- Pharo-dev - lists.pharo.org

[Pharo-project] changing the bytecodes print
by Mariano Martinez Peck March 8, 2011

March 8, 2011

Hi folks. Did you see that when you see the bytecodes of a method you see something like: 49 <70> self 50 <21> pushConstant: 'SQ_HOST32' 51 <8F 00 00 02> closureNumCopied: 0 numArgs: 0 bytes 55 to 56 55 <71> pushConstant: true 56 <7D> blockReturn ....... The bytecode number here is printed in <hex> . However, in Interpreter and its #initializeBytecodeTable they use base 10. So, each time I read a bytecode, and I want to know which is the method associated with that, I need to convert it. So, I wonder if it makes sense to change the print, so that it prints in decimal. What do you think? To change that, it seems to be in InstructionPrinter >> #print but I cannot make it work yet :( cheers Mariano

2 1

[Pharo-project] Hardware cursor vs Software Cursor
by Fernando Olivero March 8, 2011

March 8, 2011

I believe that the senders or Cursor xxx showWhile: .. are using the wrong level of abstraction, by directly referring to the Cursor class reference and more important, that all visual related cues must go thru UIManager . If we take this approach further, we see there's 1) no need to have a special Cursor class. 2) We can remove the Hardware cursors and the software cursor distinction, so any Form can be used as a "cursor" image. Leading to an understandable implementation of the drawing behind HandMorph ( if not convinced try out " Cursor wait show" , and see that nothing happens) Solution: *Refactor HandMorph, removing the distinction between hardware and software cursor. *Make the Cursor senders, to use the proper level of abstraction ( not directly using the hardware cursor, such as Cursor wait showWhile:[..]) *Remove Cursor. I will work on this, and once create an issue once working. Saludos, Fernando

2 1

Re: [Pharo-project] Pharo on top of Flash plugin
by Marcus Denker March 8, 2011

March 8, 2011

On Mar 7, 2011, at 1:29 PM, Alexandre Jasmin wrote: >>> Your image helps me a lot. >>> >>> I'm trying to get the Squeak VM compiling with Adobe Alchemy. >> >> tell us more :) > > > There are at least two solutions aiming to bring native code to the web > without installing plug-ins > - Google Native Client provides a sandboxed environment for x86 code The interesting aspect of NaCl is that it does have a JIT compiler API... (Mono is using it already). This means, it would be possible to port Cog to NaCl *and* have all the speed... Getting the stack interpreter to run should be even simpler (recompile.. plus implementing I/O via the NaCl APIs , of course). http://blog.chromium.org/2011/02/native-client-getting-ready-for-takeoff.ht… > - Alchemy (a research project at Adobe) compiles C into the bytecode > language interpreted by Flash. > ... > > Will let you know how it goes. Very interesting direction, too! The most challenging question is if it's possible to somehow "JIT" code... and how fast that then can be. Marcus -- Marcus Denker -- http://www.marcusdenker.de INRIA Lille -- Nord Europe. Team RMoD.

1 0

[Pharo-project] [ANN] ESUG supports once again SqueakDBX
by Mariano Martinez Peck March 7, 2011

March 7, 2011

We are really happy to announce that ESUG will sponsor us once again through the ESUG Summer Talk project. This means that we have reached the ESUG expectations and that they still think that relational database access is an important matter in Smalltalk. One important thing is that we are going to rename the project (we are still working on it) since SqueakDBX runs not only in Squeak but also in Pharo, and there have been even ports to Dolphin. What's the reason for this decision? Because we do not want to couple ourselves to a smalltalk dialect nor to OpenDBX, because our project is much more than that (later I will tell you about our plans). So, these are some of the possible names: ObjectPark, SmallParking, Parktalk, SmallValet, Valetalk, ValetST, NorayTalk, Ballard, Noray and Cruise. Please let us know which one is your favourite or help us find a new one. Another important subject is the team. There will be three "mentors", Esteban Lorenzano, Diogenes Moreira and myself, Mariano Martinez Peck; and three students: Guillermo Polito, Nicolas Scarcella and Santiago Bragagnolo. We are open to suggestions and ideas. In addition, we have defined a possible list of actions that I copy at the end of the email. For the moment, the url remains www.squeakdbx.org and the mailing list squeakdbx(a)lists.squeakfoundation.org Once again, we want to thank ESUG for their support and trust. Thank you very much, SqueakDBX team Possible list of actions: 1) Change SqueakDBXâs name. 2) Update GLORP version since the actual one is 3 years old. Port it again from VisualWorks, create a VW porting tool (may be). Complete support to Glorp. Today it works with PostgreSQL, Oracle and MySQL. Make it work with most databases OpenDBX supports. 3) Create a lightweight solution, alternatively to GLORP. There are some options: Make SqueakSave work with SqueakDBX. SqueakSave developers already contacted us because they wanted to do it. SqueakSave seems to be 20% slower than Glorp but you don't need to write the mappings :) Adapt Ramon Leon's active record to use an abstract database driver, and create a driver for SqueakDBX. Port the new Glorpâs kind of active record to Pharo. (included in 2). 4) Write a Pharo By Example 2 chapter based on the card game Stef built ;). 5) Cog compatibility. 6) Use Alien instead of FFI. Eliot is working on a threaded CogVM. One of the projects of the GSoC of this year was to make something similar to a threaded FFI. What the student did is a modification in Alien (I think) that can be run in a multithreaded envirorment. He worked with Eliot. The idea is when Eliot releases the threaded CogVM, this FFI would work our of the box, and would avoid locking the WHOLE vm while a C function is being invoked (as it happens today with FFI).....So....when that VM is released, we MUST migrate to that). 7) Explore performance issues (maybe with our approach of "In thread execution plugin"). 8) Complete integration with OpenDBX. For example, Oracle, for large objects (Clob, Blob, etc) use specific functions. There are specific functions in OpenDBX that have to be used if the database uses specific functions (oracle is the only one for the moment.). We don't manage those functions yet. 9) In this link http://www.squeakdbx.org/Targets%20and%20Features You can see a list of future possible features like Connection pooling (now it is done!), Prepared statement interface, Store procedures, Escape and avoid of SQL insertion, Authentication support: extends to other methods, not only user/password, Full text support, etc.

3 3

[Pharo-project] [update 1.3] #13076
by StÃ©phane Ducasse March 7, 2011

March 7, 2011

13076 ----- - repackaged MouseWheelEvent - moving AbstractResizerMorph overrides to polymorph extensions

1 0

[Pharo-project] Pharocasts: Build a XML browser
by laurent laffont March 7, 2011

March 7, 2011

Watch this to learn how to: - get and parse XML - use Polymorph to build a GUI with a tree and text editor - add a button in Pharo IDE to open the XML browser - control when your breakpoints will trigger see http://www.pharocasts.com/2011/03/build-xml-browser.html Laurent Laffont - @lolgzs <http://twitter.com/#!/lolgzs> Pharo Smalltalk Screencasts: http://www.pharocasts.com/ Blog: http://magaloma.blogspot.com/

2 3

[Pharo-project] OSProcess
by StÃ©phane Ducasse March 7, 2011

March 7, 2011

Hi guys I'm updtating the ConfigurationOfCoral and I loaded baseline spec project: 'OSProcess' with: [ spec className: 'ConfigurationOfOSProcess'; file: 'ConfigurationOfOSProcess'; version: '4.4.0'; repository: 'http://www.squeaksource.com/OSProcess' ]. version spec for: #'common' do: [ spec package: 'Filesystem' with: 'Filesystem-DamienPollet.114'; project: 'OSProcess' with: '4.4.0'; package: 'AST-Core' with: 'AST-Core-lr.87'; package: 'PetitParser' with: 'PetitParser-TudorGirba.209'; package: 'PetitTest' with: 'PetitTests-TudorGirba.28'; package: 'PetitSmaltalk' with: 'PetitSmalltalk-lr.46'.] and when I run the test I get accessor error. Any idea what would be problem? The VM? I use 5.7b3 Sted

4 9

[Pharo-project] Kernel
by Benjamin March 7, 2011

March 7, 2011

Hello guys, I finally success (with the help of Igor) to trace and create a new Smalltalk kernel based on Pharo. The new born weight 2.2Mb and includes 230 classes ;) You can try it here: Gofer new squeaksource: 'PharoTaskForces'; package: 'SystemTracing'; load. Gofer new squeaksource: 'Seed'; package: 'Seed-HazelBuilder'; load. HazelTracer2 writeImage: 'hazel.image'. If you have an error in Point>>#printOn: , you can use this fix I want to thank Igor again, because he helps me a lot :) Ben

7 20

[Pharo-project] Does FFI support WideStrings?
by Mariano Martinez Peck March 7, 2011

March 7, 2011

Hi. Someone pointed out here: http://stackoverflow.com/questions/5212790/unicode-characters-in-a-smalltal… a problem while using WideStrings with FFI. FFI throws a "Could not coerce arguments". Is this because of the usage of WideString? what we should use instead (I know that ByteString works) ? can we fix FFI so that it supports WideStrings? thanks Mariano

2 2

[Pharo-project] Fwd: [SqueakDBX] Simple object pool 0.1.1 "released".
by Mariano Martinez Peck March 7, 2011

March 7, 2011

Thanks Panu. This is awesome. We will integrate your info in our website!! :) mariano ---------- Forwarded message ---------- From: Panu Suominen <panu.j.m.suominen(a)gmail.com> Date: Mon, Mar 7, 2011 at 8:28 AM Subject: [SqueakDBX] Simple object pool 0.1.1 "released". To: The complete and open-source solution to relational database access < squeakdbx(a)lists.squeakfoundation.org> Long time ago I mailed that I have started to build new pooling solution for objects because I could not find existing one with tests. Then I promised to improve documentation. It has taken quite long time but now I have created minimal documentation that should enable people to try my implementation. Currenlty it is nothing spectacular. Project can be found at http://www.squeaksource.com/ObjectPool.html the wiki there should get you started. If one has loaded GlorpDBX one could start using pool with following code to start pooling SqueakDBX connections: Gofer new squeaksource:'ObjectPool'; package:'ConfigurationOfObjectPool'; load. ((Smalltalk at: #ConfigurationOfObjectPool) project version:'0.1.1') load:#('default' 'dbx'). SqueakDatabaseAccessor DefaultDriver: OPSqueakDBXDriver. Currently OPSqueakDBXDriver does not restrict the amount of active or pooled connections. I will improve that soon. One can load basic pooling implementation without DBX-related stuff by just executing: ((Smalltalk at: #ConfigurationOfObjectPool) project version:'0.1.1') load. -- Panu _______________________________________________ SqueakDBX mailing list SqueakDBX(a)lists.squeakfoundation.org http://lists.squeakfoundation.org/mailman/listinfo/squeakdbx

1 0