Planet Squeak Code

simpler check for extensions

by Tudor Girba (tudor.girba@gmail.com) at March 16, 2010 04:55 PM

added ClassDescription>>isExtendedInPackage:, and isDefinedInPackage:

(plus tests)

by Tudor Girba (tudor.girba@gmail.com) at March 16, 2010 04:55 PM

added ClassDescription>>isExtendedInPackage:, and isDefinedInPackage:

(plus tests)

by Tudor Girba (tudor.girba@gmail.com) at March 16, 2010 04:55 PM

added italic to identify the package extensions

by Tudor Girba (tudor.girba@gmail.com) at March 16, 2010 04:42 PM

Accepting lower case 16rff broke float reading in base 16.
Let NumberParser test auto-detect whether lowercase digit letters are allowed or not, and then disbale non-10-based floating point tests.
This makes the tests green again.

March 16, 2010 04:34 PM

empty log message

March 16, 2010 04:28 PM

empty log message

March 16, 2010 04:19 PM

- Issue 2136: Smalltalk and SmalltalkImage Rewrite Smalltalk to SmalltalkGlobals for some API.

by Stéphane Ducasse (stephane.ducasse@inria.fr) at March 16, 2010 04:15 PM

- Issue 2136: Smalltalk and SmalltalkImage Rewrite Smalltalk to SmalltalkGlobals for some API.

by Stéphane Ducasse (stephane.ducasse@inria.fr) at March 16, 2010 04:15 PM

- Issue 2136: Smalltalk and SmalltalkImage Rewrite Smalltalk to SmalltalkGlobals for some API.

by Stéphane Ducasse (stephane.ducasse@inria.fr) at March 16, 2010 04:15 PM

- Issue 2136: Smalltalk and SmalltalkImage Rewrite Smalltalk to SmalltalkGlobals for some API.

by Stéphane Ducasse (stephane.ducasse@inria.fr) at March 16, 2010 04:15 PM

- Issue 2136: Smalltalk and SmalltalkImage Rewrite Smalltalk to SmalltalkGlobals for some API.

by Stéphane Ducasse (stephane.ducasse@inria.fr) at March 16, 2010 04:15 PM

- Issue 2136: Smalltalk and SmalltalkImage Rewrite Smalltalk to SmalltalkGlobals for some API.

by Stéphane Ducasse (stephane.ducasse@inria.fr) at March 16, 2010 04:15 PM

- Issue 2136: Smalltalk and SmalltalkImage Rewrite Smalltalk to SmalltalkGlobals for some API.

by Stéphane Ducasse (stephane.ducasse@inria.fr) at March 16, 2010 04:14 PM

- Issue 2136: Smalltalk and SmalltalkImage Rewrite Smalltalk to SmalltalkGlobals for some API.

by Stéphane Ducasse (stephane.ducasse@inria.fr) at March 16, 2010 04:14 PM

- Issue 2136: Smalltalk and SmalltalkImage Rewrite Smalltalk to SmalltalkGlobals for some API.

by Stéphane Ducasse (stephane.ducasse@inria.fr) at March 16, 2010 04:14 PM

- Issue 2136: Smalltalk and SmalltalkImage Rewrite Smalltalk to SmalltalkGlobals for some API.

by Stéphane Ducasse (stephane.ducasse@inria.fr) at March 16, 2010 04:14 PM

- Issue 2136: Smalltalk and SmalltalkImage Rewrite Smalltalk to SmalltalkGlobals for some API.

by Stéphane Ducasse (stephane.ducasse@inria.fr) at March 16, 2010 04:14 PM

- Issue 2136: Smalltalk and SmalltalkImage Rewrite Smalltalk to SmalltalkGlobals for some API.

by Stéphane Ducasse (stephane.ducasse@inria.fr) at March 16, 2010 04:14 PM

- Issue 2136: Smalltalk and SmalltalkImage Rewrite Smalltalk to SmalltalkGlobals for some API.

by Stéphane Ducasse (stephane.ducasse@inria.fr) at March 16, 2010 04:13 PM

adding missing packages

March 16, 2010 02:19 PM

New version 1.0-rc3.10515

by Mariano Martínez Peck (marianopeck@gmail.com) at March 16, 2010 02:15 PM

specifying test-packages in the default

March 16, 2010 02:11 PM

fixing some Set and Dict stuff

by Camillo Bruni (camillobruni@students.unibe.ch) at March 16, 2010 02:10 PM

Moving tests in this new package

March 16, 2010 02:07 PM

moving tests in this new package

March 16, 2010 02:07 PM

moving tests in this new package

March 16, 2010 02:06 PM

Moving tests to a separated package

March 16, 2010 02:05 PM

moving tests to a separated package

March 16, 2010 02:05 PM

Moving tests to a separated package

March 16, 2010 02:04 PM

Forgot to update a package

by Mariano Martínez Peck (marianopeck@gmail.com) at March 16, 2010 01:58 PM

- Upload versions
- comments

by Mariano Martínez Peck (marianopeck@gmail.com) at March 16, 2010 01:43 PM

Some comments

by Mariano Martínez Peck (marianopeck@gmail.com) at March 16, 2010 01:43 PM

Some comments

by Mariano Martínez Peck (marianopeck@gmail.com) at March 16, 2010 01:41 PM

- correction of #mcDetectFileBlock: didn't detect particular versions. fixed.

March 16, 2010 01:38 PM

starting to build the pyramidOverview visualization . Accessible by rigth-clicking on a mooseModel in The moose panel

March 16, 2010 12:53 PM

New version 1.1 in development

by Mariano Martínez Peck (marianopeck@gmail.com) at March 16, 2010 12:26 PM

- Updated versions in 1.1.1

by Mariano Martínez Peck (marianopeck@gmail.com) at March 16, 2010 12:23 PM

Added DeltaT>>forYear: to get value at beginning of relevant year.

by Enrico Spinielli (enrico.spinielli@googlemail.com) at March 16, 2010 11:34 AM

empty log message

by Assargadon (tallman@inbox.ru) at March 16, 2010 10:38 AM

- made tests run faster

March 16, 2010 09:15 AM

addressed issue 332

by Tudor Girba (tudor.girba@gmail.com) at March 16, 2010 08:15 AM

merged

by Tudor Girba (tudor.girba@gmail.com) at March 16, 2010 08:09 AM

added mooseName

by Tudor Girba (tudor.girba@gmail.com) at March 16, 2010 08:07 AM

-corregido bug de nombres en nil al cancelar la creacion de objeto
-corregido bug al clonar objeto (si no habia seleccionado)
-corregido bug initialize

by Guillermo Polito (guillermopolito@gmail.com) at March 16, 2010 01:11 AM

empty log message

by Norberto Manzanos (nmanzanos@gmail.com) at March 16, 2010 12:45 AM

actualizada configuracion de metacello

by Guillermo Polito (guillermopolito@gmail.com) at March 16, 2010 12:33 AM

added support for keystroke preview for lists

by Chris Muller (ma.chris.m@gmail.com) at March 16, 2010 12:22 AM

latest packages

by James Foster (Squeak.Source@JGFoster.net) at March 16, 2010 12:13 AM

This slices comprises an update for two packages, Tools and ToolsTest.
The new version of Tools clean MessageTally. A user of MessageTally may now decide whether he wants to close the tally. Although it is convenient to close it (in order to not keep a reference of the compiled method and the class), this behavior is not always wished. Especially when test have to be written!
The user has now the option to not open the result window.

ToolsTest contains few and simple tests.

March 16, 2010 12:11 AM

This slices comprises an update for two packages, Tools and ToolsTest.
The new version of Tools clean MessageTally. A user of MessageTally may now decide whether he wants to close the tally. Although it is convenient to close it (in order to not keep a reference of the compiled method and the class), this behavior is not always wished. Especially when test have to be written!
The user has now the option to not open the result window.

ToolsTest contains few and simple tests.

March 16, 2010 12:11 AM

This slices comprises an update for two packages, Tools and ToolsTest.
The new version of Tools clean MessageTally. A user of MessageTally may now decide whether he wants to close the tally. Although it is convenient to close it (in order to not keep a reference of the compiled method and the class), this behavior is not always wished. Especially when test have to be written!
The user has now the option to not open the result window.

ToolsTest contains few and simple tests.

March 16, 2010 12:10 AM

Agregada una configuracion de metacello

by Guillermo Polito (guillermopolito@gmail.com) at March 16, 2010 12:03 AM

- Improved guard of authorInitialsPerSe.

by Chris Muller (ma.chris.m@gmail.com) at March 16, 2010 12:01 AM

Integrated new ListChooser from Michael Davies.

by Chris Muller (ma.chris.m@gmail.com) at March 16, 2010 12:00 AM

Integrated new ListChooser from Michael Davies.

by Chris Muller (ma.chris.m@gmail.com) at March 15, 2010 11:59 PM

- Integrated new ListChooser from Michael Davies.
- More improved guads of authorInitialsPerSe.

by Chris Muller (ma.chris.m@gmail.com) at March 15, 2010 11:59 PM

- Integrated new ListChooser from Michael Davies.

by Chris Muller (ma.chris.m@gmail.com) at March 15, 2010 11:58 PM

Added remaining significant timezones for West / US.

by Chris Muller (ma.chris.m@gmail.com) at March 15, 2010 11:57 PM

- More improved guads of authorInitialsPerSe.

by Chris Muller (ma.chris.m@gmail.com) at March 15, 2010 11:56 PM

Primera version del object browser

by Guillermo Polito (guillermopolito@gmail.com) at March 15, 2010 11:53 PM

Fix loading of image segments that are from older, but compatible image versions.

March 15, 2010 11:03 PM

previous version has something from other package

March 15, 2010 10:51 PM

message #versions added (lack in previous version and raise an error)

March 15, 2010 10:36 PM

1.0-beta.6:

- improved the GLASS.230-dkh.231 to 1.0-beta.8 upgrade script
- GemTools-Client

March 15, 2010 10:30 PM

Take 2 , now with buffering in readInto:, and next:into: calling that, as per Nicolas Cellier's suggestion.

Also, tests for all four combinations of count = / ~= bytesRead plus buffered parts or not.

March 15, 2010 10:14 PM

Take 2 , now with buffering in readInto:, and next:into: calling that, as per Nicolas Cellier's suggestion.

Also, tests for all four combinations of count = / ~= bytesRead plus buffered parts or not.

March 15, 2010 10:14 PM

Take 2 , now with buffering in readInto:, and next:into: calling that, as per Nicolas Cellier's suggestion.

Also, tests for all four combinations of count = / ~= bytesRead plus buffered parts or not.

March 15, 2010 10:14 PM

Spring layout

March 15, 2010 10:05 PM

Fix for readInto:startingAt:count: returning wrong count if count elements were successfully read.

Also includes nextLine test from Squeak.

March 15, 2010 08:09 PM

Fix for readInto:startingAt:count: returning wrong count if count elements were successfully read.

Also includes nextLine test from Squeak.

March 15, 2010 08:09 PM

Fix for readInto:startingAt:count: returning wrong count if count elements were successfully read.

Also includes nextLine test from Squeak.

March 15, 2010 08:09 PM

Let #readInto:startingAt:count: handle the buffer and answer number of byte reads and always avoid a copy, and #next:into:startingAt: use it and eventually use a copy if number of bytes read less than requested count.

by Nicolas Cellier (ncellier@ifrance.com) at March 15, 2010 08:08 PM

Minor fix

March 15, 2010 07:59 PM

Fix for readInto:startingAt:count: returning wrong number if entire count was successfully read into a buffer of size > count.

March 15, 2010 07:54 PM

Missing parantesis in new test.

March 15, 2010 07:53 PM

Fix testReadIntoStartingAtCount

by Nicolas Cellier (ncellier@ifrance.com) at March 15, 2010 07:41 PM

Fix StandardFileStream >>readInto:startingAt:count:
It did answer size of buffer rather than number of bytes read.

by Nicolas Cellier (ncellier@ifrance.com) at March 15, 2010 07:39 PM

Payments are here!

by Assargadon (tallman@inbox.ru) at March 15, 2010 06:37 PM

- fix baseline10rc22 ... use ProfStef project reference

March 15, 2010 05:50 PM

Added attributes

March 15, 2010 05:31 PM

Fixed testReadIntoStartingAtCount.

by Chris Muller (ma.chris.m@gmail.com) at March 15, 2010 05:19 PM

Moved comments in testReadIntoStartingAtCount.

by Chris Muller (ma.chris.m@gmail.com) at March 15, 2010 05:16 PM

Added FileStreamTest>>#testReadIntoStartingAtCount.

by Chris Muller (ma.chris.m@gmail.com) at March 15, 2010 05:12 PM

DeltaT works ok.
Polynomial prints and computes ok.

by Enrico Spinielli (enrico.spinielli@googlemail.com) at March 15, 2010 05:09 PM

empty log message

by Assargadon (tallman@inbox.ru) at March 15, 2010 04:58 PM

empty log message

by Assargadon (tallman@inbox.ru) at March 15, 2010 04:50 PM

Reverted method insertMethodPane:atIndex: removed by mistake (it was overridden by ShoutWhisker).

by George Herolyants at March 15, 2010 03:12 PM

Slightly more padding to allow room for titlebar icons

by Michael Davies (mykdavies@gmail.com) at March 15, 2010 03:09 PM

Improved the initial calculation of extent.

by Michael Davies (mykdavies@gmail.com) at March 15, 2010 03:02 PM

adding basic new native for Set

by Camillo Bruni (camillobruni@students.unibe.ch) at March 15, 2010 02:47 PM

Timestamps reverted for methods with just underscore assignments replaced (or formatting changed).

by George Herolyants at March 15, 2010 02:34 PM

Test Monticello

by Max-B (mbertim@gmail.com) at March 15, 2010 02:11 PM

adds support for keyboard preview for lists

by Michael Davies (mykdavies@gmail.com) at March 15, 2010 02:09 PM

added support for keystroke preview for lists

by Michael Davies (mykdavies@gmail.com) at March 15, 2010 02:08 PM

Adds ability for model to request preview of all keystroke events and decide whether to handle them itself.

by Michael Davies (mykdavies@gmail.com) at March 15, 2010 02:08 PM

To support MAD-ListChooser-MAD.8

by Michael Davies (mykdavies@gmail.com) at March 15, 2010 02:01 PM

empty log message

by Assargadon (tallman@inbox.ru) at March 15, 2010 01:59 PM

To support MAD-ListChooser-MAD.8

by Michael Davies (mykdavies@gmail.com) at March 15, 2010 01:50 PM

To support MAD-ListChooser-MAD.8

by Michael Davies (mykdavies@gmail.com) at March 15, 2010 01:50 PM

A new List Chooser

by Michael Davies (mykdavies@gmail.com) at March 15, 2010 01:46 PM

empty log message

by Assargadon (tallman@inbox.ru) at March 15, 2010 01:45 PM

replaced copy by shallowCopy

March 15, 2010 01:44 PM

added missing pequals on String

by Camillo Bruni (camillobruni@students.unibe.ch) at March 15, 2010 01:40 PM

empty log message

by Assargadon (tallman@inbox.ru) at March 15, 2010 01:34 PM

added error printout for failing assertions with should

by Camillo Bruni (camillobruni@students.unibe.ch) at March 15, 2010 01:26 PM

Replaced the call of #clone by #copy. All the 233 are green

March 15, 2010 01:22 PM

small changes; inlining makes it slightly faster

by Toon Verwaest (verwaest@iam.unibe.ch) at March 15, 2010 01:09 PM

removing all distributionMap classes

March 15, 2010 12:53 PM

adding sub-category DistributionMap containing all classes about

March 15, 2010 12:53 PM

don't throw away buckets

by Toon Verwaest (verwaest@iam.unibe.ch) at March 15, 2010 12:42 PM

avialble method sources are added to C code

by Camillo Bruni (camillobruni@students.unibe.ch) at March 15, 2010 12:28 PM

fixing minor bug

March 15, 2010 12:24 PM

refactoring

March 15, 2010 11:49 AM

refactoring

March 15, 2010 11:44 AM

bit nicer version; now ratio is always stored as smallint.

by Toon Verwaest (verwaest@iam.unibe.ch) at March 15, 2010 11:02 AM

adds support for keyboard preview for lists

March 15, 2010 10:51 AM

added support for keystroke preview for lists

March 15, 2010 10:50 AM

- added a missing subclassResponsibility: MorphicProject >> #currentVocabulary

by Levente Uzonyi (leves@elte.hu) at March 15, 2010 06:14 AM

- added compatibility method #keysAndValuesDo: to SmalltalkImage. (used by Lint)
- DataStream's TypeMap changed to a WeakIdentityKeyDictionary, because it has all variable word classes as keys, which may be unloaded from the system (for example MCMockClassG is unloaded during the tests). If that happens, the class' hash will change making the Dictionary invalid.
- don't add obsolete classes to TypeMap during DataStream >> #initialize

by Levente Uzonyi (leves@elte.hu) at March 15, 2010 05:44 AM

- added missing negation

by Levente Uzonyi (leves@elte.hu) at March 15, 2010 04:04 AM

Lint:
- removed supersends where we can be sure that no superclass implements the method
- implemented missing Collection >> #atRandom:, removed the implementation from HashedCollection, because it's the same
- removed double indexing from SequenceableCollection >> #collect:from:to: while kept the same number of bytecodes inside the loop
- unified categorization of several methods in the hierarchy

by Levente Uzonyi (leves@elte.hu) at March 15, 2010 04:04 AM

Use Character>>digitValue..
Now that it is fast enough, there is no point in replicating the algorithm.

by Nicolas Cellier (ncellier@ifrance.com) at March 14, 2010 11:18 PM

Speed-up Character digitValue PART 2.
Now get rid of class var initialization guard.
Also use value instead of charCode to reach almost a x2 speedup (5x for lowercase).

Benchmark:
['0123456789' do: [:e | e digitValue]] bench
NEW '312686.2627474505 per second.'
ORIG '197284.9430113977 per second.'

['0123456789ABCDEF' do: [:e | e digitValue]] bench
NEW '208848.6302739452 per second.
BEFORE '120782.4435112977 per second.'

['0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ' do: [:e | e digitValue]] bench
NEW '116395.9208158368 per second.'
ORIG '55508.49830033993 per second.'

['0123456789abcdefghijklmnopqrstuvwxyz' do: [:e | e digitValue]] bench
NEW '108928.8142371526 per second.'
ORIG '21273.34533093381 per second.'

by Nicolas Cellier (ncellier@ifrance.com) at March 14, 2010 11:00 PM

1) Cache Character DigitValues to gain some speed.
Warning: continue parsing lowercase digits, though it was not consensual.
Note: class var initialization testing will be removed in nxt release. It's only an upgrade guard.
2) Avoid using size == 0

Benchmark:
['0123456789' do: [:e | e digitValue]] bench
AFTER '233969.8060387922 per second.' '236418.5162967407 per second.'
BEFORE '188964.4071185763 per second.' '197284.9430113977 per second.'

['0123456789ABCDEF' do: [:e | e digitValue]] bench
AFTER '155123.375324935 per second.' '152030.1939612078 per second.'
BEFORE '120782.4435112977 per second.' '119901.4197160568 per second.'

['0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ' do: [:e | e digitValue]] bench
AFTER '73469.7060587882 per second.' '73144.3711257749 per second.'
BEFORE '55508.49830033993 per second.' '55637.2725454909 per second.'

['0123456789abcdefghijklmnopqrstuvwxyz' do: [:e | e digitValue]] bench
AFTER '71603.8792241552 per second.' '72621.875624875 per second.'
BEFORE '21194.16116776645 per second.' '21273.34533093381 per second.'

by Nicolas Cellier (ncellier@ifrance.com) at March 14, 2010 10:30 PM

Cache well known digit values in NumberParser for speed.
Testing the base as of previous implementation was not a good idea, because some Unicode characters could have a digitValue < 10.

by Nicolas Cellier (ncellier@ifrance.com) at March 14, 2010 10:19 PM

Let NumberParser test auto-detect whether lowercase digit letters are allowed or not, and then disbale non-10-based floating point tests.
This make the tests green again.

by Nicolas Cellier (ncellier@ifrance.com) at March 14, 2010 10:15 PM

Add #disabledColor to the ColorTheme (to match okColor and cancelColor)

March 14, 2010 03:24 PM

Fix trait composition test in Monticello which could fail in some circumstances.

by Andreas Raab (andreas.raab@gmx.de) at March 14, 2010 08:27 AM

- use TextCoverter to encode/decode strings

March 13, 2010 07:17 PM

- added stringencoding/decoding capabilities to TextConverter

March 13, 2010 07:17 PM

Adds ability for model to request preview of all keystroke events and decide whether to handle them itself.

March 13, 2010 03:29 PM

A one-line change to make the Magnifier Morph eat less CPU.

March 13, 2010 05:28 AM

Faster String>>convertFromWithConverter: from Pharo.

Useful in f.ex. asVMPathName.

Test:

[|converter|
converter := UTF8TextConverter new.
1 to: 50000 do: [:ix | 'abcæøåaskjdhfasiugbvsipruvnasipvunasivunapivunasieun' convertToWithConverter: converter]] timeToRun

March 13, 2010 05:27 AM

A one-line change to make the Magnifier Morph eat less CPU.

by Andreas Raab (andreas.raab@gmx.de) at March 13, 2010 05:13 AM

Faster String>>convertFromWithConverter: from Pharo.

Useful in f.ex. asVMPathName.

Test:

[|converter|
converter := UTF8TextConverter new.
1 to: 50000 do: [:ix | 'abcæøåaskjdhfasiugbvsipruvnasipvunasivunapivunasieun' convertToWithConverter: converter]] timeToRun

by Andreas Raab (andreas.raab@gmx.de) at March 13, 2010 05:09 AM

Let the debugger retain context selection when requested for fullStack.

Note: I don't like (self changed: #flash) for these reasons:
- it traverses all dependents (inefficient MVC like process)
- the text flash (not always, I don't know why), not the list from which the menu was triggered.

Maybe it would be better to disable the menu and the button when (contextStack last sender isNil).

by Nicolas Cellier (ncellier@ifrance.com) at March 12, 2010 10:02 PM

Faster String>>convertFromWithConverter: from Pharo.

Useful in f.ex. asVMPathName.

Test:

[|converter|
converter := UTF8TextConverter new.
1 to: 50000 do: [:ix | 'abcæøåaskjdhfasiugbvsipruvnasipvunasivunapivunasieun' convertToWithConverter: converter]] timeToRun

March 12, 2010 02:38 PM

A one-line change to make the Magnifier Morph eat less CPU.

March 12, 2010 08:43 AM

When removing message from a MessageTrace, maintain the selection index.

by Chris Muller (ma.chris.m@gmail.com) at March 12, 2010 06:09 AM

Fix for pruning the last message of a MessageTrace.

by Chris Muller (ma.chris.m@gmail.com) at March 11, 2010 07:55 PM

Fix SystemChangeNotification for traits.

NB, SystemNotification is broken: It requires passing in *both* old and new class/trait verbatim even though they cannot both exist at the same time and even though copying is not desirable (for the current trait implementation copying would mean recompiling all methods). So we pass in a shallowCopy to avoid excessive recompilation but of course if a client ever looks deeply inside it it will quickly find that it's not a copy. This should be fixed by passing a description of some sort (similar to MCClassDefinition) or specifically the values that were changed.

by Andreas Raab (andreas.raab@gmx.de) at March 11, 2010 04:52 PM

Adds Host WindowEvent handling

by Alexander Lazarević (Alexander@Lazarevic.de) at March 11, 2010 08:35 AM

Adds Host WindowEvent handling

by Alexander Lazarević (Alexander@Lazarevic.de) at March 11, 2010 08:33 AM

Add preference to make new version check before save optional. Also truncate newer version list to 150 characters to avoid extremely long messages.

by Andreas Raab (andreas.raab@gmx.de) at March 11, 2010 05:16 AM

Part 4 of 4 of the fix for
http://bugs.squeak.org/view.php?id=7158

by Matthew Fulmer (tapplek@gmail.com) at March 11, 2010 12:37 AM

Part 3 of 4 of the fix for
http://bugs.squeak.org/view.php?id=7158

by Matthew Fulmer (tapplek@gmail.com) at March 11, 2010 12:33 AM

Part 2 of 4 of the fix for
http://bugs.squeak.org/view.php?id=7158

by Matthew Fulmer (tapplek@gmail.com) at March 11, 2010 12:28 AM

Preload a critical method just in case of load order issues. Part 1 of 4 of the fix for
http://bugs.squeak.org/view.php?id=7158

by Matthew Fulmer (tapplek@gmail.com) at March 11, 2010 12:25 AM

Multiple automatic AmbigousSelector replacement was broken dur to subtle requestorOffset magic. Attempt a correction.

I just made the replacement work, but the second occurence selection just flash , I don't know why.

I do not really understand what I programmed... This is a door open to future bugs.
The requestorOffset inst var in Parser looks like a fragile construction to me.
I dont understand selectInvisiblyFrom:to: neither...

Not sure what happens if we mix UnsedVariable, AmbiguousSelector and other automatic corrections.

Probably more work on this subject

by Nicolas Cellier (ncellier@ifrance.com) at March 10, 2010 08:39 AM

Graphics fixes:
- Add CursorWithAlpha from OLPC. Try this:
CursorWithAlpha biggerNormal showWhile:[Sensor waitButton].
- Fix Cursor>>showWhile: to not allow changing the cursor unless being invoked from the UI process.
- Fix Form>>asFormOfDepth: to have alpha set when converting from 16 -> 32 bit.

by Andreas Raab (andreas.raab@gmx.de) at March 10, 2010 06:47 AM

Update base64 tests.

by Andreas Raab (andreas.raab@gmx.de) at March 10, 2010 06:25 AM

Give Base64MimeConverter the ability to not break lines longer than 70 characters. Useful (for example) when using long base64 encodings in urls.

by Andreas Raab (andreas.raab@gmx.de) at March 10, 2010 06:24 AM

Update test to require Dictionary>>at: to raise KeyNotFound.

by Andreas Raab (andreas.raab@gmx.de) at March 10, 2010 06:12 AM

Signal KeyNotFound instead of a generic error and provide the key in the exception and error message.

by Andreas Raab (andreas.raab@gmx.de) at March 10, 2010 06:10 AM

Add KeyNotFound error which is signaled when a key in a collection cannot be found.

by Andreas Raab (andreas.raab@gmx.de) at March 10, 2010 06:06 AM

- Comments for Collection>>min,max
- Optimized implementations for Bag>>min,max,sum

by Andreas Raab (andreas.raab@gmx.de) at March 10, 2010 06:00 AM

Provide additional support for initialize and unload.

by Andreas Raab (andreas.raab@gmx.de) at March 10, 2010 05:46 AM

- removed the postscript

by Levente Uzonyi (leves@elte.hu) at March 10, 2010 04:45 AM

- use WeakIdentityKeyDictionary instead of WeakKeyDictionary in WeakRegistry
- added a postscript to update existing WeakRegistries

by Levente Uzonyi (leves@elte.hu) at March 10, 2010 04:40 AM

Add initialize/unload support for splObj entries.

by Andreas Raab (andreas.raab@gmx.de) at March 10, 2010 03:47 AM

Get rid of a few *smbase extensions/overrides.

by Andreas Raab (andreas.raab@gmx.de) at March 10, 2010 03:36 AM

Get rid of a few *smbase extensions/overrides.

by Andreas Raab (andreas.raab@gmx.de) at March 10, 2010 03:36 AM

Get rid of a few *smbase extensions/overrides.

by Andreas Raab (andreas.raab@gmx.de) at March 10, 2010 03:35 AM

- restore lost pragma colorization

by Levente Uzonyi (leves@elte.hu) at March 10, 2010 02:53 AM

- DependencyBrowser gets it's package list from PackageInfo instead of MCWorkingCopy. This way the Tools package doesn't depend on Monticello. But the list of packages are different, that should be fixed.

by Levente Uzonyi (leves@elte.hu) at March 10, 2010 02:53 AM

Put the newer version check back into saveVersion. It prevents accidental commits when there are newer (unmerged) versions in the repository. The test got lost probably accidentally a while ago; if the removal was deliberate we should make this a preference (I'm too lazy to live without it :-)

by Andreas Raab (andreas.raab@gmx.de) at March 10, 2010 01:17 AM

Three changes:
- put Smalltalk>>at:ifAbsentPut: back (merge ar.285)
- fix for SystemDictionary>>bindingOf:
- a bit of refactoring for saveAs allowing a bit more flexibility for image packaging purposes

by Andreas Raab (andreas.raab@gmx.de) at March 10, 2010 01:08 AM

Add SystemDictionary>>at:ifAbsentPut: as member of the at:[put:|ifAbsent:|ifPresent:] family of supported methods for accessing globals. Used (for example) by Tweak.

by Andreas Raab (andreas.raab@gmx.de) at March 10, 2010 12:48 AM

Fix MessageSet filtering by author, when the author initials have not yet been specified.

by Chris Muller (ma.chris.m@gmail.com) at March 09, 2010 10:44 PM

- Fixed bug with MenuMorph>>#addStayUpIcons.
- Utilities class>>#authorInitialsPerSe is often set to an empty string, not nil. Updated some guards accordingly.
- The changed-indicator was made two-pixels thick from 3.10. Instead of putting it back to one, I softened it a bit, which looks better and more refined.

by Chris Muller (ma.chris.m@gmail.com) at March 09, 2010 09:27 PM

Fix for http://bugs.squeak.org/view.php?id=7434

by Alexander Lazarević (Alexander@Lazarevic.de) at March 09, 2010 08:38 PM

Fix for http://bugs.squeak.org/view.php?id=7438

by Alexander Lazarević (Alexander@Lazarevic.de) at March 09, 2010 04:06 PM

A fix for http://bugs.squeak.org/view.php?id=7464

by Alexander Lazarević (Alexander@Lazarevic.de) at March 09, 2010 03:02 PM

Add one more message for backward compatibility with obsolete FreeType overrides.

by Nicolas Cellier (ncellier@ifrance.com) at March 08, 2010 10:22 PM

Check that charCode < 256 before testing stopConditions
This test was absent in one method - assuming encoding = 0 ==> (charCode < 256) which is not true anymore (Unicode leadingChar = 0).
This test was charCode < stops size which now DNU since introduction of TextStopConditions, and would have groked A macron otherwise.

by Nicolas Cellier (ncellier@ifrance.com) at March 08, 2010 09:22 PM

move OutOfScopeNotification to Compiler

by Nicolas Cellier (ncellier@ifrance.com) at March 08, 2010 09:14 PM

move OutOfScopeNotification to Compiler

by Nicolas Cellier (ncellier@ifrance.com) at March 08, 2010 09:11 PM

MessageTrace fixes. When filtering the message-list, MessageTrace must know whether "Nothing would be in the list if you did that." Also, when pruning the list, the messageListIndex is now maintained.

by Chris Muller (ma.chris.m@gmail.com) at March 08, 2010 08:48 PM

move SyntaxErrorNotification where it belongs, Compiler

by Nicolas Cellier (ncellier@ifrance.com) at March 08, 2010 08:10 PM

move SyntaxErrorNotification where it belongs, Compiler
This creates a dependency on ToolSet, but Compiler did already depend on System.

by Nicolas Cellier (ncellier@ifrance.com) at March 08, 2010 08:07 PM

- #crossedX and #endOfRun is accessible again by index in TextStopConditions

March 08, 2010 03:45 PM

- use the old sized array for TextStopConditions

March 08, 2010 03:45 PM

Be backward compatible with obsolete code.
Otherwise, VMMaker update cannot be loaded safely in the image.

by Nicolas Cellier (ncellier@ifrance.com) at March 08, 2010 01:39 PM

Restore TextStopConditions backward compatibility support for FreeType overrides.
Otherwise, VMMaker update cannot be loaded safely in the image.

Add a long comment to TextStopConditions.
I cannot complain about comments if I don't put some myself.
Maybe it does not belong here, but hey, that's better than nothing.

by Nicolas Cellier (ncellier@ifrance.com) at March 08, 2010 01:37 PM

Fix from Vaidotas Did??balis

Move PluggableTextMorph >> textMorph from "*sevices-base" to "accessing" category, then after unloading all packages user can change font.

by David T. Lewis (lewis@mail.msen.com) at March 08, 2010 12:30 PM

Fix from Vaidotas Did??balis

Move PluggableTextMorph >> textMorph from "*sevices-base" to "accessing" category, then after unloading all packages user can change font.

by David T. Lewis (lewis@mail.msen.com) at March 08, 2010 12:29 PM

adapt FileList2/FileChooser to the last change

by Alexander Lazarević (Alexander@Lazarevic.de) at March 08, 2010 11:05 AM

Fixes file List display when sorted by size. (Visually this probably only works right with a fixed font)

by Alexander Lazarević (Alexander@Lazarevic.de) at March 08, 2010 09:33 AM

First part of Smalltalk/SystemDictionary conversion:
- Copy all missing methods from SystemDictionary to SmalltalkImage
- Convert Smalltalk to SmalltalkImage current globals
- Convert SmalltalkImage current to Smalltalk
This leaves the system in a state where a *lot* of methods are intentionally duplicated between SmalltalkImage and SystemDictionary to give us room to clean up things once we have the basics in place.

March 08, 2010 07:48 AM

Simple test for Collection>>#minMax

March 08, 2010 07:48 AM

Return both the minimum and maximum of the given collection in a single pass.

March 08, 2010 07:48 AM

Return both the minimum and maximum of the given collection in a single pass.

Better (certainly reads better) implementation from Randal Schwartz.

March 08, 2010 07:48 AM

Simple test for Collection>>#minMax

by Andreas Raab (andreas.raab@gmx.de) at March 08, 2010 07:42 AM

Return both the minimum and maximum of the given collection in a single pass.

Better (certainly reads better) implementation from Randal Schwartz.

by Andreas Raab (andreas.raab@gmx.de) at March 08, 2010 07:41 AM

- use the old sized array for TextStopConditions

March 08, 2010 03:59 AM

- #crossedX and #endOfRun is accessible again by index in TextStopConditions

March 08, 2010 03:59 AM

- renamed PackageDependencyTest's #testPackage:dependsOnlyOn: to #testPackage: dependsExactlyOn:, since that's the intended behavior.
- removed fixed dependencies from #testCompiler, #testExceptions and #testMorphic

by Levente Uzonyi (leves@elte.hu) at March 08, 2010 01:01 AM

Return both the minimum and maximum of the given collection in a single pass.

Better (certainly reads better) implementation from Randal Schwartz.

March 08, 2010 12:51 AM

Simple test for Collection>>#minMax

March 08, 2010 12:39 AM