Squeak4.2-10950.zip is now available at http://ftp.squeak.org/4.2, which is intended to be the actual-released 4.2 image (unless we find some problem, which we won't).
The trunk will be reopened shortly, thanks for your patience.
- Chris
Hi Chris, I'm getting this error when trying to install OB: 'Could not resolve: OCForOB [OCForOB-rr.2] in C:\FV\Dev\Squeak 4.2\package-cache http://source.wiresong.ca/ob'
Is it a know error?
Thanks you and all for his work.
FV
On Mon, Jan 31, 2011 at 2:24 PM, Chris Muller asqueaker@gmail.com wrote:
Squeak4.2-10950.zip is now available at http://ftp.squeak.org/4.2, which is intended to be the actual-released 4.2 image (unless we find some problem, which we won't).
The trunk will be reopened shortly, thanks for your patience.
- Chris
On Mon, 31 Jan 2011, Facundo Vozzi wrote:
Hi Chris, I'm getting this error when trying to install OB: 'Could not resolve: OCForOB [OCForOB-rr.2] in C:\FV\Dev\Squeak 4.2\package-cache http://source.wiresong.ca/ob'
Is it a know error?
It was a bug in ConfigurationOfOmniBrowser, but it was fixed a while ago. Try using the latest version of the configuration.
Levente
Thanks you and all for his work.
FV
On Mon, Jan 31, 2011 at 2:24 PM, Chris Muller asqueaker@gmail.com wrote:
Squeak4.2-10950.zip is now available at http://ftp.squeak.org/4.2, which is intended to be the actual-released 4.2 image (unless we find some problem, which we won't).
The trunk will be reopened shortly, thanks for your patience.
- Chris
Hi Facu, individual package maintainers will have the opportunity to add or fix their load scripts currently residing in the "Extending the System" workspace even after 4.2 is released. I will explain the details for how to do this soon, once the release details are out of the way.
Thanks, Chris
On Mon, Jan 31, 2011 at 12:08 PM, Facundo Vozzi facundov79@gmail.com wrote:
Hi Chris, I'm getting this error when trying to install OB: 'Could not resolve: OCForOB [OCForOB-rr.2] in C:\FV\Dev\Squeak 4.2\package-cache http://source.wiresong.ca/ob' Is it a know error? Thanks you and all for his work. FV
On Mon, Jan 31, 2011 at 2:24 PM, Chris Muller asqueaker@gmail.com wrote:
Squeak4.2-10950.zip is now available at http://ftp.squeak.org/4.2, which is intended to be the actual-released 4.2 image (unless we find some problem, which we won't).
The trunk will be reopened shortly, thanks for your patience.
- Chris
Nice job Chris!
Looks good for me.
I guess this is to be expected:
Image ----- H:\Squeak\Squeak4.2-10950.image Squeak4.2 latest update: #10950 Current Change Set: Unnamed1
Virtual Machine --------------- C:\Program Files\cogwin\Croquet.exe Croquet Closure Cog VM [CoInterpreter VMMaker-oscog.46] Win32 built on Jan 15 2011 10:49:27 Compiler: 3.4.4 (cygming special, gdc 0.12, using dmd 0.125)
SUnit Results ------------- 2843 run, 2832 passes, 5 expected failures, 5 failures, 1 errors, 0 unexpected passes
Failed Tests ------------ AllocationTest>>#testOneGigAllocation CompilerExceptionsTest>>#testUnusedVariable ExceptionTests>>#testHandlerFromAction MorphicUIBugTest>>#testShowAllBinParts StopwatchTest>>#testActive
Errors ------ LocaleTest>>#testLocaleChanged
Alex
2011/1/31 Chris Muller asqueaker@gmail.com
Squeak4.2-10950.zip is now available at http://ftp.squeak.org/4.2, which is intended to be the actual-released 4.2 image (unless we find some problem, which we won't).
The trunk will be reopened shortly, thanks for your patience.
- Chris
Hmm, the first two are expected. Not sure about that AllocationTest, StopwatchTest, MorphicUIBugTest in Windows. If you have a chance to look at any of those three, it'd be great, at least to see if they're VM or image-related..
This is my list of failures on Linux, which are all expected:
CompilerExceptionsTest>>#testUnusedVariable ExceptionTests>>#testHandlerFromAction MirrorPrimitiveTests>>#testMirrorAt MirrorPrimitiveTests>>#testMirrorInstVarAt ProcessTest>>#testAtomicSuspend
Thanks, Chris
On Mon, Jan 31, 2011 at 2:29 PM, Alexander Lazarević laza@blobworks.com wrote:
Nice job Chris!
Looks good for me.
I guess this is to be expected:
Image
H:\Squeak\Squeak4.2-10950.image Squeak4.2 latest update: #10950 Current Change Set: Unnamed1
Virtual Machine
C:\Program Files\cogwin\Croquet.exe Croquet Closure Cog VM [CoInterpreter VMMaker-oscog.46] Win32 built on Jan 15 2011 10:49:27 Compiler: 3.4.4 (cygming special, gdc 0.12, using dmd 0.125)
SUnit Results
2843 run, 2832 passes, 5 expected failures, 5 failures, 1 errors, 0 unexpected passes
Failed Tests
AllocationTest>>#testOneGigAllocation CompilerExceptionsTest>>#testUnusedVariable ExceptionTests>>#testHandlerFromAction MorphicUIBugTest>>#testShowAllBinParts StopwatchTest>>#testActive
Errors
LocaleTest>>#testLocaleChanged
Alex
2011/1/31 Chris Muller asqueaker@gmail.com
Squeak4.2-10950.zip is now available at http://ftp.squeak.org/4.2, which is intended to be the actual-released 4.2 image (unless we find some problem, which we won't).
The trunk will be reopened shortly, thanks for your patience.
- Chris
On Mon, 31 Jan 2011, Chris Muller wrote:
Hmm, the first two are expected. Not sure about that AllocationTest, StopwatchTest, MorphicUIBugTest in Windows. If you have a chance to
AllocationTest fails on Cog, that's "normal". MorphicUIBugTest fails because of a bug which Eliot already fixed, but he didn't release new VMs with the fix. Both of them should pass using SqueakVM.
StopwatchTest is suspicious. Why does it fail?
Levente
look at any of those three, it'd be great, at least to see if they're VM or image-related..
This is my list of failures on Linux, which are all expected:
CompilerExceptionsTest>>#testUnusedVariable ExceptionTests>>#testHandlerFromAction MirrorPrimitiveTests>>#testMirrorAt MirrorPrimitiveTests>>#testMirrorInstVarAt ProcessTest>>#testAtomicSuspend
Thanks, Chris
On Mon, Jan 31, 2011 at 2:29 PM, Alexander Lazarević laza@blobworks.com wrote:
Nice job Chris!
Looks good for me.
I guess this is to be expected:
Image
H:\Squeak\Squeak4.2-10950.image Squeak4.2 latest update: #10950 Current Change Set: Unnamed1
Virtual Machine
C:\Program Files\cogwin\Croquet.exe Croquet Closure Cog VM [CoInterpreter VMMaker-oscog.46] Win32 built on Jan 15 2011 10:49:27 Compiler: 3.4.4 (cygming special, gdc 0.12, using dmd 0.125)
SUnit Results
2843 run, 2832 passes, 5 expected failures, 5 failures, 1 errors, 0 unexpected passes
Failed Tests
AllocationTest>>#testOneGigAllocation CompilerExceptionsTest>>#testUnusedVariable ExceptionTests>>#testHandlerFromAction MorphicUIBugTest>>#testShowAllBinParts StopwatchTest>>#testActive
Errors
LocaleTest>>#testLocaleChanged
Alex
2011/1/31 Chris Muller asqueaker@gmail.com
Squeak4.2-10950.zip is now available at http://ftp.squeak.org/4.2, which is intended to be the actual-released 4.2 image (unless we find some problem, which we won't).
The trunk will be reopened shortly, thanks for your patience.
- Chris
2011/1/31 Alexander Lazarević laza@blobworks.com
Failed Tests
AllocationTest>>#testOneGigAllocation
Well, this might be because my machine just has 2GB installed. :) This fails regardless if I use a stack vm or a cog vm. What I find strange is, that there is no OutOfMemory raised and array just will be an array of size 0. array := [Array new: sz] on: OutOfMemory do:[:ex| failed := true].
MorphicUIBugTest>>#testShowAllBinParts
StopwatchTest>>#testActive
Errors
LocaleTest>>#testLocaleChanged
All of the above pass if I run each of them as a single test ...
Alex
On Mon, Jan 31, 2011 at 11:24:46AM -0600, Chris Muller wrote:
Squeak4.2-10950.zip is now available at http://ftp.squeak.org/4.2, which is intended to be the actual-released 4.2 image (unless we find some problem, which we won't).
The trunk will be reopened shortly, thanks for your patience.
Yay!
Really nice work Chris. To my eye, the visual appearance is crisp and welcoming, and gives a very good impression on first opening the image. You have done a great job of preparing a positive experience for new Squeakers and old hands alike. Well done.
Dave
That's really nice Dave, thanks. A big part of the look is attributable to the great icon work from Michael Davies. That was a nice change Michael, and that new push-pin jazzed it up. :) Thanks for that.
On Mon, Jan 31, 2011 at 6:46 PM, David T. Lewis lewis@mail.msen.com wrote:
On Mon, Jan 31, 2011 at 11:24:46AM -0600, Chris Muller wrote:
Squeak4.2-10950.zip is now available at http://ftp.squeak.org/4.2, which is intended to be the actual-released 4.2 image (unless we find some problem, which we won't).
The trunk will be reopened shortly, thanks for your patience.
Yay!
Really nice work Chris. To my eye, the visual appearance is crisp and welcoming, and gives a very good impression on first opening the image. You have done a great job of preparing a positive experience for new Squeakers and old hands alike. Well done.
Dave
Really nice, Chris. Thanks.
-C
-- Craig Latta www.netjam.org/resume + 31 020 894 6247 + 1 415 287 3547
Hi,
On a fresh image, opening the "Extending the system" workspace and trying to install the omnibrowser results in an error:
Could not resolve: OCForOB
Top of the trace is MetacelloFetchingMCSpecLoader>>error.
I apologize for not producing more detail but I am writing this from my phone and it is fairly tedious as it is. If you need more details please let me know and I'll try to get on a computer and produce them.
Thanks, Jeff
On 2011-02-01 9:47 AM, "Craig Latta" craig@netjam.org wrote:
Really nice, Chris. Thanks.
-C
-- Craig Latta www.netjam.org/resume + 31 020 894 6247 + 1 415 287 3547
Jeff,
The best I can tell you is that Metacello was not able to download the OCForOB package from the the omnibrowser repository ... I looked at http://source.wiresong.ca/ob and there is no OCForOB package in that repository either the package was removed or the Metacello configuration is in error.
Looking a little closer, it looks like 4.2 is supposed to use the same set of packages as Pharo which implies that the repository should be 'http://source.lukas-renggli.ch/omnibrowser' in the configuration baseline and not http://source.wiresong.ca/ob. My guess is that the packages were in the package-cache when the configuration was tested.
You could edit the config and make the change repository change yourself (method baseline121:) and try to load it again...
I've cc'ed Guilllermo ...
Dale
On 02/01/2011 11:35 AM, Jeff G wrote:
Hi,
On a fresh image, opening the "Extending the system" workspace and trying to install the omnibrowser results in an error:
Could not resolve: OCForOB
Top of the trace is MetacelloFetchingMCSpecLoader>>error.
I apologize for not producing more detail but I am writing this from my phone and it is fairly tedious as it is. If you need more details please let me know and I'll try to get on a computer and produce them.
Thanks, Jeff
On 2011-02-01 9:47 AM, "Craig Latta" <craig@netjam.org mailto:craig@netjam.org> wrote:
Really nice, Chris. Thanks.-C
-- Craig Latta www.netjam.org/resume http://www.netjam.org/resume
- 31 020 894 6247
- 1 415 287 3547
On Tue, 1 Feb 2011, Dale Henrichs wrote:
Jeff,
The best I can tell you is that Metacello was not able to download the OCForOB package from the the omnibrowser repository ... I looked at http://source.wiresong.ca/ob and there is no OCForOB package in that repository either the package was removed or the Metacello configuration is in error.
There used to be the following in the #squeakCommon section of #basline12: in ConfigurationOfOmniBrowser-GuillermoPolito.70:
package: 'OCForOB' with: [ spec repository: 'http://www.squeaksource.com/OCompletion'; requires: #('OCompletion' 'OmniBrowser' 'OB-Standard')].
But now it's just
package: 'OCForOB' with: [ spec requires: #('OCompletion' 'OmniBrowser' 'OB-Standard')];
The repository is specified for Pharo in it's own section, but not for Squeak. Btw both Pharo and Squeak should use the latest version of OCForOB from the OCompletion repository on squeaksource. But first the configuration of OCompletion should be updated and my Metacello knowledge is not enough to do that properly (adding stable versions for both Squeak 4.1 and 4.2).
Looking a little closer, it looks like 4.2 is supposed to use the same set of packages as Pharo which implies that the repository should be 'http://source.lukas-renggli.ch/omnibrowser' in the configuration baseline and not http://source.wiresong.ca/ob. My guess is that the packages were in the package-cache when the configuration was tested.
I doubt that would work. Lukas' version of OB has Pharo specific code which is not compatible with Squeak.
Levente
You could edit the config and make the change repository change yourself (method baseline121:) and try to load it again...
I've cc'ed Guilllermo ...
Dale
On 02/01/2011 11:35 AM, Jeff G wrote:
Hi,
On a fresh image, opening the "Extending the system" workspace and trying to install the omnibrowser results in an error:
Could not resolve: OCForOB
Top of the trace is MetacelloFetchingMCSpecLoader>>error.
I apologize for not producing more detail but I am writing this from my phone and it is fairly tedious as it is. If you need more details please let me know and I'll try to get on a computer and produce them.
Thanks, Jeff
On 2011-02-01 9:47 AM, "Craig Latta" <craig@netjam.org mailto:craig@netjam.org> wrote:
Really nice, Chris. Thanks.-C
-- Craig Latta www.netjam.org/resume http://www.netjam.org/resume
- 31 020 894 6247
- 1 415 287 3547
On 2 February 2011 02:19, Levente Uzonyi leves@elte.hu wrote:
On Tue, 1 Feb 2011, Dale Henrichs wrote:
Looking a little closer, it looks like 4.2 is supposed to use the same set of packages as Pharo which implies that the repository should be ' http://source.lukas-renggli.ch/omnibrowser' in the configuration baseline and not http://source.wiresong.ca/ob. My guess is that the packages were in the package-cache when the configuration was tested.
I doubt that would work. Lukas' version of OB has Pharo specific code which is not compatible with Squeak.
Just in case anyone is still having problems getting OB into a 4.2 image...
I put a pragmatic solution for this into the inbox as Morphic-MAD.512, but the problem with the unexpected changes to the postscript and pre-amble seem to have derailed its adoption, so for anyone having problems with the loading of OmniBrowser, here's a more reliable version of that part of the "Extending the system" samples:
"Omnibrowser including Refactoring engine and OCompletion. Version 1.2" Installer ss project: 'MetacelloRepository'; install: 'ConfigurationOfOmniBrowser'. ((Smalltalk at: #ConfigurationOfOmniBrowser) project perform: #version: with: '1.2') load: #(Dev).
"Omnibrowser latest release - may include Pharo-specific code!!" (Installer ss project: ''MetacelloRepository'') install: ''ConfigurationOfOmniBrowser''. ((Smalltalk at: #ConfigurationOfOmniBrowser) project perform: #lastVersion) load: #(Dev).
I've just tested it with a virgin 4.2 install, and the first snippet loads OB fine.
Hi guys,
To change the squeak.org website, it is 4.2 out or not? I just saw a Weekly Squeak article but on the other side I see here a lot of problem reports...
Best regards Janko
On 03. 02. 2011 20:51, Michael Davies wrote:
On 2 February 2011 02:19, Levente Uzonyi <leves@elte.hu mailto:leves@elte.hu> wrote:
On Tue, 1 Feb 2011, Dale Henrichs wrote: Looking a little closer, it looks like 4.2 is supposed to use the same set of packages as Pharo which implies that the repository should be 'http://source.lukas-renggli.ch/omnibrowser' in the configuration baseline and not http://source.wiresong.ca/ob. My guess is that the packages were in the package-cache when the configuration was tested. I doubt that would work. Lukas' version of OB has Pharo specific code which is not compatible with Squeak.Just in case anyone is still having problems getting OB into a 4.2 image...
I put a pragmatic solution for this into the inbox as Morphic-MAD.512, but the problem with the unexpected changes to the postscript and pre-amble seem to have derailed its adoption, so for anyone having problems with the loading of OmniBrowser, here's a more reliable version of that part of the "Extending the system" samples:
"Omnibrowser including Refactoring engine and OCompletion. Version 1.2" Installer ss project: 'MetacelloRepository'; install: 'ConfigurationOfOmniBrowser'. ((Smalltalk at: #ConfigurationOfOmniBrowser) project perform: #version: with: '1.2') load: #(Dev).
"Omnibrowser latest release - may include Pharo-specific code!!" (Installer ss project: ''MetacelloRepository'') install: ''ConfigurationOfOmniBrowser''. ((Smalltalk at: #ConfigurationOfOmniBrowser) project perform: #lastVersion) load: #(Dev).
I've just tested it with a virgin 4.2 install, and the first snippet loads OB fine.
At 11:24 AM -0600 1/31/11, Chris Muller apparently wrote:
Squeak4.2-10950.zip is now available at http://ftp.squeak.org/4.2, which is intended to be the actual-released 4.2 image (unless we find some problem, which we won't).
The trunk will be reopened shortly, thanks for your patience.
- Chris
In a fresh Squeak4.2-10950 running on Cogr2349 on Mac, the Save from Preference Browser does not appear to do anything. Same deal when running on Squeak 5.8b12.app
I find the powerful grey desktop background pretty depressing, could it have a little color? I like that the image apparently ships with Colorful Windows turned on.
Thx, Ken
On 1/31/11 9:24 AM, Chris Muller wrote:
Squeak4.2-10950.zip is now available at http://ftp.squeak.org/4.2,
Thanks! I've been away for a while and a quick test showed lots to like. Here's what I tried:
- run it with standard (3.8) vm on a two year old macbook, Help -> About this System -> Tiny Benchmarks:
492781520 bytecodes/sec; 10904180 sends/sec
- as above but with latest mac Cog vm found at http://www.mirandabanda.org/files/Cog/VM/ :
618731117 bytecodes/sec; 57063019 sends/sec. FWIW although these numbers are higher, the tiny benchmark took longer in wall clock time to complete (about 6s in cog vs 4s with the standard vm.)
In both vms mousing around feels very snappy, also there's a noticeable slight lag in unhighlighting the old item when mousing around on the world menu.
It was strange to find the Toggle full screen option in Projects menu, I kept looking in the Windows or squeak menus.
Best, -Simon
On 2/1/2011 10:12 PM, Simon Michael wrote:
On 1/31/11 9:24 AM, Chris Muller wrote:
Squeak4.2-10950.zip is now available at http://ftp.squeak.org/4.2,
Thanks! I've been away for a while and a quick test showed lots to like. Here's what I tried:
- run it with standard (3.8) vm on a two year old macbook, Help -> About
this System -> Tiny Benchmarks:
492781520 bytecodes/sec; 10904180 sends/sec
- as above but with latest mac Cog vm found at
http://www.mirandabanda.org/files/Cog/VM/ :
618731117 bytecodes/sec; 57063019 sends/sec. FWIW although these numbers are higher, the tiny benchmark took longer in wall clock time to complete (about 6s in cog vs 4s with the standard vm.)
That can happen. If you look at the code, the benchmark adjusts itself to ensure it runs for a certain amount of time otherwise the results would be all over the place. That's why the Cog version can take more total time - it simply has run a *lot* more of the benchmark code ;-)
Cheers, - Andreas
In both vms mousing around feels very snappy, also there's a noticeable slight lag in unhighlighting the old item when mousing around on the world menu.
It was strange to find the Toggle full screen option in Projects menu, I kept looking in the Windows or squeak menus.
Best, -Simon
Chris Muller <asqueaker <at> gmail.com> writes:
Squeak4.2-10950.zip is now available at http://ftp.squeak.org/4.2,
Thanks for your work! A small remark: I follow the "How to extend your system" information and load OmniBrowser. But I get an error saying: 'Could not resolve: OCForOB [OCForOB-rr.2] in /..../sq4.2-10950/package-cache http://source.wiresong.ca/ob'.
What could be the problem?
Kind regards, Pascal
On 2 February 2011 00:11, Pascal Vollmer pascal.vollmer@nexgo.de wrote:
Chris Muller <asqueaker <at> gmail.com> writes:
Squeak4.2-10950.zip is now available at http://ftp.squeak.org/4.2,
Thanks for your work! A small remark: I follow the "How to extend your system" information and load OmniBrowser. But I get an error saying: 'Could not resolve: OCForOB [OCForOB-rr.2] in /..../sq4.2-10950/package-cache http://source.wiresong.ca/ob'.
What could be the problem?
See Dale's comment.
Kind regards, Pascal
With fresh Squeak4.2-10950, doing cmd-click no longer pulls up the halos on Mac, does not seem to matter which VM.
Ken
At 11:24 AM -0600 1/31/11, Chris Muller apparently wrote:
Squeak4.2-10950.zip is now available at http://ftp.squeak.org/4.2, which is intended to be the actual-released 4.2 image (unless we find some problem, which we won't).
The trunk will be reopened shortly, thanks for your patience.
- Chris
squeak-dev@lists.squeakfoundation.org
-
Alexander Lazarević -
Andreas Raab -
Chris Muller -
Chris Muller -
Craig Latta -
Dale Henrichs -
David T. Lewis -
Facundo Vozzi -
Igor Stasenko -
Janko Mivšek -
Jeff G -
Ken G. Brown -
Levente Uzonyi -
Michael Davies -
Pascal Vollmer -
Simon Michael