June 2012 - Pharo-dev - lists.pharo.org

Re: [Pharo-project] fulltext method search

by phil＠highoctane.be

Ok, removed it. So now, I see a "method indexing" progress bar popping for a while repeatedly :-). Well, it works. And I tried it out: "this is an instantaneous (~2ms!) search over all method sources" Search default matchSentence: 'Convert to a String with Characters' {(ByteArray>>#asString "a CompiledMethod(108265472)")} I am looking into how to integrate this with Spotlight [which is really fantastic and improves productivity a lot!] (I guess I'll do another binding like Ctrl-Enter instead of Shift-Enter). So, how do you get the results? Is there a quick way to open the browser on a given CompliedMethod right away? (Ah! So much to learn.... Sound is working nicely now BTW, got a couple clues!) Phil 2012/6/25 Camillo Bruni <camillobruni(a)gmail.com> > > not that I would do something fancy :) (and yeas 2.0) > > do:displayingProgress: > > is the only thing I do basically (and you can remove the display... part > so it should even work under 1.4 I think) > > > > On 2012-06-25, at 00:38, phil(a)highoctane.be wrote: > > Error is due to the Progress Morph (guess you use the new thing in 2.0) > > > > 2012/6/25 Camillo Bruni <camillobruni(a)gmail.com> > > > >> I was bored today (after yet another kernel panic) so I implemented a > quick > >> fulltext search for methods. > >> > >> > >> "this takes around 1.5mins" > >> Search default buildIndex. > >> > >> "this is an instantaneous (~2ms!) search over all method sources" > >> Search default matchSentence: 'Convert to a String with > Characters' > >> > >> > >> I think we should integrate that into some kind of "SpotLight" search > >> for accessing the sources much much faster... > >> > >> The current draft is available here: > >> http://smalltalkhub.com/#!/~dh83/spotlight > >> > >> > >> > > >

June 25, 2012

Re: [Pharo-project] Opal ?

by Goubier Thierry

Le 25/06/2012 09:35, Marcus Denker a Ã©crit : > > On Jun 21, 2012, at 4:38 PM, Goubier Thierry wrote: > >>> >> >> Ok, this is where I may see things. >> >> To insert things in the AST, I can subclass the RBProgramNodeVisitor ? >> > > Or you could use the RBParseTreeRewriter... Well, I used the RBProgramNodeVisitor : worked perfectly, since the original code was based on VW's ProgramNodeVisitor. Reimplemented part of the RBParseTreeRewriter rewritter, in fact, but then it avoided reimplementing all the rewritting rules as RB patterns and blocks. So far, Opal works fine, thanks for the work done. Except for one thing : recompiling an instrumented method erase the original source in the method (and force a decompilation); I could not find how to force it to still use the original (non modified) source. Regards, Thierry -- Thierry Goubier CEA list Laboratoire des Fondations des SystÃ¨mes Temps RÃ©el EmbarquÃ©s 91191 Gif sur Yvette Cedex France Phone/Fax: +33 (0) 1 69 08 32 92 / 83 95

June 25, 2012

Re: [Pharo-project] Debugging code with morphic leads to freezes and a weird message

by phil＠highoctane.be

Oh, okay, thanks a lot!. Lesson taken: check for implementers :) 2012/6/25 Marcus Denker <marcus.denker(a)inria.fr> > > On Jun 24, 2012, at 9:39 PM, phil(a)highoctane.be wrote: > > > While debugging actions behind buttons, I do get Project > class>>spawnNewProcessIfThisIsUI: not being understood. > > > > This prevents from debugging interactions properly. > > > > And sometimes I get the dreaded "Computation Terminated" message, > effectively locking me out of the image (nothing moves anymore, I've got to > kill the process). > > > > Ideas on how to fix? This is 1.4 14438 > > > > > > #spawnNewProcessIfThisIsUI: has been moved to UIManager.... it seems that > OB still calls it on Project. > > Marcus > > > > -- > Marcus Denker -- http://marcusdenker.de > > >

June 25, 2012

Re: [Pharo-project] Debugging code with morphic leads to freezes and a weird message

by Marcus Denker

On Jun 24, 2012, at 9:39 PM, phil(a)highoctane.be wrote: > While debugging actions behind buttons, I do get Project class>>spawnNewProcessIfThisIsUI: not being understood. > > This prevents from debugging interactions properly. > > And sometimes I get the dreaded "Computation Terminated" message, effectively locking me out of the image (nothing moves anymore, I've got to kill the process). > > Ideas on how to fix? This is 1.4 14438 > > #spawnNewProcessIfThisIsUI: has been moved to UIManager.... it seems that OB still calls it on Project. Marcus -- Marcus Denker -- http://marcusdenker.de

June 25, 2012

Re: [Pharo-project] Opal ?

by Marcus Denker

On Jun 21, 2012, at 4:38 PM, Goubier Thierry wrote: >> > > Ok, this is where I may see things. > > To insert things in the AST, I can subclass the RBProgramNodeVisitor ? > Or you could use the RBParseTreeRewriter... -- Marcus Denker -- http://marcusdenker.de

June 25, 2012

Re: [Pharo-project] Array size in inspector

by Davide Varvello

Hi Doru, I gave a glance to Glamorous inspector some weeks ago I think it could be better if I could change instance variables like old inspector, i.e. on the right panel without the need of sending a message on self. Moreover, today I reloaded your inspector on my image but I couldn't remember how to run it, after a search on Pharo's list I foundÂ GTInspector openOn: someObject It can be useful if you can add it onÂ http://www.humane-assessment.com/blog/glamorous-inspector-for-smalltalkÂ Â Cheers Â Davide ---- -Â Cerchi un bravo Dentista, Avvocato, Commercialista? Un buon Hotel, Ristorante, Pizzeria? Io l'ho trovato su Oltre il Passaparola - Blog: Cambia il Tempo >________________________________ > From: Tudor Girba-2 [via Smalltalk] <ml-node+s1294792n4636329h45(a)n4.nabble.com> >To: Davide Varvello <varvello(a)yahoo.com> >Sent: Saturday, June 23, 2012 11:04 PM >Subject: Re: Array size in inspector > > >You can also give the Glamorous Inspector a try: >http://www.humane-assessment.com/blog/glamorous-inspector-for-smalltalk > >Cheers, >Doru > > > >On 19 Jun 2012, at 16:19, Davide Varvello wrote: > > >> Hi, >> I've a very long array I'd like to inspect, unfortunately inspector shows >> only first 100 elements and few last elements of my array. How can change >> that behavior? >> Thanks >> Davide >> >> -- >> View this message in context: http://forum.world.st/Array-size-in-inspector-tp4635484.html >> Sent from the Pharo Smalltalk mailing list archive at Nabble.com. >> >-- >www.tudorgirba.com > >"Value is always contextual." > > > > > > >>________________________________ > >If you reply to this email, your message will be added to the discussion below: http://forum.world.st/Array-size-in-inspector-tp4635484p4636329.html >To unsubscribe from Array size in inspector, click here. >NAML > > -- View this message in context: http://forum.world.st/Array-size-in-inspector-tp4635484p4636488.html Sent from the Pharo Smalltalk mailing list archive at Nabble.com.

June 25, 2012

Re: [Pharo-project] Problem with Monticelo package and corrupted sourcecode/zip

by Pavel Krivanek

Can you try to file-in the uncompressed file? Or at least look if the file is complete. Because I discovered this issue when trying to load Fuel tests into the Pharo Kernel (the download and unzip of the package is done directly by Pharo) and I had no problem with unzip of this file on Linux but the resultant file was not complete and the loading finished with syntax error. -- Pavel On Sun, Jun 24, 2012 at 11:58 PM, phil(a)highoctane.be <phil(a)highoctane.be> wrote: > I downloaded it, renamed the .mcz as a .zip and used the unarchiver on it. > No issue. > > 3271 lines inÂ source.st > > OSX Lion over here. > > Was able to get it back from my 1.4 on CogVM and browse the downloaded > package. > > Seems to be a OS X related issue. Rebooting a Mac may be rare, but may > help... Network problems? Sometimes my NAS needs a kick in the butt.. > > Phil > > 2012/6/24 Mariano Martinez Peck <marianopeck(a)gmail.com> >> >> Hi guys. I have a problem and I don't understand. If you download >> FuelTests-MarianoMartinezPeck.269 from http://ss3.gemstone.com/ss/Fuel and >> try to unzip it, in my MacOSX I have an error saying the fie is corrupted. >> If I proceed anyway, and I open the .st it is CUT in the middle (it does not >> finish properly). I thought it was a problem with Traits (because FuelTest >> has traits), but I tried other packages even with traits and they are fine. >> Then I thought that that version of the package could be corrupted, but I >> went several versions ago in history and same problem. ThenÂ thoguht it was >> Monticello problem. I went to Pharo 2.0 20020 and again, same problem. >> >> So I am lost. I remember something about the zip compression problem in >> Monticello.... could this be related? >> >> thanks >> >> -- >> Mariano >> http://marianopeck.wordpress.com >> > > > > -- > Philippe Back > Dramatic Performance Improvements > Mob: +32(0) 478 650 140 | Fax: +32 (0) 70 408 027 Mail: phil(a)highoctane.be | > Web: http://philippeback.eu | Blog: http://philippeback.be > > High Octane SPRL > rue cour Boisacq 101 > 1301 Bierges > Belgium

June 25, 2012

[Pharo-project] Is it possible to open up two / multiple window surfaces on demand

by S Krish

I am looking for a VM support to open up on configuration atleast 2 if not more as possible windows ( surfaces to paint morphs on). Viz: I run the dev code / browsers etc on one panel and designate the other window as the runtime panel to popup runtime windows only

June 25, 2012

[Pharo-project] [CTI] News :)

by Santiago Bragagnolo

Hi all! i just wrote a couple of posts about the Concrete type inference project! I hope you enjoy my adventures & misadventures over this fantastic world :D http://concretetypeinference.blogspot.com.ar/2012/06/about-mechanisms-part-… [ About Error contexts ] http://concretetypeinference.blogspot.com.ar/2012/06/about-mechanisms-part-… [ About Error management ] http://concretetypeinference.blogspot.com.ar/2012/06/about-mechanisms-part-… About primitive analysis improvements ] http://concretetypeinference.blogspot.com.ar/2012/06/concrete-type-inferenc… About progress :) ] Good week, and keep tunned :)

June 25, 2012

Re: [Pharo-project] fulltext method search

by Camillo Bruni

not that I would do something fancy :) (and yeas 2.0) do:displayingProgress: is the only thing I do basically (and you can remove the display... part so it should even work under 1.4 I think) On 2012-06-25, at 00:38, phil(a)highoctane.be wrote: > Error is due to the Progress Morph (guess you use the new thing in 2.0) > > 2012/6/25 Camillo Bruni <camillobruni(a)gmail.com> > >> I was bored today (after yet another kernel panic) so I implemented a quick >> fulltext search for methods. >> >> >> "this takes around 1.5mins" >> Search default buildIndex. >> >> "this is an instantaneous (~2ms!) search over all method sources" >> Search default matchSentence: 'Convert to a String with Characters' >> >> >> I think we should integrate that into some kind of "SpotLight" search >> for accessing the sources much much faster... >> >> The current draft is available here: >> http://smalltalkhub.com/#!/~dh83/spotlight >> >> >>

June 24, 2012