Discussion:
Daily Commit Log
(too old to reply)
commits at source.squeak.org ()
2012-12-28 05:55:13 UTC
Permalink
Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours:

http://lists.squeakfoundation.org/pipermail/packages/2012-December/005603.html

Name: Morphic-dtl.633
Ancestors: Morphic-fbs.632

Fix for: Project current chooseNaturalLanguage ==> DNU

=============================================
commits at source.squeak.org ()
2012-12-31 05:55:21 UTC
Permalink
Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours:

http://lists.squeakfoundation.org/pipermail/packages/2012-December/005604.html

Name: System-nice.505
Ancestors: System-bf.504

Correct MessageTally spyOn: [] MNU on new COG VM

=============================================
commits at source.squeak.org ()
2013-01-01 05:55:06 UTC
Permalink
Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours:

http://lists.squeakfoundation.org/pipermail/packages/2012-December/005605.html

Name: System-fbs.506
Ancestors: System-nice.505

Update the Copyright notice.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2012-December/005606.html

Name: ReleaseBuilder-fbs.89
Ancestors: ReleaseBuilder-fbs.88

Transferring packages to the new update stream requires valid credentials. This lets the release manager transfer things without changing source, and without storing credentials in the image.

=============================================
commits at source.squeak.org ()
2013-01-02 05:55:04 UTC
Permalink
Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours:

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005607.html

Name: Kernel-nice.723
Ancestors: Kernel-nice.720

Speed-up LargeInteger asFloat when highBit > Float precision.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005608.html

Name: Kernel-nice.722
Ancestors: Kernel-nice.720

Let 0.0 negated answer Float negativeZero.

This let following code work:
self assert: (Compiler evaluate: '-0.0') hex = Float negativeZero hex

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005609.html

Name: Kernel-nice.724
Ancestors: Kernel-nice.723, Kernel-nice.722

Merge 0.0 negated

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005610.html

Name: Kernel-nice.725
Ancestors: Kernel-nice.724

Fast-up large integer modulo operations (\\ and rem:)

Implementation notes:
Quotient and remainder are computed in a single LargeIntegersPlugin primitive (see digitDiv:neg:) so it's faster to just use it.
For LargeInteger with 64 bits or less, LargeInteger primitives (31 32 33) are faster than the plugin (especially in COG) so try them first.
This results in a 2x speed up of modulo operations in 4.2.5 VM (whatever bit length), and a 2x speed up in COG VM for bit length > 64.
There is a penalty of 15% in COG for #rem: when bit length <= 64 because there is no primitiveRem...
Well, I added primitiveRem and it is in both VM branches, but it has no primitive number assigned.
If we assign a primitive number (20 ?) we can expect a 5x speed up for rem and bitLength <= 64.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005611.html

Name: ReleaseBuilder-fbs.90
Ancestors: ReleaseBuilder-fbs.89

Basic ReleaseBuilder for Squeak 4.5. Note that its releaseRepository is trunk. This is so that the 4.5 alpha builds will update from the right place. At release time, this must be changed to 'squeak45'.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005612.html

Name: Tests-fbs.178
Ancestors: Tests-ar.176

Now that we're in 4.5, it's time to think about fixing this.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005613.html

Name: Compiler-cwp.245
Ancestors: Compiler-eem.243

Environments bootstrap - stage 1

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005614.html

Name: Kernel-cwp.726
Ancestors: Kernel-nice.725

Environments bootstrap - stage 1

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005615.html

Name: System-cwp.507
Ancestors: System-fbs.506

Environments bootstrap - stage 1

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005616.html

Name: Environments-cwp.7
Ancestors: Environments-cwp.6

Environments bootstrap - stage 1

=============================================
commits at source.squeak.org ()
2013-01-03 05:55:09 UTC
Permalink
Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours:

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005617.html

Name: Compiler-cwp.246
Ancestors: Compiler-cwp.245

Environments bootstrap - stage 2

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005618.html

Name: Compiler-cwp.247
Ancestors: Compiler-cwp.246

Environments bootstrap - stage 3

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005619.html

Name: Environments-cwp.8
Ancestors: Environments-cwp.7

Environments bootstrap - stage 3

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005620.html

Name: Kernel-cwp.727
Ancestors: Kernel-cwp.726

Environments bootstrap - stage 3

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005621.html

Name: System-cwp.508
Ancestors: System-cwp.507

Environments bootstrap - stage 3

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005622.html

Name: Tools-cwp.433
Ancestors: Tools-eem.431

Environments bootstrap - stage 3

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005623.html

Name: Traits-cwp.292
Ancestors: Traits-ul.290

Environments bootstrap - stage 3

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005624.html

Name: Tests-fbs.179
Ancestors: Tests-fbs.178

Update PackageDependencyTest to reflect the new relationships between Environments and other packages.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005625.html

Name: Tests-fbs.180
Ancestors: Tests-fbs.179

This does _not_ fix the failing test. I'm renaming in a separate commit so that the actual fix is apparent (when someone does so).

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005626.html

Name: Collections-nice.497
Ancestors: Collections-bf.496

Thu shalt not invoke super in vain

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005627.html

Name: Tests-fbs.181
Ancestors: Tests-fbs.180

While valuable enough on its own, this test really aims to try nail down/understand the failure in testClassRemovalAndRecompilationWontCreateDuplicateVariableBindings.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005628.html

Name: Kernel-nice.728
Ancestors: Kernel-cwp.727

Fix http://bugs.squeak.org/view.php?id=7605.
Creating a new Float of size ~= 2 may crash the image.
With this change it is safe to create a Float new (0.0), no use to forbid it.
Thanks Levente

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005629.html

Name: EToys-nice.90
Ancestors: EToys-bf.89

Use at:ifAbsentPut: instead of at:ifAbsent: [ at:put: ]
This is an update of http://bugs.squeak.org/view.php?id=5667

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005630.html

Name: Kernel-nice.729
Ancestors: Kernel-nice.728

Use at:ifAbsentPut: instead of at:ifAbsent: [ at:put: ]
This is an update of http://bugs.squeak.org/view.php?id=5667

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005631.html

Name: System-nice.509
Ancestors: System-cwp.508

Use at:ifAbsentPut: instead of at:ifAbsent: [ at:put: ]
This is an update of http://bugs.squeak.org/view.php?id=5667

=============================================
commits at source.squeak.org ()
2013-01-04 05:55:05 UTC
Permalink
Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours:

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005632.html

Name: Sound-dtl.33
Ancestors: Sound-ul.32

Bitshift optimizations by Nicolas Cellier. Use #>> or #<&lt; operator rather than #bitShift: in cases where shift direction is known in order to eliminate runtime check for shift direction in compiled VM methods.

These changes affect ADPCMCodec methods that are translated as primitives in the ADPCMCodecPlugin and are thus part of the compiled VM. See VMMaker-dtl.293 in the VMMaker repository for related changes affecting other methods in the VM.

Discussion thread: http://lists.squeakfoundation.org/pipermail/vm-dev/2012-December/011688.html

Original change sets: http://code.google.com/p/cog/issues/detail?id=111

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005633.html

Name: Environments-fbs.9
Ancestors: Environments-cwp.8

The ReleaseSqueakTrunk CI job fails because it raises a debugger. This avoids that problem.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005634.html

Name: Tests-fbs.182
Ancestors: Tests-fbs.181

Environment >> #at:ifPresent:ifAbsent:.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005635.html

Name: Environments-fbs.10
Ancestors: Environments-fbs.9

Environment >> #at:ifPresent:ifAbsent:.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005636.html

Name: Environments-fbs.10
Ancestors: Environments-fbs.9

Environment >> #at:ifPresent:ifAbsent:.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005637.html

Name: Environments-fbs.11
Ancestors: Environments-fbs.10

For the moment - until we're sure about the proper API for environments - put things that mimic Dictionary methods in the compatibility protocol.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005638.html

Name: ToolsTests-fbs.57
Ancestors: ToolsTests-fbs.56

Test that when you select a system category in a new Browser, you see what you expect to see: a template for creating a new class.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005639.html

Name: Tools-fbs.434
Ancestors: Tools-cwp.433

Pulling these bits out makes new kinds of Browsers able to partially override behaviour.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005640.html

Name: ToolsTests-fbs.58
Ancestors: ToolsTests-fbs.57

Basic labelString tests. These don't test the multiwindow labelStrings though.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005641.html

Name: Tools-fbs.435
Ancestors: Tools-fbs.434

Don't use Class objects when all you care about is a name.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005642.html

Name: ToolsTests-fbs.59
Ancestors: ToolsTests-fbs.58

More characterisation tests. Note that #testClassCommentAnnotation is 'forward looking', in that it defines the soon-to-be behaviour of Browser >> #annotationForClassCommentFor:.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005643.html

Name: Tools-fbs.436
Ancestors: Tools-fbs.435

We already have the organization in classOrganizer. CodeHolder's version goes and looks it up.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005644.html

Name: ToolsTests-fbs.60
Ancestors: ToolsTests-fbs.59

Much better testing for Browser >> #annotationForClassCommentFor:.

=============================================
commits at source.squeak.org ()
2013-01-05 05:55:08 UTC
Permalink
Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours:

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005645.html

Name: Tools-fbs.437
Ancestors: Tools-fbs.436

We already have the organizer. No need to look it up again.

=============================================
commits at source.squeak.org ()
2013-01-08 05:55:13 UTC
Permalink
Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours:

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005646.html

Name: System-bf.510
Ancestors: System-nice.509

Wouldn't it be useful to know how old an object is? In our images there are objects that were created more than 30 years ago, after all. With this, we will know when objects were created, starting from today. See ObjectHistory's class comment for details.

=============================================
commits at source.squeak.org ()
2013-01-10 05:55:12 UTC
Permalink
Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours:

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005647.html

Name: KernelTests-nice.240
Ancestors: KernelTests-nice.238

Test expectations of Euclidean division with quotient rounded toward zero (quo: rem:) or toward negative infinity (// \\) for various range of Integer.

=============================================
commits at source.squeak.org ()
2013-01-11 05:55:10 UTC
Permalink
Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours:

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005648.html

Name: Tools-bf.438
Ancestors: Tools-fbs.437, Tools-bf.429

Fix changes in debugger getting lost when reverting to method from an inner block. As reported by Aran Lunzer:

Set up the following method on some class:

bugger
| a1 |
a1 := 0.
[ 2 / a1 ] value

then send <class> new bugger

In the walkback, try changing the last line to

[ b1:=2. 2 / a1 ] value

(oops - said b1 instead of a1...)

Debugger asks if it's ok to revert to the method in which the block was defined. Say ok. Then it asks about adding b1 as a new variable (NB: at this stage, the code edits have already disappeared). Then cancel, intending to fix the variable name. But your edits are lost.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005649.html

Name: Environments-bf.12
Ancestors: Environments-fbs.11

Environment>>bindingOf: must not answer undeclared bindings. E.g., in my case a binding in undeclared shadowed a perfectly fine class variable higher up in the class hierarchy.

=============================================
commits at source.squeak.org ()
2013-01-12 05:55:12 UTC
Permalink
Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours:

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005650.html

Name: Kernel-bf.730
Ancestors: Kernel-nice.729

Cache source files during class mutation for noticeable speedup with larger hierarchies.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005651.html

Name: Tools-bf.439
Ancestors: Tools-bf.438

When debugging a process from ProcessBrowser, do not try to resume it (see http://lists.squeakfoundation.org/pipermail/squeak-dev/2012-December/166749.html)

=============================================
commits at source.squeak.org ()
2013-01-16 05:55:15 UTC
Permalink
Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours:

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005652.html

Name: GraphicsTests-nice.31
Ancestors: GraphicsTests-nice.30

Some degenerated rectangles (lines) can intersects: even though they have a null area.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005653.html

Name: Graphics-nice.196
Ancestors: Graphics-bf.195

Some degenerated rectangles (lines) can intersects: even though they have a null area.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005654.html

Name: Morphic-dtl.634
Ancestors: Morphic-dtl.633

Fix for Morphic layout bug reported by Bob Arning in http://lists.squeakfoundation.org/pipermail/squeak-dev/2013-January/167757.html

Invalidate changed regions in Morph>>layoutBounds: in the same manner as in Morph>>doLayoutIn:

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005655.html

Name: Tests-fbs.183
Ancestors: Tests-fbs.182

Fail with a more meaningful description of the problem.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005656.html

Name: Tests-fbs.184
Ancestors: Tests-fbs.183

OrderedCollection's printString is inadequate for logging this error: rather, manually join the offending unbound globals with a comma.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005657.html

Name: Tests-fbs.185
Ancestors: Tests-fbs.184

#assert:equals: (and even more so #assert:equals:description:) is just so much nicer than a bare #assert:.

=============================================
commits at source.squeak.org ()
2013-01-18 05:55:04 UTC
Permalink
Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours:

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005658.html

Name: System-fbs.511
Ancestors: System-bf.510

Filenames like "3+4.st" are perfectly valid, but need to be URL encoded.

I would be happy to see a simpler implementation.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005659.html

Name: Kernel-nice.730
Ancestors: Kernel-nice.729

1) connect primitive 20 for large integer modulo (in range 31 to 64 bits)
2) simplify #digitDiv:neg: senders since it can cope with a negative SmallInteger as first parameter

WARNING: do not use with COG 2662 because primitive 20 (rem) is broken for negative case.
This will have to wait a few days in the inbox...

NOTE: the fallback code for Euclidean division #digitDiv:neg: is awfully dated with its nibbles... our (positive) SmallInteger range is 30bits.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005660.html

Name: Kernel-nice.731
Ancestors: Kernel-bf.730, Kernel-nice.730

merge nice+bf 730

=============================================
commits at source.squeak.org ()
2013-01-21 05:55:25 UTC
Permalink
Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours:

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005661.html

Name: System-dtl.512
Ancestors: System-fbs.511

Only the instance of SystemOrganizer that is the global SystemOrganization should send change notifications to the singleton SystemChangeNotifier.

Addresses the case of a FileContentsBrowser that uses a new SystemOrganization to load code for browsing and should not initiate a system change notification for the class being browsed, as reported by Bob Arning on squeak-dev http://lists.squeakfoundation.org/pipermail/squeak-dev/2013-January/167941.html

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005662.html

Name: ST80-nice.140
Ancestors: ST80-eem.139

Use degreesToRadians where due

=============================================
commits at source.squeak.org ()
2013-01-22 05:55:05 UTC
Permalink
Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours:

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005663.html

Name: ReleaseBuilder-nice.91
Ancestors: ReleaseBuilder-fbs.90

Reduce level of superstition: 1 garbageCollect is worth 3.
Don't collect garbage before Symbol compactSymbolTable., because this one will invoke garbageCollect by itself.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005664.html

Name: ToolsTests-nice.61
Ancestors: ToolsTests-fbs.60

Reduce level of superstition: 1 garbageCollect is worth 3.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005665.html

Name: ScriptLoader-nice.334
Ancestors: ScriptLoader-ul.333

Reduce level of superstition: 1 garbageCollect is worth 3.

=============================================
commits at source.squeak.org ()
2013-01-23 05:55:10 UTC
Permalink
Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours:

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005666.html

Name: Collections-bf.498
Ancestors: Collections-nice.497, Collections-topa.366

Merge Collections-topa.366:
Provide #asCamelCase for Strings.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005667.html

Name: Monticello-nice.528
Ancestors: Monticello-bf.527

Let MCTool inherit from Model.
This is to avoid nasty garbage to survive in DependentsFields with kilo bytes of un-released MCDefinition cache.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005668.html

Name: Morphic-bf.635
Ancestors: Morphic-dtl.634

DockingBarUpdatingMenuMorph: forget references held in menu items as soon as menu is closed

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005669.html

Name: Monticello-hjh.530
Ancestors: Monticello-hjh.529

Updated comments in MCDefinition, MCPackage and MCSnapshot after review by Bert Freudenberg.

=============================================
commits at source.squeak.org ()
2013-01-24 05:55:03 UTC
Permalink
Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours:

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005670.html

Name: Morphic-bf.636
Ancestors: Morphic-bf.635, Morphic-kb.619

Merge Morphic-kb.619:
- a fix to avoid DockingBarMorph holding onto windows

=============================================
commits at source.squeak.org ()
2013-01-25 05:55:13 UTC
Permalink
Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours:

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005671.html

Name: Monticello-bf.531
Ancestors: Monticello-hjh.530

Add "search history" to version history browser, too.

=============================================
commits at source.squeak.org ()
2013-01-26 05:55:20 UTC
Permalink
Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours:

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005672.html

Name: Multilingual-fbs.157
Ancestors: Multilingual-ul.156

Lars' fix for Mantis report #7709: Image Segments can't be written to file anymore.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005673.html

Name: Tests-fbs.186
Ancestors: Tests-fbs.185

Tests for Lars' Mantis report #7709: Image Segments can't be written to file anymore.

=============================================
commits at source.squeak.org ()
2013-01-29 05:55:19 UTC
Permalink
Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours:

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005674.html

Name: Tests-fbs.187
Ancestors: Tests-fbs.186

Mantis #7726: tests for Installer >> #packageAndVersionFrom:

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005675.html

Name: Installer-Core-fbs.361
Ancestors: Installer-Core-bp.360

Mantis #7726: Installer should sanitise user input. Now we can handle whitespace around/between package names/versions.

=============================================
commits at source.squeak.org ()
2013-01-30 05:55:15 UTC
Permalink
Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours:

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005676.html

Name: Collections-cmm.499
Ancestors: Collections-bf.498

FloatCollection inherits the convenient add:/remove: API of OrderedCollection, but using an internal FloatArray for better efficiency.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005677.html

Name: CollectionsTests-cmm.195
Ancestors: CollectionsTests-ul.194

Tests for FloatCollection.

=============================================
commits at source.squeak.org ()
2013-01-31 05:55:18 UTC
Permalink
Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours:

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005678.html

Name: Collections-cmm.500
Ancestors: Collections-cmm.499

Categorize overridden method.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-January/005679.html

Name: CollectionsTests-cmm.196
Ancestors: CollectionsTests-cmm.195

fix

=============================================
commits at source.squeak.org ()
2013-02-03 05:55:13 UTC
Permalink
Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours:

http://lists.squeakfoundation.org/pipermail/packages/2013-February/005680.html

Name: Monticello-fbs.532
Ancestors: Monticello-bf.531

Installer repository failover #1 of 2: Make MCRepositoryGroup behave more like a repository, finding the repository/repositories containing a particular version.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-February/005681.html

Name: Installer-Core-TestRunner.363
Ancestors: Installer-Core-fbs.361, Installer-Core-fbs.362

Installer repository failover #2 of 2: InstallerMonticello uses an MCRepositoryGroup, so that attempting to load a particular version will try each repository in turn. #basicView will open _all_ the InstallerMonticello's repositories.

=============================================
commits at source.squeak.org ()
2013-02-06 05:55:14 UTC
Permalink
Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours:

http://lists.squeakfoundation.org/pipermail/packages/2013-February/005682.html

Name: SUnit-cmm.90
Ancestors: SUnit-ul.89

- Do not wipe out pre-initialized state on TestCases. Tests do not need to be prevented from running more than once if proper setUp/tearDown methods are present.

=============================================
commits at source.squeak.org ()
2013-02-07 05:55:12 UTC
Permalink
Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours:

http://lists.squeakfoundation.org/pipermail/packages/2013-February/005683.html

Name: SUnit-eem.91
Ancestors: SUnit-cmm.90

Robust testCase printing. i.e. TestCase new asString should
not raise an error.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-February/005684.html

Name: Kernel-eem.732
Ancestors: Kernel-nice.731

Make CompiledMethod>>removeProperty:[ifAbsent:] get rid
of the AdditionalMethodState if the resulting state is empty.

Streamline BlockClosure>>newProcess[With:] which does
not need to run unwinds if control reaches the end of
the process.

Use the abstract blockCreationBytecodeMessage to compute
BlockClosure's size instead of accessing bytecodes directly.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-February/005685.html

Name: GetText-eem.26
Ancestors: GetText-nice.25

Fix clearAllDomains. It used to use the non-existent
allCompiledMethodsDo:.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-February/005686.html

Name: SUnit-cmm.92
Ancestors: SUnit-eem.91

Remove obsolete test.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-February/005687.html

Name: ScriptLoader-eem.335
Ancestors: ScriptLoader-nice.334

Update ScriptLoader>>cleanUpMethods.
Add voidMapCache and use startTimerEventLoop instead
of obsolete startTimerInterruptWatcher

=============================================
commits at source.squeak.org ()
2013-02-08 05:55:08 UTC
Permalink
Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours:

http://lists.squeakfoundation.org/pipermail/packages/2013-February/005688.html

Name: Kernel-dtl.733
Ancestors: Kernel-eem.732

Remove EventSensor>>wait2ms

Use of EventSensor>>wait2ms in sensor direct access to mouse and keyboard causes slow and incorrect keyboard handling in MVC. In newer Squeak images (3.9 and later), removing the delay restores normal MVC keyboard input.

This change removes the wait2ms delay entirely. If the problem described in http://lists.squeakfoundation.org/pipermail/squeak-dev/2007-May/116185.html should reappear, a different solution will be required.

Notes:

In Squeak trunk, the wait2ms delay is invoked only when in MVC. However, Scratch (and possibly other applications) makes use of direct sensor access from Morphic (arguably this is an error in Scratch, but the usage exists nonetheless).

Background:

Recent thread in which St&#233;phane Rollandin identified the source of the problem:
http://lists.squeakfoundation.org/pipermail/squeak-dev/2013-February/168526.html

Rationale for introducing the delay in earlier versions of Squeak:
http://lists.squeakfoundation.org/pipermail/squeak-dev/2007-May/116185.html

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-February/005689.html

Name: Exceptions-fbs.43
Ancestors: Exceptions-cmm.37
Backported From: Exceptions-fbs.42

Stay in the debugger for longer. Part 2 of 3.

New exception hierarchy lets interesting parties catch NotImplemented errors to encourage developers to implement missing parts.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-February/005690.html

Name: Morphic-fbs.638
Ancestors: Morphic-bf.636
Backported From: Morphic-fbs.637

Stay in the debugger for longer. Part 1 of 3.

UI hooks to create overriding methods.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-February/005691.html

Name: Tools-fbs.445
Ancestors: Tools-bf.439
Backported From: Tools-fbs.444

Stay in the debugger for longer. Part 4 of _5_.

Implement the Debugger side of things. Actually _remove_ the frame that signaled the NotImplemented from the call stack.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-February/005692.html

Name: Kernel-fbs.739
Ancestors: Kernel-dtl.733, Kernel-fbs.738

Stay in the debugger for longer. Part 3 of 5.

Stack introspection so the Debugger doesn't have to. Signal the new exceptions when necessary.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-February/005693.html

Name: KernelTests-fbs.243
Ancestors: KernelTests-nice.240
Backported From: KernelTests-fbs.242

Stay in the debugger for longer. Part 5 of _5_.

Tests showing the signalling of the new style marker methods.

=============================================
commits at source.squeak.org ()
2013-02-09 05:55:06 UTC
Permalink
Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours:

http://lists.squeakfoundation.org/pipermail/packages/2013-February/005694.html

Name: Morphic-tfel.639
Ancestors: Morphic-fbs.638

Avoid redrawing the main docking bar all the time. This is a huge resource hog if you reach the threshold in the DamageRecorder and the docking bar damage rectangle is merged with the others, effectively forcing a huge chunk of the World redrawn

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-February/005695.html

Name: Tools-fbs.446
Ancestors: Tools-fbs.445

Fixes http://bugs.squeak.org/view.php?id=7730

If you're browsing class Foo, there's no need to go back to Foo, so don't show it in the list of recently visited classes.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-February/005696.html

Name: Exceptions-fbs.44
Ancestors: Exceptions-fbs.43

Move FTPConnectionException to Network-Protocols.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-February/005697.html

Name: Network-fbs.139
Ancestors: Network-dtl.138

Move FTPConnectionException to Network-Protocols.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-February/005698.html

Name: Compiler-eem.248
Ancestors: Compiler-cwp.247

Slightly less circuitous generation of methods with temp
names. Since schematic temp names string is computed
during generation, best to generate a method and copy
it with the computed temp names than generate two
separate methods.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-February/005699.html

Name: Compiler-eem.249
Ancestors: Compiler-eem.248

Simplify Parser>expression. braceExpression is handled in
primaryExpression, as it should be. The extra invocation in
expression is unnecessary, and harmlessly erroneous.

Fix snafu with Compiler-eem.248 (slightly less circuitous generation of methods with temp names).

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-February/005700.html

Name: Compiler-eem.250
Ancestors: Compiler-eem.249

Fix incomplete fix for decompilation of #to:do: loops in
Compiler-eem.242. Fixes decompilation of e.g. compilation
of SoundRecorder>>segmentsAbove:normalizedVolume:
when compiled with Compiler-nice.224 or later.

=============================================
commits at source.squeak.org ()
2013-02-10 05:55:07 UTC
Permalink
Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours:

http://lists.squeakfoundation.org/pipermail/packages/2013-February/005701.html

Name: ST80-dtl.141
Ancestors: ST80-nice.140

Fix ParagraphEditor>>zapSelectionWithCompositionWith: failing to reset setMark or setPoint values after converting to Unicode in the case of an input string of length > 1. The problem was apparent only in the case of keyboard entry from MVC when a wait2ms delay causing two or more characters to be buffered prior to handling the zap selection. Problem was best exhibited using a Linux interpreter VM with poor delay time resolution. This update fixes the problem in MVC but has not been tested with Unicode input (author is using US English keyboard and locale).

Problem identified by Bob Arning: http://lists.squeakfoundation.org/pipermail/squeak-dev/2013-February/168666.html

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-February/005702.html

Name: ST80-dtl.142
Ancestors: ST80-dtl.141

Revert last change to zapSelectionWithCompositionWith: because original implementation is required for multilingual support.

Change ParagraphEditor>>readKeyboard to always process one character at a time per suggestion from Bob Arning.

Background and explanation at http://lists.squeakfoundation.org/pipermail/squeak-dev/2013-February/168725.html

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-February/005703.html

Name: ReleaseBuilder-fbs.92
Ancestors: ReleaseBuilder-nice.91

Implement Eliot Miranda's request to clear all text domains, and recompile everything prior to release.

Technically we only need to recompile when we change the Compiler, but doing it every time we release something is much simpler/easier: I can't imagine how we'd write a ReleaseBuilder that would know (without introspecting on some source of version history somewhere) that the Compiler had just changed.

=============================================
commits at source.squeak.org ()
2013-02-12 05:55:16 UTC
Permalink
Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours:

http://lists.squeakfoundation.org/pipermail/packages/2013-February/005704.html

Name: Compiler-eem.251
Ancestors: Compiler-eem.250

Yet another fix for decompilation of to:do: loops compiled
with Compiler-nice.224 or later. If the stack is non-empty
when convertDoDoLoop is invoked, the stack must be left
non-empty. This fixes decompilation of e.g.
^nil to: 3 do: [:i| i class]

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-February/005705.html

Name: Tests-eem.188
Ancestors: Tests-fbs.187

Provde a test for Compiler-eem.251 which shows its still
broken. I fixed decompilation of
true ifTrue: [^thisContext method decompile].
^nil to: 3 do: [:i| i class]
but /not/ decompilation of
^nil to: 3 do: [:i| i class]

tsk, tsk.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-February/005706.html

Name: Compiler-eem.252
Ancestors: Compiler-eem.251

Yet another fix for decompilation of to:do: loops compiled
with Compiler-nice.224 or later. If the stack is non-empty
when convertDoDoLoop is invoked, the stack must be left
non-empty. This really does fix decompilation of e.g.
^nil to: 3 do: [:i| i class]
and not just
true ifTrue: [^thisContext method decompile].
^nil to: 3 do: [:i| i class]

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-February/005707.html

Name: Installer-Core-fbs.364
Ancestors: Installer-Core-TestRunner.363

InstallerSqueakMap should use Installer's user input sanitising methods.

=============================================
commits at source.squeak.org ()
2013-02-13 05:55:06 UTC
Permalink
Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours:

http://lists.squeakfoundation.org/pipermail/packages/2013-February/005708.html

Name: Collections-fbs.501
Ancestors: Collections-cmm.500

Typo correction.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-February/005709.html

Name: CollectionsTests-fbs.197
Ancestors: CollectionsTests-cmm.196

Missing tests for String >> #asCamelCase.

http://bugs.squeak.org/view.php?id=7725

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-February/005710.html

Name: Installer-Core-fbs.365
Ancestors: Installer-Core-fbs.364

http://bugs.squeak.org/view.php?id=7722

Use FileDirectory's abstractions rather than relying on a platform specific path separator.

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-February/005711.html

Name: Graphics-bf.197
Ancestors: Graphics-nice.196

Make grabbing color cursor from screen work. E.g. (CursorWithAlpha fromUser showWhile: [Sensor waitButton])

=============================================

http://lists.squeakfoundation.org/pipermail/packages/2013-February/005712.html

Name: Environments-fbs.13
Ancestors: Environments-bf.12

Max Leske's fix to Environments to support Fuel (hence the methods being in the compatibility category).

=============================================

Continue reading on narkive:
Loading...