[squeak-dev] Daily Commit Log

commits at source.squeak.org commits at source.squeak.org
Sun Sep 16 23:55:05 UTC 2012


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

http://lists.squeakfoundation.org/pipermail/packages/2012-September/005510.html

Name: Tests-ul.164
Ancestors: Tests-ul.163

Fixing Monticello tests:
- restore #packageName for MCMockPackageInfo, MCEmptyPackageInfo, MCDirtyPackageInfo
- re-register these PackageInfos in postscript
- renamed the virtual #systemCategory of MCMockPackageInfo to Tests-MonticelloMocks, so the extension method MCSnapshotTest >> #mockClassExtension can be in the Tests package instead of Monticello

MCMethodDefinitionTest enhancements:
- updated #ownPackage, because it's Tests now
- save and restore the timestamp of #override
- remove the empty system category and the workingCopy of the mock package after the tests

These changes break MCWorkingCopyTest's #testRenamePrefix and #testRenameSuffix and some of  tests in MCSnapshotBrowserTest.

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

http://lists.squeakfoundation.org/pipermail/packages/2012-September/005511.html

Name: Tests-ul.165
Ancestors: Tests-ul.164

Break out Monticello package renaming tests into a separate TestCase, and use the existing category Tests-Monticello-Mock as its mock package name. This is needed because package renaming depends on the contract that package names are prefixes of their contained category names.

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

http://lists.squeakfoundation.org/pipermail/packages/2012-September/005512.html

Name: Tests-ul.167
Ancestors: Tests-ul.165

Since Monticello-bf.476 MCSnapshotBrowser shows empty categories too, so include the systemCategories of the MCMockPackageInfo in MCSnapshotBrowserTest >> #allCategories in order to make MCSnapshotBrowserTest pass.

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

http://lists.squeakfoundation.org/pipermail/packages/2012-September/005513.html

Name: KernelTests-ul.233
Ancestors: KernelTests-ul.232

Speed-up for CompiledMethodComparisonTest >> #testHash:
- instead of comparing all methods with all other, only compare methods which have the same size and header
- only compare each pair of methods at most once (assuming #= is symmetric)
It now takes less than 30 seconds using the interpreter VM on a 4 years old notebook.
Also added a progress bar and decreased the timeout.

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

http://lists.squeakfoundation.org/pipermail/packages/2012-September/005514.html

Name: Tests-ul.168
Ancestors: Tests-ul.167

Removed #expectedFailures from MirrorPrimitiveTests because the new interpreter VM for Windows supports them.

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

http://lists.squeakfoundation.org/pipermail/packages/2012-September/005515.html

Name: CollectionsTests-ul.192
Ancestors: CollectionsTests-fbs.191

- marked RWBinaryOrTextStreamTest >> #testExisting as an expected failure, because it expects a different behavior than the current one

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

http://lists.squeakfoundation.org/pipermail/packages/2012-September/005516.html

Name: KernelTests-ul.234
Ancestors: KernelTests-ul.233

#testPerformCanExecutelongMethodWithTemps and #testPerformInSuperclassCanExecutelongMethodWithTemps from CompiledMethodTest are expected to fail on the interpreter VMs.

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


More information about the Squeak-dev mailing list