Planet Squeak Code

empty log message

by Nicolas Passerini (npasserini@gmail.com) at January 28, 2012 01:32 AM

empty log message

by Nicolas Passerini (npasserini@gmail.com) at January 28, 2012 12:59 AM

empty log message

by Nicolas Passerini (npasserini@gmail.com) at January 28, 2012 12:58 AM

empty log message

by Nicolas Passerini (npasserini@gmail.com) at January 28, 2012 12:39 AM

empty log message

by Nicolas Passerini (npasserini@gmail.com) at January 28, 2012 12:24 AM

empty log message

by Nicolas Passerini (npasserini@gmail.com) at January 28, 2012 12:23 AM

empty log message

by Nicolas Passerini (npasserini@gmail.com) at January 28, 2012 12:23 AM

-optimizations

by Esteban Lorenzano (estebanlm@gmail.com) at January 27, 2012 11:21 PM

-DateAndTime is a valid mongo type
-added a small optimization on UTF8 codec

by Esteban Lorenzano (estebanlm@gmail.com) at January 27, 2012 11:19 PM

Added calculation for centering text inside the tooltip.

by Matti J. Kärki (mjk@iki.fi) at January 27, 2012 11:07 PM

Initial version. A rectangle tooltip. Subclass of BalloonMorph. To change the tooltip for the Squeak environment, edit the showBalloon:hand: method in the Morph class.

by Matti J. Kärki (mjk@iki.fi) at January 27, 2012 10:15 PM

empty log message

by Sebastian Sastre (seb@sebastianconcept.com) at January 27, 2012 06:58 PM

add mappings for window change in mac

by Debora Fortini (deborafortini@gmail.com) at January 27, 2012 06:56 PM

VSAManager has been given the ability to know if its operations have comparable achievement numbers. It will test for this when running calculations. Calculations will return a nil if the subordinate operations do not have comparable achievement. VSA>>operationsCompare: has been modified to run the myManager instance operationsCompare: method so both the VSA and related myManager instance have the same setting.

by Robert Bruce Carleton (rbc@rbcarleton.com) at January 27, 2012 05:49 PM

added lost inst methods

January 27, 2012 04:58 PM

empty log message

by Sebastian Sastre (seb@sebastianconcept.com) at January 27, 2012 04:44 PM

Added OARiakSession class >> newDefault

by nicolas petton (petton.nicolas@gmail.com) at January 27, 2012 04:39 PM

Removed OakSession>>connection

by nicolas petton (petton.nicolas@gmail.com) at January 27, 2012 04:38 PM

Trivial printon for buckets.

by nicolas petton (petton.nicolas@gmail.com) at January 27, 2012 04:36 PM

empty log message

by nicolas petton (petton.nicolas@gmail.com) at January 27, 2012 04:30 PM

empty log message

by nicolas petton (petton.nicolas@gmail.com) at January 27, 2012 04:29 PM

Removed old unused exception classes

by nicolas petton (petton.nicolas@gmail.com) at January 27, 2012 04:27 PM

initial version for pharo

January 27, 2012 04:26 PM

Removed OARiakSession>>serverError

by nicolas petton (petton.nicolas@gmail.com) at January 27, 2012 04:25 PM

January 27, 2012 04:25 PM

initial version for pharo

January 27, 2012 04:25 PM

Added OASession>>commit

by nicolas petton (petton.nicolas@gmail.com) at January 27, 2012 04:25 PM

Use operations in transactions: save, update and delete

by nicolas petton (petton.nicolas@gmail.com) at January 27, 2012 04:23 PM

empty log message

by nicolas petton (petton.nicolas@gmail.com) at January 27, 2012 04:22 PM

Cleanup

by nicolas petton (petton.nicolas@gmail.com) at January 27, 2012 03:26 PM

Check for oid not found

by nicolas petton (petton.nicolas@gmail.com) at January 27, 2012 03:19 PM

First commit

by nicolas petton (petton.nicolas@gmail.com) at January 27, 2012 03:02 PM

Riak backend :)

by nicolas petton (petton.nicolas@gmail.com) at January 27, 2012 03:01 PM

Test FileDirectory>relativeNameIfAbsoluteFor:

by Eliot Miranda (eliot.miranda@gmail.com) at January 27, 2012 01:41 PM

Provide FileDirectory>relativeNameIfAbsoluteFor:,
which does what I expect relativeNameFor: to do.

by Eliot Miranda (eliot.miranda@gmail.com) at January 27, 2012 01:40 PM

working on AprovaFacil

by Sebastian Sastre (seb@sebastianconcept.com) at January 27, 2012 01:06 PM

Re-implemented some metrics using moose chef

January 27, 2012 01:02 PM

Re-implemented some metrics using moose chef

January 27, 2012 01:01 PM

empty log message

by jimmyscratchlab (jimmyscratchlab@gmail.com) at January 27, 2012 12:56 PM

Give Get implemented

AccountView does not reflect it yet

Development Dump

by Ramshreyas Rao (ramshreyas.rao@gmail.com) at January 27, 2012 12:23 PM

Give - Get implemented

Account not reflecting this yet.

Development Dump

by Ramshreyas Rao (ramshreyas.rao@gmail.com) at January 27, 2012 12:22 PM

added handled stuff + fixed copy paste error

January 27, 2012 12:21 PM

event model changes to support new keydown and keyup

January 27, 2012 12:14 PM

- add menus on lists

by Benjamin Van Ryseghem (benjamin.vanryseghem@gmail.com) at January 27, 2012 11:33 AM

- minor fixes
- add some protocols to improve the communocation between the elements

by Benjamin Van Ryseghem (benjamin.vanryseghem@gmail.com) at January 27, 2012 11:11 AM

changes path

by Andre Hora (andrehoraa@gmail.com) at January 27, 2012 10:01 AM

+ comments.

by Andre Hora (andrehoraa@gmail.com) at January 27, 2012 09:58 AM

Fixed string escaping.

by Michael Rüfenacht (m.ruefenacht@students.unibe.ch) at January 27, 2012 09:54 AM

2 image test successful

January 27, 2012 09:47 AM

Beginning work to upgrade implementation to STOMP 1.1 by using new ActiveMQ Apollo as reference server.

January 27, 2012 09:16 AM

This version adds more testing changes. Temporary files created during unit testing are cleaned up.

by Robert Bruce Carleton (rbc@rbcarleton.com) at January 27, 2012 07:10 AM

empty log message

by Nicolas Passerini (npasserini@gmail.com) at January 27, 2012 05:52 AM

empty log message

by Nicolas Passerini (npasserini@gmail.com) at January 27, 2012 05:52 AM

empty log message

by Nicolas Passerini (npasserini@gmail.com) at January 27, 2012 05:51 AM

empty log message

by Nicolas Passerini (npasserini@gmail.com) at January 27, 2012 03:17 AM

This version provides a symetric equality
(1 i = 1 j k) = (1 j k = 1 i) and:
[(-1 k k = 1) = (1 = -1 k k)]

Due to implementation of Number>>#=, this is at the price that every number extension (Complex Quaternion Octonion ...) answer true to isNumber.

This is incompatible with most senders of isNumber which don't expect an imaginary one and thus highly questionable.

by nicolas cellier (nicolas.cellier.aka.nice@gmail.com) at January 27, 2012 02:02 AM

Correct hash so that it fits equal Number / Complex hash.
Let 1 j = 2 answer false rather than DNU.

by nicolas cellier (nicolas.cellier.aka.nice@gmail.com) at January 27, 2012 01:54 AM

Test equality and hash.
We want aQuaternion = aQuaternion reduce,
that is a quaternion with null unreal part shall equal its real part,
and a quaternion with null j and k imaginary parts shall equal corresponding complex.

If they can equal a complex or a number, they also must have an equal hash.

Note that due to current implementation of Number>>#= and Complex>>#=,
a Quaternion should answer true to isNumber to make equality a symetric relation.
This sounds bad, doesn't it ?

by nicolas cellier (nicolas.cellier.aka.nice@gmail.com) at January 27, 2012 01:46 AM

Since the tests tend to break upon small changes i decided to split the bigger stuff up to get smaller chunks. Even methods that worked got their own tests now to see where the reimplementation fails.

by Michael Rüfenacht (m.ruefenacht@students.unibe.ch) at January 27, 2012 01:28 AM

Fixed the for statement and tried to find a workaround for escaped strings but obviously wasnt able to. So i modified the testSymfonyHttpRequestPrepareBasePathIfStatement2
Test to make all green before quitting today.

by Michael Rüfenacht (m.ruefenacht@students.unibe.ch) at January 27, 2012 01:27 AM

Add tests for the functions sin, cos, tan, exp, ln, sinh, cosh, tanh, raisedTo:

Notably, the quaternion functions are extensions of the complex functions.
(1 + 2 i) is a complex
(1 + 2 j k) is a quaternion = (1 + 2 i)
Hence we must have f(1 + 2 i) = f(1 + 2 j k)

by nicolas cellier (nicolas.cellier.aka.nice@gmail.com) at January 27, 2012 01:01 AM

Correct functions #sin and #cos
Remove #exp dependency on #sinc function

by nicolas cellier (nicolas.cellier.aka.nice@gmail.com) at January 27, 2012 12:52 AM

Correct bug: a return was missing in most functions (sin cos tan exp ln)

by nicolas cellier (nicolas.cellier.aka.nice@gmail.com) at January 26, 2012 11:55 PM

Added some tests (some where unnecessary and failed because of incorrect string formatting with ' or ").

by Michael Rüfenacht (m.ruefenacht@students.unibe.ch) at January 26, 2012 10:58 PM

Fixed a lot of small stuff that tend to break the expression parser.

by Michael Rüfenacht (m.ruefenacht@students.unibe.ch) at January 26, 2012 10:57 PM

Initial commit.

by Falko Wiese (fwiese.wse@gmail.com) at January 26, 2012 10:34 PM

And when you load the config package, actually load SqueakCheck!

by Frank Shearar (frank@angband.za.org) at January 26, 2012 09:36 PM

Widget adaptations needed for Aida 6.4

by Janko Mivšek (janko.mivsek@eranova.si) at January 26, 2012 08:27 PM

Added some keywords and some dumb ass stuff i forgot before.

by Michael Rüfenacht (m.ruefenacht@students.unibe.ch) at January 26, 2012 07:54 PM

- fix a bug getting the current sender

by Daniel Galdames (lemuus@gmail.com) at January 26, 2012 07:54 PM

- fix a bug getting the current sender

by Daniel Galdames (lemuus@gmail.com) at January 26, 2012 07:54 PM

- fix a bug getting the current sender

by Daniel Galdames (lemuus@gmail.com) at January 26, 2012 07:54 PM

Forgot some stuff in the language constructs which broke the new tests. Fixed now.

by Michael Rüfenacht (m.ruefenacht@students.unibe.ch) at January 26, 2012 07:53 PM

another fix for 2-image test

January 26, 2012 07:31 PM

MethodWrappers for Pharo1.4

by Daniel Galdames (lemuus@gmail.com) at January 26, 2012 06:37 PM

2-image example corrected

January 26, 2012 06:32 PM

Once again fix outputDir for windows builds. DO NOT CHANGE THAT.

by Mariano Martinez Peck (marianopeck@gmail.com) at January 26, 2012 06:32 PM

Use RGMethodDefinition instead of MethodReference

by Daniel Galdames (lemuus@gmail.com) at January 26, 2012 06:27 PM

Added comments.

by Andre Hora (andrehoraa@gmail.com) at January 26, 2012 06:10 PM

Added #FBDirectoryRepository

by Andre Hora (andrehoraa@gmail.com) at January 26, 2012 06:09 PM

Examples corrected for 2-image test

January 26, 2012 06:00 PM

New version 3.9 that fixes the sound package to load in latest Pharo 1.4 and fixes OSProcess dependencies in baseline 2.0

January 26, 2012 05:52 PM

Another day another test. Today: the symfony2 event dispatcher.

by Michael Rüfenacht (m.ruefenacht@students.unibe.ch) at January 26, 2012 05:32 PM

kongruent zum Arbeitspapier Januar 2012
aufgeräumt
Tests verbessert

January 26, 2012 05:24 PM

empty log message

by Michael Rüfenacht (m.ruefenacht@students.unibe.ch) at January 26, 2012 05:17 PM

added isStatic property

by Michael Rüfenacht (m.ruefenacht@students.unibe.ch) at January 26, 2012 05:16 PM

Startet refining the properties. I am not shure if it wouldnt be better to refactor at a deeper level to get objects out of the expressions.

by Michael Rüfenacht (m.ruefenacht@students.unibe.ch) at January 26, 2012 05:16 PM

empty log message

by Marc Mooser (marcmooser@bluewin.ch) at January 26, 2012 05:12 PM

empty log message

by Marc Mooser (marcmooser@bluewin.ch) at January 26, 2012 05:11 PM

empty log message

by Marc Mooser (marcmooser@bluewin.ch) at January 26, 2012 05:11 PM

installTagsForPresent:

by Marc Mooser (marcmooser@bluewin.ch) at January 26, 2012 05:11 PM

new version with the browser

January 26, 2012 04:45 PM

empty log message

January 26, 2012 04:42 PM

Adding version of rules in the manifest

January 26, 2012 04:41 PM

updated rules.

by Andre Hora (andrehoraa@gmail.com) at January 26, 2012 04:15 PM

Comments on EMMAExtractor class

by Usman Bhatti (usman.bhatti@gmail.com) at January 26, 2012 03:59 PM

empty log message

January 26, 2012 03:48 PM

Hocked the babel entities to MooseEntity

by Fabrizio Perin (fabrizio.perin@gmail.com) at January 26, 2012 03:46 PM

added entities: MJDPAssociation and MJDPEntity to organize the element of the model

by Fabrizio Perin (fabrizio.perin@gmail.com) at January 26, 2012 03:45 PM

-small bugfixes
-added nicedit component to rich text editors

January 26, 2012 03:44 PM

-hack to allow passing blocks on containers (not just views), so a simple "refresh" would change the contents (no need to remove/add refreshed subcomponents)
I'm not happy with this way to solve this problem, but is a first step.
-added #call: and #show: protocols to reef views... they just pass the call to container seaside component

January 26, 2012 03:42 PM

Add test for EMMAExtractor

by Usman Bhatti (usman.bhatti@gmail.com) at January 26, 2012 03:23 PM

make a special conf for pharo.1.3.x, because apparently Pharo 1.4.x already includes AST-Core

by Daniel Galdames (lemuus@gmail.com) at January 26, 2012 02:58 PM

Added primGetSDCardRoot primitive

by Dmitry Golubovsky (golubovsky@gmail.com) at January 26, 2012 02:46 AM

Added getSDCardRoot primitive

by Dmitry Golubovsky (golubovsky@gmail.com) at January 26, 2012 02:42 AM

Fix for package renaming.

by Chris Muller (ma.chris.m@gmail.com) at January 25, 2012 06:58 PM

move asStringOrText conversion from drawOn: (the caller)
to display:atRow:on: (the called) to avoid destroying column
info used by the MulticolumnLazyListMorph.

by Chris Muller (ma.chris.m@gmail.com) at January 25, 2012 06:57 PM

Fix for package renaming.

by Chris Muller (ma.chris.m@gmail.com) at January 25, 2012 06:56 PM

Merged.

by Chris Muller (ma.chris.m@gmail.com) at January 25, 2012 06:56 PM

Merge fix for MulticolumnLazyListMorph from Christopher Oliver.

by Chris Muller (ma.chris.m@gmail.com) at January 25, 2012 06:54 PM

When creating new Forms, use "self species" rather than "self class" so that Forms behind a WriteBarrier will create new instances of the correct class.

by Chris Muller (ma.chris.m@gmail.com) at January 25, 2012 06:45 PM

Resave merged version for highest version number.

by Chris Muller (ma.chris.m@gmail.com) at January 25, 2012 06:44 PM

Add some coverage analysis to check for unimplemented opcodes
in the Cogit. Implement AndCwR & OrCwR in CogIA32Compiler.

by Eliot Miranda (eliot.miranda@gmail.com) at January 24, 2012 10:50 PM

- Fixed Integer >> #isPowerOfTwo which returned true for 0. Added optimized versions to Integer's subclasses.
- Updated MethodDictionary class >> #sizeFor: to be error resistant if the semantics of #asLargerPowerOfTwo would change.

by Levente Uzonyi (leves@elte.hu) at January 23, 2012 02:53 PM

- Fixed and enhanced IntegerTest >> #testIsPowerOfTwo.
- Removed IntegerTest >> #testIsPowerOfTwoM6873, because #testIsPowerOfTwo now includes the same checks.

by Levente Uzonyi (leves@elte.hu) at January 23, 2012 02:46 PM

merge and correct Kernel-nice.643
(3/2)/(-3/4) was bogus

Reminder: these changes should speed up mixed Integer-Fraction arithmetic
(they avoid creating intermediate Fraction)

January 22, 2012 06:22 PM

Add a few tests about fraction (denominator is allways positive and arithmetic results are automatically reduced)

by Nicolas Cellier (nicolas.cellier.aka.nice@gmail.com) at January 22, 2012 06:16 PM

add a test for bitCount

by Nicolas Cellier (nicolas.cellier.aka.nice@gmail.com) at January 22, 2012 05:23 PM

merge Kernel-nice.644

Add Integer>>bitCount to count the bits set (to 1) in an Integer.

by Nicolas Cellier (nicolas.cellier.aka.nice@gmail.com) at January 22, 2012 05:13 PM

Add Integer>>bitCount to count the bits set (to 1) in an Integer.

by Nicolas Cellier (nicolas.cellier.aka.nice@gmail.com) at January 22, 2012 05:09 PM

Update reference to obsolete TextMorphEditor with a plausible looking replacement, replacing "self class name = #TextMorphEditor" with "self isKindOf: TextEditor". This eliminates the last direct reference to TextMorphEditor in the image.

by David T. Lewis (lewis@mail.msen.com) at January 21, 2012 11:29 PM

Provide a stub implementation of TextMorphEditor, subclassed from its replacement TextEditor, for the benefit of external packages such as Connectors that depend on it.

Note that an existing Connectors installation may have been damaged in the trunk update process. Check for this by opening a Text Rectangle or similar from the Connectors tab, then hit <esc> while focused on the text region. If problems are experienced, reload Connectors and evaluate 'NCTextMorphEditor initialize'.

by David T. Lewis (lewis@mail.msen.com) at January 21, 2012 11:05 PM

merge Kernel-nice.656

Remove one unused message: now that Integer is polymorphic to Fraction he never has to #adaptToFraction:andSend:

by Nicolas Cellier (nicolas.cellier.aka.nice@gmail.com) at January 21, 2012 03:45 PM

Remove one unused message: now that Integer is polymorphic to Fraction he never has to #adaptToFraction:andSend:

by Nicolas Cellier (nicolas.cellier.aka.nice@gmail.com) at January 21, 2012 03:42 PM

removed defaultToLF send

by Igor Stasenko (siguctua@gmail.com) at January 17, 2012 06:35 PM

* Modified the way the locale changes are notified to morphs. Basically, Project>>#updateLocaleDependentsWithPreviousSupplies:gently: now sends #localeChanged to all morphs and the default implementation in Morph does nothing.
* Added Scott fix to retain the old order (first all morphs that are not top-level ScriptEditorMorphs)

by Ricardo Moran (richi.moran@gmail.com) at January 13, 2012 06:28 PM

* Added #localeChanged to Morph with an empty implementation. Subclasses that want to get notified of a locale change should override this method.

by Ricardo Moran (richi.moran@gmail.com) at January 13, 2012 06:20 PM

http://tracker.squeakland.org/browse/SQ-1005
"Open script" halo handle for ScriptActivationButton

by Karl Ramberg (karlramberg@gmail.com) at January 13, 2012 05:09 PM

CogUnixNoGLConfig now does not include B3DAcceleratorPlugin.

by Mariano Martinez Peck (marianopeck@gmail.com) at January 13, 2012 01:13 PM

Nuke TextMorphEditor which is not used anymore and inherits
from ParagraphEditor (hence avoiding problems when
ParagraphEditor is unloaded as part of the ST80 package).

by Eliot Miranda (eliot.miranda@gmail.com) at January 11, 2012 09:23 PM

http://tracker.squeakland.org/browse/SQ-984

by Karl Ramberg (karlramberg@gmail.com) at January 11, 2012 10:28 AM

-merged

by Esteban Lorenzano (estebanlm@gmail.com) at January 10, 2012 03:36 PM

Added primitives to get and set VM idle timer interval.

by Dmitry Golubovsky (golubovsky@gmail.com) at January 10, 2012 04:02 AM

Android VM idle timer can now be configured via Pharo settings browser.

by Dmitry Golubovsky (golubovsky@gmail.com) at January 10, 2012 03:55 AM

Enhanced pointer tracing:
- handle SmallIntegers correctly (they don't point to any object)
- handle instances of compact classes correctly (they don't point to their class)
- weak references are ignored, because they don't stop the garbage collector in collecting objects
These changes modify the behavior of #pointsTo:, #outboundPointersDo: and #inboundPointersExcluding:.

January 09, 2012 01:47 AM

VMMaker 4.7.20

Define HAVE_INTERP_H in the generated interp.h file to enable check for included file, e.g. from sqMemoryAccess.h.

Check argument count in primitivePerformInSuperclass, fail primitive if not 3 or 4.

by David T. Lewis (lewis@mail.msen.com) at January 08, 2012 03:15 PM

asColorref was using the color's green element twice, instead of red, green, and blue. Also, Color did not have a symmetrical constructor to create a Color from a colorref. Finally, converting from a float to an integer by using asInteger, rather than round seemed like the wrong thing to do.

I tested this by iterating through Color class>>indexedColors, and converting them to colorrefs, then creating a Color from the colorref, and comparing the original with the new color. Without rounding I received ~239 failures and with rounding I received ~21 (still to be expected as we are converting a float in the range 0.0 - 1.0 to an integer in the range 0 - 255, so there will still be some error).

January 06, 2012 09:38 PM

- add missing Win32WindowProxy

by Bert Freudenberg (bert@freudenbergs.de) at January 06, 2012 01:44 PM

Pick up halo to behave the same as a red-button grab.

January 06, 2012 04:12 AM

Update method comment for a test that documents a limitation of the interpreter VM (resolved in Cog)

by David T. Lewis (lewis@mail.msen.com) at January 06, 2012 02:09 AM

Add host window support. This is based on the stubs found in Pharo, with various fixes and extensions by me. In particular, I added event handling support. See examplePaint in DisplayHostWindow class.

by Bert Freudenberg (bert@freudenbergs.de) at January 05, 2012 05:49 PM

- dispatch events to HostWindowProxy
- add EventSensorConstants for window events and touch events

by Bert Freudenberg (bert@freudenbergs.de) at January 05, 2012 05:41 PM

VMMaker 4.7.19

Reference Mantis 7429: Add Mirror Primitives to the VM

Update primitivePerformInSuperclass to support ContextPart>>object:perform:withArguments:inClass:

Implementation differs from that of oscog in that the original primitivePerformAt: is retained unmodified, and the necessary stack adjustments are done in primitivePerformInSuperclass for the special case of argumentCount 4 (mirror primitive call) rather than 3. The oscog approach may be prefered (not least for its clearer method naming), but making the change in primitivePerformInSuperclass is low risk and more easily implemented by a Sunday Squeaker.

All MirrorPrimitiveTests pass.

by David T. Lewis (lewis@mail.msen.com) at January 05, 2012 05:41 AM

Included LocalePlugin in the VM configuration

by Dmitry Golubovsky (golubovsky@gmail.com) at January 05, 2012 04:52 AM

Fix.

by Chris Muller (ma.chris.m@gmail.com) at January 05, 2012 02:17 AM

-working on CMakeMaker for iPhone (Still not working)
-small changes

by Esteban Lorenzano (estebanlm@gmail.com) at January 02, 2012 06:57 PM

fix unpleasant linebreak

by Chris Cunnington (smalltalktelevision@gmail.com) at January 02, 2012 03:36 PM

- when updating from a new repository of add-on packages (e.g. VMMaker), skip ahead to the latest config

by Chris Cunnington (smalltalktelevision@gmail.com) at January 02, 2012 03:36 PM

deprecate Number class>>#readExactlyFrom:
#readFrom: now does the same job.
This was just an interim message for smooth transition introduced in 3.9. Anyway, who would request a readInexactly feature?

by Chris Cunnington (smalltalktelevision@gmail.com) at January 02, 2012 03:36 PM

- fix: SocketStream >> #upTo: returned at most 100000 bytes/characters.
- fix: Added a workaround to Socket >> #waitForDataIfClosed: for the Cog bug, which causes lost signals of Sockets' readSemaphore .

by Chris Cunnington (smalltalktelevision@gmail.com) at January 02, 2012 03:36 PM

minor refactorings: use #anySatisfy: #allSatisfy: #noneSatisfy: where it simplifies

by Chris Cunnington (smalltalktelevision@gmail.com) at January 02, 2012 03:35 PM

- use #becomeForward: instead of #become:

by Chris Cunnington (smalltalktelevision@gmail.com) at January 02, 2012 03:35 PM

Added test for handling Notification inside a progress block.

by Chris Cunnington (smalltalktelevision@gmail.com) at January 02, 2012 03:35 PM

Removed unnecessary blocks.

by Chris Cunnington (smalltalktelevision@gmail.com) at January 02, 2012 03:35 PM

Add Juan's factorial example to illustrate variable scoping for block closures

by Chris Cunnington (smalltalktelevision@gmail.com) at January 02, 2012 03:35 PM

Updated DependencyBrowserTest >> #testClassList, because String >> #numArgs now also depend on String.

by Chris Cunnington (smalltalktelevision@gmail.com) at January 02, 2012 03:35 PM

Force local cache refresh when saving a new package release.

by Chris Cunnington (smalltalktelevision@gmail.com) at January 02, 2012 03:35 PM

Clean up after Smalltalk/SystemDictionary refactoring.

by Chris Cunnington (smalltalktelevision@gmail.com) at January 02, 2012 03:35 PM

Add a test for the default clipRect in a FormCanvas.

by Chris Cunnington (smalltalktelevision@gmail.com) at January 02, 2012 03:35 PM

Force local cache refresh when saving a new package release.

by Chris Cunnington (smalltalktelevision@gmail.com) at January 02, 2012 03:35 PM

Cosmetic: move or remove a few temps inside closures

by Chris Cunnington (smalltalktelevision@gmail.com) at January 02, 2012 03:35 PM

- show floats when exploring transforms

by Chris Cunnington (smalltalktelevision@gmail.com) at January 02, 2012 03:35 PM

- cache read only source files during some tests

by Chris Cunnington (smalltalktelevision@gmail.com) at January 02, 2012 03:35 PM

I'm cleaning house before the release of 4.3. I have a dirty package and I'm pushing it out to the trunk before final build. I added nothing to this, but it must be different somehow, as it's dirty.

by Chris Cunnington (smalltalktelevision@gmail.com) at January 02, 2012 03:35 PM

Make PreferenceBrowser window wider so even the longest default preference will not be cut off

by Chris Cunnington (smalltalktelevision@gmail.com) at January 02, 2012 03:35 PM

Improved UUIDTest >> #testComparison.

by Chris Cunnington (smalltalktelevision@gmail.com) at January 02, 2012 03:35 PM

Add test documenting a problem with PNGReadWriter.

by Chris Cunnington (smalltalktelevision@gmail.com) at January 02, 2012 02:56 PM

Cover changes in Files-eem.115 with some simple test

by Chris Cunnington (smalltalktelevision@gmail.com) at January 01, 2012 11:25 PM

Copied to source.squeak.org: will now host here.

by Dmitry Golubovsky (golubovsky@gmail.com) at January 01, 2012 03:19 AM

Copy to source.squeak.org: will now host here.

by Dmitry Golubovsky (golubovsky@gmail.com) at January 01, 2012 03:15 AM

Added primClipboardString: obtains a string from Android clipboard if there is any. Returns an empty string if nothing is available.

by Dmitry Golubovsky (golubovsky@gmail.com) at January 01, 2012 03:10 AM

Fix weak finalization thrashing, root cause of user interrupt issues.

Perform the check for VM support of new finalization once at image startUp time. This prevents the weak finalization process from creating new weak references that result in thrashing between VM and image as the VM signals the the image to clean up weak references, and the finalization process produces additional weak to be removed.

With this change the finalization process is much less active, and user interrupt handling works as intended such that any of the following can interrupted in the expected way:

"[true] whileTrue"
"[[true] whileTrue] forkAt: Processor userSchedulingPriority + 1"
"Smalltalk createStackOverflow"
"[Smalltalk createStackOverflow] forkAt: Processor userSchedulingPriority + 1"

by David T. Lewis (lewis@mail.msen.com) at December 30, 2011 05:59 AM

Fix weak finalization thrashing, root cause of user interrupt issues.

Move the check for VM support of new finalization out of the finalization process and perform the check once at image startUp time. This prevents the weak finalization process from creating new weak references that result in thrashing between VM and image as the VM signals the the image to clean up weak references, and the finalization process produces additional weak to be removed.

With this change the finalization process is much less active, and user interrupt handling works as intended such that any of the following can interrupted in the expected way:

"[true] whileTrue"
"[[true] whileTrue] forkAt: Processor userSchedulingPriority + 1"
"Smalltalk createStackOverflow"
"[Smalltalk createStackOverflow] forkAt: Processor userSchedulingPriority + 1"

by David T. Lewis (lewis@mail.msen.com) at December 30, 2011 05:59 AM

Fix weak finalization thrashing, root cause of user interrupt issues.

Perform the check for VM support of new finalization once at image startUp time. This prevents the weak finalization process from creating new weak references that result in thrashing between VM and image as the VM signals the the image to clean up weak references, and the finalization process produces additional weak to be removed.

With this change the finalization process is much less active, and user interrupt handling works as intended such that any of the following can interrupted in the expected way:

"[true] whileTrue"
"[[true] whileTrue] forkAt: Processor userSchedulingPriority + 1"
"Smalltalk createStackOverflow"
"[Smalltalk createStackOverflow] forkAt: Processor userSchedulingPriority + 1"

by David T. Lewis (lewis@mail.msen.com) at December 30, 2011 05:58 AM

SystemOrganizer>>classify:under: now uses #at:ifAbsent: to look up a class when generating system change notifications. This prevents bogus change notifications and walk backs when running tests against SystemOrganizer instances.

by Colin Putney (cputney@wiresong.ca) at December 29, 2011 06:44 AM

Allow subclasses to override the internal 'array' of a HashedCollection.

by Chris Muller (ma.chris.m@gmail.com) at December 26, 2011 06:57 PM

Merged.

by Chris Muller (ma.chris.m@gmail.com) at December 26, 2011 06:56 PM

- When using the new list-filtering, if the user's keystroke will cause filtering to an empty list, then flash rather than present an empty list.
- Preserve 300ms "slow-keystroke" timing for when the legacy preference is engaged for filterableLists. That way, in case someone wants to press, t -- t -- t to go to the third in the list that starts with 't', they will only have to wait 300ms in between presses.
- Order the items with front-matching items at the top of the list, and substring-matches below that.

by Chris Muller (ma.chris.m@gmail.com) at December 24, 2011 08:49 PM

Fix weak finalization thrashing, root cause of user interrupt issues.

Perform the check for VM support of new finalization once at image startUp time. This prevents the weak finalization process from creating new weak references that result in thrashing between VM and image as the VM signals the the image to clean up weak references, and the finalization process produces additional weak to be removed.

With this change the finalization process is much less active, and user interrupt handling works as intended such that any of the following can interrupted in the expected way:

"[true] whileTrue"
"[[true] whileTrue] forkAt: Processor userSchedulingPriority + 1"
"Smalltalk createStackOverflow"
"[Smalltalk createStackOverflow] forkAt: Processor userSchedulingPriority + 1"

by David T. Lewis (lewis@mail.msen.com) at December 23, 2011 03:39 PM

Install a rice paper pattern as the default morphic project background.

This is based on http://webtreats.mysitemyway.com/seamless-paper-patterns/

From the readme.txt:

"This resource is free for personal and commercial use ~ no attribution necessary."

by Alexander Lazarević (Alexander@Lazarevic.de) at December 23, 2011 10:06 AM

fix to invocation of file-specific "services" from contents pane of a FileList

December 23, 2011 08:32 AM

Fix needed to make removeElement: work in subclass XMLElement

December 23, 2011 06:10 AM

Fix weak finalization thrashing, root cause of user interrupt issues.

Move the check for VM support of new finalization out of the finalization process and perform the check once at image startUp time. This prevents the weak finalization process from creating new weak references that result in thrashing between VM and image as the VM signals the the image to clean up weak references, and the finalization process produces additional weak to be removed.

With this change the finalization process is much less active, and user interrupt handling works as intended such that any of the following can interrupted in the expected way:

"[true] whileTrue"
"[[true] whileTrue] forkAt: Processor userSchedulingPriority + 1"
"Smalltalk createStackOverflow"
"[Smalltalk createStackOverflow] forkAt: Processor userSchedulingPriority + 1"

by David T. Lewis (lewis@mail.msen.com) at December 22, 2011 09:19 PM

Final spacing for Welcome windows.

by Chris Cunnington (smalltalktelevision@gmail.com) at December 22, 2011 09:18 PM

- Don't attempt to resize screen until we're using the correct primitives to do so!
- Revert standard size to 800x600.

by Chris Muller (ma.chris.m@gmail.com) at December 22, 2011 07:48 PM

Force local cache refresh when saving a new package release.

by Chris Muller (ma.chris.m@gmail.com) at December 22, 2011 06:38 PM

Force local cache refresh when saving a new package release.

by Chris Muller (ma.chris.m@gmail.com) at December 22, 2011 06:38 PM

fix unpleasant linebreak

by Alexander Lazarević (Alexander@Lazarevic.de) at December 22, 2011 04:18 PM

Add Google+ Page

by Alexander Lazarević (Alexander@Lazarevic.de) at December 22, 2011 08:59 AM

How to extend the system -> OpenGL

by Alexander Lazarević (Alexander@Lazarevic.de) at December 22, 2011 08:50 AM

Stupid mistake on my part. Wrote 1024@758 when it should have been 1024@768 for #setDisplayExtent: in #prepareNewBuild

by Chris Cunnington (smalltalktelevision@gmail.com) at December 22, 2011 02:18 AM

Make all the Help>>Welcome Workspaces accessed by TheWorldMainDockingBar appear as a result of ReleaseBuilderTrunk>>prepareNewBuild.

by Chris Cunnington (smalltalktelevision@gmail.com) at December 22, 2011 01:31 AM

I'm cleaning house before the release of 4.3. I have a dirty package and I'm pushing it out to the trunk before final build. I added nothing to this, but it must be different somehow, as it's dirty.

by Chris Cunnington (smalltalktelevision@gmail.com) at December 21, 2011 08:49 PM

Merged heads

by Alexander Lazarević (Alexander@Lazarevic.de) at December 21, 2011 07:59 AM

fix (received from Scott Wallace) to long-standing bug in which a newly opened message list would not scroll the code to show the first search-term match

by Alexander Lazarević (Alexander@Lazarevic.de) at December 21, 2011 07:53 AM

fix to protocol browser's filtering by class-variable ref

by Alexander Lazarević (Alexander@Lazarevic.de) at December 21, 2011 07:53 AM

fix to protocol browser's filtering by class-variable ref

by Alexander Lazarević (Alexander@Lazarevic.de) at December 21, 2011 07:50 AM

fix (received from Scott Wallace) to long-standing bug in which a newly opened message list would not scroll the code to show the first search-term match

by Alexander Lazarević (Alexander@Lazarevic.de) at December 21, 2011 07:44 AM

- fix: even empty MethodDictionaries should have one slot

by Levente Uzonyi (leves@elte.hu) at December 21, 2011 03:30 AM

A pair of Welcome Workspaces added to TheWorldMainDockingBar. Welcome To Squeak 4.3. Future Directions.

by Levente Uzonyi (leves@elte.hu) at December 20, 2011 05:43 PM