May 2016 - Pharo-dev - lists.pharo.org

[pharo-project/pharo-core]

by GitHub

Branch: refs/tags/60027 Home: https://github.com/pharo-project/pharo-core

May 21, 2016

[pharo-project/pharo-core] d0636f: 60027

by GitHub

Branch: refs/heads/6.0 Home: https://github.com/pharo-project/pharo-core Commit: d0636f96268209a1251cd9dd3227c94455395619 https://github.com/pharo-project/pharo-core/commit/d0636f96268209a1251cd9dd… Author: Jenkins Build Server <board(a)pharo-project.org> Date: 2016-05-21 (Sat, 21 May 2016) Changed paths: M AST-Core.package/RBBlockNode.class/instance/accessing/statementComments.st A BlueInk-Core.package/BIConfigurableFormatter.class/instance/private-formatting/formatBlockCommentFor_.st M BlueInk-Core.package/BIConfigurableFormatter.class/instance/private-formatting/formatBlock_.st A BlueInk-Core.package/BIConfigurableFormatter.class/instance/private/isNonEmptySingleLineBlock_.st M Refactoring-Core.package/RBAbstractClass.class/README.md M Refactoring-Core.package/RBClass.class/README.md M Refactoring-Core.package/RBMetaclass.class/README.md M Refactoring-Core.package/RBMethod.class/README.md M Refactoring-Core.package/RBNamespace.class/README.md R ScriptLoader60.package/ScriptLoader.class/instance/pharo - scripts/script60026.st A ScriptLoader60.package/ScriptLoader.class/instance/pharo - scripts/script60027.st R ScriptLoader60.package/ScriptLoader.class/instance/pharo - updates/update60026.st A ScriptLoader60.package/ScriptLoader.class/instance/pharo - updates/update60027.st M ScriptLoader60.package/ScriptLoader.class/instance/public/commentForCurrentUpdate.st Log Message: ----------- 60027 18253 formatter misplaces some block comments https://pharo.fogbugz.com/f/cases/18253 18267 Add Class comments to Refactoring-Core-Model classes https://pharo.fogbugz.com/f/cases/18267 18284 SmallFloat64 identityHash is answering bad value https://pharo.fogbugz.com/f/cases/18284 http://files.pharo.org/image/60/60027.zip

May 21, 2016

Re: [Pharo-dev] [Vm-dev] Minutes of the Meeting to Discuss Moving Cog to Github

by Esteban Lorenzano

> On 21 May 2016, at 03:22, Torsten Bergmann <astares(a)gmx.de> wrote: > > 3. Is there any effect on http://files.pharo.org/vm/ <http://files.pharo.org/vm/> ? there should not be any change thereâ¦ once reunified sources we will continue producing Pharo VMs and offering them in the usual mechanisms :) Esteban

May 21, 2016

Re: [Pharo-dev] [Vm-dev] Minutes of the Meeting to Discuss Moving Cog to Github

by Esteban Lorenzano

> On 20 May 2016, at 23:49, Eliot Miranda <eliot.miranda(a)gmail.com> wrote: > > Indeed, I hope that with the new repository it will be much easier for the people who got the VM working on iPad to integrate their changes. It should also be possible to have Linux ARM v6 (Pi) and Linux ARM v7 (Android) VMs. For iPad, sources were working a couple of years ago and we never comebackâ¦ make it work should be not hard but once integrated with Johnâs work on cocoa we'll need to have a new lookâ¦ Esteban

May 21, 2016

Re: [Pharo-dev] Turning off Spotter using internet!

by Esteban Lorenzano

> On 21 May 2016, at 08:43, stepharo <stepharo(a)free.fr> wrote: > > Thanks sven. > > I'm working on image without personal setting and often in places with no internet (or full of wifis but that I cannot connect to :). > > > Stef > >>> On 19 May 2016, at 13:41, stepharo <stepharo(a)free.fr> wrote: >>> >>> Hi guys >>> >>> This is hyper annoying that Spotter look for the net by default. >>> >>> We should turn it off. >> This was 'fixed' in the sense that it now only gets it once every 24 hours and then caches it in the image. >> >> The real problem is that NetNameResolver (dns name -> IP address resolution, using of course opaque primitives) sometimes hangs. This makes it very hard for the image to decide if it has network connection or not. And thus hard to react in an intelligent way. this means we need better support for networking, I think :) Esteban >> >> This situation is also hard to debug, because if I turn off my wireless network, NetNameResolver fails immediately (which is perfect). It is hard to reproduce the case where it hangs, at least for me. >> >> That is the issue that we should fix. >> >> Assuming someone has no internet, and force them to manage the connection state themselves manually, seems very backward to me - in this century. >> >> So I vote against turning it off. >> >>> We should also improve the comment of the catalog setting because it is unclear >>> >>> that this is this setting that should be turned off. >> Yes that would be good. >> >>> Stef >>> >>> >> >> > >

May 21, 2016

Re: [Pharo-dev] GT inspector: how to get side by side panes

by Nicolai Hess

2016-05-21 9:03 GMT+02:00 stepharo <stepharo(a)free.fr>: > I would like to see different views of a compiled method (source, > bytecode) side by side > > > I don't know if this is possible for all gt presentations, but for Bytecode/Sourcecode view you can use GTBytecodeBrowser new openOn:(Morph>>#drawOn:) > Is there a way to see a new pane (describing a different aspect) on the > right without been force to click on > > an element of the objects? > > I tried option alt control + click > > > Stef > > > >

May 21, 2016

Re: [Pharo-dev] GT inspector: how to get side by side panes

by Tudor Girba

Hi, This is not really possible in the current implementation. Cheers, Doru > On May 21, 2016, at 9:03 AM, stepharo <stepharo(a)free.fr> wrote: > > I would like to see different views of a compiled method (source, bytecode) side by side > > > Is there a way to see a new pane (describing a different aspect) on the right without been force to click on > > an element of the objects? > > I tried option alt control + click > > > Stef > > > -- www.tudorgirba.com www.feenk.com "To utilize feedback, you first have to acquire it."

May 21, 2016

GT inspector: how to get side by side panes

by stepharo

I would like to see different views of a compiled method (source, bytecode) side by side Is there a way to see a new pane (describing a different aspect) on the right without been force to click on an element of the objects? I tried option alt control + click Stef

May 21, 2016

Re: [Pharo-dev] Turning off Spotter using internet!

by stepharo

Thanks sven. I'm working on image without personal setting and often in places with no internet (or full of wifis but that I cannot connect to :). Stef >> On 19 May 2016, at 13:41, stepharo <stepharo(a)free.fr> wrote: >> >> Hi guys >> >> This is hyper annoying that Spotter look for the net by default. >> >> We should turn it off. > This was 'fixed' in the sense that it now only gets it once every 24 hours and then caches it in the image. > > The real problem is that NetNameResolver (dns name -> IP address resolution, using of course opaque primitives) sometimes hangs. This makes it very hard for the image to decide if it has network connection or not. And thus hard to react in an intelligent way. > > This situation is also hard to debug, because if I turn off my wireless network, NetNameResolver fails immediately (which is perfect). It is hard to reproduce the case where it hangs, at least for me. > > That is the issue that we should fix. > > Assuming someone has no internet, and force them to manage the connection state themselves manually, seems very backward to me - in this century. > > So I vote against turning it off. > >> We should also improve the comment of the catalog setting because it is unclear >> >> that this is this setting that should be turned off. > Yes that would be good. > >> Stef >> >> > >

May 21, 2016

Re: [Pharo-dev] Pharo IDE in Amber?

by Pierce Ng

On Fri, May 20, 2016 at 02:58:21PM +0200, Siemen Baader wrote: > If this can be overcome, I think the options would be to either: > a) implement a Morphic raster rendering backend based on HTML5 Canvas, or: > 2) create DOM & CSS based Morphic widgets, like the Lively Kernel does. iii) Take a look at the Shampoo protocol: https://github.com/dmatveev/shampoo. With a Shampoo client in Pharo and a Shampoo server in Amber, you can use Pharo as your Amber IDE. Of course initially this is just using the Pharo IDE as an editor for programming Amber and doesn't offer much integration, but once the connectivity is there, I'm sure other possibilities open up. Btw, I cooked up what I called MorphIDE in 2011 when Amber was Jtalk. But eventually I didn't use Jtalk for anything so MorphIDE didn't grow beyond the proof of concept. See http://www.samadhiweb.com/blog/2011.09.03.morphide.html. Also, the Jtalk/Amber examples on my blog may not work because the JS files were lost when I moved the blog across machines. Pierce

May 21, 2016