- Pharo-dev - lists.pharo.org

[Pharo-project] Can I use GDB with SqueakVM ?
by Mariano Martinez Peck March 27, 2010

March 27, 2010

(cross posting) Hi folks. I have a SqueakVM with an image that uses SqueakDBX to talk to a database. SqueakDBX talks to a external library (written in C) called OpenDBX. At the same time, OpenDBX then talks to another library which is the database client library. When....I am having a problem with Firebird and it seems to be related with threads management. So, what I want to do is to start the SqueakVM with GDB, run my tests and try to get as much information as possible when it crash. So far I am in Ubuntu 9.10 and I have tried the following: ubuntu@ubuntu-desktop:~/opendbx/trunk/test$ gdb /home/ubuntu/squeak/pharo-vm-0.15.2f-linux/squeak GNU gdb (GDB) 7.0-ubuntu Copyright (C) 2009 Free Software Foundation, Inc. License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html > This is free software: you are free to change and redistribute it. There is NO WARRANTY, to the extent permitted by law. Type "show copying" and "show warranty" for details. This GDB was configured as "i486-linux-gnu". For bug reporting instructions, please see: <http://www.gnu.org/software/gdb/bugs/>... Reading symbols from /home/ubuntu/squeak/pharo-vm-0.15.2f-linux/squeak...done. (gdb) set args /home/ubuntu/squeak/pharo1.0-10500-rc1dev09.12.1/pharo1.0-10500-rc1dev09.12.dbx.1.image (gdb) run Starting program: /home/ubuntu/squeak/pharo-vm-0.15.2f-linux/squeak /home/ubuntu/squeak/pharo1.0-10500-rc1dev09.12.1/pharo1.0-10500-rc1dev09.12.dbx.1.image [Thread debugging using libthread_db enabled] Cannot find new threads: generic error (gdb) As you can see there the error "[Thread debugging using libthread_db enabled] Cannot find new threads: generic error" I am not an expert in C or VM so I don't know if I am doing the things correctly. I tried to google a bit, and I found different solutions. I copy exactly what I found: "I've find a workaround, i've link the debugging app with libpthread and now gdb feels happy, and i can run it with gdb !" "But however, if I compile the program with -pthread option, am able to debug the program: $ cc -W -Wall -Wextra -g 20.c `pkg-config --cflags --libs gtk+-2.0` -pthread" "The problem appears to be that something goes wrong with dynamically loading libpthread in case the program has not been linked to it at compile time. That would explain why it works when you use msethread, because that one explicitly links to libpthread at compile time (via the imports in mselibc, I guess). Why dynamically loading libpthread causes problems is another question. Maybe somehow the cthreads unit loads a different ("wrong") copy of libpthread than the one the static linker uses when linking the program. Or, alternatively, maybe the static linker selects a different version of libc to link against if libpthread is also used (libc and libpthread are quite interwoven on GNU systems, even though they are separate libraries)." So...I don't understand exactly how can I fix it. Thanks for any help. Mariano

1 0

[Pharo-project] bug in Pharo RC3 , OB related
by Mariano Martinez Peck March 27, 2010

March 27, 2010

In PharoDev RC3 open a class browser and go to any class that overrides a method. For example, TestCase >> assert: Now, click on the green narrow to go to the super class implementor: MessageNotUnderstood: TestCase class>>withAllSuperAndSubclassesDo: The method withAllSuperAndSubclassesDo: is not in Pharo 1.0 but it is in 1.1 Cheers Mariano

2 1

[Pharo-project] How to easily intercept messages sent to ALL objects ?
by Mariano Martinez Peck March 27, 2010

March 27, 2010

Hi folks. I want to intercept (and do something) for all objects. I know few approaches, for example: 1) using method wrappers (implementing run: aSelector with: arguments in: aReceiver) I can wrap all compiled methods of all classes and there I have the receiver. 2) become all objects to some kind of hacky object that stores the original object and that redefines the doesNotUnderstand: But in 1) I need to change all CompiledMehtod of all classes. in 2) I have to become all the objects to another object Then my question is, is there an easier way to intercept all messages send to all objects from the image side (not going to vm) and with a lower cost ? And if going to vm, how ? Thanks. Mariano

8 13

[Pharo-project] [update 1.1] #11298
by Marcus Denker March 27, 2010

March 27, 2010

11298 ----- Issue 2216: remove 21 unsent methods Issue 2215: testFontFullName recursively deletes all files in the image directory - Undo a change from Issue 2030 Marcus -- Marcus Denker -- http://www.marcusdenker.de INRIA Lille -- Nord Europe. Team RMoD.

1 0

[Pharo-project] Ubuntu users
by Schwab,Wilhelm K March 27, 2010

March 27, 2010

Hello all, I hope this is simply irresponsible FUD, but one possible explanation for my image meltdown yesterday is that it was fried in copying to a USB stick. I have also been messing around with FFI, which is another way things could have gone bad. Attempts to load the bad image end in "Read failed or premature end of image file." I am asking myself whether unmounting USB drives is working as well as it did prior to some updates that came out last week. Any thoughts? Backups are always a good idea, and you might want to be a little extra careful until experience proves me wrong. Bill

2 1

[Pharo-project] [GSoC] Problems, Students bio and projects without mentor
by Mariano Martinez Peck March 27, 2010

March 27, 2010

Hi everybody. I hope everybody is fine. As you may have noticed, the ESUG GSoC has been quite busy the last weeks. Students started to register in the website, discuss with mentors, etc... A little problem that we saw few times: when you register (as a student or mentor) you should receive a confirmation email. I don't know why but some people reported this mail was filtered in the spam folder. So, if you are mentor or student and didn't receive a confirmation email, please check you spam folder. Students, when you edit your profile there is a tab called "Biography". Please, take the time to complete that. There are multiple students per project and that may help the mentor to choose the student. Another problem is that we have two projects without mentors. 1. Native windows for Squeak and Pharo http://gsoc2010.esug.org/projects/native-windows 2. OpenSSL Smalltalk wrapper http://gsoc2010.esug.org/projects/openssl Fortunately no student yet expressed interest on any of them. There is a deadline in Monday to remove them if there won't be any mentor found. That's all for the moment. Mariano ESUG GSoC admin esug.gsoc.adm2010(a)gmail.com http://gsoc2010.esug.org/

1 0

[Pharo-project] [update 1.1] #11297
by Marcus Denker March 27, 2010

March 27, 2010

11297 ----- Issue 2214: tests for shadowing Issue 2212: some refactoring related to Smalltalk globals Issue 2206: Rename JPEGReadWriter2 to PluginBasedJPEGReadWriter Issue 2166: MethodProperties pending tasks -- Marcus Denker -- marcus(a)2denker.de http://www.2denker.de 2denker UG (haftungsbeschrÃ¤nkt) Sitz KÃ¶ln Amtsgericht - Registergericht - KÃ¶ln, HRB 65065 GeschÃ¤ftsfÃ¼hrer: Christian Denker und Dr. Marcus Denker

2 1

Re: [Pharo-project] [squeak-dev] A new Smalltalk mailing-list
by Serge Stinckwich March 27, 2010

March 27, 2010

Yes can we wait until everyone migrate to the new mailing-list ? Thank you. On Sat, Mar 27, 2010 at 5:25 PM, stephane ducasse <stephane.ducasse(a)gmail.com> wrote: > Ken I imagine that serge was not aware of that possibility, Â close it and we keep the archive. > > Stef > > On Mar 26, 2010, at 5:18 PM, Ken Causey wrote: > >> OK, but that's a separate issue. Â Deleting the list does not necessarily >> delete the archive, it simply eliminates the membership list and of >> course the ability to receive and relay new messages. Â If you are not >> going to use the list any more I really do need to close/delete it, at >> which point I can leave the archive in place or send it to you. >> >> Ken >> >> On Fri, 2010-03-26 at 09:30 +0700, Serge Stinckwich wrote: >>> No, we can keep the archive for historical purpose ;-) >>> >>> On Fri, Mar 26, 2010 at 6:02 AM, Ken Causey <ken(a)kencausey.com> wrote: >>>> Does that mean the old list >>>> >>>> http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/squeak-fr >>>> >>>> should now be deleted? >>>> >>>> Ken >>>> >>>> On Thu, 2010-03-25 at 22:16 +0700, Serge Stinckwich wrote: >>>>> Hi all, >>>>> french Smalltalkers (Squeakers, Pharoers, VWers, GnuSTers, ...) have >>>>> now a new mailing list: http://groups.google.com/group/smalltalk-fr >>>>> >>>>> The archive of the mailing-list is available on nabble : >>>>> http://n3.nabble.com/smalltalk-fr-f471484.html >>>>> >>>>> This mailing-list replace the old Squeak-fr mailing list. >>>>> >>>>> Best regards, >>>> >>>> >>> >>> >>> >> >> > > -- Serge Stinckwich UMI UMMISCO 209 (IRD/UPMC), Hanoi, Vietnam Smalltalkers do: [:it | All with: Class, (And love: it)] http://doesnotunderstand.org/

1 0

[Pharo-project] Hello squeakers
by stephane ducasse March 27, 2010

March 27, 2010

Dear Squeakers I want to send you a message because I estimate Squeakers and I want to open the door to see how new relationship can be build. So consider that as an open hand - even if my english may let you think otherwise. First, Pharo is not against Squeak. We forked because we believed that we could not make Squeak move in any coherent direction. We are sure that you understand our reasons. Just think a moment about the amount of time and energy we invested in Squeak in the past (I wrote more than anybody else books on squeak, build tutorials, lecture support, videos....- with esug over the years we spent more than 30 kEuros in Squeak related actions) so deciding to go for Pharo was not an easy choice but a necessary one: At one point I was thinking to quit Smalltalk and go to see Ruby and Python for real. Pharo is the only way that I get back my fun in Smalltalk. So what are my dreams? For Pharo ====== We want a clean, lean and fast Smalltalk. An implementation that makes other dynamic language jealous. We want a place where we/you can innovate. We want people to be able to invent THEIR future. We want a place where people can make money with it and build robust applications. Being able to experiment fast is important but for that the system should be clean, robust and flexible. Having a platform for experimentation requires that the platform is not experimental. About innovation I mean in no order: Support for multitouch screen, bootstrappable Smalltalk, immutability bit and its impact, ephemerons, new module system?, first class instance variables, using traits for real (like in ruby where any class can be a model without inheriting from model), VAT-like system?, event system like in AmbiantTalk?, I put ? because some of these should be implemented assessed tested... and understood deeply. About clean Clean network, clean event system, clean object kernel, better compiler (open - we got first class instance variable with no runtime penalties in one afternoon). Clean class builder... Now enough about Pharo. http://www.pharo-project.org/ About Etoys ======== I love Etoys (we translated the book and did more presentations of etoys than most squeakers) but I do not like its implementation. Why? Because it is bad. Any body that looks at it knows it. When I removed Etoys part from Pharo I'm sad but there is no other choices. Now it does not mean that I'm against Etoys and Etoys has the Etoys 40 image (note that we collected in 3.9 most of the etoys fixes with little support from Etoyers which forked way before, we did the same with the fixes of diego of Smalltalk). But again you can judge otherwise. About the ranting or the little war between Squeak and Pharo ====================================== Frankly I'm tired about us ranting against Squeak/andreas/... and the inverse. For example Traits are cool, Javascript and PHP will probably have them as Perl-6, Scala, Fortress. Now Squeak can remove them. I have no problem with that. Seriously this is your decision. People in squeak-dev can freely say negative points about me if this helps. I decided that I will not rant nor get negative feelings about that. I found the red pill :) *I* decided that I want to head to the future. So we will not rant nor make any bad statement about the past anymore. Not even report history or on old facts: if you were there you should remember, else there is the archive :). This is my last mail on the past. Frankly I have the best job I can dream about. I'm lucky just check my h-index for the fun, I have more than most researchers I know. In addition, I loved working and learning from people like lukas, adrian, nicolas, levente, marcus, ..... The next 10 years should be the best of my life and I will take advantage of that. I want to have **10/15** years of pure fun and I will do it. I want and will create positive energy. Look at ESUG we are doing a great job. About cross dialect energy ================= Now the key point of this mail. I **deeply** appreciate the attitude of people like nicolas, levente, and igor that do not bash us and help Pharo but also Squeak. I sent this message mainly because of their attitude. I'm sad to see all this (their) energy duplicated. We cleaned and improved a lot Pharo over the last two years (more than you may think) and we will continue. Squeak could have benefitted from it. Nicolas luckily for you pushed a lot of our fixes in Squeak already. I'm getting the fixes of Squeak that are interesting for Pharo. Now depending on the Squeak vision we could share some common things. May be we can build a better future together but not at all price. You see our goals is to get a clean, lean, flexible and robust system. If you want to share something with us let us know. You know now the vision of Pharo. Squeak may want to compete with us too. This is ok too. If you want to help us building our vision you are welcome. Our logo is a lighthouse and it means that it will stand and last long because it has to guide boats. So we will continue Pharo against tempests and giant waves :) https://gforge.inria.fr/frs/download.php/26678/pharoVideo.zip http://www.youtube.com/watch?v=m2LeNBY_5gk The video is really cool (dan this is the one you wanted on waves). Stef

6 6

[Pharo-project] Stateful Traits Literature
by Stefan Marr March 26, 2010

March 26, 2010

Hi: A bit off-topic but I think the experts are here, so: Is there somewhere an overview of all traits proposal related to the handling of state? Possibly with a nice taxonomy and visualization? Thanks Stefan -- Stefan Marr Software Languages Lab Vrije Universiteit Brussel Pleinlaan 2 / B-1050 Brussels / Belgium http://soft.vub.ac.be/~smarr Phone: +32 2 629 2974 Fax: +32 2 629 3525

5 11