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

Re: [Pharo-project] Unresponsive 1.4 image

by Janko MivÅ¡ek

Hi guys, This code reproduces the blockade in fresh 1.4 OneClick (with or without Transcript window open): [ (Delay forSeconds: 5) wait. SmalltalkImage current saveSession. ] forkAt: Processor userBackgroundPriority Note that fork on default priority does not block while on user background priority it blocks. It does not always block immediately (sometimes yes sometimes not) but it after image restart it is always blocked. But if you periodically write to Transcript from process on user background priority, it does not block, even if you heavily browse around, write to transcript, do manual snapshot: [ [true] whileTrue: [(Delay forMilliseconds: 500) wait. Transcript show: '.tst.'] ] forkAt: Processor userBackgroundPriority So, it seems not only thread-safe Transcript but something with snapshot (from a background priority, why?) is in play here. SmalltalkImage>>snapshot:andQuit: , shouldnât run it on highest priority to block all other processes during the preparation for snapshot? Hope this helps a bit more Janko Dne 21. 04. 2012 11:48, piÅ¡e Igor Stasenko: > So, this is related to thread-safe transcript code.. > which using semaphore to synchronize access to transcript stream. > I guess what happens is that a semaphore is blocked in another > process, but for some strange reason never released. This leads to > situation, that any other process which will attempt to write to > transcript will be blocked as well.. forever. Why is input not blocked immediately but after image restart? More investigation: 1. correction: image blocks in any case after restart, regardless if there is additional writing to Transcript or not. Snapshot namely writes to Transcript by itself and this is enough to block. 1. even if Transcript is closed the image blocks after restart This simulated snapshot from the background does NOT block after restart: [ (Delay forSeconds: 5) wait. SmalltalkImage current saveSession. ] fork This one also not: [ | elapsed | (Delay forSeconds: 5) wait. elapsed := Time millisecondsToRun: [SmalltalkImage current saveSession]. Transcript show: ' in ', (elapsed // 1000) printString, 's '. ] forkAt: Processor userBackgroundPriority > > 2012/4/21 Janko MivÅ¡ek <janko.mivsek(a)eranova.si>: >> Hi guys, >> >> It seems that writing to Transcript from background process causes my >> freeze after restart. And yes, I have Transcript window open. >> >> This is a code snippet in question: >> >> AIDASite class>>imageSnapshot >> ... >> elapsed := Time millisecondsToRun: >> [SmalltalkImage current saveSession]. >> Transcript show: ' in ', (elapsed // 1000) printString, 's '. >> >> >> If I comment out the last line, snapshot doesn't block the image after >> restart. >> >> This is not necessary 1.4 specific problem, because I noticed similar >> problem in 1.3 also. Just that on 1.3 input freezes immediately while on >> 1.4 it freezes after restart. >> >> Let me investigate a bit further, what if Transcript window is not open? >> >> Best regards >> Janko >> >> >> >> Dne 20. 04. 2012 21:42, piÅ¡e StÃ©phane Ducasse: >>> Ok may be nautilus bug. It should be different than janko problem. >>> >>> Stef >>> >>> On Apr 20, 2012, at 9:39 PM, Sean P. DeNigris wrote: >>> >>>> >>>> StÃ©phane Ducasse wrote >>>>> >>>>> on which os? >>>>> >>>> >>>> Mac Lion 10.7.3 >>>> >>>> -- >>>> View this message in context: http://forum.world.st/Unresponsive-1-4-image-tp4574693p4575049.html >>>> Sent from the Pharo Smalltalk mailing list archive at Nabble.com. >>>> >>> >>> >>> >> >> -- >> Janko MivÅ¡ek >> Aida/Web >> Smalltalk Web Application Server >> http://www.aidaweb.si >> > > > -- Janko MivÅ¡ek Aida/Web Smalltalk Web Application Server http://www.aidaweb.si

April 21, 2012

Re: [Pharo-project] The video about Athens

by Sean P. DeNigris

Sean P. DeNigris wrote > > I get Generic failure > fnSpec: #(#bool #ioRegisterSurface #(#ulong #aCairoSurfaceHandle #, > #void #* #sqSurfaceDispatchPtr #, #int #* #idHolder)) > module: #SurfacePlugin > errorCode: 1 > Igor, I guess you didn't run into this on Mac... any ideas why it's still not working? Sean -- View this message in context: http://forum.world.st/The-video-about-Athens-tp4571100p4576317.html Sent from the Pharo Smalltalk mailing list archive at Nabble.com.

April 21, 2012

[Pharo-project] Feedback about SmallLint

by Mariano Martinez Peck

Hi guys. After our meeting at Lille, I did a pass of SmallLint over Fuel project. I have to say that I enjoyed the tool and that. I love the manifiesto idea. I love to store metada and to be able to reject rules and scenarios to avoid having false positives again in the future. In addition, it looks like I had less false positive than with other tools. However, there were a couple of rules that I don't understand or may suggest something: 1) First, apart form showing the package between parenthesis, I would also show the protocol (category). Because some rules play with that, so it is handy if you can directly see them there, otherwise you need to browse the code. 2) RBContainsRule is not clear for me. *" 'Checks for the common code fragment: "(aCollection detect: [:each | ''some condition''] ifNone: [nil]) ~= nil". contains: can simplify this code to "aCollection contains: [:each | ''some condition'']". Not only is the contains: variant shorter, it better signifies what the code is doing.'"* That is not very true. #contains answers true or false, while detect:ifNone:[] answers whatever, and in your example it is nil. So, the answer will be different and hence it is not the same. Ok, in your example, you compare after with == nil... but... I don't really see a use case. For example, in my case it detected: classNamed: className ^ (migrations detect: [:m | m sourceClassName = className ] ifNone: [ ^ self globalClassNamed: className ]) targetClass. 3) RBLawOfDemeterRule: maybe you can exclude string concatenation? (#,) because it detected a lot of messages like this: signalWith: aGlobalName and: aSelector ^ self signal: 'Method ', aGlobalName, '>>#', aSelector, ' not found.' Also, what about excluding messages like #asXXX. For example, here it detected "self signature asByteArray". Is that completly breaking the law of demeter? mmm I don't think so. verifySignatureFrom: aDecoder | streamSignature | streamSignature := ByteArray new: self signature size. aDecoder nextEncodedBytesInto: streamSignature. (self signature asByteArray = streamSignature) ifFalse: [ FLBadSignature signalCurrentSignature: self signature streamSignature: streamSignature ]. 4) RBSentNotImplementedRule: I didn't understand why it has selected: largeIdentityHash <primitive: 75> self primitiveFailed both messages are implemented :( 5) RBClassInstVarNotInitializedRule could be smarter and search not only for #initialize but for #initializeXXX. Lots of times, I initialize my objects with data, using a particular nitialize method, such as: initializeWithBehaviors: someBehaviors extensionMethods: someExtensionMethods so the rule says there is no initialize...but I have that one ;) That's all the feedback so far. It helped me to find some things to improve so thanks! -- Mariano http://marianopeck.wordpress.com

April 21, 2012

Re: [Pharo-project] The video about Athens

by StÃ©phane Ducasse

Cool I was thinking exactly about using platform. Srf On Apr 20, 2012, at 2:54 PM, Igor Stasenko wrote: > i uploaded the new version of Athens-Cairo, which takes the 'default' path > depending on platform it running, also fixed it, so you don't have to > change multiple places. > also , updated config version 1.0 to load this package. > > On 20 April 2012 14:45, Igor Stasenko <siguctua(a)gmail.com> wrote: >> As i suspected, it works out of the box on my ubuntu-32 bit installation, >> i just had to change the cairo library path to >> /usr/lib/libcairo.so >> >> :) >> >> On 20 April 2012 14:30, Fernando Olivero <fernando.olivero(a)usi.ch> wrote: >>> Athens is an API, the rendering is actually done by the backends. >>> Currently Ballon and Cairo are supported. >>> >>> So, the answer is that Athens is supported in all platforms using the >>> Ballon backend, (note that ballon does not have all the features of >>> Cairo, so some of the features in the demo wont work), and the Cairo >>> backend as well. >>> >>> You just need to tell NativeBoos were to locate the cairo library >>> (built using MacPorts, so the path starts at /opt/...). >>> >>> For now we just hardcoded the library path (you can change them by >>> hand for now, to point to your local Cairo lib) >>> >>> TCairoLibrary class>>nbLibraryNameOrHandle >>> ^ '/opt/local/lib/libcairo.2.dylib' >>> >>> AthensSurface class>>nbLibraryNameOrHandle >>> ^ '/opt/local/lib/libcairo.2.dylib' >>> >>> This should be changed, perhaps a settings? >>> >>> Fernando >>> >>> On Fri, Apr 20, 2012 at 12:25 PM, Alexis Parseghian <a.parseg(a)gmail.com> wrote: >>>>> People should start to use Athens, apps around Pharo would start be >>>>> look waaaayyy cooler. >>>> >>>> About that. >>>> >>>> Is it supposed to work outside of the Mac platform ? >>>> >>>> Tried the demo, and AthensDemoMorph new openInWorld crashes on >>>> "function unavailable". That backtraces into Block in >>>> NBFFICallout>>generateCall:module: where module seems to be >>>> '/opt/local/lib/libcairo.2.dylib' which is obviously Mac-ish but makes >>>> no sense on Linux. >>>> >>>> Where is the library name and location set/picked up ? I guess this >>>> question could be generalized to how the platform-specific >>>> initialization happens. >>>> >>>> As an aside, all NativeBoost-Tests are green. >>>> >>>> Platform : debian testing 64-bit with ia32 libs installed. Latest >>>> NBCogVM from jenkins, blank Pharo 1.4 #14414 with Athens 1.0 >>>> (explicitly) installed from SS:Athens/ConfigurationOfAthens. >>>> >>>> I also look forward to discovering whatever new, inventive stuff these >>>> new projects enable. >>> >> >> >> >> -- >> Best regards, >> Igor Stasenko. > > > > -- > Best regards, > Igor Stasenko. >

April 21, 2012

Re: [Pharo-project] MethodWrappers

by Marcus Denker

On Apr 21, 2012, at 1:48 PM, StÃ©phane Ducasse wrote: > > On Apr 20, 2012, at 7:29 PM, Paul DeBruicker wrote: > >> If you were to integrate MethodWrappers into Pharo today would you still use >> these packages: >> >> Compiler-eem.214 & Compiler-eem.215, >> Kernel-619 to Kernel-eem.6 > > what are the changes these packages introduce? This was this report: http://code.google.com/p/pharo/issues/detail?id=4813 All integrated. Marcus -- Marcus Denker -- http://marcusdenker.de

April 21, 2012

Re: [Pharo-project] MethodWrappers

by StÃ©phane Ducasse

On Apr 20, 2012, at 7:29 PM, Paul DeBruicker wrote: > If you were to integrate MethodWrappers into Pharo today would you still use > these packages: > > Compiler-eem.214 & Compiler-eem.215, > Kernel-619 to Kernel-eem.6 what are the changes these packages introduce? > > or use the latest from Squeak trunk? > > I'm also trying to build a DTrace enabled unix VM based on Adrian Lienhards > work here: http://www.adrian-lienhard.ch/blog?dialog=smalltak-meets-dtrace. > Has anyone already made one for linux that I could just use? > > > Thanks > > > > Eliot Miranda-2 wrote >> >> On Wed, Sep 21, 2011 at 3:48 AM, Torsten Bergmann <astares@> wrote: >> >>> Hi Eliot, >>> >>> at [1] you mentioned a port of MethodWrappers to Squeak 4.2. >>> Is there also a port for Pharo available or do you plan to >>> port it? >>> >> >> I was hoping someone with focus on Pharo would integrate the necessary >> compiler and kernel changes into Pharo. The Squeak trunk packages are >> Compiler-eem.214 & Compiler-eem.215, and Kernel-619 to Kernel-eem.623. >> I've >> attached the diffs. Notionally file in Compiler-213-214.st & >> Kernel-618-619.st, then file-in Compiler-214-215.st & Kernel-619-623.st. >> These are generally useful, a) restoring the ability to create subclasses >> of CompiledMethod, and b) fixing various subtle copying-related >> state-sharing bugs with AdditionalMethodState (the objects that add >> additional state to compiled methods, pragmas, properties and "extra inst >> vars"). >> >> The MethodWrappers4.2 package should then just load. It serves as a model >> for adding inst vars to CompiledMethod. >> >> >>> For those who dont know about MethodWrappers: look at [2] >>> >> >> Indeed. Yes, there's an alternative implementation above >> objects-as-methods, but that implementation is far from complete. It >> would >> be good to see he same kind of completeness. e.g. with MethodWrappers you >> can write >> >> | indent wrapper | >> indent := 0. >> wrapper := MwBlockMethodWrapper on: #benchFib inClass: Integer. >> wrapper >> beforeBlock: [:rcvr :args| >> Transcript crtab: indent; print: rcvr; space; nextPutAll: #benchFib; >> flush. >> indent := indent + 1]; >> afterBlock: [:rcvr :args| indent := indent - 1]. >> [wrapper install. >> 7 benchFib] ensure: [wrapper uninstall] >> >> whereas there are few if any examples with the objects-as-methods AFAICS. >> >>> >>> Thanks >>> Torsten >>> >>> >>> [1] >>> http://lists.squeakfoundation.org/pipermail/squeak-dev/2011-September/16156… >>> [2] http://www.refactory.com/Software/MethodWrappers/ >>> -- >>> NEU: FreePhone - 0ct/min Handyspartarif mit Geld-zurÃ¼ck-Garantie! >>> Jetzt informieren: http://www.gmx.net/de/go/freephone >>> >> >> >> >> -- >> best, >> Eliot >> > > > -- > View this message in context: http://forum.world.st/MethodWrappers-tp3829576p4574768.html > Sent from the Pharo Smalltalk mailing list archive at Nabble.com. >

April 21, 2012

[Pharo-project] InnoSetup was: RE: One-click Windows comment

by Schwab,Wilhelm K

InnoSetup is very non-intrusive and flexible. The resulting installers have options for silent and very silent installation, so yes, you probably can do what you want with them. Of course, Inno and resulting setups don't do anything (or not much anyway) that can't be done from any software, but it does a nice job of putting files in known locations, updating always, only if newer, running software, etc. As an aside, it's interesting to see the term "ActiveX" *still* being used. This was something that MS was trying to kill a LONG time ago. I'm increasingly glad that I started tuning them out and focusing on the simple and reliable vs. getting wound up in every new "technology" they tried to jam down my throat. I'm much happier on Linux. But I digress. One can use Smalltalk to generate inno scripts (for a good while, I used inno to repackage MS' monthly patch madness (you had to be there...). I had code that abstracted adding files to be copied, executed, etc., or at least something like that - haven't looked at it for a few years now. With the resulting script, run the inno compiler over it, and a nicely behaved installer emerges. It was preferable to what MS was doing at the time for various reasons. And of course, one can simply write a script using Inno's or another IDE, or just a plain text editor. I took that route for installing my Dolphin-deployed executables. The script generally did a "copy if newer" type of installation, which worked well. With every update, would compile the inno scripts and send the installers on their way. The various target machines could then be told to look for and install the updates. After some nail-biting, the machines generally would re-appear on the "grid." Every so often I'd have to unstick a box, but that was rare. The real fear was hardware failures, which was the most common reason that a machine failed to return to service. Didn't happen often, but it was bad when it did. Bill ________________________________________ From: pharo-project-bounces(a)lists.gforge.inria.fr [pharo-project-bounces(a)lists.gforge.inria.fr] on behalf of Igor Stasenko [siguctua(a)gmail.com] Sent: Saturday, April 21, 2012 5:43 AM To: Pharo-project(a)lists.gforge.inria.fr Subject: Re: [Pharo-project] One-click Windows comment On 21 April 2012 10:48, phil(a)highoctane.be <phil(a)highoctane.be> wrote: > Inno Setup would be a good thing to use for creating a Windows installer. > > http://www.jrsoftware.org/isinfo.php > > I use it regularly. Interested in a sample? > Since we're using automated builds, it is important that install packaging tools can run from command line and don't require any user interaction(s) in order to create an output. is inno installer allows to do that? > Phil > > > 2012/4/21 Marcus Denker <marcus.denker(a)inria.fr> >> >> >> On Apr 20, 2012, at 11:49 AM, Herby VojÄÃk wrote: >> >> > The ideal way would probably that .bat (or .cmd or .vbs) creates the >> > .lnk by some ActiveX magic, runs it and quits; subseuqntly .lnk could be >> > clicked directly. >> > >> We are planning of having installers for the three platforms... so we can >> make them more in the style that people expect. >> >> But this will be done step by step and slowly (like everything :-) >> >> For the meantime, I added a catch-all tracker for ideas how to improve the >> one-click: >> >> http://code.google.com/p/pharo/issues/detail?id=5643 >> >> >> -- >> Marcus Denker -- http://marcusdenker.de >> >> > > > > -- > Philippe Back > "Helping you hit the top 3 outcomes you really want to achieve" > > 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 > -- Best regards, Igor Stasenko.

April 21, 2012

Re: [Pharo-project] strange logic

by StÃ©phane Ducasse

enablementSelector: aSelector enablementSelector := aSelector isBlock ifTrue: [aSelector copyForSaving] ifFalse: [aSelector] may be it would be good to get straight on the accepted parameter: because after we get this kind of logicâ¦.

April 21, 2012

Re: [Pharo-project] How can I do literate programming?

by StÃ©phane Ducasse

We need to improve on that level but this requires to fix the formatter and we do not cycle for that. Now we would love that. Guys have been working on a mark down parser so that class comments and help can be optional displayed nicely. Now the first thing is to write class and methods comments. > Hello all, > > having a very short memory I need to write quite a lot of comments. I loved > the way to do this in Haskell, where you essentially write a LaTeX document > which happens to contains some code snippets and which still is an executable > program. So in Smalltalk you can write full expressions in comments and class example methods. > > In squeak I used to write "pretty" class comments, using sections with titles > in bold fonts and hypertext links where another class or method was referenced > in the text. > > In Pharo and in the latest squeaks this no longer seems to work. Bold fond > silently disappears when saving the comment, hypertext seems to be gone > entirely in Pharo. > > Comments in methods are moved aound by the configurable formatter, i.e. I > found no way to place comments BETWEEN statements. > > Now I certaninly don't want to complain. But I would be happy if someone could > give me some advice on how to comment code in a stylish way. > > -- > Martin >

April 21, 2012

[Pharo-project] How can I do literate programming?

by Martin Drautzburg

Hello all, having a very short memory I need to write quite a lot of comments. I loved the way to do this in Haskell, where you essentially write a LaTeX document which happens to contains some code snippets and which still is an executable program. In squeak I used to write "pretty" class comments, using sections with titles in bold fonts and hypertext links where another class or method was referenced in the text. In Pharo and in the latest squeaks this no longer seems to work. Bold fond silently disappears when saving the comment, hypertext seems to be gone entirely in Pharo. Comments in methods are moved aound by the configurable formatter, i.e. I found no way to place comments BETWEEN statements. Now I certaninly don't want to complain. But I would be happy if someone could give me some advice on how to comment code in a stylish way. -- Martin

April 21, 2012