Squeak-dev March 2008

squeak-dev@lists.squeakfoundation.org

157 participants
229 discussions

Using SQLite3 package
by Giuseppe Luigi Punzi 03 Mar '08

03 Mar '08

Hi list, At the moment, I need to "talk" with some M$Access and SQLITE DB. With ODBC from Diego Gomez Deck, seems to work like a charm with this 2 DB's, but I found SQLITE3 package on squeakmap: http://map.squeak.org/package/b396aec0-e9cd-4e70-8746-eb38284f75af and I thought it is better to avoid ODBC use. I loaded it in my testing image (now 3.9.1), but when I try to evaluate this block(1) I get an error. "Error: Unable to find function address" |con s1 s2 s3 | con := SQLiteConnection fileNamed: 'E:\Desarrollo\InciGestFPC\datos.s3db'. con inspect. s1 := 'create table employee (id, name)'. s2 := 'insert into employee values (1, ''Adam'')'. s3 := 'insert into employee values (2, ''Eve'')'. (Array with: s1 with: s2 with: s3) do: [:sql | con executeQuery: sql]. SqueakLog is attached. Thanks for your help and time. -- Giuseppe Luigi Punzi <glpunzi(a)lordzealon.com> <http://www.lordzealon.com>

1 0

Election 2008: Answers from Yoshiki
by Yoshiki Ohshima 03 Mar '08

03 Mar '08

Hello, 1. Approximately, how much time do you plan on spending on Squeak during the coming year (in any kind of unit)? I will be still spending some chunk of my time constantly on the development of OLPC Etoys. Also, for making prototypes of new UI and end-user application, I'll be using Squeak (probably the OLPC Etoys image). In the longer term, though, VPRI trys to roll out new system that are not based on Squeak. I'll be working on these projects. 2. What are in your mind the three most important issues (not necessarily technical) we need to address in the coming year? Giving somebody(s) the authority of making technical dicisions would be important to move the community. 3. What is your view on fund raising and how any such collected money should be dealt with? For small chunks, simple donations from individuals members in the community would be good. But honestly I don't see some entity that offers bigger funding toward Squeak itself. There will be funding for projects that may use Squeak, but only if these projects are attractive. 4. What is your view on the ongoing process of making SqueakFoundation a not-for-profit legal entity? It is making progress and I hope it goes through. At the same time, even if it doesn't go through, that wouldn't be a disaster. It is people's willingness that drives the community. 5. Do you think the Team model is appropriate for organising our efforts or should we come up with something else? Team would work better if there is still a leader. 6. Do you have any specific views on how the Squeak board and the Squeak community should work together with the Squeak satellite communities (Croquet, Seaside, Sophie, Squeakland, Scratch etc), also referred to as "stakeholder communities"? Share the contributions and insights. Something like DeltaStream, which increases the visibility of other people's patches and let us cherry pick them would be good. I don't see a real merging effort of them will happen, though. I wouldn't even think about such effort with the OLPC Etoys image and the mainstream image. 7. The squeak.org release is our most important asset. How do you see it evolving over the next few years? For various reasons such as licensing and people's feeling toward minimum systems, I'd think that we would have two systems. One is based on a kind of small system that remotely related to current "Squeak". And another is maintenance versions of current image(s). The former should have cleaner license, but I think it is ok to distribute the latter even if some people think that the license is murky. 8. Do you have any thoughts on the current relicensing effort? I personally think that the license issue is overrated in the open source software community. Anybody can sue anybody after all for anything, so there is no way to achieve "100%" cleanness in this world. Nonetheless, we should try to do some "reasonable effort" to relicensing. And, like the Japanese versions and other languages version of Squeak and Squeakland, people wouldn't stop using, and that wouldn't be a problem. 9. How would you like Squeak to be positioned in the open source world in year 2012? An honest answer is that something much better comes out from the project at VPRI and the many members of Squeak community move on to it. 10. What do you see as the overall role of the board? Sometimes being at the Viewpoints helps, and knowing the POV of other language communities such as the largeer Japanese Squeak community helps. -- Yoshiki

7 9

Alternate BlockContext>>forkAt: for Andreas, was #fork and deterministic resumption of the resulting process
by Louis LaBrunda 03 Mar '08

03 Mar '08

Hi Andreas, I'm sorry to be resurrecting this thread but my mind often thinks about these things while trying to get to sleep. Anyway I hope my suggestions will be of value. I am presenting two similar suggestions. They both accept your desire to have the new process (or at least the block of the new process) wait to run until after the calling process can run and save the result of the #fork message. The first is closest to your suggestion. Instead of a helper process it wraps the block being forked in a block that delays running of the forked block until there are no processes ready to run for the same priority. I think this is equivalent to the helper process of a lower priority waiting to run. The second suggestion also wraps the block being forked in a block that delays running of the forked block but does so by pushing itself to the back of the ready to run list for its priority twice. This should be enough to insure the calling process has run and saved the result of the #fork message. I'm not looking for any kind of a fight/trouble, I just thought these might be an interesting alternative. Lou "Suggestion 1 from Lou" BlockContext>>forkAt: priority "Create and schedule a Process running the code in the receiver at the given priority. Answer the newly created process." | forkedProcess helperBlock | helperBlock := [ [(Processor waitingProcessesAt: priority) notEmpty] whileTrue: [Processor yield]. self value. ]. forkedProcess := Process forContext: helperBlock priority: priority. ^forkedProcess resume. "Suggestion 2 from Lou" BlockContext>>forkAt: priority "Create and schedule a Process running the code in the receiver at the given priority. Answer the newly created process." | forkedProcess helperBlock | helperBlock := [ Processor yield. Processor yield. self value. ]. forkedProcess := Process forContext: helperBlock priority: priority. ^forkedProcess resume. "Suggestion from Andreas" BlockContext>>forkAt: priority "Create and schedule a Process running the code in the receiver at the given priority. Answer the newly created process." | forkedProcess helperProcess | forkedProcess := self newProcess. forkedProcess priority: priority. priority = Processor activePriority ifTrue:[ helperProcess := [forkedProcess resume] newProcess. helperProcess priority: priority-1. helperProcess resume. ] ifFalse:[ forkedProcess resume ]. "Original code" BlockContext>>forkAt: priority "Create and schedule a Process running the code in the receiver at the given priority. Answer the newly created process." | forkedProcess | forkedProcess := self newProcess. forkedProcess priority: priority. ^ forkedProcess resume ----------------------------------------------------------- Louis LaBrunda Keystone Software Corp. SkypeMe callto://PhotonDemon mailto:Lou@Keystone-Software.com http://www.Keystone-Software.com

6 11

Summer of Code 2008
by giovanni 03 Mar '08

03 Mar '08

Hi Janko, Esteban, I'm at work at the moment - I'll send more details regarding the SoC this evening. Giovanni

1 0

100 Million objects
by Herbert König 03 Mar '08

03 Mar '08

Hi, out of curiosity (forgot the reason) in a Workspace I did a:= Array new: 100000000 withAll: 1.0 Which gave me an 80% cpu usage. To verify I browsed a MessageTally and found that incremental GC used 88%. One Smalltalk garbageCollect brought things back to normal. As far as I remember my swiki readings this is because everything surviving a GC is moved to old space which is ignored by incremental GC. Another interesting observation: Doing the allocation with immediate GC is quite fast while doing a:= nil with immediate GC takes *much* longer. BTW, if you don't have the latest CPU 10 million objects will do. Win 3.10.6 VM, 3.8.2, 3.9 final and latest 3.10 images. Do I have to learn from this, that when creating lots of objects I intend to keep for a while, it's good to force a GC instead of waiting that time will move my objects to old space? Cheers Herbert mailto:herbertkoenig@gmx.net

8 22

Re: [squeak-dev] Copying a file from Squeak on Linux
by Bill Schwab 03 Mar '08

03 Mar '08

Dave, Thanks for the read on OSProces; I will try it a few more times and see what happens. As an aside, I have long thought that the error log should append vs. truncate by default; this is by no means the first time I have lost data due to the current behavior. Whenever I have a stubborn problem, I alter the logging code, but it is often too late for something (hopefully) transient like this. Back to my immediate goal, a Linux-specific solutoin is perfectly adequate, as I already have a working solution on Windows. The exchanges occur between a "floppy" (now typically a flash drive) and the local file system. So, I can get Windows to copy to the stick, and then have Squeak copy from the stick to the file system and back. The copy plugin appears not to be in my VM. To be safe though, how would I test that? So far, all I have found is to call a function in the plugin, which could fail for various reasons besides the code not being present. Thanks, Bill ============================ "David T. Lewis" <lewis@...> lewis@... wote: On Sun, Mar 02, 2008 at 12:31:55PM -0500, Bill Schwab wrote: > I thought to load OS Process, but Squeak exited silently (sadly, I did > not think to look for an error until after the log was overwritten by a > subsequent error). I tried again and it loaded normally; is OSProcess > stable? Bill, OSProcess should be stable on Unix/Linux, and I think that most of its functions work on OS/X. It's not much use on Windows. If you encounter this problem again on a Linux system, I would be grateful for any details you can provide. I'm not sure that I understand what you are trying to do, but if you mean to move or copy files using unix commands driven by Squeak, then the combination of OSProcess and CommandShell will help, but this would be a platform-specific solution. Note also that Squeak has the FileCopyPlugin (you may need to build it yourself with VMMaker if it's not in your VM, I'm not sure). This is attempt to provide some of the file copy behavior that you are trying to achieve, and it should work on multiple platforms, so possibly that would be of some help. See FileList>>primitiveCopyFileNamed:to: which uses the FileCopyPlugin. Dave Wilhelm K. Schwab, Ph.D. University of Florida Department of Anesthesiology PO Box 100254 Gainesville, FL 32610-0254 Email: bschwab(a)anest.ufl.edu Tel: (352) 846-1285 FAX: (352) 392-7029

3 2

Re: [squeak-dev] Setting 3.10 Dev image up for VM compiling
by Ken G. Brown 03 Mar '08

03 Mar '08

Further: When generating SerialExtendedUnixPlugin as an external plugin, need to add the following two files in /Mac OS/Plugins/SerialExtendedUnixPlugin to Xcode SerialExtendedUnixPlugin.h sqMacSerialExtended.c Ffenstri stuff also includes: Ffenestri-b-3-MultiHostWindows-Sunits.1.cs HostMenus-Fixes-1.1.cs sq3.10-7159dev08.02.3 with level playing field and Squeak 3.8.18beta1U.app on Mac. -------------------------------------------------------------------------- I am able to build a vm now after also resolving the following issues: SerialExtendedPlugin was not loading into 3.10 Dev refer to mantis bug M6966 where I have uploaded the new version including fixes: EnhancedSerialPorts-JMM.11wSmalltalkFixesFor3.10-kgb.cs Klatt is wanted by the Xcode project from the svn tree but is missing from the VMMaker generated source tree. Klatt is in Speech on SqueakMap but does not load in 3.10. for now need to remove Klatt from build. MacMenuBarPlugin.c was wanted by the Xcode project but was missing. The following change sets are required to resolve this. Ffenstri stuff from http://homepage.mac.com/WebObjects/FileSharing.woa/67/wo/dpDUKrp5QoOxiKIu.1… Ffenestri-b-1-HostWindowPlugin.2.cs Ffenestri-b-2-WindowProxies.3.cs Ffenestri-b-3-MultiHostWindows.1.cs Ffenestri-b-4-Events-Morphic.1.cs HostMenus-Base.2.cs HostMenus-Events.1.cs HostMenus-Mac-MacPlugin.1.cs HostMenus-Mac-SUnits.1.cs Need to change SerialExtendedPlugin folder to SerialExtendedUnixPlugin in Mac OS/Plugins so Xcode can find it Compile error Line 270 MacMenuBarPlugin.c = #if TARGET_API_MAC_CARBON; Needs the semi-colon removed This error appears in the generated file. Ken G. Brown At 4:00 AM -0800 3/2/08, squeak-dev-request(a)lists.squeakfoundation.org apparently wrote: >Date: Sat, 1 Mar 2008 18:50:57 -0700 >From: "Ken G. Brown" <kbrown(a)mac.com> >Subject: [squeak-dev] Setting 3.10 Dev image up for VM compiling > >sq3.10-7159dev08.02.3 and Squeak 3.8.18beta1U.app on Mac. > >While following the instructions in the readme which lives in the sourcetree at >http://squeakvm.org/svn/squeak/trunk/platforms/Mac%20OS/vm/Documentation/re… > >and after loading Keith's Level Playing Field, Balloon3D, FFI, and VMMaker 3.8b6, I have found the following items of interest: > >As noted in the readme, >1. VMM38-gc-instrument-image.1.cs is already there in 3.10 Dev > >2. VMM38-64bit-imageUpdates.1.cs is already there in 3.10 Dev > >As well, >3. JMM-fixBiasToGrow.1.cs, and bigCursor-bf.1.cs load in with with VMMaker-tpr.58 changeSet which comes with VMMaker 3.8b6 > >However, >4. VmUpdates-dtl stuff is not there, and needs loaded > >Maybe this will help save someone some time and uncertainty. > >Ken G. Brown

1 0

Delay time question
by Robert F. Scheer 03 Mar '08

03 Mar '08

I compared with Jon Hylands and discovered a big difference between my Ubuntu and his Windows results. Time millisecondsToRun: [10 timesRepeat: [(Delay forMilliseconds: 1000) wait]] 10015 Time millisecondsToRun: [100 timesRepeat: [(Delay forMilliseconds: 100) wait]] 10015 Time millisecondsToRun: [1000 timesRepeat: [(Delay forMilliseconds: 10) wait]] 13405 Time millisecondsToRun: [10000 timesRepeat: [(Delay forMilliseconds: 1) wait]] 45728 He did the third test and got something like 10250 instead of my 13405. Is this behavior well-known and understood or is there something wrong with my install? Thanks for any help. It's a big problem for my robot (and therefore me) at the moment. - Robert

10 46

Re: [squeak-dev] Setting 3.10 Dev image up for VM compiling
by Ken G. Brown 03 Mar '08

03 Mar '08

sq3.10-7159dev08.02.3 with level playing field and Squeak 3.8.18beta1U.app on Mac. -------------------------------------------------------------------------- I am able to build a vm now after also resolving the following issues: SerialExtendedPlugin was not loading into 3.10 Dev refer to mantis bug M6966 where I have uploaded the new version including fixes: EnhancedSerialPorts-JMM.11wSmalltalkFixesFor3.10-kgb.cs Klatt is wanted by the Xcode project from the svn tree but is missing from the VMMaker generated source tree. Klatt is in Speech on SqueakMap but does not load in 3.10. for now need to remove Klatt from build. MacMenuBarPlugin.c was wanted by the Xcode project but was missing. The following change sets are required to resolve this. Ffenstri stuff from http://homepage.mac.com/WebObjects/FileSharing.woa/67/wo/dpDUKrp5QoOxiKIu.1… Ffenestri-b-1-HostWindowPlugin.2.cs') fileIn. Ffenestri-b-2-WindowProxies.3.cs') fileIn. Ffenestri-b-3-MultiHostWindows.1.cs') fileIn. Ffenestri-b-4-Events-Morphic.1.cs') fileIn. HostMenus-Base.2.cs') fileIn. HostMenus-Events.1.cs') fileIn. HostMenus-Mac-MacPlugin.1.cs') fileIn. HostMenus-Mac-SUnits.1.cs') fileIn. Need to change SerialExtendedPlugin folder to SerialExtendedUnixPlugin in Mac OS/Plugins so Xcode can find it Compile error Line 270 MacMenuBarPlugin.c = #if TARGET_API_MAC_CARBON; Needs the semi-colon removed This error appears in the generated file. Ken G. Brown At 4:00 AM -0800 3/2/08, squeak-dev-request(a)lists.squeakfoundation.org apparently wrote: >Date: Sat, 1 Mar 2008 18:50:57 -0700 >From: "Ken G. Brown" <kbrown(a)mac.com> >Subject: [squeak-dev] Setting 3.10 Dev image up for VM compiling > >sq3.10-7159dev08.02.3 and Squeak 3.8.18beta1U.app on Mac. > >While following the instructions in the readme which lives in the sourcetree at >http://squeakvm.org/svn/squeak/trunk/platforms/Mac%20OS/vm/Documentation/re… > >and after loading Keith's Level Playing Field, Balloon3D, FFI, and VMMaker 3.8b6, I have found the following items of interest: > >As noted in the readme, >1. VMM38-gc-instrument-image.1.cs is already there in 3.10 Dev > >2. VMM38-64bit-imageUpdates.1.cs is already there in 3.10 Dev > >As well, >3. JMM-fixBiasToGrow.1.cs, and bigCursor-bf.1.cs load in with with VMMaker-tpr.58 changeSet which comes with VMMaker 3.8b6 > >However, >4. VmUpdates-dtl stuff is not there, and needs loaded > >Maybe this will help save someone some time and uncertainty. > >Ken G. Brown

1 0

Copying a file from Squeak on Linux
by Bill Schwab 02 Mar '08

02 Mar '08

Hello all, I am starting to move files back and forth between Windows and Linux. On Windows, I do this using an evolved version of a command line program I wrote in the win3.1 days, and is now a GUI app written in Dolphin. I now need to create a Squeak version of it to idiot proof my copying to and from Linux. My question is in two parts. First, is there any trick to the modification times on Linux vs. Windows. I vaguely recall hearing that a forgotten difference in behavior of the two systems lead Sun to cause some trouble in the early days of Java. Cautions will be gladly researched. Ubunutu at least seems to do what I expect when I copy "manually." The second question is what is the best way to copy files in/from Squeak, given that I care about preserving modification times? I thought to load OS Process, but Squeak exited silently (sadly, I did not think to look for an error until after the log was overwritten by a subsequent error). I tried again and it loaded normally; is OSProcess stable? Bill Wilhelm K. Schwab, Ph.D. University of Florida Department of Anesthesiology PO Box 100254 Gainesville, FL 32610-0254 Email: bschwab(a)anest.ufl.edu Tel: (352) 846-1285 FAX: (352) 392-7029

2 1

Jump to page: