Planet Squeak Code

Cheat to support strange situations

by Assargadon (tallman@inbox.ru) at May 17, 2012 12:04 AM

2.156 : bug fixed that reorder the nodes when laying out with the tree layout. Thanks Ben and Doru

May 16, 2012 10:57 PM

2.156 : bug fixed that reorder the nodes when laying out with the tree layout. Thanks Ben and Doru

May 16, 2012 10:55 PM

* Changed PESerialPort>>readByteArray to return an empty byte array if the portNumber is nil.

by Ricardo Moran (richi.moran@gmail.com) at May 16, 2012 10:52 PM

* Added DuinoBotTestSketch.

by Ricardo Moran (richi.moran@gmail.com) at May 16, 2012 10:48 PM

* Added WString.h to the c dependencies in Win32 makefile

by Ricardo Moran (richi.moran@gmail.com) at May 16, 2012 10:45 PM

Style for autocompleter incorporated into SPAutocompleteTextInputComponent

by Assargadon (tallman@inbox.ru) at May 16, 2012 10:22 PM

Updating the default

May 16, 2012 08:24 PM

* Renamed DuinoBotFirmware to DuinoBotDirectMode (note to self: remember to add an empty DuinoBotFirmware class to the currently inexistent backwards-compatibility package).
* Added DuinoBotDirectModeSketch, which is essentially the same as ArduinoDirectModeSketch but including a different library.
* Changed DuinoBotMorph to automatically install the DuinoBotDirectMode.

by Ricardo Moran (richi.moran@gmail.com) at May 16, 2012 07:51 PM

Fixed the order preserving graph layout bug. The problem stemmed from Collection>>difference: having changed the behavior from Pharo 1.3 to Pharo 1.4.

Thanks Ben!

May 16, 2012 07:43 PM

Let #floorLog: be exact in the base used by underlying Float implementation (currently 2 is hardcoded because Float class>>base was not implemented).
In other cases, warn about possible rounding errors.

by Nicolas Cellier (nicolas.cellier.aka.nice@gmail.com) at May 16, 2012 07:33 PM

Add a test to demonstrate that despite being internally represented in base 2, a Float cannot compute floorLog: 2 exactly if it only rely on inexact log: implementation.

Indeed, an error on math functions values up to 3 ulp is compliant with IEEE 754 floating point standard.

by Nicolas Cellier (nicolas.cellier.aka.nice@gmail.com) at May 16, 2012 07:24 PM

* Changed ArduinoMorph>>installFirmwareIfPossible to avoid asking the ArduinoDirectMode to upload itself and instead translating and uploading the ArduinoDirectModeSketch.

by Ricardo Moran (richi.moran@gmail.com) at May 16, 2012 07:08 PM

* Added ArduinoDirectModeSketch, which is just an almost empty sketch (for now) that includes the "ArduinoDirectMode.h" library and calls its "loop" and "setup" functions.

by Ricardo Moran (richi.moran@gmail.com) at May 16, 2012 07:07 PM

tests for state and message/variable replacement

by Manuel Leuenberger (manuel.leu@student.unibe.ch) at May 16, 2012 06:49 PM

removed unused stuff

by Manuel Leuenberger (manuel.leu@student.unibe.ch) at May 16, 2012 06:48 PM

- Update for newer VMMaker:
don't use the result of #argumentsAddByte:toIter: anymore as there is no more results returned

by Damien Cassou (damien.cassou@gmail.com) at May 16, 2012 05:10 PM

Fix Color printing to show at most 3 decimal places

by Bert Freudenberg (bert@freudenbergs.de) at May 16, 2012 04:53 PM

Add Stream>>print:maxDecimalPlaces:

by Bert Freudenberg (bert@freudenbergs.de) at May 16, 2012 04:51 PM

Provides a naive #printShowingMaxDecimalPlaces: and #printOn:maxDecimalPlaces:

To be tested...
{
0.7 printShowingMaxDecimalPlaces: 3.
2.0001 printShowingMaxDecimalPlaces: 3.
-1.9994 printShowingMaxDecimalPlaces: 3.
}

by Bert Freudenberg (bert@freudenbergs.de) at May 16, 2012 04:36 PM

threw away the scope stuff, it just has to many problems. introduced manual reference management via aliasing and replacing

by Manuel Leuenberger (manuel.leu@student.unibe.ch) at May 16, 2012 03:40 PM

introduced replacing of message sends and variables for method commands

by Manuel Leuenberger (manuel.leu@student.unibe.ch) at May 16, 2012 03:38 PM

added isMethod to BFMethodCommand, for the sake of consistency

by Manuel Leuenberger (manuel.leu@student.unibe.ch) at May 16, 2012 03:37 PM

added the writer to the baseline so we can download in 1 step

by Guillermo Polito (guillermopolito@gmail.com) at May 16, 2012 02:47 PM

the startup process just logs with a little more detail

by Guillermo Polito (guillermopolito@gmail.com) at May 16, 2012 02:45 PM

some cleansing to the initialization process

by Guillermo Polito (guillermopolito@gmail.com) at May 16, 2012 02:34 PM

deprecate mirror folders 2

by Johannes Rasche (johannes.rasche@myNeat.de) at May 16, 2012 01:55 PM

deprecate mirror folders

by Johannes Rasche (johannes.rasche@myNeat.de) at May 16, 2012 11:34 AM

deprecate mirror folders

by Johannes Rasche (johannes.rasche@myNeat.de) at May 16, 2012 11:33 AM

- Fixed path for resources under Sugar plateform
- Open virtual keyboard when providing the name of a sketch to keep.

by Hilaire Fernandes (hilaire@ofset.org) at May 16, 2012 10:17 AM

tag registry

by Johannes Rasche (johannes.rasche@myNeat.de) at May 16, 2012 10:10 AM

tag registry

by Johannes Rasche (johannes.rasche@myNeat.de) at May 16, 2012 10:10 AM

added a nextPut: Character lf; to REPHudson>>printConcern:

May 16, 2012 09:44 AM

added MooseJEEScripts>>transformToArray:

by Fabrizio Perin (fabrizio.perin@gmail.com) at May 16, 2012 09:25 AM

neat folders redesigned 2

by Johannes Rasche (johannes.rasche@myNeat.de) at May 16, 2012 08:56 AM

neat folders redesigned 2

by Johannes Rasche (johannes.rasche@myNeat.de) at May 16, 2012 08:55 AM

scoped buffering works, but it's a pain. will switch to templating to define used interfaces

by Manuel Leuenberger (manuel.leu@student.unibe.ch) at May 16, 2012 08:13 AM

May 15, 2012 11:23 PM

Provides a naive #printShowingMaxDecimalPlaces: and #printOn:maxDecimalPlaces:

To be tested...
{
0.7 printShowingMaxDecimalPlaces: 3.
2.0001 printShowingMaxDecimalPlaces: 3.
-1.9994 printShowingMaxDecimalPlaces: 3.
}

May 15, 2012 09:21 PM

Consoles for output (end-user) messages

by Hervé Verjus (herve.verjus@univ-savoie.fr) at May 15, 2012 09:18 PM

0.160 : have used a moValue: instead of a roValue:

May 15, 2012 08:28 PM

0.160 : have used a moValue: instead of a roValue:

May 15, 2012 08:26 PM

0.160 : have used a moValue: instead of a roValue:

May 15, 2012 08:26 PM

empty log message

by Hervé Verjus (herve.verjus@univ-savoie.fr) at May 15, 2012 07:08 PM

NITraceMsgManager class for displaying trace messages in Transcript

by Hervé Verjus (herve.verjus@univ-savoie.fr) at May 15, 2012 06:46 PM

* Added classes to compile sketches into the DuinoBotCDC. It requires the new Arduino sketch directory tree, and only works in windows for now.

by Ricardo Moran (richi.moran@gmail.com) at May 15, 2012 06:33 PM

* Modified ArduinoWin32Uploader class>>#makefile to use the new Arduino sketch directory tree based on the DuinoPack. Take into account that this new directory only works in Win32 for now. The linux makefile still uses the old dir.

by Ricardo Moran (richi.moran@gmail.com) at May 15, 2012 06:31 PM

Fix decompilation of inlined to:do: loops whose value is used
(e.g. ^55 to: 57 do: [:i| self]). Avoid using the sugared checkBlock:as:
in the ifnil transformers that also use the unsugared checkBlock:as:maxArgs:

by Eliot Miranda (eliot.miranda@gmail.com) at May 15, 2012 05:04 PM

Provide a more helpful comment for Class>>copy.

by Eliot Miranda (eliot.miranda@gmail.com) at May 15, 2012 04:46 PM

refactored unused test classes

by Fabrizio Perin (fabrizio.perin@gmail.com) at May 15, 2012 04:16 PM

refactored unused test classes

by Fabrizio Perin (fabrizio.perin@gmail.com) at May 15, 2012 04:15 PM

refactored classes to unused classes

by Fabrizio Perin (fabrizio.perin@gmail.com) at May 15, 2012 04:05 PM

refactored unused classes

by Fabrizio Perin (fabrizio.perin@gmail.com) at May 15, 2012 04:03 PM

changes to error handler post mark

May 15, 2012 03:03 PM

changes to error handler post mark

May 15, 2012 03:02 PM

Error handler using post mark

by HAW (hernan.wilkinson@gmail.com) at May 15, 2012 02:29 PM

Error handler using post mark

by HAW (hernan.wilkinson@gmail.com) at May 15, 2012 02:28 PM

Support for DBXTalk

by Olivier Auverlot (olivier@auverlot.fr) at May 15, 2012 01:52 PM

Fixed file extention checking code

by Assargadon (tallman@inbox.ru) at May 15, 2012 01:39 PM

empty log message

by Assargadon (tallman@inbox.ru) at May 15, 2012 01:37 PM

new error handler

May 15, 2012 01:36 PM

New Error handler

May 15, 2012 01:35 PM

properly support bit test operations under 64 bit

by Camillo Bruni (camillo.bruni@inria.fr) at May 15, 2012 01:20 PM

Small refactorings and bugfixes

by Dennis Schenk (d.schenk@students.unibe.ch) at May 15, 2012 12:28 PM

Small refactorings and bugfixes

by Dennis Schenk (d.schenk@students.unibe.ch) at May 15, 2012 12:27 PM

64 and 63 combined

by Assargadon (tallman@inbox.ru) at May 15, 2012 12:24 PM

431 and 430 combined

by Assargadon (tallman@inbox.ru) at May 15, 2012 12:23 PM

minor fix to ZnStaticFileServerDelegate

by Sven Van Caekenberghe (sven@beta9.be) at May 15, 2012 12:16 PM

Some tests for the annotation stuff.

by Michael Rüfenacht (m.ruefenacht@students.unibe.ch) at May 15, 2012 12:03 PM

Implemented annotation grammar of doctrine.

by Michael Rüfenacht (m.ruefenacht@students.unibe.ch) at May 15, 2012 12:03 PM

Lot of improvements for second try of lunch

by Assargadon (tallman@inbox.ru) at May 15, 2012 11:08 AM

empty log message

by Assargadon (tallman@inbox.ru) at May 15, 2012 11:07 AM

- just a try

by Benjamin Van Ryseghem (benjamin.vanryseghem@gmail.com) at May 15, 2012 10:49 AM

Removed doc block parser tests.

by Michael Rüfenacht (m.ruefenacht@students.unibe.ch) at May 15, 2012 10:33 AM

Deleted the additional docBlock parser (functionality is added to the core parser).

by Michael Rüfenacht (m.ruefenacht@students.unibe.ch) at May 15, 2012 10:32 AM

Updated #lineBarCompositeDiagram2

May 15, 2012 09:54 AM

Fixing to allow regular axis (#regularAxis) on diagram when max value is set.

May 15, 2012 09:53 AM

after big crash :)

by Johannes Rasche (johannes.rasche@myNeat.de) at May 15, 2012 09:03 AM

Fix Mac OS X libc name. In 10.5 it changed to libgcc_s.1.dylib from libc.dylib

by Eliot Miranda (eliot.miranda@gmail.com) at May 14, 2012 11:16 PM

= on BaseUnit

May 14, 2012 09:35 PM

= on BaseUnit

May 14, 2012 09:35 PM

A proof of concept I did some time ago. The idea was to subclass of NuancedTestCase our serialization tests, and define different stream strategies as nuances. It's quite finished but finally I didn't get convinced of adopting it.

May 14, 2012 07:25 PM

Use Iliad-Core0.9.1.2-JohnnyT.68 for GemStone compatability

May 14, 2012 04:33 PM

For GemStone compatability changed the fork method selector from forkAt:named: to forkAt: in ILSessionManager >> startSessionExpirationProcess

May 14, 2012 04:27 PM

replaced direct usage of ZnValueDelegate with indirect usage through the #onRequestRespond: method

by Sven Van Caekenberghe (sven@beta9.be) at May 14, 2012 04:21 PM

May 14, 2012 03:26 PM

May 14, 2012 03:25 PM

May 14, 2012 03:05 PM

after big restore

by Johannes Rasche (johannes.rasche@myNeat.de) at May 14, 2012 02:19 PM

after big restore

by Johannes Rasche (johannes.rasche@myNeat.de) at May 14, 2012 02:18 PM

after big restore

by Johannes Rasche (johannes.rasche@myNeat.de) at May 14, 2012 02:17 PM

after big restore

by Johannes Rasche (johannes.rasche@myNeat.de) at May 14, 2012 02:15 PM

- Create a changeset preamble from the commit messages when loading a version

by Bert Freudenberg (bert@freudenbergs.de) at May 14, 2012 12:21 PM

added state aliasing, not yet tested

by Manuel Leuenberger (manuel.leu@student.unibe.ch) at May 14, 2012 11:01 AM

init talented streams using layers in specific scopes

by Manuel Leuenberger (manuel.leu@student.unibe.ch) at May 14, 2012 11:00 AM

refactoring, reduced code duplication by creating common superclasses, fixed nulling of state when adapted after state definition

by Manuel Leuenberger (manuel.leu@student.unibe.ch) at May 14, 2012 10:55 AM

empty log message

by Paul RICHTER (RichterPN2002@yahoo.co.jp) at May 14, 2012 10:18 AM

added ZnStaticFileServerDelegateTests for
- expiration
- cache-control
- if-modified-since, not-modified

May 14, 2012 10:12 AM

refactored ZnStaticFileServerDelegate a bit:
- store expiration times as seconds in #defaultMimeTypeExpiration & #mimeTypeExpiration map
- add not only Cache-Control but Expires header as well
- removed unused #responseForFile:
fixed a bug in ZnMessage>>#clearEntity (didn't #close streaming entities with HEAD requests)

May 14, 2012 10:00 AM

Without package requirement.

May 14, 2012 03:12 AM

2.156 : Merged with the code of Ben.

May 14, 2012 02:59 AM

2.156 : Merged with the code of Ben.

May 14, 2012 02:58 AM

2.156 : Merged with the code of Ben.

May 14, 2012 02:58 AM

-tweaks

by John McKeon (p3anoman@gmail.com) at May 14, 2012 01:17 AM

move utility method to tools extensions

May 13, 2012 07:33 PM

Make sure to hide virtual keyboard after loading a sketch

by Hilaire Fernandes (hilaire@ofset.org) at May 13, 2012 05:57 PM

- add GemStone verions

by marschal (philippe.marschall@gmail.com) at May 13, 2012 05:34 PM

- add REST
- add Welcome

by marschal (philippe.marschall@gmail.com) at May 13, 2012 05:30 PM

- Issue 5875: Too Many Options

May 13, 2012 05:17 PM

Added primOnScreenKbd to show/hide onscreen Android keyboard upon Smalltalk code request.

by Dmitry Golubovsky (golubovsky@gmail.com) at May 12, 2012 01:00 PM

Added primitive onScreenKbd: anInteger. When called with 1, onscreen keyboard appears on screen, with 0 it disappears.

by Dmitry Golubovsky (golubovsky@gmail.com) at May 12, 2012 12:51 PM

Merge and improve hash tests from pad.210

by Nicolas Cellier (nicolas.cellier.aka.nice@gmail.com) at May 11, 2012 09:09 PM

If Nicolas Cellier's new DateAndTime>>#hash is accepted then these tests need to be changed

DateAndTimeEpochTest>>#testHash
DateAndTimeLeapTest>>#testHash
TimespanTest>>#testHash

And I added the test he used to show how the other hash I proposed was suboptimal.

by Nicolas Cellier (nicolas.cellier.aka.nice@gmail.com) at May 11, 2012 09:00 PM

Nicolas Cellier proposed this hash which has fewer collisions than the one I proposed (~1 per 10,000) and is still plenty fast (100x faster than the original). It also can handle the following test:

| date1 date2 |
date1 := DateAndTime new ticks: (DateAndTime unixEpoch + 1 hours)
ticks offset: 0 hours.
date2 := DateAndTime new ticks: (DateAndTime unixEpoch - 2 hours)
ticks offset: -3 hours.
self assert: (date1 = date2) ==> [date1 hash = date2 hash]

by Nicolas Cellier (nicolas.cellier.aka.nice@gmail.com) at May 11, 2012 08:53 PM

Includes an updated version of DateAndTime>>#hash which is faster than the current version (~130x) and also relatively conflict free (~4 per 10,000)

by Nicolas Cellier (nicolas.cellier.aka.nice@gmail.com) at May 11, 2012 08:53 PM

Improve DateAndTime hash speed by ~2 order of magnitude with a collision rate of ~1 per 10,000, following suggestion of pad, reviewed by nice.

Care to HashedCollection rehashAll in postscript.

by Nicolas Cellier (nicolas.cellier.aka.nice@gmail.com) at May 11, 2012 08:49 PM

It allows to accept a method definition from DependencySorter.

by Bert Freudenberg (bert@freudenbergs.de) at May 11, 2012 06:56 PM

Allows to accept method definitions in the DependencyBrowser. Works with class-side methods as well.

by Yoshiki Ohshima (yoshiki@vpri.org) at May 11, 2012 06:53 PM

If Nicolas Cellier's new DateAndTime>>#hash is accepted then these tests need to be changed

DateAndTimeEpochTest>>#testHash
DateAndTimeLeapTest>>#testHash
TimespanTest>>#testHash

And I added the test he used to show how the other hash I proposed was suboptimal.

May 11, 2012 12:15 AM

Nicolas Cellier proposed this hash which has fewer collisions than the one I proposed (~1 per 10,000) and is still plenty fast (100x faster than the original). It also can handle the following test:

| date1 date2 |
date1 := DateAndTime new ticks: (DateAndTime unixEpoch + 1 hours)
ticks offset: 0 hours.
date2 := DateAndTime new ticks: (DateAndTime unixEpoch - 2 hours)
ticks offset: -3 hours.
self assert: (date1 = date2) ==> [date1 hash = date2 hash]

May 11, 2012 12:10 AM

Includes an updated version of DateAndTime>>#hash which is faster than the current version (~130x) and also relatively conflict free (~4 per 10,000)

May 10, 2012 07:47 PM

Show newest change sets first in ChangeSorter (as intended by The Creator; cf. comment in reorderChangeSets)

by Bert Freudenberg (bert@freudenbergs.de) at May 10, 2012 06:52 PM

Add an exception handler to the emit phase of method
generation so that emitting too much code answers the
same error as emitting too little, instead of an ugly error
from the bowels of CompiledMethod class>>new:.

by Eliot Miranda (eliot.miranda@gmail.com) at May 08, 2012 06:25 PM

Avoid spurious decimals in MessageTally reports.

Indeed,

45.8 roundTo: 0.1
-> 45.800000000000004

(45.8 roundTo: 1/10) asFloat
-> 45.8

by Nicolas Cellier (nicolas.cellier.aka.nice@gmail.com) at May 06, 2012 08:02 PM

Avoid assigning to a block argument in inlineCaseStatement-
BranchesIn:localizingVars:
Fix a comment.

by Eliot Miranda (eliot.miranda@gmail.com) at May 05, 2012 10:34 PM

Add editMethod: convenience. Recategorize class-side methods.

by Eliot Miranda (eliot.miranda@gmail.com) at May 05, 2012 08:30 PM

Guard the case when there is a reference to obsolete classes. One could imagine to have a pseudo package to show obsolete classes referenced from this package, but guarding againt an error is one step toward it anyway.

by Bert Freudenberg (bert@freudenbergs.de) at May 05, 2012 12:19 PM

Guard the case when there is a reference to obsolete classes. One could imagine to have a pseudo package to show obsolete classes referenced from this package, but guarding againt an error is one step toward it anyway.

May 05, 2012 04:11 AM

Remove now unused changes file scanning method.

by Bert Freudenberg (bert@freudenbergs.de) at May 04, 2012 03:33 PM

Fix MC override tests: you cannot override a method in your own package

by Bert Freudenberg (bert@freudenbergs.de) at May 04, 2012 03:26 PM

Fix reverting an overloaded method. Uses new changes file scanning method from PackageInfo-Base-bf.61.

by Bert Freudenberg (bert@freudenbergs.de) at May 04, 2012 03:24 PM

Factor out override scanning into changeRecordsForMethod:do: to allow use by MC, too.

by Bert Freudenberg (bert@freudenbergs.de) at May 04, 2012 03:20 PM

Rename Number class>>#readSqueakSyntaxFrom: into #readSmalltalkSyntaxFrom:

This is to provide a common API across dialects for external packages since both VW7.x and Pharo2.0 have #readSmalltalkSyntaxFrom:

Note that Number syntax may vary slightly across dialects and #readSmalltalkSyntaxFrom: is to be understood "in local dialect"...

#readSqueakSyntaxFrom: was quite recent and not used in trunk, so I removed rather than deprecate. If you don't agree, raise your voice.

by Nicolas Cellier (nicolas.cellier.aka.nice@gmail.com) at May 04, 2012 08:51 AM

Stricter methodsWithUnboundGlobals

by Eliot Miranda (eliot.miranda@gmail.com) at May 03, 2012 04:15 PM

Plugins:
Fix ThreadedFFIPlugin's preambleCCode for Pharo's mingw config.
Merge fix for JPEGReadWriter2Plugin (http://bugs.squeak.org/view.php?id=7691)
Merge with VMMaker version s of plugins
JPEGReaderPlugin,
KlattSynthesizerPlugin
Matrix2x3Plugin
Mpeg3Plugin

by Eliot Miranda (eliot.miranda@gmail.com) at May 02, 2012 06:42 PM

-added AioPlugin and SqueakSSLPlugin for cocoa builds

by Esteban Lorenzano (estebanlm@gmail.com) at May 01, 2012 10:46 AM

Allow the system to be extended with concatenation semantics.

by Chris Muller (ma.chris.m@gmail.com) at May 01, 2012 12:49 AM

Merged cmm.471 to allow, but not require, the system to support concatenation semantics.

by Chris Muller (ma.chris.m@gmail.com) at May 01, 2012 12:48 AM

Sort the results of browse references... for inst vars.

by Eliot Miranda (eliot.miranda@gmail.com) at April 30, 2012 11:34 PM

added internetConfigurationPlugin as external for the Unix config

by Guillermo Polito (guillermopolito@gmail.com) at April 30, 2012 05:19 PM

A reimplementation of Travis Griggs' piecesCutWhere:[do:]

by Eliot Miranda (eliot.miranda@gmail.com) at April 29, 2012 12:03 AM

empty log message

by Eliot Miranda (eliot.miranda@gmail.com) at April 28, 2012 11:30 PM

Separately loadable Balloon-Engine shared pool used by both image-leverl code and the Balloon plugin.

by Eliot Miranda (eliot.miranda@gmail.com) at April 28, 2012 11:30 PM

VMMaker 4.9

Add StackInterpreter and related classes, with additional changes to support ObjectMemory / Interpreter refactoring. The organization of the traditional Interpreter and ObjectMemory are now better aligned with the oscog branch of VMMaker, while the generated code for a standard VM remains equivalent to prior versions.

NewObjectMemory and StackInterpreter are not yet ready for use, and as always Cog and StackInterpreter VMs should be build from the oscog branch.

Upgrade note - If problems are experienced in generating a standard VM after loading this version of VMMaker, try recompiling the VMClass hierarchy, and reinitializing ObjectMemory and Interpreter. Pool variables are now used extensively, and in some cases a recompile may be needed to support the updates.

Changes that affect generated code:

- Allow inlining of ObjectMemory>>accessibleObjectAfter:

- All other changes are cosmetic, related to variable ordering, temp variable names, extraneous white space, etc.

- Generated code for the standard interpreter is equivalent to previous version, except that #accessibleObjectAfter: is now inlined.

Interpreter / ObjectMemory refactorings:

- Add StackInterpreter, StackInterpreterPrimitives, InterpreterStackPages, VMStructType and ReadOnlyArrayWrapper from oscog.

- Add CogClass and Cogit stub classes to support the refactoring.

- Add ObjectMemory subclass ClassicObjectMemory. Move methods that were overridden in NewObjectMemory from ObjectMemory to ClassicObjectMemory, and retain abstract implementations in ObjectMemory. Methods with 'self subclassResponsibility' are abstract and are not translated to C.

- The traditional object memory used with Interpreter is now ClassicObjectMemory, and the new object memory for StackInterpreter is NewObjectMemory, both of which are subclassed from ObjectMemory.

by David T. Lewis (lewis@mail.msen.com) at April 28, 2012 08:23 PM

- added config options for SSL plugin on unix

by Igor Stasenko (siguctua@gmail.com) at April 27, 2012 05:27 PM

Add tests for BlockClosure>isClean

by Eliot Miranda (eliot.miranda@gmail.com) at April 26, 2012 07:02 PM

Implement endPC numArgs and numTemps for closures
and contexts (both block and method activations).
Provide CompiledMethod>abstractBytecodeMessagesDo:
et al.
Implement BlockClosure>isClean to identify self-contained blocks.

by Eliot Miranda (eliot.miranda@gmail.com) at April 26, 2012 06:07 PM

Merge with nice.652 (Float unambiguous printing).

by Nicolas Cellier (nicolas.cellier.aka.nice@gmail.com) at April 26, 2012 05:06 PM

PURPOSE:

Any Float now prints with the minimal number of digits that describes it unambiguously.
This way, every two different Float will have a different printed representation.
More over, every Float can be reconstructed from its printed representation with #readFrom:.

self assert: ([:f | f isNaN or: [(Float readFrom: f printString) = f]] value: Float someInstance).

Note that Float nan, Float infinity and Float infinity negated still print as 'NaN' 'Infinity' and '-Infinity' which are compatible with #readFrom:.

RATIONALE:

the old behaviour was obscuring our data like for example:
0.1 successor printString = 0.1 printString.
1.0e-100 printString = '9.99999999999999e-101'.

The old behaviour was returning many digits without any guaranty of exactness which is useless.

The old behaviour was faster (x4) but this is less relevant than exactness.
Similar or better speed should be obtained by controlling number of printed digits if we can afford inexactness.

IMPLEMENTATION:

The essential change was to use #absPrintExactlyOn:base: in #printOn:base:
Side note: this is really a bad name, because it prints the shortest base-representation, not the exact one.
Anyway, we can only print the exact one in even bases.
For example, the exact representation of 0.1 in base 10 is:
0.1 asFraction asScaledDecimal = 0.1000000000000000055511151231257827021181583404541015625s55.

The second change was to marginally fast-up #absPrintExactlyOn:base: main loop by avoiding a #not send and piping decimal point test.
Avoiding the #not makes the intention a tiny bit clearer.
Also of few formatting has been performed in the last lines.

REJECTED CHANGES:

It is possible to move self > 0.0 before self isNaN to statistically reduce the number of tests performed.
This works because Float nan > 0.0 = false.
But this speed-up is quite marginal.

Similarly (self = Infinity) could replace (self isInfinite) and save another send and also another test (because NegativeInfinity can't happen at this stage).
This would be at the price of a class var reference leak.

It could be more interesting to move this #isInfinite test in #printOn:base: in order to gather print rules for exceptional values.
I didn't to avoid duplicating the test in the two branches > 0.0 and < 0.0.

A far more efficient speed-up would be to optimize LargeInteger arithmetic.
I think there is room, the VM is still using byte operations (thus at most 16 bits).

by Nicolas Cellier (nicolas.cellier.aka.nice@gmail.com) at April 26, 2012 05:01 PM

Simple history search for the working copy browser.
Open a Workspace containing all the history (version
summaries) for the current package.

by Eliot Miranda (eliot.miranda@gmail.com) at April 26, 2012 04:41 PM

using Eliot's indentation

by Esteban Lorenzano (estebanlm@gmail.com) at April 26, 2012 12:56 PM

merged ar.5 and tbn.5

by Torsten Bergmann (Torsten.Bergmann@astares.de) at April 26, 2012 12:16 PM

use cursor form since star is not in Pharo

by Torsten Bergmann (Torsten.Bergmann@astares.de) at April 26, 2012 12:14 PM

#clone is not in Pharo, so copy it from Squeak to make at least the FFIPluginTest green again - all of this needs review/reworked

by Torsten Bergmann (Torsten.Bergmann@astares.de) at April 26, 2012 12:12 PM

Faster non-local return through unwind-protects by not
ignoring the firstUnwindContext supplied by the VM.

by Eliot Miranda (eliot.miranda@gmail.com) at April 25, 2012 05:59 PM

added ffi compiler flags

by Esteban Lorenzano (estebanlm@gmail.com) at April 25, 2012 12:32 PM

fix on FFI for mingw

by Esteban Lorenzano (estebanlm@gmail.com) at April 25, 2012 12:30 PM

Add missing initialization in Decompiler>>decompileBlock:

by Nicolas Cellier (nicolas.cellier.aka.nice@gmail.com) at April 22, 2012 02:35 PM

Use new classify:under:suppressIfDefault:logged: method after config map ensured it is loaded.

by Bert Freudenberg (bert@freudenbergs.de) at April 20, 2012 01:31 AM

When manually recategorizing a method, make an entry in the changes file. Otherwise the methpd's latest change record in the changes file will not match what's in the image, confusing Monticello.

by Bert Freudenberg (bert@freudenbergs.de) at April 20, 2012 01:26 AM

When manually recategorizing a method, make an entry in the changes file. Otherwise the methpd's latest change record in the changes file will not match what's in the image, confusing Monticello.

by Bert Freudenberg (bert@freudenbergs.de) at April 20, 2012 01:24 AM

When looking up the change record for an overridden method, ignore all versions from the overriding package, to really get at the original method. (Amended: use category not method for lookup. And fix category inclusion test)

by Bert Freudenberg (bert@freudenbergs.de) at April 20, 2012 01:14 AM

When looking up the change record for an overridden method, ignore all versions from the overriding package, to really get at the original method.

by Bert Freudenberg (bert@freudenbergs.de) at April 19, 2012 11:10 PM

Fix intermittant load problem. We sometimes would get a 'changes file not found' in the middle of updating, but retrying found it just fine. I finally tracked it down to too many files being open. This change re-uses the same read-only copy during a package load (which additionally might speed up loading a tiny bit).

by Bert Freudenberg (bert@freudenbergs.de) at April 19, 2012 04:36 PM

Fix loading of CompiledMethod subclasses. The old code
just used CompiledMethod's format, which includes its
compact class index, hence resulting in instantiating the
subclass returning an instance of CompiledMethod. The
new code masks out CompiledMethod's compact class imdex (if any) and masks in the right compact class index.

by Eliot Miranda (eliot.miranda@gmail.com) at April 18, 2012 05:40 PM

Provide a kbdTest for the EventSensor.

by Eliot Miranda (eliot.miranda@gmail.com) at April 18, 2012 05:29 PM

Stack/CoInterpreter/Cogit:
Implement proper bounds checking for byte access to compiled
methods. Raise errors for accesses outside initialPC to size.

Simulator:
Nuke an old breakpojnt method in the simulator.

by Eliot Miranda (eliot.miranda@gmail.com) at April 18, 2012 06:09 AM

Fix DNU in event recorder playback

by Bert Freudenberg (bert@freudenbergs.de) at April 17, 2012 11:35 PM

Provide a short-cut (cmd-shift-C) for copyReference.

by Eliot Miranda (eliot.miranda@gmail.com) at April 16, 2012 08:53 PM

Provide a short-cut (cmd-shift-C) for copyReference.

Nuke null method in TimeProfileBrowser's menu processing.

Make the same bounds check in the indirect temp ref side
of privateDereference:in:.

by Eliot Miranda (eliot.miranda@gmail.com) at April 16, 2012 08:52 PM

Fix another potential bounds violation in namedTempAt:in:,
e.g. in [ |a| a:=2 ] asContext tempsAndValues

by Eliot Miranda (eliot.miranda@gmail.com) at April 16, 2012 08:29 PM

Fix keyboard filtering in PluggableListMorphOfMany (e.g., the TestRunner's class list)

by Bert Freudenberg (bert@freudenbergs.de) at April 13, 2012 11:34 PM

Better menu augmentation In the working copy browser's
repository list; store diffs/full versions applies to a specific
repository.

by Eliot Miranda (eliot.miranda@gmail.com) at April 13, 2012 05:30 PM

In the working copy browser's repoisitory list, make
load/save repositories & flush cached versions available
when no repository is selected (since these apply to all
repositories). Also group them at the end of the menu
when a repository is selected to indicate they apply to all.
Add a line to be nice.

by Eliot Miranda (eliot.miranda@gmail.com) at April 13, 2012 05:23 PM

CoInterpreter:
Provide a thorough flush primitive for CompiledMethods that discards
all machine code and makes sure that any contexts using the method
have bytecode pcs. Primitive #215 (same as 116 in the Stack VM).
This is much slower than 116 (flushCache) since it has to enumerate
over all heap contexts.

Provide an xray primitive for CompiledMethod that answers if a
method has machine code, and if so if it's machine code is frameless,
and/or refers to a young object. No primitive number. Used to test
the above.

Make printOopShort: print Association keys. Useful for
longPrintOop:, and hence printReferencesTo: etc.

Nuke theFP ~= framePointer assert in externalDivorceFrame:
andContext: since both clients are careful enough to either avoid
or deal with this case.

Slang:
Avoid annoying objectRepresentation already removed warnings on
generation.

Reformat some Slang code in trying to understand mis-inlining
of [inlinedSend] whileTrue. [I know, IIABDFI, but sometmes one has
to format to read and comprehend].

Collapse generateWhileForeverBreakFalse/TrueLoop:on:indent:
methods to one taking a parameter.

by Eliot Miranda (eliot.miranda@gmail.com) at April 13, 2012 04:58 PM

IOSPlugin recategorized

by Esteban Lorenzano (estebanlm@gmail.com) at April 13, 2012 09:26 AM

empty log message

by Esteban Lorenzano (estebanlm@gmail.com) at April 13, 2012 09:23 AM

Fix 2 DNUs with MCOrganizationDefinitions

by Bert Freudenberg (bert@freudenbergs.de) at April 11, 2012 06:17 PM

Revert to a non-thread-safe TranscriptStream>contents to
avoid deadlock in things like Compiler>recompileAll.

by Eliot Miranda (eliot.miranda@gmail.com) at April 11, 2012 04:37 PM

Fix TranscriptStream>ensureEndsWith: & ensureCr

by Eliot Miranda (eliot.miranda@gmail.com) at April 09, 2012 12:16 AM

Hack fix for shift-enclose-' which at least gives enclose for
the comment character ". But still no shift-enclose for
| (which maybe shifted \) or < (which may be shifted ,).

by Eliot Miranda (eliot.miranda@gmail.com) at April 08, 2012 10:31 PM

allMethodsSelect:localTo: aClass

by Eliot Miranda (eliot.miranda@gmail.com) at April 08, 2012 09:10 PM

Make MessageNames pattern text enlargeable (needed now
MessageNames can take many semi-colon separated patterns)

by Eliot Miranda (eliot.miranda@gmail.com) at April 08, 2012 08:19 PM

Refactor bytecode table initialization in the Cogits to mimic initialization
in the Stack/CoInterpreter. Refactor Cogit class initialization so that
only the concrete class need be initialized.

by Eliot Miranda (eliot.miranda@gmail.com) at April 06, 2012 10:59 PM

Moved TempScopeEditor to its own package
(http://source.squeak.org/trunk/TemporaryVariableScopeEditor)

by Eliot Miranda (eliot.miranda@gmail.com) at April 05, 2012 05:00 PM

A useful utility to move method-level temps used in blocks
into their smallest enclosing block scope. See the class side
of TempScopeEditor. Extracted from the Cog package.

by Eliot Miranda (eliot.miranda@gmail.com) at April 05, 2012 04:57 PM

Provide add repository... to package list in Monticello Browser

by Eliot Miranda (eliot.miranda@gmail.com) at April 05, 2012 04:53 PM

- fix windows crashing by setting -O1 for gcc3x-cointerp.c

by Igor Stasenko (siguctua@gmail.com) at April 05, 2012 04:26 PM

merged with latest from Eliot (emm.154)

by Esteban Lorenzano (estebanlm@gmail.com) at April 05, 2012 08:58 AM

Fix bad conceptual bug with become on methods. Unlike full and
incremental GC, the reference from a Cog method to its method
object must not be remapped since they're two halves of the same
object.

by Eliot Miranda (eliot.miranda@gmail.com) at April 03, 2012 08:42 PM

browseAllSelect:localToPackage:

by Eliot Miranda (eliot.miranda@gmail.com) at April 03, 2012 04:06 PM

Added a default value for CompositeStub>>children.

by Colin Putney (cputney@wiresong.ca) at April 03, 2012 12:01 AM

Introduced a state dictionary to WidgetStub. This lets us simulate a widget more easily and correctly: we maintain local state for thinks like button labels or list contents, so that we don't query the model unless something changes, like a real widget would do.

by Colin Putney (cputney@wiresong.ca) at April 03, 2012 12:01 AM

Added icon support.

by Colin Putney (cputney@wiresong.ca) at April 03, 2012 12:00 AM

Extended PluggableTextMorphPlus to support a getMenuSelector with 3 arguments. The third argument, if present, will be the current selection interval.

by Colin Putney (cputney@wiresong.ca) at April 03, 2012 12:00 AM

Added icon support for ToolBuilder-Morphic.

by Colin Putney (cputney@wiresong.ca) at April 02, 2012 11:59 PM

merge

by Colin Putney (cputney@wiresong.ca) at April 02, 2012 11:58 PM

Added icon support to PluggableListMorph and LazyListMorph.

by Colin Putney (cputney@wiresong.ca) at April 02, 2012 11:57 PM

small fix on IOSPlugin

by Esteban Lorenzano (estebanlm@gmail.com) at April 02, 2012 10:29 PM

locale plugin integrated

by Esteban Lorenzano (estebanlm@gmail.com) at April 02, 2012 06:56 PM