Daily Commit Log

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view
|

Daily Commit Log

commits-2
Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours:

http://lists.squeakfoundation.org/pipermail/packages/2015-April/008190.html

Name: Collections-nice.622
Ancestors: Collections-mt.621

#toBraceStack: is not used for compiling { } for so long that it's really time to get rid of it.

Symbol>>numArgs: does not need to copy self into a temp var.

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

http://lists.squeakfoundation.org/pipermail/packages/2015-April/008191.html

Name: GetText-nice.36
Ancestors: GetText-kfr.35

Though it can't really be tested in current trunk image state (InternalTranslator is not initialized) , modernize the way to utf8+gzip+base64 encode...

This how how simpler the world could be without those Multi RW Binary or Text things, no need to send those reset/ascii/binary incantations.

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

http://lists.squeakfoundation.org/pipermail/packages/2015-May/008192.html

Name: CollectionsTests-tfel.240
Ancestors: CollectionsTests-mt.239

Add a test case to check that ByteString>>findSubstring:in:startingAt:matchTable: behaves the same way as the primitive implementation (esp when using a negative index for startingAt:)

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

http://lists.squeakfoundation.org/pipermail/packages/2015-May/008193.html

Name: Collections-tfel.623
Ancestors: Collections-nice.622

fix fallback code for ByteString>>findSubstring:in:startingAt:matchTable: when passing a starting index <= 0

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

http://lists.squeakfoundation.org/pipermail/packages/2015-May/008194.html

Name: Morphic-mt.953
Ancestors: Morphic-mt.952

Text editor hooks modified to be more descriptive on model side. And to not confuse Tweak.

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

http://lists.squeakfoundation.org/pipermail/packages/2015-May/008195.html

Name: MorphicTests-mt.29
Ancestors: MorphicTests-mt.28

Tests added for text editor's model callbacks.

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

http://lists.squeakfoundation.org/pipermail/packages/2015-May/008196.html

Name: EToys-mt.124
Ancestors: EToys-topa.123

Tiles-view check moved from Tools package.

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

http://lists.squeakfoundation.org/pipermail/packages/2015-May/008197.html

Name: MorphicExtras-mt.162
Ancestors: MorphicExtras-mt.161

Added from Morph package.

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

http://lists.squeakfoundation.org/pipermail/packages/2015-May/008198.html

Name: Morphic-mt.954
Ancestors: Morphic-mt.953

Deprecations updated and some dependencies to MorphicExtras removed.

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

http://lists.squeakfoundation.org/pipermail/packages/2015-May/008199.html

Name: Tools-mt.613
Ancestors: Tools-mt.612

Deprecations updated.

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

http://lists.squeakfoundation.org/pipermail/packages/2015-May/008200.html

Name: 46Deprecated-mt.1
Ancestors:

Some deprecations for 4.6 added.

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

http://lists.squeakfoundation.org/pipermail/packages/2015-May/008201.html

Name: Kernel-nice.921
Ancestors: Kernel-topa.920

Provide an exact version for #floorLog: for those Number using exact arithmetic
(as proposed in inbox Kernel-nice.721 / 11 December 2012).

Previous version was not exact, otherwise this count would have been zero:
    (-300 to: -1) count: [:n | n ~= ((10 raisedTo: n) floorLog: 10)].

Note that this won't make #log: exact, and could lead to disagreement between the two functions.
However, old behavior compatible with log: is still possible by passing the radix argument asFloat, so this is not a real problem.

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

http://lists.squeakfoundation.org/pipermail/packages/2015-May/008202.html

Name: KernelTests-nice.293
Ancestors: KernelTests-mt.292

Tests asserting floorLog: exactness for Number performing exact arithmetic.

This was inbox KernelTests-nice.239 / 11 December 2012

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

http://lists.squeakfoundation.org/pipermail/packages/2015-May/008203.html

Name: Kernel-nice.922
Ancestors: Kernel-nice.921

Remove an un-necessary inst. var. shadowing (inbox Kernel-nice.745 7 March 2013)

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

http://lists.squeakfoundation.org/pipermail/packages/2015-May/008204.html

Name: Kernel-nice.923
Ancestors: Kernel-nice.922

Use an intermediate abstract Float representation (sign, exponent, significand) in order to convert single precision bits to double precision bits rather than attempting the conversion by direct bit manipulations.

This lead to simpler code, because it is like a semantic bit manipulation rather than a raw bit manipulation.

And this also lead to faster code on 32 bits interpreter/COG/Spur VM thanks to avoidance of LargeInteger

(was Kernel-nice.893 / 24 December 2014)

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

http://lists.squeakfoundation.org/pipermail/packages/2015-May/008205.html

Name: Collections-tfel.624
Ancestors: Collections-tfel.623

remove overrides for beginsWith: in ByteStrings and ByteSymbols, because these actually perform worse now than the generic implementation in String

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

http://lists.squeakfoundation.org/pipermail/packages/2015-May/008206.html

Name: Collections-mt.625
Ancestors: Collections-tfel.624

Speed-up #endsWith: for strings.

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

http://lists.squeakfoundation.org/pipermail/packages/2015-May/008207.html

Name: Morphic-mt.955
Ancestors: Morphic-mt.954

Flag about a performance issue in scroll bars added. We should improve that at some point.

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

http://lists.squeakfoundation.org/pipermail/packages/2015-May/008208.html

Name: Tools-mt.614
Ancestors: Tools-mt.613

Some descriptive balloon texts added to browser's switch buttons.

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

http://lists.squeakfoundation.org/pipermail/packages/2015-May/008209.html

Name: SMLoader-mt.82
Ancestors: SMLoader-cmm.81

Fixed layout problems and selection bugs in SqueakMap tools.

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

http://lists.squeakfoundation.org/pipermail/packages/2015-May/008210.html

Name: Tools-mt.615
Ancestors: Tools-mt.614

Use code panes in inspector and object explorer (was text panes).

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

http://lists.squeakfoundation.org/pipermail/packages/2015-May/008211.html

Name: ShoutCore-ul.46
Ancestors: ShoutCore-ul.45

SHParserST80 performance improvements:
- store the allowUnderscoreAssignments and allowUnderscoreSelectors preferences in instance variables
- don't honor per class allowUnderscoreAssignments preferences anymore (just like how allowUnderscoreSelectors works)
- ensure that instanceVariables is always an Array
- reordered branches in some methods based on the methods of the Morph class as sample
- use quick returns in frequently used testing methods
- release the errorBlock in #parse, instead of an #ensure: block in #parse:
- decreased the initial size of ranges to 40
- use Symbol >> #lookup: instead of Symbol >> #hasInterned:ifTrue:
- use the fact that classOrMetaClass is nil when it's not a Behavior
- use a fast primitive in #scanComment instead of a loop

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

http://lists.squeakfoundation.org/pipermail/packages/2015-May/008212.html

Name: ShoutCore-ul.47
Ancestors: ShoutCore-ul.46

Removed encodedCharSet from SHParserST80. All methods referencing it are sending the corresponding method to the character again.

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

http://lists.squeakfoundation.org/pipermail/packages/2015-May/008213.html

Name: Collections-mt.626
Ancestors: Collections-mt.625

Added the (configurable) possibility to disable forced transcript updates as well as to enable a generic redirection to stdout.

Only works when using the #show: or #showln: interface in Transcript.

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

http://lists.squeakfoundation.org/pipermail/packages/2015-May/008214.html

Name: ST80-mt.183
Ancestors: ST80-dtl.182

Some assumptions on model interface removed.

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

http://lists.squeakfoundation.org/pipermail/packages/2015-May/008215.html

Name: Morphic-mt.956
Ancestors: Morphic-mt.955

Some assumptions on model interface removed from text fields.

Missing #appendEntryLater added to text fields, which is needed for deferred transcript updates.

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