- Pharo-dev - lists.pharo.org

[Pharo-project] FS-Bug render Pharo 2.0 useless on Win (Issue 6531)

by Torsten Bergmann

Anyone with deeper knowledge on FS code able to have a look at issue 6531 - "Not possible to save one click image on Windows" http://code.google.com/p/pharo/issues/detail?id=6531 I found the original cause - it crashes since the code will open the changes file for a second time. The reason is that FileStream(class)>>onHandle: will not really work on an existing handle - and instead opens the file again. It's easy to debug (see the bug entry what to evaluate in a workspace). I'm not that familiar with the design and implementation of the new file system. To me it looks like it is not a bug but merely just unfinished in the transition from new to old filesystem. Maybe someone with better knowledge of the intentions of the new file system can help fixing it - and assure it works on all platforms afterwards. Thanks in advance! Bye T.

Sept. 19, 2012

Re: [Pharo-project] [Screencast] Roassal, an agile visualization engine

by Frank Shearar

On 19 September 2012 20:13, laurent laffont <laurent.laffont(a)gmail.com> wrote: > Roassal graphically renders objects using short and expressive Smalltalk > expressions. A large set of interaction are offered for a better user > experience. Painting, brushing, interconnecting, zooming, drag and dropping > will just make you more intimate with any arbitrary object model. > > Thank you Vanessa PeÃ±a ! > > http://www.pharocasts.com/2012/09/roassal.html That's a really nice screencast. Thanks, Laurent and Vanessa! I like the implication towards the end of the screencast that the same code one used in a Smalltalk image would run quite happily in a browser as well. I guess that depends on how much UI-specific code there is. How much commonality would you say is there between what runs in Pharo and what runs in Amber, Laurent? frank > Laurent

Sept. 19, 2012

[Pharo-project] [Screencast] Roassal, an agile visualization engine

by laurent laffont

Roassal graphically renders objects using short and expressive Smalltalk expressions. A large set of interaction are offered for a better user experience. Painting, brushing, interconnecting, zooming, drag and dropping will just make you more intimate with any arbitrary object model. Thank you Vanessa PeÃ±a ! http://www.pharocasts.com/2012/09/roassal.html Laurent

Sept. 19, 2012

Re: [Pharo-project] [update 2.0] #20297

by Camillo Bruni

perfect! thanks a lot esteban :) On 2012-09-19, at 18:45, Esteban Lorenzano <estebanlm(a)gmail.com> wrote: > 20297 > ----- > > Issue 6639: FileSystem enhancements > http://code.google.com/p/pharo/issues/detail?id=6639 > > Issue 6695: RPackageSet speedup > http://code.google.com/p/pharo/issues/detail?id=6695 > > Issue 6700: missing method in RPackageSet > http://code.google.com/p/pharo/issues/detail?id=6700 >

Sept. 19, 2012

Re: [Pharo-project] nautilus speedup?

by Camillo Bruni

On 2012-09-19, at 19:48, Stephan Eggermont <stephan(a)stack.nl> wrote: > Camillo wrote: >> I did a quick nautilus improvement: >> >> https://code.google.com/p/pharo/issues/detail?id=6667 >> >> CAVEAT: One thing I did was not showing all methods when opening the browser on a class. >> >> For me this makes perfectly sense since I use the CMD-F + CMD-M shortcut >> for searching for a method. Or go over all the protocol using the >> arrow-keys on the selected protocol list. >> >> what do you think? > > It sounds like a perfect idea to get some traction for getting rid of method categories. > > Why would anyone want not to look at all methods in a class by default? > In well refactored code, the number of methods is so limited that no scrolling > should be needed. Not all code is like that, of course. > > Reading code is the most important activity of developing, and it is essential to > limit the interaction needed to browse through code. > > Much better would be to show in the code pane which instance variables have accessors > (r/w) and show all without accessors by default. > >> >> However I can see that a "classical" user might be disturbed. This cut's down >> the opening lag of a new browser from ~400ms to ~200ms on my machine, to me >> thats quite worth it :). > > If the problem is performance, focus on performance. Why can not showing 30 items > in a list make a difference of 200 ms? well yeah, that would be nice, but takes too much time for now: - PluggableListMorph does many stupid things to kill performance - TrueType has 3 levels of caches which make every operation on it super slow I know where the problems are on the lower levels, however they are not to be solved that quickly... so not showing 30 items for me is valid right now. adding an option so I can disable them on my machine is also valid. but then again, most people out there use the mouse... so that is one piece of time loss :D, and this time in the range of seconds, not milliseconds.

Sept. 19, 2012

Re: [Pharo-project] nautilus speedup?

by Stephan Eggermont

Camillo wrote: > I did a quick nautilus improvement: > > https://code.google.com/p/pharo/issues/detail?id=6667 > > CAVEAT: One thing I did was not showing all methods when opening the browser on a class. > > For me this makes perfectly sense since I use the CMD-F + CMD-M shortcut > for searching for a method. Or go over all the protocol using the > arrow-keys on the selected protocol list. > > what do you think? It sounds like a perfect idea to get some traction for getting rid of method categories. Why would anyone want not to look at all methods in a class by default? In well refactored code, the number of methods is so limited that no scrolling should be needed. Not all code is like that, of course. Reading code is the most important activity of developing, and it is essential to limit the interaction needed to browse through code. Much better would be to show in the code pane which instance variables have accessors (r/w) and show all without accessors by default. > > However I can see that a "classical" user might be disturbed. This cut's down > the opening lag of a new browser from ~400ms to ~200ms on my machine, to me > thats quite worth it :). If the problem is performance, focus on performance. Why can not showing 30 items in a list make a difference of 200 ms? Stephan

Sept. 19, 2012

[Pharo-project] [update 2.0] #20297

by Esteban Lorenzano

20297 ----- Issue 6639: FileSystem enhancements http://code.google.com/p/pharo/issues/detail?id=6639 Issue 6695: RPackageSet speedup http://code.google.com/p/pharo/issues/detail?id=6695 Issue 6700: missing method in RPackageSet http://code.google.com/p/pharo/issues/detail?id=6700

Sept. 19, 2012

Re: [Pharo-project] ss3 down?

by Esteban Lorenzano

yeah, a network problem here :( On Sep 19, 2012, at 6:34 PM, Dale Henrichs <dhenrich(a)vmware.com> wrote: > The site is up and running and we have had no alerts from our data center so there must be something else going on if you are having trouble accessing the site ... > > Dale > > ----- Original Message ----- > | From: "Usman Bhatti" <usman.bhatti(a)gmail.com> > | To: Pharo-project(a)lists.gforge.inria.fr > | Sent: Wednesday, September 19, 2012 7:11:37 AM > | Subject: Re: [Pharo-project] ss3 down? > | > | For me, its down since morning. Cannot access the website too. > | > | > | On Wed, Sep 19, 2012 at 3:24 PM, Nicolas Petton < > | petton.nicolas(a)gmail.com > wrote: > | > | > | > | Works for me. > | > | Nico > | > | > | > | On Wed, 19 Sep 2012 15:17:19 +0200, Esteban Lorenzano < > | estebanlm(a)gmail.com > wrote: > | > is down, or just for me? > | > > | > > | > | -- > | Nicolas Petton > | http://nicolas-petton.fr > | > | > | >

Sept. 19, 2012

Re: [Pharo-project] Gofer meets Metacello

by Dale Henrichs

Sept. 19, 2012

Re: [Pharo-project] ss3 down?

by Dale Henrichs

Sept. 19, 2012