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

Re: [Pharo-project] "Do not feed the trolls"

by StÃ©phane Ducasse

Thanks a lot Christophe! I love your trust :) It puts such responsibility on us but we accept it and we will do our best to make sure you can deliver robust and cool applications. For your information, Pinesoft's CEO told me that they are full of Pharo projects - a great news. Stef > It is not up to me whether our CMSBOX is an important project in a general > context. But to answer your question: almost 500 customers in Switzerland > are using this Pharo-based application until now and for them, this > application and the underlying very stable technology with ongoing > development IS important. Besides, we at netstyle.ch created several > business critical web-based applications, from automatic inventory > management to a large CRM for a Swiss insurance broker. Although you > probably never heard of these applications I guess we are nevertheless an > important component of the Pharo eco system and we use it because we believe > it to be the right way of the ongoing development. > > Although there is always room for improvement, the migration to Pharo was > and is an important step within our company strategy. On this basis, we > built a company from scratch with more than 10 employees whereof five > software engineers are doing nothing else than working with the Pharo IDE. > It is due to the ongoing development of a small community that our personal > "success" became possible. Could not be more thankful for that! Keep the > work going! > > To now improve whatever needs to improved in your opinion, one should not > issue orders to or offend the community but to collaborate constructively on > suggestions. Honestly, this was not our strength within the past months too, > but we are willing to support this lively community as much as we can in the > future although we are not Wikipedia or similar. I am convinced that a > constructive and appreciative collaboration is the only way to improve > Pharo so please consider my words when writing your next message. > > Cheers, > > Chris Wysseier > Founder and CEO of netstyle.ch and cmsbox > > P.S: Also wanted to say hi to everybody as I am new on the list. Be aware > that I am not an *experienced* software engineer (let's say "anymore" ;) ) > and therefore will not be able to discuss on technical strategies or tasks. > But I would like to know what is going on and to show support and presence > to all of you. > > -- > View this message in context: http://forum.world.st/Do-not-feed-the-trolls-tp4405636p4406708.html > Sent from the Pharo Smalltalk mailing list archive at Nabble.com. >

Feb. 21, 2012

Re: [Pharo-project] Nautilus Refactorings

by StÃ©phane Ducasse

Ben I like the suggestion of mariano and luc. I discussed with them. - opening by default on packages is better because it does not force you to display the full hierarchy. - could you try to see if you can just display the packages and hierarchy without swapping them? Stef > > One last thing: when you type the name of a class somewhere and you do a cmd+b I think it would be much better if Nautilus open the "normal" browser and not "hierarchy" one. > > having cmd+h for hierarchy would be great :) > > +9999999 > > to open a hierarchy when you do cmd+b is bad for me beacuse: > > 1) 99% of the times I browse that way, I want to see that class, and 99% also the rest of the classes of that package, not the hierarchy > 2) I am used to that since the last..mm 3 years ;) > 3) Because it seems the hierarchy windows is slow in big cases [Object browse] timeToRun 19225 so it took me 20 seconds to open the window ;) > > Because of that, I much prefer cmd + to for a normal browse. I don't pretend all agree, so I would be very happy when you have the setting. Please let me know so that I can set it :) > > Cheers! > > > > Ben we should produce a configuration because this is important. > What you can do is at least load the baseline and snapshot only when you will release on version for 1.4 and 1.5 > > > > > > On Mon, Feb 20, 2012 at 8:38 PM, Benjamin <benjamin.vanryseghem.pharo(a)gmail.com> wrote: > > To be honest, if it works, I do not want to lose my time on Configuration > > > > > > It shouldn't be done by hand, but generated by a tool :) > > > > > > Ben > > > > On Feb 20, 2012, at 8:06 PM, Mariano Martinez Peck wrote: > > > >> > >> > >> On Tue, Feb 14, 2012 at 6:39 PM, Benjamin <benjamin.vanryseghem.pharo(a)gmail.com> wrote: > >> Hello guys, > >> > >> I know a some of you are waiting for the integration of the RB engine into Nautilus. > >> > >> Some of them are already implemented, so if you wanna test them, here is the gofer script > >> > >> Gofer new > >> url: 'http://ss3.gemstone.com/ss/Nautilus'; > >> package: 'ConfigurationOfNautilus'; > >> load. > >> > >> (Smalltalk at: #ConfigurationOfNautilusRefactoring) perform: #loadDefault > >> > >> > >> > >> > >> So...that's the way of loading Nautilus with refactorings? Wouldn't it be better to include RB dependencis in ConfigurationOfNautilus and have a specific group there. So you can do for example > >> > >> ConfigurationOfNautilus project stableVersion load: 'WithRefactorings'. > >> > >> ? > >> > >> BTW, I checked the ConfigurationOfNautilusRefactoring and it looks you miss a critical feature of Metacello: project references :) From ConfigurationOfNautilusRefactoring you can directly depend on ConfigurationOfRB or ConfigurationOfNautilus rather than duplicating everything in there. Don't tell Stef you didn't read the chapter hahahaha > >> > >> > >> > >> It loads everything. > >> > >> For now I have only done some refactorings for classes, so if you popup the menu on a class, a new item named Refactoring should appear :) > >> > >> Moreover, the 'rename class' method is replaced by the RB one :) > >> > >> > >> Thanks in advance for your feedback, and if you want to participate, you are more than welcome :) > >> > >> > >> Ben > >> > >> > >> > >> -- > >> Mariano > >> http://marianopeck.wordpress.com > >> > > > > > > > > > > -- > > Mariano > > http://marianopeck.wordpress.com > > > > > > > > -- > Mariano > http://marianopeck.wordpress.com >

Feb. 21, 2012

Re: [Pharo-project] Nautilus Refactorings

by StÃ©phane Ducasse

On Feb 21, 2012, at 1:59 PM, Tudor Girba wrote: > Hi Ben, > > He means that ConfigurationOfNautilus should specify all dependencies. > For example, it should specify RPackage as dependency. yes that for example. we should sit when I'm not running like hell this week > > Cheers, > Doru > > > > On Tue, Feb 21, 2012 at 10:06 AM, Benjamin > <benjamin.vanryseghem.pharo(a)gmail.com> wrote: >> Something like ConfigurationOfNautilus ? >> >> >> I do not get it >> >> >> Ben >> >> On Feb 21, 2012, at 9:53 AM, StÃ©phane Ducasse wrote: >> >>> >>> On Feb 21, 2012, at 9:46 AM, Benjamin wrote: >>> >>>> A configuration for ? >>> >>> nautilus so that it loads automatically everything it needs in a core 1.4 >>> >>>> >>>> >>>> Ben >>>> >>>> On Feb 21, 2012, at 8:21 AM, StÃ©phane Ducasse wrote: >>>> >>>>> >>>>> On Feb 20, 2012, at 11:33 PM, Mariano Martinez Peck wrote: >>>>> >>>>>> One last thing: when you type the name of a class somewhere and you do a cmd+b I think it would be much better if Nautilus open the "normal" browser and not "hierarchy" one. >>>>> >>>>> having cmd+h for hierarchy would be great :) >>>>> >>>>> Ben we should produce a configuration because this is important. >>>>> What you can do is at least load the baseline and snapshot only when you will release on version for 1.4 and 1.5 >>>>>> >>>>>> >>>>>> On Mon, Feb 20, 2012 at 8:38 PM, Benjamin <benjamin.vanryseghem.pharo(a)gmail.com> wrote: >>>>>> To be honest, if it works, I do not want to lose my time on Configuration >>>>>> >>>>>> >>>>>> It shouldn't be done by hand, but generated by a tool :) >>>>>> >>>>>> >>>>>> Ben >>>>>> >>>>>> On Feb 20, 2012, at 8:06 PM, Mariano Martinez Peck wrote: >>>>>> >>>>>>> >>>>>>> >>>>>>> On Tue, Feb 14, 2012 at 6:39 PM, Benjamin <benjamin.vanryseghem.pharo(a)gmail.com> wrote: >>>>>>> Hello guys, >>>>>>> >>>>>>> I know a some of you are waiting for the integration of the RB engine into Nautilus. >>>>>>> >>>>>>> Some of them are already implemented, so if you wanna test them, here is the gofer script >>>>>>> >>>>>>> Gofer new >>>>>>> url: 'http://ss3.gemstone.com/ss/Nautilus'; >>>>>>> package: 'ConfigurationOfNautilus'; >>>>>>> load. >>>>>>> >>>>>>> (Smalltalk at: #ConfigurationOfNautilusRefactoring) perform: #loadDefault >>>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>>>> So...that's the way of loading Nautilus with refactorings? Wouldn't it be better to include RB dependencis in ConfigurationOfNautilus and have a specific group there. So you can do for example >>>>>>> >>>>>>> ConfigurationOfNautilus project stableVersion load: 'WithRefactorings'. >>>>>>> >>>>>>> ? >>>>>>> >>>>>>> BTW, I checked the ConfigurationOfNautilusRefactoring and it looks you miss a critical feature of Metacello: project references :) From ConfigurationOfNautilusRefactoring you can directly depend on ConfigurationOfRB or ConfigurationOfNautilus rather than duplicating everything in there. Don't tell Stef you didn't read the chapter hahahaha >>>>>>> >>>>>>> >>>>>>> >>>>>>> It loads everything. >>>>>>> >>>>>>> For now I have only done some refactorings for classes, so if you popup the menu on a class, a new item named Refactoring should appear :) >>>>>>> >>>>>>> Moreover, the 'rename class' method is replaced by the RB one :) >>>>>>> >>>>>>> >>>>>>> Thanks in advance for your feedback, and if you want to participate, you are more than welcome :) >>>>>>> >>>>>>> >>>>>>> Ben >>>>>>> >>>>>>> >>>>>>> >>>>>>> -- >>>>>>> Mariano >>>>>>> http://marianopeck.wordpress.com >>>>>>> >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> -- >>>>>> Mariano >>>>>> http://marianopeck.wordpress.com >>>>>> >>>>> >>>>> >>>> >>>> >>> >>> >> >> > > > > -- > www.tudorgirba.com > > "Every thing has its own flow" >

Feb. 21, 2012

Re: [Pharo-project] How to cook Mr Popatoe Process?

by Jimmie Houchin

On 2/21/2012 2:37 AM, Noury Bouraqadi wrote: > > On 21 fÃ©vr. 2012, at 07:44, StÃ©phane Ducasse wrote: > >> Hi guys >> >> we have practically two solutions to cook a troll: >> - either I remove his email address from the list (2 min) >> a bit rude and unfriendly but this is a troll after all ;) >> > +1 Everybody is fed up with spams. > > Noury > Unfortunately, this may not solve the problem unless we play the game of "Whack a Troll". We would have to delete his email address every time he creates a new one. I think ultimately we as a community need to simply learn not to feed the trolls. When they are trolling, ignore them. If he actually provides something usefully such as a bug report with reproducible actions, or a patch, an appropriate approved actionable suggestion. Then we can treat those emails with the respect due such emails. But if he simply continues to Parrot the same old thing. Plonk! Ignore. Or as Stef said, >> - or we all create a nice rule: as Read >> >> I suggest you all to do 2 and keep you good energy for helping and >> not arguing endless with Mr Potato Process. Jimmie

Feb. 21, 2012

Re: [Pharo-project] Package TextLint for Debian?

by Jorge Ressia

Hi Thomas, I agree with Damien. If you need a hand I am willing to help. Cheers, On Tue, Feb 21, 2012 at 9:47 AM, Damien Cassou <damien.cassou(a)gmail.com>wrote: > Dear Thomas, > > On Tue, Feb 21, 2012 at 9:40 AM, Thomas Koch <thomas(a)koch.ro> wrote: > > I've just found textlint for Emacs[1] and wondered, whether it could be > > packaged for Debian (or any other Distro). Unfortunately the Git repo > does not > > contain the source code of textlint but only several binaries. So I > wondered > > whether all the textlint logic happens locally or whether the plugin > contacts > > a textlint server? > > > > [1] https://github.com/DamienCassou/textlint > > > > Even if the textlint source code would be available, it seems that > textlint > > depends on Pharo, which also isn't packaged for Debian. Is that right? I > don't > > know anything about Smalltalk, so I can't package Pharo myself, but I'd > be > > happy to help if somebody of you would be interested to package it. > > I'm really interested in getting TextLint distributed as a debian and > ubuntu packages. If you can lead this, I will help you as much as I > can. I don't think you need a package for Pharo itself. Just try to > create a package from what is on github: the pharo image + VM + bash > and emacs scripts. You can send me private emails. > > > -- > Damien Cassou > http://damiencassou.seasidehosting.st > > "Lambdas are relegated to relative obscurity until Java makes them > popular by not having them." James Iry > > -- Jorge Ressia www.jorgeressia.com

Feb. 21, 2012

Re: [Pharo-project] Why do we use Pharo? a.k.a. rethinking Pharo marketing

by Esteban Lorenzano

Hi, while I see many of your points, I disagree with some of them: We have a tool that allow us to deliver complete and powerful applications in 2-6 weeks (and even less): Pharo. Right now my small company (of myself :) ) is finishing (bah, doing some change requests) for an application for managing thousands of accounts and data. Application was made on Glamour+Magritte+MongoTalk+Some persistence and reporting tool made by myself, based on magritte. Customer is really happy, because now they can obtain reports of any kind instead the shit they have before (made in Delphi, but there is not a problem of delphi itself but the programmer using it) The time to develop from scratch this? Four weeks-man. And if we took off all the work I made to make MongoTalk work, reporting and some personal adjusts to glamour, it would be less than three weeks (is important to know, because that work does not need to be repeated...). (btw, I could do the same for the web, using Seaside) Also, we are working (not yet there, but almost), to accomplish several of the issues not completely done for enterprise: 1) we have DBXTalk, to RDBMS. Is slow, but that's a problem of FFI that we think it will be solved soon (with Threaded FFI) 1.1) we also have a growing set of packages for working with NoSQL (Mongo, Riak, Tokio) 2) we have UI components and powerful frameworks like Glamour, but I agree there is a lot more work needed there 3) we have ways to talk to outside world (FFI, NB)... but also I agree more work is needed there (for instance, I would like to embed pharo into other applications, not just call libraries from pharo) 3.1) we also have SOAP support (never tried, not sure it's status) 4) For remote: we have Ubiquitalk 5) authentication: not sure what you mean but we have cryptography and work for secure connections is on the way (with Zodiac) 6) reporting. Ok, not much here... but there is HPDF package and I will release my HPDFReport and CSVReport classes soon (they are very easy classes anyway, don't expect too much :) for... bach modules... not sure what are you meaning with that... So yes, many of this issues are not really finished or stable (and that's why I see your concerns)... but just think where we were last year. I'm really optimistic about the future. What we really, really need is a book: "Pharo for the Enterprise". Something like the JEE specification, explaining how enterprise customers can use Pharo for their business, right now is a bit confusing, I admit that. Other concern (but this is a no-point for discussion, I just wanted to say it :P): Java and C# are not easy at all to learn. They are hard and ugly and new programmers spend years learning it (if they ever learn it completely)... also it is uncomfortable for starters (look how much you need to have a simple "hello, world" there). You think preparing an image is hard? give a new programer a maven script and ask him to build an application with that. Or worst, ask him to add a new dependency... Industry uses those languages and not Smalltalk for many-many reasons... but this is not one of them. (Also... there is a lot of big industries still developing on COBOL. sadly) best, Esteban El 21/02/2012, a las 2:16p.m., Krishsmalltalk escribiÃ³: > > The choice is ours to make and define. > > Yes enterprise is boring for those wanting life on the edge: tech or anything alike... I would love to jump over to the other side... > > Its like the freedom of being Picasso vs being a commercial artist in a studio. > > Enterprise typically will use software that is stable and capable of being easily usable by newbies. It cares less for the edge of technology, which can be achieved by just a minimal subset of the developer community. Yes no new larger enterprise stuff will happen in COBOL or ilk... But java and c# have proven to be easier to train in freshers, capable of adapting and calling any of the new technology if required etc. The very facts we tout as major winners in smalltalk is actually dimly viewed by the bulk of managers in enterprise: dynamic typing to begin with. They love the fact that the junior dev cannot commit many blunders that smalltalk can carry. Image based runtime, packaging , source code mgmt in st, etc are amongst stuff few comprehend well. Also we lack any decent libraries to for many a task, including reporting, DBMS , soap and many other stuff which when required only the ingenous smalltalker can cobble it, not our general developer base. > > I can be specific with tons of example, that we need to push a clean rock stable kernel and winning platform like rails, that make it hugely possible for a small firm to push out completed apps in 2 to 6 weeks. It's like the construction industry, be able to pick all components of the software: authentication, DBMS / ORM , reporting, UI components, messaging, interfaces to outside world viz soap, FFI , Remote , batch modules, ... One can push in a long list... > > All this is important only if we want to succeed statistically and that enriches the community to then indulge even more in research, it's symbiotic in that way. The more success you get in the industry the more funding there will be to do fundamental research eventually, the more happier all levels of smalltalkers from the expert to the beginners will be. > > Modified Quote from Kung Fu panda: master oogway "enterprise .. No enterprise .. Who cares", what we do care for is the eventual adoption of smalltalk as an equal partner in the landscape of languages, platforms in the world, gives enough money for everyone around. Unequal if it must eventually.. > > > > On Feb 21, 2012, at 1:02 PM, Marcus Denker <marcus.denker(a)inria.fr> wrote: > >> >> On Feb 21, 2012, at 3:11 AM, Krishsmalltalk wrote: >> >>> Yes, >>> >>> Do not emphasize on one or few keywords. Malleable is better than plastic (can be off putting for any environment conscious) >>> >>> Adopt the Pharo motto: >>> >>> "Pharo shall be one of the best enterprise platform in 3 years." >> >> >> Isn't "Enterprise Plattform" another word for "boring and complicated" ?? >> >> Marcus >> >> >> -- >> Marcus Denker -- http://marcusdenker.de >> >> >

Feb. 21, 2012

Re: [Pharo-project] Netstyle experience

by Krishsmalltalk

Having talked about this many times earlier, i have now codified for the system that has impediment to using toggle breakpoint, halts in polling methods, kernel , system classes etc.. saving the file after any runtime execution of Pharo.. I do have personally a stable env now with 1.3 on ubuntu but on windows though a big improvement now, it still stack overflows more due to a basic mistake no experienced small talker commits viz putting a halt in a polling method, System classes etc.. , modifying a code that is invoked every millisecond in say mouseDown: or the ilk. What is needed is a very simple Pharo kernel, that has this safe mode on, which disallows for a fresher the mistakes which bites.... Not Feb 21, 2012, at 8:24 PM, Sven Van Caekenberghe <sven(a)beta9.be> wrote: > Hi Dave, > > On 21 Feb 2012, at 15:34, Dave Mason wrote: > >> This is the biggest thing for me. I'm trying to get my Analysis of Algorithms students to recognise and value a powerful environment like Pharo Smalltalk, but some of them are experiencing so many crashes that they've taken to entering their code in Notepad and then pasting it into Pharo!!! This is not a strong advert for Smalltalk or Pharo. >> >> I suspect many of these crashes are stack overflows. I *thought* that Squeak/Pharo was supposed to recognise stack overflow, but I just wrote a method: >> crash >> ^self crash + 4 >> and my image grew to 500+MB and then crashed. This is with Pharo 1.3 one-click. With 1.1 one-click, I *do* get the Low space window popping up at 516MB (although I don't appear able to do much... it doesn't even seem to quit-no-save); I eventually did a force-quit. This is all on Mac OS X 10.6.8. >> The 1.3 says: >> VM: Mac OS - intel - 1068 - Croquet Closure Cog VM >> [CoInterpreter VMMaker-oscog- IgorStasenko.123] 21.0 >> Image: Pharo1.3 [Latest update: #13307] >> >> How can I get more stability? Is it time to move to 1.4? Is there something I or my students can do to help? >> >> Thanks for everybody's efforts! >> >> ../Dave > > Out of the box experience is indeed very important. > > There was an important fix related to stopping infinite loops recently, I am sure the latest 1.4 includes it. > > Sven > >

Feb. 21, 2012

Re: [Pharo-project] Why do we use Pharo? a.k.a. rethinking Pharo marketing

by Krishsmalltalk

The choice is ours to make and define. Yes enterprise is boring for those wanting life on the edge: tech or anything alike... I would love to jump over to the other side... Its like the freedom of being Picasso vs being a commercial artist in a studio. Enterprise typically will use software that is stable and capable of being easily usable by newbies. It cares less for the edge of technology, which can be achieved by just a minimal subset of the developer community. Yes no new larger enterprise stuff will happen in COBOL or ilk... But java and c# have proven to be easier to train in freshers, capable of adapting and calling any of the new technology if required etc. The very facts we tout as major winners in smalltalk is actually dimly viewed by the bulk of managers in enterprise: dynamic typing to begin with. They love the fact that the junior dev cannot commit many blunders that smalltalk can carry. Image based runtime, packaging , source code mgmt in st, etc are amongst stuff few comprehend well. Also we lack any decent libraries to for many a task, including reporting, DBMS , soap and many other stuff which when required only the ingenous smalltalker can cobble it, not our general developer base. I can be specific with tons of example, that we need to push a clean rock stable kernel and winning platform like rails, that make it hugely possible for a small firm to push out completed apps in 2 to 6 weeks. It's like the construction industry, be able to pick all components of the software: authentication, DBMS / ORM , reporting, UI components, messaging, interfaces to outside world viz soap, FFI , Remote , batch modules, ... One can push in a long list... All this is important only if we want to succeed statistically and that enriches the community to then indulge even more in research, it's symbiotic in that way. The more success you get in the industry the more funding there will be to do fundamental research eventually, the more happier all levels of smalltalkers from the expert to the beginners will be. Modified Quote from Kung Fu panda: master oogway "enterprise .. No enterprise .. Who cares", what we do care for is the eventual adoption of smalltalk as an equal partner in the landscape of languages, platforms in the world, gives enough money for everyone around. Unequal if it must eventually.. On Feb 21, 2012, at 1:02 PM, Marcus Denker <marcus.denker(a)inria.fr> wrote: > > On Feb 21, 2012, at 3:11 AM, Krishsmalltalk wrote: > >> Yes, >> >> Do not emphasize on one or few keywords. Malleable is better than plastic (can be off putting for any environment conscious) >> >> Adopt the Pharo motto: >> >> "Pharo shall be one of the best enterprise platform in 3 years." > > > Isn't "Enterprise Plattform" another word for "boring and complicated" ?? > > Marcus > > > -- > Marcus Denker -- http://marcusdenker.de > >

Feb. 21, 2012

Re: [Pharo-project] syntax error in StartupLoader script

by Camillo Bruni

check the latest slice, I think I fixed the debugger issue. SLICE-Issue-5290-Cleanup-Startup-Error-Handling-CamilloBruni.7 best cami On 2012-02-21, at 13:10, Camillo Bruni wrote: > thanks, I will have a look at it > > > On 2012-02-21, at 06:10, Ben Coman wrote: >> >> I loaded SLICE-Issue-5290 onto a freshly unzipped Pharo-1.4-14341 + cogwin_r2522.zip >> This caused a problem with the debugger as follows. In workspace I execute... >> self halt. >> x := 1. >> y := 2. >> z := x+y. >> >> I click the second line of the stack "UndefinedObject>>DoIt" which now shows "halt" highlighted. >> I click <Into> and " := 1" is now highlighted. >> I click <Into> again and nothing happens. There is no further response from any of the debugger buttons, although <Full Stack> causes the window to flash once. >> >> I can temporarily fix this misbehaviour by forcing #isPostMortem to always return false. >> >> >> Camillo Bruni wrote: >>> you can try to load >>> SLICE-Issue-5290-Cleanup-Startup-Error-Handling-CamilloBruni.2 >>> >>> from the pharoInbox which will significantly improve the error responses during startup. >>> >>> best >>> cami >>> >>> >>> On 2012-02-18, at 00:19, Ben Coman wrote: >>> >>> >>>> When using the following StartupLoader script >>>> >>>> StartupLoader default executeAtomicItems: { >>>> StartupAction name: 'Shared Package Cache' code: [ MCCacheRepository default directory setPathName: 'C:\Apps\Smalltalk\!Shared-Package-Cache'.]. >>>> }. >>>> >>>> I was getting a "syntax error in startup" that I could not identify (is it obvious to others?) >>>> This was due to the debugger not opening at the point of error since SmalltalkImage>>executeDeferredStartupActions catches the errors. So I hacked this as follows... >>>> deferredStartupActions do: [:each | >>>> [ each cull: resuming ] value. >>>> "btcdebug >>>> on: Halt >>>> do: [ :err| >>>> Smalltalk >>>> logError: '==== Startup Error: ', err description >>>> inContext: err signalerContext. >>>> errors add: err]"]. >>>> >>>> and then after a quit and restart the dialog "Syntax error: Unmatched string quote" appeared with... >>>> StartupLoader default executeAtomicItems: { >>>> StartupAction name: 'Shared Package Cache' code: [ MCCacheRepository default directory setPathName: _Unmatched string quote_ -> 'C:\Apps\Smalltalk\ >>>> >>>> So after removing "!" from the folder name the syntax error went away. >>>> >>>> So now a few questions... >>>> >>>> 1. Is there a simpler way to debug the startup scripts that I have missed? >>>> If not, can a StartupLoader>>test or similar be implemented which runs the same as at image start time but without the error catching in executeDeferredStartupActions, rather than needing to hack executeDeferredStartupActions as I did. >>>> >>>> 2. Is it reasonable to set a shared package cache in the startup like I have? Or are there some obvious problems with this that I don't see? >>>> I would like to avoid downloading all packages every time unzip a fresh image for testing. >>>> >>>> 3. Is the invalid "!" fixable (though of course I've worked around it)? >>>> >>>> I'll open any required bug entries. This was with Pharo-14333 / cogwin_r2522.zip / Windows 7. >>>> >>>> cheers, -ben >>>> >>>> >>>> >>>> >>> >>> >>> >>> >> >> >

Feb. 21, 2012

Re: [Pharo-project] Proposal: coverage tests

by Sven Van Caekenberghe

Pavel, This is incredible ! This is why we love Smalltalk, in what other language could you write it like that ? I love it. Sven On 21 Feb 2012, at 16:03, Pavel Krivanek wrote: > Hi, > > what about to intrroduce tests that will test coverage of tests? Some > improved version of the code below. Are there some tool for Smalltalk > showing uncovered lines of code? There are such tools in Java world > :-) One little problem is that this code do not work on Linux because > of a VM error that cause test failures on CI server too. > > Cheers, > Pavel > > | class test trace classTrace result | > > class := Time. > test := TimeTest. > > trace := Dictionary new. > > thisContext runSimulated: [test suite run] contextAtEachStep: [ :current | > | cls sel methods bytecodes | > cls := current method methodClass. > sel := current method selector. > methods := trace at: cls ifAbsentPut: Dictionary new. > bytecodes := methods at: sel ifAbsentPut: Set new. > bytecodes add: current pc.] > > trace . > > result := String streamContents: [:s | > > classTrace := trace at: class ifAbsent: nil. > classTrace > ifNil: [ > s nextPutAll: 'Class ', class name, ' not called'; cr. ] > ifNotNil: [ > s nextPutAll: class name; cr. > class selectorsDo: [:sel | > | m usedBytecodes allBytecodes | > m := class >> sel. > usedBytecodes := (classTrace at: sel ifAbsent: Set new) size. > allBytecodes := 0. > m symbolicLinesDo: [:pc :lineForPC | allBytecodes := allBytecodes + 1 ].. > s tab; nextPutAll: sel; nextPutAll: ' - '; nextPutAll: > (usedBytecodes / allBytecodes * 100) asFloat asString; nextPutAll: > '%'; cr. ]]. > ]. > result.. >

Feb. 21, 2012