March 2010 - Pharo-dev - lists.pharo.org

Re: [Pharo-project] Inspecting an ExternalData

by HernÃ¡n Morales Durand

Using Pharo-1.0-10508-rc2 #10508 and FFI-Kernel-ar.12 you should see something like the attached inspector on a ExternalData. Cheers, HernÃ¡n 2010/3/27 Schwab,Wilhelm K <bschwab(a)anest.ufl.edu>: > Hello all, > > At least in RC1, I am unable to inspect instances of ExternalData; #inspectorClassOf: is returning nil (Standard tool set). > > Any ideas? Â I am getting the external data instances from funtions returning void pointers. Â Exploring the instances might be working, but I'm not sure what to expect just yet. > > Bill > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >

March 29, 2010

Re: [Pharo-project] Squeak VM+FT2Plugin, all Pharo 1.0 rc3 tests green

by laurent laffont

2010/3/29 John M McIntosh <johnmci(a)smalltalkconsulting.com> > So what Ghz is this machine? > model name : Intel(R) Core(TM)2 CPU T7200 @ 2.00GHz cpu MHz : 1994.987 cache size : 4096 KB Laurent Laffont > > On 2010-03-28, at 10:42 AM, laurent laffont wrote: > > Thank you ! > > Here the result: > > Squeak VM 3.11.3 -r 2135 + FT2Plugin + --CFLAGS="-O3 -fomit-frame-pointer > -mfpmath=sse -march=native -s -mtune=generic" > ========================================== > 339072847 bytecodes/sec; 9560580 sends/sec > 338848444 bytecodes/sec; 9552573 sends/sec > 338177014 bytecodes/sec; 9473233 sends/sec > 337953795 bytecodes/sec; 9496896 sends/sec > 338848444 bytecodes/sec; 9496896 sends/sec > > so the best performance I have so far. > > Laurent Laffont > > > -- > =========================================================================== > John M. McIntosh <johnmci(a)smalltalkconsulting.com> Twitter: > squeaker68882 > Corporate Smalltalk Consulting Ltd. http://www.smalltalkconsulting.com > =========================================================================== > > > > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >

March 29, 2010

Re: [Pharo-project] The Windows VM

by Henrik Johansen

Ehm, that was a change to allow old image segments to load correctly, squeakmap just happened to be the broken usecase. Cheers, Henry Den 29. mars 2010 kl. 00.44 skrev Mariano Martinez Peck <marianopeck(a)gmail.com >: > The new version 4.1 doesn't seem to be improvements in speed and > from what I read about Andreas comment, it only fixes problems for > SqueakMap, which was removed in Pharo, and a change in the close > event of the windows.

March 29, 2010

Re: [Pharo-project] squeakToUTF-8 and related?

by John M McIntosh

On 2010-03-28, at 10:02 AM, StÃ©phane Ducasse wrote: > >> You should ask Sophie team, their knowledge certainly is far more >> advanced than mine. > > The problem is that most of them disappeared after the java rewrite announce. Ok, well it's not as if we disappeared from the planet... The UTF8 work came out of work done on the OLPC/eToys, check with Bert You need to carefully check that some of the UTF8 encoders/decoder methods, or classes actually DO make UTF8 One of them lied and did a UTF8ToMacRoman or was that UTF8ToLatin1 conversion when it claimed a UTF8ToUnicode32. In Sophie because we wrote it's own storage subsystem we could ensure that all textual data going to storage was UTF8, and all text coming in was UTF8->WideStrings. Also all references to external resources were made into URI that were UTF8 and http encoded safe. Mind we did have a tiny disagreements at times about unicode characters and encoding in URLs between Safari and FireFox, never was quite sure who had the bug. -- =========================================================================== John M. McIntosh <johnmci(a)smalltalkconsulting.com> Twitter: squeaker68882 Corporate Smalltalk Consulting Ltd. http://www.smalltalkconsulting.com ===========================================================================

March 29, 2010

Re: [Pharo-project] Squeak VM+FT2Plugin, all Pharo 1.0 rc3 tests green

by John M McIntosh

So what Ghz is this machine? On 2010-03-28, at 10:42 AM, laurent laffont wrote: > Thank you ! > > Here the result: > > Squeak VM 3.11.3 -r 2135 + FT2Plugin + --CFLAGS="-O3 -fomit-frame-pointer -mfpmath=sse -march=native -s -mtune=generic" > ========================================== > 339072847 bytecodes/sec; 9560580 sends/sec > 338848444 bytecodes/sec; 9552573 sends/sec > 338177014 bytecodes/sec; 9473233 sends/sec > 337953795 bytecodes/sec; 9496896 sends/sec > 338848444 bytecodes/sec; 9496896 sends/sec > > so the best performance I have so far. > > Laurent Laffont -- =========================================================================== John M. McIntosh <johnmci(a)smalltalkconsulting.com> Twitter: squeaker68882 Corporate Smalltalk Consulting Ltd. http://www.smalltalkconsulting.com ===========================================================================

March 29, 2010

[Pharo-project] The Windows VM

by Mariano Martinez Peck

Hi. I have just tested the Windows VM availables and tested the 1 tinyBenchmarks and running all tests of PharoDev RC3. I tested with this Windows VMs: 1) The one that it is in pharo website (PharoVM-Win32-3.11.8): 2) The other one from pharo website (exupery-vm-0.15-win32): 3) THe new one that Andreas released today (SqueakVM-Win32-4.0.1-bin) 1) vm\SqueakVM-Win32-4.0.1-bin\ has freetype 1 tinyBenchmarks '404104183 bytecodes/sec; 11914092 sends/sec' '363120567 bytecodes/sec; 10946039 sends/sec' '390839694 bytecodes/sec; 12597221 sends/sec' '346179851 bytecodes/sec; 9996295 sends/sec' 1 gofer test fails 2) exupery-vm-0.15-win32 doesnt have freetype. Unacceptable for Pharo. 1 tinyBenchmarks '365714285 bytecodes/sec; 9935342 sends/sec' '365975696 bytecodes/sec; 10435290 sends/sec' '349011588 bytecodes/sec; 16777208 sends/sec' '339973439 bytecodes/sec; 11292844 sends/sec' Two RB test with errors 3) SqueakVM-Win32-4.0.1-bin has freetype 1 tinyBenchmarks '353835521 bytecodes/sec; 11294326 sends/sec' '415922014 bytecodes/sec; 12319712 sends/sec' '369675090 bytecodes/sec; 11149338 sends/sec' '373177842 bytecodes/sec; 9858402 sends/sec' all tests geen We suggestion is to remove the Exupery Vm from the pharo website. The new version 4.1 doesn't seem to be improvements in speed and from what I read about Andreas comment, it only fixes problems for SqueakMap, which was removed in Pharo, and a change in the close event of the windows. I would only let for download the PharoVM-Win32-3.11.8 Cheers Mariano

March 28, 2010

Re: [Pharo-project] Squeak VM+FT2Plugin, all Pharo 1.0 rc3 tests green

by Mariano Martinez Peck

Ok....I after a little time, I run both: benchmars and all tests in PharoDev RC3 in 3 vms: 1) the one from Pharo download: http://squeakvm.org/unix/release/Squeak-3.11.3.2135-linux_i386.tar.gz 2) the other one from Pharo Download (exupery): http://gforge.inria.fr/frs/download.php/24736/pharo-vm-0.15.2f-linux.zip 3) A new one, compiled as Laurent said with Squeak VM 3.11.3 -r 2135 + FT2Plugin + --CFLAGS="-O3 -fomit-frame-pointer -mfpmath=sse -march=native -s -mtune=generic" Results 1) Has no FreeType. Unacceptable for Pharo. Should be remove from website. '241965973 bytecodes/sec; 9441865 sends/sec' '248302618 bytecodes/sec; 8876087 sends/sec' '248062015 bytecodes/sec; 9426258 sends/sec' '241965973 bytecodes/sec; 9272986 sends/sec' All test pass. 2) Has freetype '467579908 bytecodes/sec; 11130837 sends/sec' '467579908 bytecodes/sec; 11687732 sends/sec' '462929475 bytecodes/sec; 11845268 sends/sec' '466727438 bytecodes/sec; 10483247 sends/sec' Has two failing tests in TestIdenting 3) Has freetype '423490488 bytecodes/sec; 12494874 sends/sec' '426666666 bytecodes/sec; 12580047 sends/sec' '422442244 bytecodes/sec; 11204099 sends/sec' '414574898 bytecodes/sec; 12157397 sends/sec' All PharoRC3 green. So...In summary, the new compiled VM has almost the same performance as Expuery, but it is more cleaner, easy to understand and install, it is repeatable, and has the sources. I strongly to remove the TWO current Linux VM links from Pharo download site, and upload this new compiled VM but doing some work: more testing, readme.txt, rename it to PharoXXX, etc. We should have this VM for Pharo 1.0 and include it in the one click. Cheers Mariano On Sun, Mar 28, 2010 at 6:20 PM, Mariano Martinez Peck < marianopeck(a)gmail.com> wrote: > > > 2010/3/28 laurent laffont <laurent.laffont(a)gmail.com> > >> >>> >>>> In debian (I don't really know ubuntu), you have apt-file to search for >>>> a package's file: >>>> apt-file search asoundlib.h >>>> >>>> And a really great command is "auto-apt run" which installs >>>> automagically all needed packages for a given command. You can try: >>>> auto-apt run ../unix/cmake/configure >>>> >>>> when it detects a missing header, it will search the corresponding >>>> package and prompt you to install it. >>>> >>>> >>>> >>> >>> I tried both tools but don't seem to work...I don't know why. They just >>> do nothing. But anyway, I do it manually :) >>> >> >> >> you have to run >> >> apt-file update >> >> in order to populate / refresh apt file cache. >> >> > Yes, exactly there was the problem...it couldn't update....start to loop > and didn't find the files to download. > Let me try it again and I paste you the console if you want. > > > >

March 28, 2010

Re: [Pharo-project] how to contribute test to pharo

by StÃ©phane Ducasse

very simple. You picked some tests and check them or write new ones and you published either in the inbox or in the PharoTaskForces if you need some times to hack with friends when this is ok you tell us and open a ticket. Stef On Mar 28, 2010, at 10:34 PM, Hilaire Fernandes wrote: > I can't find any information in the website > > Hilaire > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

March 28, 2010

Re: [Pharo-project] Error parsing XML File

by Fabrizio Perin

@Alex: sorry but i cannot send you the file (i shouldn't even have it). Thanks a lot for the support and the explanation :) Fabrizio 2010/3/28 StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> > tx! > On Mar 28, 2010, at 9:24 PM, jaayer wrote: > > > ---- On Sun, 28 Mar 2010 11:47:43 -0700 HernÃ¡n Morales Durand < > hernan.morales(a)gmail.com> wrote ---- > > > >> Hi Alex, > >> Try with the attached XML file in your Pharo directory > >> > >> XMLDOMParser > >> parseDocumentFromFileNamed: > >> (FileDirectory default fullNameFor: 'likelySubtags.xml') > >> > >> however, it seems to be fixed in XML-Parser-JAAyer.72 > >> Cheers, > > > > The problem was due to XMLStreamReader>>nextMatchAll: relying on > #position: and #position and MultiByteFileStream understanding stream > position in terms of bytes rather than characters. That's why it worked when > Alexandre tried parsing a string containing multi-byte UTF-8 characters--the > position of a ReadStream on a string corresponds to the position of > characters in the string regardless of their width--but failed when Fabrizio > tried to parse a file containing those same characters. It is fixed now and > a little faster, too. > > > > > > _______________________________________________ > > Pharo-project mailing list > > Pharo-project(a)lists.gforge.inria.fr > > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >

March 28, 2010

Re: [Pharo-project] [Preview] Help System with new model

by StÃ©phane Ducasse

Cool who is impatient :) I will look at that tomorow evening... tomorrow meeting wihout internet (can you imagine that?) stef On Mar 28, 2010, at 10:39 PM, Torsten Bergmann wrote: > Hi, > > sorry for the delay on the help system - but time is > rare these days. > > I know some of you are impatient - therefore I've commited my > restructured code for the help system for you to look what > is coming - even when it is not fully finished. But the restructured > new model is now in place - I have to add more tests and > adapt the "help on the help system" to document my ideas. > I also want to work on compatibility with Squeak as the > first version had. Currently the new stuff only work > in Pharo. > > What is new > =========== > While we had 3 model classes in my initial version (and after > Dannys additions even more classes) I now reduced the model > to a very, very simple approach with just one model class. > > We now have a HelpBrowser that is "just" displaying topics. > A topic (see class HelpTopic) is > > - providing a title and optionally an icon > - able to have subtopics, > - provides contents (that is displayed be the browser), > - can be identified by a unique key (so we can later open the > help browser on a specific topic from within other tools) > > You can create help topics from each and anything. > > > To try it out just load the baseline (in Pharo 1.0 or 1.1): > > -------------------------------------------- > Gofer new > squeaksource: 'MetacelloRepository'; > package: 'ConfigurationOfHelpSystem'; > load. > > ((Smalltalk at: #ConfigurationOfHelpSystem) project version: '1.0-baseline') load > -------------------------------------------- > > > You can open the help browser from the world tool menu or via > code: > > HelpBrowser open > > You can open the help browser on any object that is able to > be converted into a help topic (via method #asHelpTopic). > > So you can write for instance: > > HelpBrowser openOn: Integer > > opening a short API help/system reference on the Integer class. > The above expression is the short form for: > > HelpBrowser openOn: (SystemReference forClass: Integer) > > If you want you can include the subclasses: > > HelpBrowser openOn: (SystemReference hierarchyFor: Integer) > > or even methods > > HelpBrowser openOn: (SystemReference hierarchyWithMethodsFor: Integer) > > You can browse the whole system reference documentation using: > > HelpBrowser openOn: SystemReference > > But these are only a few examples what we can extract from the > system. > > However - the major goal is NOT an API browser, the idea is to > provide a simple architecture to provide browsable help contents > depending on the context. For instance it should also be possible > to use the help system to provide end user help on any commercial > application that is written with Pharo. > > As I said you can convert any text data into topics to be displayed > by the help system: the browser is able to open on anything that > is convertable to a root topic using #asHelpTopic. More on this later. > > When you open the help browser it provides help for the > whole system, so > > HelpBrowser open > > is the same as > > HelpBrowser openOn: SystemHelp > > which internally uses the class CustomHelp and subclasses to provide > written text and API help to form a system wide help system. > > If you want you can also load "Metacello-Help" and "Squeak-Help" > from the http://www.squeaksource.com/HelpSystem repository to see > how the new topics are automatically used when they were just loaded. > > > Yes, I would like to see a richer text support (using markup > language) and written text documentation accessible directly > from within the image. But the first step was to get a simple > model, the next step would be to see if the extension mechanism > is good enough (CustomHelp, HelpBuilder with own builders, ...) > > > Some more notes on the architecture and extension ideas: > ======================================================== > 1. the model as I said before is now reduced to a single class > (HelpTopic) > > 2. When you open the help browser on an object the method #asHelpTopic > is sent to the object by the browser to get the root node of > the displayed topic hierarchy: > > HelpBrowser openOn: myObject > > Typically the object does not convert itself to a help topic structure, > usually it dispatches to a builder (see HelpBuilder and subclasses) > who does all this. > > So the help text can be already in the image even when the help > system itself is NOT YET THERE/LOADED and if you want to provide help text > in whatever form you are not dependent on the help system itself!!! > > Later a builder can extract and convert the data. You can have builders > for anything - even a builder who gets informations from the web or > hard disk. > > But note: if you want to have your own help shown up in the > standard help then you have to at least use the "CustomHelp" subclass > mechanism. > > 3. The help browser is a simple tree view and is currently limited to > ASCII text. One step in the future could be to provide a "contentType" > (similar to MIME) to support other content types (for instance active > morphs displayed in clean up book morphs, HTML help with scamper in > the help browser, ...) > > 4. I dont expect the basic model to change much further - maybe > > Take care: all of this is still unfinished but in a usable state. > The "Help on Help" is not yet updated. The tests are not reviewed, > just brouht to green. > > So you have to rely on the implementation, all the informations > in this mail or directly ask questions. > > Bye > Torsten > > > > > > > > > > > > > -- > GMX DSL: Internet, Telefon und Entertainment fÃ¼r nur 19,99 EUR/mtl.! > http://portal.gmx.net/de/go/dsl02 > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

March 28, 2010