they make that could be addressed by improvements to syntax highlighting. Just a suggestion, Jeff -- Jochen "Jeff" Rick, Ph.D. http://www.je77.com/ Skype ID: jochenrick

5 8

Bug in SSL
by Norbert Hartl April 8, 2014

April 8, 2014

As you might know a sever bug in the openssl library har been reported to the public. This bug is pretty nasty and allows in an easy fashion to access protected data. Please consider upgrading your products and operating systems to a fixed version of openssl. https://www.openssl.org/news/secadv_20140407.txt regards, Norbert

2 1