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

[Pharo-project] Fwd: Are Objects really hard?

by Janko MivÅ¡ek

-------- Prvotno sporoÄilo -------- Datum: Sun, 12 Feb 2012 09:55:08 +0000 Od: Bob Calco <bob.calco(a)ellelministries.org> Hi everyone, On Feb 11, 2012, at 1:58 PM, Janko MivÅ¡ek wrote: > Hi Stef, > > S, stephane ducasse piÅ¡e: > >> Frankly I do not care about what other people are thinking. >> OOP is a success look at Java, C#. >> >> Now let us keep our energy to build better Smalltalks. > Well, after hard work it is good from time to time to make a > retrospection and let our thoughts to think broader, to look from > a distance to our work. To see the forest and not just trees. > > So such debate from time to time is certainly refreshing and needed, > specially if is started from a outsider's perspective.Every wise man > listen to the opinion of others. Well, of course wisely :) > > In this case I see wise thinking about weaknesses of OO, Smalltalk > and how to overcome it by better "best practices". For instance, the > newcommers are asking where to find a guidelines for modeling OO > models in pure OO way. In this guidelines we can emphasise above > mentioned best practices, then author's claim that "no one really > understands to this day how to do them right" won't be valid > much anymore. > > Best regards > Janko As a certified noob to Smalltalk hailing from the Ruby, Delphi, .NET and Java worlds where RDBMs rule, I must say I'm amused how anyone can say Smalltalk has *failed.* I had a funny experience when doing mainly Rails development a couple years ago. I had an interview at a PHP shop, where the main technical guru there insisted that Ruby/Rails had "failed" and wanted me to tell him why. I told him, actually, there were bugs in the early ActiveRecord API that exposed Rails to scalability problems, not the framework or design concept itself. I pointed out that if Rails failed, so did Symfony, and Django, ASP.NET <http://ASP.NET> MVC, and countless other copy-cat frameworks in other languages/platforms. If one measure of failure is industry adoption, then there was no way Rails failed. And where did MVC come from? Well, we know that answer. The whole discussion is pointless. Engineers sometimes get the idea that because they or a project they were on failed that happened to be using a particular tool, the tool is a failure. At bottom, this guy was bitten by Rails and so he had a kind of vendetta against it. As if PHP is the way, the truth and the light. By the way they offered me a job anyway, but I didn't take it. :) Some tools make certain things harder or easier. I find Smalltalk liberating as it makes it easy to decompose hard domain problems sans all the syntactic cruft of the so-called modern languages -- which, funny enough, are constantly trying to find new and more complicated ways to express what Smalltalk has always done simply: generic collections, metadata, reflection, etc. So, I'm going against the grain and getting into Smalltalk. I have been a bit disappointed by the state of certain libraries and the relatively small community, but I enjoy the language and feel more productive in it, so...that's what matters to me. - Bob -- Janko MivÅ¡ek Aida/Web Smalltalk Web Application Server http://www.aidaweb.si

Feb. 12, 2012

Re: [Pharo-project] Are Objects really hard?

by Milan Mimica

2012/2/11 Janko MivÅ¡ek <janko.mivsek(a)eranova.si> > > In this case I see a wise thinking about weaknesses of OO and Smalltalk > and how to overcome it by better "best practices". For instance, the > newcommers are asking where to find a guidelines for modeling OO domain > models in pure OO way. In this guidelines we can emphasise above > mentioned best practices, then author's claim that "no one really > understands to this day how to do them right" won't be valid much anymore. > There are many books written on OO domain modeling and there is so much knowledge floating around. The claim that no one understands OO it just so stupid that it's not worth citing it. I for one do know how do it right, in any language. You don't even need an OO laguage. It has nothing do to with Smalltalk. It's about how you model real-life problems in code. I find it easy, straightforward and natural. -- Milan Mimica http://sparklet.sf.net

Feb. 12, 2012

Re: [Pharo-project] Regular expression

by Lukas Renggli

VB-Regexp does not support any of the following common regular expression features: lookahead, lookbehind, recursion, multiplicity ranges, non-greedy quantifier, shy groups, directives, conditionals, named captures, back references, and comments. You can (conceptually) do all this with PetitParser, but then you are not writing a regular expression anymore but a (PEG) parser. There was once a plugin that called an external regular expression library (I think it used the library Perl is using) that supported the features you are looking for. Lukas On 11 February 2012 23:27, <recursive68(a)gmail.com> wrote: > Hi, > > Is it possible to use the VB-Regexp package to construct a single regular > expression that ensures a string is alphanumeric and contains both at least > one number and alphabetic character. > > Looking at the regexp package and documentation it doesn't appear so. In > javascript or perl you use something like this with lookahead assertions: > > ^(?=.*[a-zA-Z])(?=.*[0-9]).*$ > > I have something working using multiple regexp's but wondered if a more > typical regexp was possible using the package ? > > Thanks > > -- Lukas Renggli www.lukas-renggli.ch

Feb. 12, 2012

Re: [Pharo-project] Regular expression

by StÃ©phane Ducasse

http://rmod.lille.inria.fr/pbe2/ https://gforge.inria.fr/scm/viewvc.php/*checkout*/PharoByExampleTwo-Eng/Reg… I would be surprised that we call a regexp package something that cannot handle Stef On Feb 11, 2012, at 11:27 PM, recursive68(a)gmail.com wrote: > Hi, > > Is it possible to use the VB-Regexp package to construct a single regular expression that ensures a string is alphanumeric and contains both at least one number and alphabetic character. > > Looking at the regexp package and documentation it doesn't appear so. In javascript or perl you use something like this with lookahead assertions: > > ^(?=.*[a-zA-Z])(?=.*[0-9]).*$ > > I have something working using multiple regexp's but wondered if a more typical regexp was possible using the package ? > > Thanks > > > >

Feb. 12, 2012

Re: [Pharo-project] Need help with OpenGL visual creation on linux

by Igor Stasenko

ok, so If no conforming visual exists, NULL is returned. It is hard to imagine, that these attributes not supported {GLX_RGBA. GLX_DEPTH_SIZE. 24. GLX_DOUBLEBUFFER. 0} the code seems to be fine. This is what i found: ------- http://www.opengl.org/wiki/Programming_OpenGL_in_Linux:_GLX_and_Xlib If glXChooseVisual returns with success, the visual's id will be output. If NULL is returned, there is no visual that fulfills your needs. In that case, check the output of glxinfo again. Maybe you have to use a different depth buffer size (GLX_DEPTH_SIZE, 16 instead of GLX_DEPTH_SIZE, 24), or you could even have to remove the GLX_DEPTH_SIZE (or the GLX_DOUBLEBUFFER) entry. This should be considered especially if you want to create programs not only for your computer, but for other ones: You should code your program in a way that it can check a list of different combinations of visual attributes, because the capabilities depend heavily on the hardware. ----- They also using vi = glXChooseVisual(dpy, 0, att); second argument = 0. not #defaultScreen. On 10 February 2012 19:59, Javier PimÃ¡s <elpochodelagente(a)gmail.com> wrote: > yes, notice that it is in a different > repo:Â http://www.squeaksource.com/NBXLib > > > On Fri, Feb 10, 2012 at 3:53 PM, Igor Stasenko <siguctua(a)gmail.com> wrote: >> >> On 10 February 2012 19:48, Javier PimÃ¡s <elpochodelagente(a)gmail.com> >> wrote: >> > To load NBOpenGL-X package you need to load NBXLib-core first >> > >> i don't see it. did you uploaded it? >> >> > >> > On Fri, Feb 10, 2012 at 3:39 PM, Igor Stasenko <siguctua(a)gmail.com> >> > wrote: >> >> >> >> Btw, javier i cannot load your package to check what you did: >> >> >> >> This package depends on the following classes: >> >> Â NBXLibConstants >> >> You must resolve these dependencies before you will be able to load >> >> these definitions: >> >> Â NBGLXContextDriver >> >> Â supportsCurrentPlatform >> >> Â createContext: >> >> >> >> >> >> >> >> >> >> On 10 February 2012 19:34, Igor Stasenko <siguctua(a)gmail.com> wrote: >> >> > On 10 February 2012 19:19, Lawson English <lenglish5(a)cox.net> wrote: >> >> >> Is that code even ready for consumption? >> >> >> >> >> >> I was told that until there is a new ConfigurationOfNBOpenGL ready, >> >> >> to >> >> >> not >> >> >> load the new packages. >> >> >> >> >> > >> >> > you could try :) >> >> > >> >> > but this is what i working on now (slowly integrating the parts of >> >> > Javier's code, >> >> > because not everything which he did i like ;) >> >> > >> >> > >> >> >> L. >> >> >> >> >> >> >> >> >> On 2/10/12 10:12 AM, Javier PimÃ¡s wrote: >> >> >> >> >> >> I'm trying to create an OpenGL context on linux but glxChooseVisual >> >> >> fails. >> >> >> Maybe there's someone there experienced with this that can help. >> >> >> >> >> >> The code I'm trying is: >> >> >> >> >> >> display := NBXLibDisplay open. >> >> >> window := display defaultRootWindow. >> >> >> visualInfo := NBXLibVisualInfo fromPointer: (gl chooseVisual: >> >> >> display >> >> >> screen: display defaultScreen attributes: {GLX_RGBA. GLX_DEPTH_SIZE. >> >> >> 24. >> >> >> GLX_DOUBLEBUFFER. 0} asWordArray). >> >> >> ... >> >> >> >> >> >> but chooseVisual returns a null pointer. I even tried putting an >> >> >> array >> >> >> with >> >> >> only 0 on attributes but didn't work either (and these attributes >> >> >> should be >> >> >> supported). >> >> >> >> >> >> Any idea of what could be wrong? The code is available to test in >> >> >> squeaksource, you need nativeboost+NBXLib to try >> >> >> >> >> >> Cheers, >> >> >> Javier >> >> >> >> >> >> -- >> >> >> Lic. Javier PimÃ¡s >> >> >> Ciudad de Buenos Aires >> >> >> >> >> >> >> >> > >> >> > >> >> > >> >> > -- >> >> > Best regards, >> >> > Igor Stasenko. >> >> >> >> >> >> >> >> -- >> >> Best regards, >> >> Igor Stasenko. >> >> >> > >> > >> > >> > -- >> > Lic. Javier PimÃ¡s >> > Ciudad de Buenos Aires >> >> >> >> -- >> Best regards, >> Igor Stasenko. >> > > > > -- > Lic. Javier PimÃ¡s > Ciudad de Buenos Aires -- Best regards, Igor Stasenko.

Feb. 12, 2012

Re: [Pharo-project] load testing a database

by Ben Coman

Just thinking laterally (since I don't have the experience to apply to the analysis), a practical business solution is to throw hardware at the problem - particularly for a legacy application on old hardware. One site I worked at was having issues with the overnight posting of the days accounts taking upwards of 12 hours. The application had outgrown its initial design. This was brought down to 15 minutes by upgrading the RAM and also to Serial Attached SCSI RAID10. The other thing to consider is the decreasing cost of solid state storage (which I've heard of good stuff but haven't used myself yet) http://www.mcobject.com/in_memory_database - non-Oracle but good overview http://www.dba-oracle.com/t_ram_disk_performance.htm http://www.ramsan.com/files/oracle_performance_tuning_with_ssd.pdf - "The dramatic price/performance ratio of SSD is changing the way that Oracle databases are tunes. Sub-optimal Oracle databases no longer have to undergo expensive and time-consuming re-design. SSD technology is no competing head-on with Oracle consulting dervices." cheers, -ben Tudor Girba wrote: > Hi, > > On 11 Feb 2012, at 21:13, Philippe Marschall wrote: > > >> On 11.02.2012 19:30, Tudor Girba wrote: >> >>> Hi, >>> >>> Sorry for the late reply, and thanks everyone for the suggestions. >>> >>> I did not provide much details because I am new to this domain and I wanted to see from the reactions if maybe I am not missing some relevant direction. >>> >>> We are trying to measure how an Oracle database can cope with an increase in usage (basically, there will be more users for the application). >>> >> Just more users or also more data? Will the users access the database through one application or does each have it's own application? >> > > Both, but users is the more pressing problem. We already have enough data to expose problems. It's one application which is a legacy two-tier-Delphi-based wrapped with a three-tier-JEE. > > > >>> We are basing our analysis on typical SQL statements coming from the application. We are currently doing load testing by: >>> - recording sql statements from some use cases that are considered to be important >>> - generalizing them by replacing actual values with generic variables >>> >> That shouldn't be necessary, the queries should already contain bind variables (unless your database layer is crap). >> > > Well, indeed, this should not happen, but legacy is never clean :). Anyway, there aren't that many cases. > > > >>> - providing meaningful values for the generic variables >>> - replaying them against the database from several client machines >>> - consuming the first record from the responses >>> >> Why only the first? >> > > Because we are not interested in the response. Only to check that something gets returned. Is this a problem? > > > >>> - reporting the timing of the statements >>> - recording the CPU, memory and I/O load of the server >>> >> Oracle already provides tools for many of these things. >> >> >>> However, I am interested in pitfalls, and in the way people interpret the results given that it is hard to determine what is a typical usage in terms of what statements to trigger and at what delays. >>> >> The yourself a competent Oracle-DBA and probably sysadmin and storage guy as well. No seriously, you wouldn't want to have GemStone benchmarked by someone who has never used Smalltak before, would you? >> > > Thanks, we do have a competent Oracle specialist :). But, this being a tricky job, I thought of asking around for other experiences. > > > Cheers, > Doru > > > > >> Cheers >> Philippe >> >> >> > > -- > www.tudorgirba.com > > Innovation comes in least expected form. > That is, if it is expected, it already happened. > > > >

Feb. 12, 2012

Re: [Pharo-project] Regular expression

by Schwab,Wilhelm K

I've used the package to good effect, but you are beyond me - hopefully someone else can offer some real help. You might something in the Reg ex chapter on http://pharobyexample.org/ Look under "Pharo by Example 2" on the right side of the page. HTH (some). Bill ________________________________ From: pharo-project-bounces(a)lists.gforge.inria.fr [pharo-project-bounces(a)lists.gforge.inria.fr] on behalf of recursive68(a)gmail.com [recursive68(a)gmail.com] Sent: Saturday, February 11, 2012 5:27 PM To: pharo-project(a)lists.gforge.inria.fr Subject: [Pharo-project] Regular expression Hi, Is it possible to use the VB-Regexp package to construct a single regular expression that ensures a string is alphanumeric and contains both at least one number and alphabetic character. Looking at the regexp package and documentation it doesn't appear so. In javascript or perl you use something like this with lookahead assertions: ^(?=.*[a-zA-Z])(?=.*[0-9]).*$ I have something working using multiple regexp's but wondered if a more typical regexp was possible using the package ? Thanks

Feb. 11, 2012

[Pharo-project] Regular expression

by recursive68＠gmail.com

Hi, Is it possible to use the VB-Regexp package to construct a single regular expression that ensures a string is alphanumeric and contains both at least one number and alphabetic character. Looking at the regexp package and documentation it doesn't appear so. In javascript or perl you use something like this with lookahead assertions: ^(?=.*[a-zA-Z])(?=.*[0-9]).*$ I have something working using multiple regexp's but wondered if a more typical regexp was possible using the package ? Thanks

Feb. 11, 2012

[Pharo-project] SocketPlugin: ignoring unknown option 'TCP_CORK'

by Philippe Marschall

Hi I had this crazy idea, what instead of buffering at the application level I use the native buffer of the socket? I would be doing more or less this: set TCP_CORK to 1 set TCP_NODELAY to 0 do individual writes set TCP_CORK to 0 set TCP_NODELAY to 1 do last write goto 1 (use TCP_NOWAIT and TCP_NOPUSH on BSD/MacOS) I hacked something together and tried to push it through some benchmarks but the SocketPlugin was not very impressed: SocketPlugin: ignoring unknown option 'TCP_CORK' This is on Linux with Cog. Cheers Philippe

Feb. 11, 2012

Re: [Pharo-project] load testing a database

by Tudor Girba

Hi, On 11 Feb 2012, at 21:13, Philippe Marschall wrote: > On 11.02.2012 19:30, Tudor Girba wrote: >> Hi, >> >> Sorry for the late reply, and thanks everyone for the suggestions. >> >> I did not provide much details because I am new to this domain and I wanted to see from the reactions if maybe I am not missing some relevant direction. >> >> We are trying to measure how an Oracle database can cope with an increase in usage (basically, there will be more users for the application). > > Just more users or also more data? Will the users access the database through one application or does each have it's own application? Both, but users is the more pressing problem. We already have enough data to expose problems. It's one application which is a legacy two-tier-Delphi-based wrapped with a three-tier-JEE. >> We are basing our analysis on typical SQL statements coming from the application. We are currently doing load testing by: >> - recording sql statements from some use cases that are considered to be important >> - generalizing them by replacing actual values with generic variables > > That shouldn't be necessary, the queries should already contain bind variables (unless your database layer is crap). Well, indeed, this should not happen, but legacy is never clean :). Anyway, there aren't that many cases. >> - providing meaningful values for the generic variables >> - replaying them against the database from several client machines >> - consuming the first record from the responses > > Why only the first? Because we are not interested in the response. Only to check that something gets returned. Is this a problem? >> - reporting the timing of the statements >> - recording the CPU, memory and I/O load of the server > > Oracle already provides tools for many of these things. > >> However, I am interested in pitfalls, and in the way people interpret the results given that it is hard to determine what is a typical usage in terms of what statements to trigger and at what delays. > > The yourself a competent Oracle-DBA and probably sysadmin and storage guy as well. No seriously, you wouldn't want to have GemStone benchmarked by someone who has never used Smalltak before, would you? Thanks, we do have a competent Oracle specialist :). But, this being a tricky job, I thought of asking around for other experiences. Cheers, Doru > > Cheers > Philippe > > -- www.tudorgirba.com Innovation comes in least expected form. That is, if it is expected, it already happened.

Feb. 11, 2012