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

Re: [Pharo-dev] Moldable Pharo

by stepharo

did you check the nautilus plugin? Because you cannot compose everything but you can add plugin and I clean them today. Stef Le 8/7/15 13:28, Torsten Bergmann a Ã©crit : > Hi, > > with GT Inspector/Spotter we now have more moldable tools and Pharo becomes again a more flexible Tool and IDE. > The settings framework also allows to use the settings browser for own purposes and in own applications > - which is nice. One can even have a custom world menu in Pharo for the own app (see [1]) > > Reminds me on Eclipse (RCP, ...) where one can extend the IDE at specific extension > points to provide own custom tools and even build applications that are not IDE's. > > What about a moldable browser? For instance it would be nice if I could extend > the browser with own panes as replacement for the regular code pane. > > So for instance a method that returns an XML string could be shown as > a clickable tree instead of the code pane, ... > > foo > <xml> > ^'<someXML></someXML> > > With a flexible code editor like Rubric/TxText and Petit infrastructure or > other one could even provide moldable code panes with syntax highlighting for > different languages (PHP, CSS, HTML, ... > > Even the browser panes (packages, classes, ...) could be moldable - you display in a tree > an object with relations to other objects - when you click on them more customizable > details (depending on the displayed object) open up in new panes. > > The more (moldable) infrastructure Pharo provides the more cases one will find where it > can be applied. > > So the question is how modable is Pharo already (beside the mentioned cases) and > how moldable will Pharo be in the future? What is the next planned step in this area (if any)? > > Thx > T. > > > [1] http://lists.gforge.inria.fr/pipermail/pharo-project/2010-September/032670.… > >

July 8, 2015

Re: [Pharo-dev] Moldable Tools are Cooooool!

by Tudor Girba

I am so happy to see this moldability virus spreading like that :). Nice work. As Andrei pointed out, you might want to take a look at the GTDebugger. Here is a paper describing it: http://scg.unibe.ch/scgbib?query=Chis14b&display=abstract Cheers, Doru On Wed, Jul 8, 2015 at 7:06 PM, Sean P. DeNigris <sean(a)clipperadams.com> wrote: > Wow, things are getting way more fun around here :) > > Pharo Debugger Customization for Phexample > https://www.youtube.com/watch?v=Bq6RK_z3_Is > > In just a few minutes, Pharo's moldable debugger let's us turn a useless > error message into a much better format - diffed even! In a few minutes, > we've made our IDE experience much better! > > Happy customizing :) > > > > ----- > Cheers, > Sean > -- > View this message in context: > http://forum.world.st/Moldable-Tools-are-Cooooool-tp4836594.html > Sent from the Pharo Smalltalk Developers mailing list archive at > Nabble.com. > > -- www.tudorgirba.com "Every thing has its own flow"

July 8, 2015

Re: [Pharo-dev] Moldable Pharo

by Tudor Girba

Hi, Moldability is the very center part of the philosophy behind GT and Pharo. We want the whole environment to become moldable. Essentially, we bring dynamism into the IDE (the language has long been dynamic, but the tools were much less so). Indeed, right now you see it materialized by default in the inspector and spotter. In the Moose image, the debugger also supports the same concept. Next, it will be materialized in the playground. And yes, the code browser will also be affected, but probably at the end. We specifically did not start with the code browser exactly because by not doing that we forced ourselves to reinvent the IDE from the basics. We started with the inspector because Pharo is object-oriented not class-oriented :). We are currently working on the Playground because this is where objects start. I think we are on a good path to reinvent that little interface as well. We will likely have news on that soon. Cheers, Doru

July 8, 2015

Re: [Pharo-dev] ZnUrl>>#withPathSegments: Surprise

by Sean P. DeNigris

Sven Van Caekenberghe-2 wrote > where do you find proof for the #withX: meaning 'replace X in a copy of > the receiver ' ? I guess from the collection #with:... messages that start from a fresh copy, but also quickly browsing the image, I see other such uses e.g. #withAngle:, #withNewSelection:, all of which return a new instance with the argument passing through to it independent of the original value. ----- Cheers, Sean -- View this message in context: http://forum.world.st/ZnUrl-withPathSegments-Surprise-tp4836593p4836616.html Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com.

July 8, 2015

Evaluating Confidential Info

by Sean P. DeNigris

Now that Playground is saving all snippets for later use, how would one evaluate something for which there should be no record e.g. logging into a server? ----- Cheers, Sean -- View this message in context: http://forum.world.st/Evaluating-Confidential-Info-tp4836615.html Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com.

July 8, 2015

Re: [Pharo-dev] ZnUrl>>#withPathSegments: Surprise

by Sven Van Caekenberghe

> On 08 Jul 2015, at 19:13, Sean P. DeNigris <sean(a)clipperadams.com> wrote: > > Sven Van Caekenberghe-2 wrote >> I do remember adding this especially for you ;-) > > Ha ha, yes I remember... the #/ is certainly useful. My question is just > about the name #withPathSegments:. Maybe something like > #withAddedPathSegments: might be clearer? Also, there is the #copyXyz > convention used elsewhere in Pharo. Maybe#copyAddingPathSegments: for > uniformity? Maybe, but where do you find proof for the #withX: meaning 'replace X in a copy of the receiver ' ? I mean, it does not strike me as confusing. Of course, the longer #copyAddingPathSegments: is long and clear. > ----- > Cheers, > Sean > -- > View this message in context: http://forum.world.st/ZnUrl-withPathSegments-Surprise-tp4836593p4836596.html > Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com. >

July 8, 2015

Re: [Pharo-dev] Moldable Tools are Cooooool!

by Andrei Chis

On Wed, Jul 8, 2015 at 8:07 PM, Sean P. DeNigris <sean(a)clipperadams.com> wrote: > Andrei Chis wrote > > have the diff view appear dynamically in the debugger > > when you get an assertion failed due to PhexMatcher > > Is that possible? I mean "does the framework allow for that capability", > not > "could it be hacked in". > That's almost possible but using the Glamour interface of the debugger. The Spec interface is not really optimized to be customizable. Though it should be doable with a bit more work. You would do that by creating a custom debugger extension and customizing the interface with the widgets that you want/need. Right now the framework does not automatically switch between extensions when it detects that a more specific extension is available, though incorporating these feature into the framework is not that difficult. The SUnit debugger gives an example of how to customize the glamour interface (http://scg.unibe.ch/research/moldabledebugger) Cheers, Andrei > > > ----- > Cheers, > Sean > -- > View this message in context: > http://forum.world.st/Moldable-Tools-are-Cooooool-tp4836594p4836600.html > Sent from the Pharo Smalltalk Developers mailing list archive at > Nabble.com. > >

July 8, 2015

Re: [Pharo-dev] Moldable Tools are Cooooool!

by Andrei Chis

Very nice :) Though an even cooler think would be to have the diff view appear dynamically in the debugger when you get an assertion failed due to PhexMatcher :) Cheers, Andrei On Wed, Jul 8, 2015 at 7:06 PM, Sean P. DeNigris <sean(a)clipperadams.com> wrote: > Wow, things are getting way more fun around here :) > > Pharo Debugger Customization for Phexample > https://www.youtube.com/watch?v=Bq6RK_z3_Is > > In just a few minutes, Pharo's moldable debugger let's us turn a useless > error message into a much better format - diffed even! In a few minutes, > we've made our IDE experience much better! > > Happy customizing :) > > > > ----- > Cheers, > Sean > -- > View this message in context: > http://forum.world.st/Moldable-Tools-are-Cooooool-tp4836594.html > Sent from the Pharo Smalltalk Developers mailing list archive at > Nabble.com. > >

July 8, 2015

Re: [Pharo-dev] Moldable Tools are Cooooool!

by Sean P. DeNigris

Andrei Chis wrote > have the diff view appear dynamically in the debugger > when you get an assertion failed due to PhexMatcher Is that possible? I mean "does the framework allow for that capability", not "could it be hacked in". ----- Cheers, Sean -- View this message in context: http://forum.world.st/Moldable-Tools-are-Cooooool-tp4836594p4836600.html Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com.

July 8, 2015

Re: [Pharo-dev] ZnUrl>>#withPathSegments: Surprise

by Sven Van Caekenberghe

I don't know. If you see where it is used, you'll see it makes sense to add: 'http://host.com:80' asZnUrl / 'files' / 'readme.txt' Where #/ is equivalent to #withPathSegments. I do remember adding this especially for you ;-) > On 08 Jul 2015, at 18:30, Sean P. DeNigris <sean(a)clipperadams.com> wrote: > > I expected #withPathSegments: to return a copy with path segments replaced by > the argument, but in fact they were added to the existing path. Just me? > > > > ----- > Cheers, > Sean > -- > View this message in context: http://forum.world.st/ZnUrl-withPathSegments-Surprise-tp4836593.html > Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com. >

July 8, 2015