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

Re: [Pharo-dev] Pharo configurations via git unloadable behind proxy

by Sven Van Caekenberghe

Pharo only knows about HTTP proxies, not proxies for any other protocol. > On 26 Apr 2015, at 16:42, Andreas Wacknitz <a.wacknitz(a)gmx.de> wrote: > >> >> Am 26.04.2015 um 16:36 schrieb Max Leske <maxleske(a)gmail.com>: >> >> >>> On 26 Apr 2015, at 15:42, Andreas Wacknitz <a.wacknitz(a)gmx.de> wrote: >>> >>> Trying to load a configuration that involves git (eg. QualityAssistant) with the configuration browser fails if I am behind a proxy while others (without git) work fine. >>> Do I miss a configuration for git? >> >> What exactly fails? Do you get an exception? Is it a connection problem? > The contents of downloaded files (eg. github-UkoQualityAssistantv041531859342567419945494505101053.zip) is > just > > <HTML></HTML> > > and thus decompressing fails. > This problem is reproducible behind a proxy and works without a proxy. Other configurations (e.g. NoCSV) work properly behind the proxy.

April 26, 2015

Re: [Pharo-dev] Pharo configurations via git unloadable behind proxy

by Andreas Wacknitz

> Am 26.04.2015 um 16:36 schrieb Max Leske <maxleske(a)gmail.com>: > > >> On 26 Apr 2015, at 15:42, Andreas Wacknitz <a.wacknitz(a)gmx.de> wrote: >> >> Trying to load a configuration that involves git (eg. QualityAssistant) with the configuration browser fails if I am behind a proxy while others (without git) work fine. >> Do I miss a configuration for git? > > What exactly fails? Do you get an exception? Is it a connection problem? The contents of downloaded files (eg. github-UkoQualityAssistantv041531859342567419945494505101053.zip) is just <HTML></HTML> and thus decompressing fails. This problem is reproducible behind a proxy and works without a proxy. Other configurations (e.g. NoCSV) work properly behind the proxy.

April 26, 2015

Re: [Pharo-dev] brickValue: in Glamour-Morphic-Brick

by Nicolai Hess

2015-04-26 1:35 GMT+02:00 Aliaksei Syrel <alex.syrel(a)gmail.com>: > Can we remove #brickValue: ? >> I don't see how this is different from #cull: >> And the name is strange, I don't see how this depends an a brick. > > > Not really, it is implemented also in Object. > Ah, ok, I missed that one. > It is used to script brick layout properties. So user can say: > > brick height: 10. > brick height: #approximate. > brick height: [ :b | b width * 2]. > > or even > > brick height: HeightStrategy new. > > so brickValue: is sent to any kind of object that is used as height. So > without changing brick itself user can implement its own class and override > brickValue: and have custom strategy to calculate height. > > Such practice is used in many places in brick. > > I saw the call to brickValue: in GTSpotter and GLMAnimation, is it only used for brick attributes, like in your example above? nicolai > > > Cheers, > Alex > > On Sun, Apr 26, 2015 at 12:16 AM, Nicolai Hess <nicolaihess(a)web.de> wrote: > >> Can we remove #brickValue: ? >> I don't see how this is different from #cull: >> And the name is strange, I don't see how this depends an a brick. >> >> nicolai >> > >

April 26, 2015

Re: [Pharo-dev] Pharo configurations via git unloadable behind proxy

by Max Leske

> On 26 Apr 2015, at 15:42, Andreas Wacknitz <a.wacknitz(a)gmx.de> wrote: > > Trying to load a configuration that involves git (eg. QualityAssistant) with the configuration browser fails if I am behind a proxy while others (without git) work fine. > Do I miss a configuration for git? What exactly fails? Do you get an exception? Is it a connection problem? > > Regards > Andreas > >

April 26, 2015

Pharo configurations via git unloadable behind proxy

by Andreas Wacknitz

Trying to load a configuration that involves git (eg. QualityAssistant) with the configuration browser fails if I am behind a proxy while others (without git) work fine. Do I miss a configuration for git? Regards Andreas

April 26, 2015

Re: [Pharo-dev] CI Still Down

by Marcus Denker

Yes, sadly this needs the admins of ci.inria.fr <http://ci.inria.fr/> and it is weekend... > On 26 Apr 2015, at 01:05, Sean P. DeNigris <sean(a)clipperadams.com> wrote: > > ... since yesterday! > > > > ----- > Cheers, > Sean > -- > View this message in context: http://forum.world.st/CI-Still-Down-tp4821894.html > Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com. >

April 26, 2015

Re: [Pharo-dev] Issue 15394: (Launcher) Safer Image Deletion

by stepharo

Yes! Stef Le 24/4/15 15:56, Sean P. DeNigris a Ã©crit : > https://pharo.fogbugz.com/default.asp?15394 > > Especially as we move to push the launcher as the default download for new > users, this will be crucial. Right now, we are deleting via an `rm -fr` like > approach, which is quite harsh for a GUI tool, especially when deleting an > image, which is akin to an operating system. It would be better to remove > the confirmation dialog and move the image to the trash, with an undo if > possible. > > > > ----- > Cheers, > Sean > -- > View this message in context: http://forum.world.st/Issue-15394-Launcher-Safer-Image-Deletion-tp4821635.h… > Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com. > >

April 26, 2015

Re: [Pharo-dev] SystemWindow vs. StandardWindow

by Ben Coman

+1... * Outside of Pharo, I've only ever know it be called a "window title". Never heard of "window label". A "label" would be for panes, tabs and widgets within the window. * Historical context + implementor-count... * StandardWindow>>title: ~2006 +25-most-window-related * StandardWindow>>title ~2006 +170 (most return constant strings - window related?) * SystemWindow>>label ~1998 +370-many-unrelated-to-windows * SystemWindow>>setLabel: ~2011 +1 * Not sure how useful it is, but I compiled a table for comparison https://docs.google.com/spreadsheets/d/1YvMgH8uSbOsSf30T7lyaCCcnDLy4tzXnM39… * Would need to consider the 8 methods of SystemWindow "label" protocol * One random observation EyeCollectionInspector>>title and EyeCollectionInspector-class>>label cheers -ben On Fri, Apr 24, 2015 at 10:53 PM, Sean P. DeNigris <sean(a)clipperadams.com> wrote: > Is this distinction necessary? > > I started digging when I noticed that SystemWindow>>#label and #setLabel: > becomes StandardWindow>>#title and #title. Wouldn't it be better to just > push #title up if we like that API better? It's exhausting to manage all > these tiny (and possibly unnecessary) differences... > > > > ----- > Cheers, > Sean > -- > View this message in context: > http://forum.world.st/SystemWindow-vs-StandardWindow-tp4821655.html > Sent from the Pharo Smalltalk Developers mailing list archive at > Nabble.com. > >

April 26, 2015

Re: [Pharo-dev] brickValue: in Glamour-Morphic-Brick

by Aliaksei Syrel

> > Can we remove #brickValue: ? > I don't see how this is different from #cull: > And the name is strange, I don't see how this depends an a brick. Not really, it is implemented also in Object. It is used to script brick layout properties. So user can say: brick height: 10. brick height: #approximate. brick height: [ :b | b width * 2]. or even brick height: HeightStrategy new. so brickValue: is sent to any kind of object that is used as height. So without changing brick itself user can implement its own class and override brickValue: and have custom strategy to calculate height. Such practice is used in many places in brick. Cheers, Alex On Sun, Apr 26, 2015 at 12:16 AM, Nicolai Hess <nicolaihess(a)web.de> wrote: > Can we remove #brickValue: ? > I don't see how this is different from #cull: > And the name is strange, I don't see how this depends an a brick. > > nicolai >

April 25, 2015

CI Still Down

by Sean P. DeNigris

... since yesterday! ----- Cheers, Sean -- View this message in context: http://forum.world.st/CI-Still-Down-tp4821894.html Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com.

April 25, 2015