Daily Commit Log

Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours:

http://lists.squeakfoundation.org/pipermail/packages/2010-December/003983.html

Name: Kernel-ul.518
Ancestors: Kernel-dtl.517

- introduced Object >> #shouldBePrintedAsLiteral as a replacement for #isLiteral during printing and storing

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

http://lists.squeakfoundation.org/pipermail/packages/2010-December/003984.html

Name: Collections-ul.411
Ancestors: Collections-ul.410

- use #shouldBePrintedAsLiteral instead of #isLiteral when printing or storing characters and arrays

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

http://lists.squeakfoundation.org/pipermail/packages/2010-December/003985.html

Name: MorphicExtras-ul.96
Ancestors: MorphicExtras-fbs.95, MorphicExtras-edc.95

merged

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

http://lists.squeakfoundation.org/pipermail/packages/2010-December/003986.html

Name: Kernel-ul.519
Ancestors: Kernel-ul.518

- copied another two methods from BlockContext to BlockClosure. One of them fixes http://bugs.squeak.org/view.php?id=7579 .

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

http://lists.squeakfoundation.org/pipermail/packages/2010-December/003987.html

Name: Kernel-ul.520
Ancestors: Kernel-ul.519

A few changes to Semaphore:
- save a block creation + activation + a possible context switch in #critical:ifCurtailed:
- save a block creation + activation in #critical:ifError:, this changes the behavior when mutuallyExcludedBlock doesn't understand #ifError:
- #critical:ifLocked: is guaranteed to not get locked. The earlier implementation sent #critical: which could cause a context switch, so another process could enter the critical section before the current. Also fixed the comments of this method.

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

http://lists.squeakfoundation.org/pipermail/packages/2010-December/003988.html

Name: Kernel-ul.521
Ancestors: Kernel-ul.520

Implemented CompiledMethod >> #asString the same way Object does.

Why?
Because it raised an error, which it shouldn't.

Why doesn't it return a string with the bytes as characters like ByteArray does it?
Because it's not very useful to get an unreadable string with random looking characters. Also CompiledMethod is a subclass of ByteArray only to avoid code duplication, it could be a variableByteSubclass of Object.

Why doesn't it return the source code like Pharo does it?
Because that requires several changes in the way the sources and changes files are used (concurrency issues). And it's a lot slower.

The related mantis issue is http://bugs.squeak.org/view.php?id=7570 .

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

http://lists.squeakfoundation.org/pipermail/packages/2010-December/003989.html

Name: Morphic-laza.488
Ancestors: Morphic-laza.487

Try to prevent empty progress morphs after updates when SystemProgressMorph reset is used in a Postscript

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

http://lists.squeakfoundation.org/pipermail/packages/2010-December/003990.html

Name: Morphic-laza.489
Ancestors: Morphic-laza.488

Provide a recipe for loading OCompletion in the "Extending the system" workspace

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

http://lists.squeakfoundation.org/pipermail/packages/2010-December/003991.html

Name: System-ul.398
Ancestors: System-ul.397

- fix: SystemDictionary >> #traitNames returns a sorted collection. It also uses the cached nonClassNames which is faster.
- added SystemDictionary >> #classAndTraitNames which returns a sorted collection of all class and trait names

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

http://lists.squeakfoundation.org/pipermail/packages/2010-December/003992.html

Name: Tools-fbs.283
Ancestors: Tools-ul.282

When a method changes out from under your feet in the Debugger (e.g., you remove an instvar triggering a recompile of a method through which you've stepped), the Debugger blanks out contents. In the UI, this manifests as a blank code pane.

This change resets the contents ivar by setting it to the new, recompiled, CompiledMethod as a side effect of calling self selectedMessage.

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

http://lists.squeakfoundation.org/pipermail/packages/2010-December/003993.html

Name: Tools-fbs.284
Ancestors: Tools-ul.282

When a method changes out from under your feet in the Debugger (e.g., you remove an instvar triggering a recompile of a method through which you've stepped), the Debugger blanks out contents. In the UI, this manifests as a blank code pane.

This change resets the contents ivar by setting it to the new, recompiled, CompiledMethod as a side effect of calling self selectedMessage.

This version is Eliot Miranda's improved version of my (fbs) initial submission.

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