darcs patch: Find all methods, in order of decreasing... (and 4 more)

View: New views
1 Messages — Rating Filter:   Alert me  

darcs patch: Find all methods, in order of decreasing... (and 4 more)

by Tony Garnock-Jones-2 :: Rate this Message:

Reply to Author | View Threaded | Show Only this Message

Wed Jun  7 21:00:55 BST 2006  Tony Garnock-Jones <tonyg@...>
  * Find all methods, in order of decreasing specificity, for a selector.

Wed Jun  7 21:07:30 BST 2006  Tony Garnock-Jones <tonyg@...>
  * Remove pun of #as: for encoding/decoding - needs a better codec framework

Wed Jun  7 21:07:47 BST 2006  Tony Garnock-Jones <tonyg@...>
  * Replace pun of #as: for ASCII-encoding with {en,de}codeASCII
 
  Using #as: to convert strings into byte-arrays and vice versa is
  problematic for two reasons:
 
  1. it defaults to ASCII encoding without allowing specification of the
  particular codec that should be used; and
 
  2. it inadvertently overrides (String as: Symbol), causing problems
  with non-interning.
 
  Removing the punning of #as: for String codecing and replacing it with
  a more precise API takes the first step along the road toward a proper
  codec framework, and also avoids the nasty side-effect preventing
  Symbol interning in some places where it's needed (e.g. CObject's
  parseDefinitionsFrom:into:).

Wed Jun  7 21:45:58 BST 2006  Tony Garnock-Jones <tonyg@...>
  * Update Directory primitives to use encodeASCII/decodeASCII

Wed Jun  7 21:50:33 BST 2006  Tony Garnock-Jones <tonyg@...>
  * Use encodeASCII/decodeASCII in extlib.slate


New patches:

[Find all methods, in order of decreasing specificity, for a selector.
Tony Garnock-Jones <tonyg@...>**20060607200055] {
hunk ./src/lib/symbol.slate 129
+s@(Symbol traits) findAllMethodsOn: args
+[| w m |
+  w: (ExtensibleArray new writer).
+  m: (s findOn: args).
+  [m isNotNil] whileTrue: [
+     w nextPut: m.
+     m: (s findOn: args after: m).
+  ].
+  w collection
+].
+
}

[Remove pun of #as: for encoding/decoding - needs a better codec framework
Tony Garnock-Jones <tonyg@...>**20060607200730] {
hunk ./src/i18n/encodings-unicode.slate 492
-
-b@(ByteArray traits) as: s@(UnicodeString traits)
-[
- b collect: [| :each | each as: UnicodeCharacter] into: (s newSizeOf: b)
-].
-
-s@(UnicodeString traits) as: a@(ByteArray traits)
-[
- s collect: [| :each | each as: ASCIICharacter] into: (a newSizeOf: s)
-].
-
-s1@(UnicodeString traits) as: _@(UnicodeString traits)
-[
-  s1
-].
}

[Replace pun of #as: for ASCII-encoding with {en,de}codeASCII
Tony Garnock-Jones <tonyg@...>**20060607200747
 
 Using #as: to convert strings into byte-arrays and vice versa is
 problematic for two reasons:
 
 1. it defaults to ASCII encoding without allowing specification of the
 particular codec that should be used; and
 
 2. it inadvertently overrides (String as: Symbol), causing problems
 with non-interning.
 
 Removing the punning of #as: for String codecing and replacing it with
 a more precise API takes the first step along the road toward a proper
 codec framework, and also avoids the nasty side-effect preventing
 Symbol interning in some places where it's needed (e.g. CObject's
 parseDefinitionsFrom:into:).
] {
hunk ./src/lib/string.slate 150
-a@(ByteArray traits) as: s@(String traits)
-"FIXME: calls to this method should be either converted to direct ASCIIString calls,
-or an optional hint should be used to decide on the encoding in the ByteArray."
-[
-  s == String
-    ifTrue: [a as: ASCIIString]
-    ifFalse: [resend]
-].
-
hunk ./src/lib/string.slate 426
-s@(ASCIIString traits) as: a@(ByteArray traits)
-"Since ByteArrays cannot hold objects, they must store character codes."
+s@(ASCIIString traits) encodeASCII
+"FIXME: Eventually this should become something like (String encode: stringcodec ASCII)"
hunk ./src/lib/string.slate 429
-  newA: (a newSizeOf: s).
+  newA: (ByteArray newSizeOf: s).
hunk ./src/lib/string.slate 431
-    [| :each :index | newA at: index put: (each as: a elementType)].
+    [| :each :index | newA at: index put: (each as: newA elementType)].
hunk ./src/lib/string.slate 435
-a@(ByteArray traits) as: s@(ASCIIString traits)
-"Since ByteArrays cannot hold objects, they must store character codes."
+a@(Sequence traits) decodeASCII
+"FIXME: Eventually this should become something like (Sequence decode: stringcodec ASCII)"
hunk ./src/lib/string.slate 438
-  newS: (s newSizeOf: a).
+  newS: (ASCIIString newSizeOf: a).
hunk ./src/lib/string.slate 440
-    [| :each :index | newS at: index put: (each as: s elementType)].
+    [| :each :index | newS at: index put: (each as: newS elementType)].
hunk ./src/lib/string.slate 444
+
}

[Update Directory primitives to use encodeASCII/decodeASCII
Tony Garnock-Jones <tonyg@...>**20060607204558] {
hunk ./src/lib/directory.slate 66
-  result: (dir primitiveOpen: dirname).
+  result: (dir primitiveOpen: dirname encodeASCII).
hunk ./src/lib/directory.slate 86
-    [(dir primitiveOpen: (dir locator as: String)) isNegative not
+    [(dir primitiveOpen: (dir locator as: String) encodeASCII) isNegative not
hunk ./src/lib/directory.slate 100
- ifFalse: [buffer copyFrom: 0 to: len - 1]]
+ ifFalse: [(buffer copyFrom: 0 to: len - 1) decodeASCII]]
hunk ./src/lib/directory.slate 148
-  dir newNamed: ((buffer first: dirLength) as: String)
+  dir newNamed: ((buffer first: dirLength) decodeASCII)
hunk ./src/lib/directory.slate 155
-  returnCode: (dir primitiveSetCurrentDirectory: ((dir locator as: String) as: ByteArray)).
+  returnCode: (dir primitiveSetCurrentDirectory: ((dir locator as: String) encodeASCII)).
}

[Use encodeASCII/decodeASCII in extlib.slate
Tony Garnock-Jones <tonyg@...>**20060607205033] {
hunk ./src/lib/extlib.slate 23
-  buffer: (String newSize: 256).
-  buffer first: (l primitiveErrorInto: buffer)
+  buffer: (ByteArray newSize: 256).
+  (buffer first: (l primitiveErrorInto: buffer)) decodeASCII
hunk ./src/lib/extlib.slate 94
-  (d primitiveOpen: d name handle: d handle)
+  (d primitiveOpen: d name encodeASCII handle: d handle)
hunk ./src/lib/extlib.slate 220
-  (d primitiveLookup: m name handle: d handle into: m handle)
+  (d primitiveLookup: m name encodeASCII handle: d handle into: m handle)
}

Context:

[sdl-keyboard-fix
timmy@...**20060605055503
 
 The windows.slate and sdl-windows plugin
 didn't include support for keyboard button
 presses. I'm not sure why it wasn't there,
 but I went ahead and added it.
 
]
[pass-keyboard-events-to-children
timmy@...**20060606040547
 
 A patch to add an easy, perhaps temporary, way to
 pass the keyboard events to all the children
 SceneElements because I'm not sure inputHandler
 would be for more specific cases I think. There probably
 should be an easier interface for using that feature.
 I also fixed some bugs in the text buffer and
 now pressing a button actually adds text to the buffer.
 
]
[textbox beginnings
timmy@...**20060601015957
 
 In this patch I write a generic interface for textbuffers and then I
 fill out an example one that is based on an array of arrays. I
 extended the Command object for textbuffer manipulations so when
 things need to modify a textbuffer, they can group a bunch of commands
 in a recording and then execute them at once. Those can then be
 replayed with #undo to undo the changes. But where should I store
 these recordings so that they can be played back and stuff? I created
 a file called focus.slate for the beginnings of a way to determine
 what morphs have focus and I think I'll try to figure out in there a
 way that user keyboard input is redirected to the right
 components. But I'm not really sure how I should design that. Right
 now nothing is getting keypresses, so the textbox in the ui.slate file
 is pretty much a dud.
 
 I also did some cleaning in listbox.slate and put the behavior in
 scrollpane where it belongs. Perhaps in the future I can make it so
 scrollpanes don't crash and burn in the drawing method when the
 scrollbars are nil.
 
 I also started a file called fundamental-mode. This would ideally be a
 command list that takes basic key presses and sends the
 textbuffercommands to the buffer. Right now though I have no idea how
 the defined CommandMap is supposed to work (it's a dictionary but what
 maps to what? what does translators do? what kind of object is an
 accelerator? etc)
 
]
[activators-and-listbox-menus
timmy@...**20060528222506
 
 I added menus based on listboxes and centralized
 some functionality in an activator object. I added
 some code to demonstrate menus in ui.slate. In the future,
 I suppose I could make it easier for objects to have
 menus via a mixin and it'd probably be a good idea
 to make them look better when slate speeds up. On
 the other hand, I think it would be interesting to
 dump the traditional vertical menus for a web/pie-menu-like
 visible mouse-gesture menu.
 
 
]
[scrollbar-pagedelta-cleanup
timmy@...**20060522044932]
[scrolling-and-buttons
timmy@...**20060522042106
 
 I added scrollbars and extended the listbox to use them.
 I also made a simple button class and put demos of using
 both components in the ui.slate file. At this point,
 the UI is really slow so it's too much trouble to make
 the scrollbar cool.
 
]
[listbox-and-themeable
timmy@...**20060517030537
 
 I added a basic listbox component to make listboxes,
 I cleaned up the ui.slate file to not use
 #addSlotAndReturnValue, and I also added a basic
 theme framework thingy to centralize the location
 of color information, etc.
 
]
[Bug-fix for unexpected block header tokens in the Parser.
Brian T. Rice <water@...>**20060516015716]
[Update time plugin to support gettimeofday.
Tony Garnock-Jones <tonyg@...>**20060514234717
 
 - Support for Doubles would be nice, since then we could use a
   double-precision float to represent fractions-of-a-second in UTC
   since the epoch.
 
 - When calling GetTimeOfDay applyWith:, initially I just passed the
   CObject itself, expecting that the storage field of the CObject
   would be magically extracted and passed through the FFI to C. This
   turns out not to be the case: I had to say "GetTimeOfDay applyWith:
   tv storage" explicitly.
 
 - We should consider making socket_Select take a CTimeval.
]
[Added #+- and #+/- as Number methods to create Intervals.
Brian T. Rice <water@...>**20060513231219]
[Add explicit .boring file
Tony Garnock-Jones <tonyg@...>**20060504144956]
[Avoid stomping on lib/time.slate's Time from plugins/time/time.slate
Tony Garnock-Jones <tonyg@...>**20060503234342]
[Fix several problems with Time and Duration.
Tony Garnock-Jones <tonyg@...>**20060503234304]
[Factor truncateTo:paddedBy: and truncateTo: out into Sequence
Tony Garnock-Jones <tonyg@...>**20060503225037
 
 The code was duplicated between EncodedString and String. Now it's in
 Sequence, where it kind of belongs.
]
[Note the fact the counting semaphore now works.
Tony Garnock-Jones <tonyg@...>**20060503225020]
[New bug (?) report in BUGS
Tony Garnock-Jones <tonyg@...>**20060503224945]
[Remove debugging print statements from sharedqueue.
Tony Garnock-Jones <tonyg@...>**20060503191625]
[Fix semaphore implementation to *not* share a global waiters queue. Sigh.
Tony Garnock-Jones <tonyg@...>**20060503191002
 
 Mutable state considered harmful? :-( This bug has been present since
 my first submission of the semaphore code. All the hypothetical faults with
 ensure: and friends were red herrings!
]
[less cryptic examples at beginning of documentation
wkornew@...**20060415100454]
[UI: Fixed bitrot
Todd Fleming <todd@...>**20060411230540]
[Simplified Makefile newboot component targets, using a new imagemaker.image explicit target.
Brian T. Rice <water@...>**20060406083010]
[Fixed PrettyPrinter usage in the Debugger.
Brian T. Rice <water@...>**20060330171932]
[Bug-fix in #Cairo namespace setup code.
Brian T. Rice <water@...>**20060409183116]
[Added `erOnKey: as a variant of `er which calls the first selector on all results of applying the key selector to the input arguments.
Brian T. Rice <water@...>**20060406081749]
[Bug-fix for the Cairo binding.
Brian T. Rice <water@...>**20060406010330]
[Cleanups and a fix for SDL-windows plugin Slate code.
Brian T. Rice <water@...>**20060405161132]
[Removed old and cruddy SDL code.
Brian T. Rice <water@...>**20060331061029]
[Used ANSI strchr instead of POSIX index string function in boot.c, fixing native Windows builds.
Brian T. Rice <water@...>**20060403151104]
[Made usage of VARIANT in the BOOTIMG Makefile variable, and changed the blessnew target to bless, where VARIANT customizes its usage.
Brian T. Rice <water@...>**20060401000258]
[Added a VARIANT configuration variable to common.mk so that bootstraps of new VM/image sets can proceed without special command-line futzing.
Brian T. Rice <water@...>**20060331063924]
[Added a `disable macro for the purpose of "commenting-out".
Brian T. Rice <water@...>**20060331062253]
[Use of &slots: in non-core library code.
Brian T. Rice <water@...>**20060331061808]
[Added whereIs: as a convenience method for finding Paths to objects, similar to POSIX `where`.
Brian T. Rice <water@...>**20060330175918]
[Use of the style of recent REPL namespace code to improve the Debugger design.
Brian T. Rice <water@...>**20060330170334]
[Moved the SmallInteger prototype printOn: fix into print.slate.
Brian T. Rice <water@...>**20060330083629]
[Added #addImmutableSlot: for #addImmutableSlot:valued:.
Brian T. Rice <water@...>**20060329203801]
[Added a trivial #defineOddball: method to make new Oddball objects.
Brian T. Rice <water@...>**20060329202214]
[Made the REPL use an interactive parse by default, for ease of use.
Brian T. Rice <water@...>**20060329031249]
[Fixed a typo in File Locator #copy method.
Brian T. Rice <water@...>**20060329030631]
[More pidginPrimitive adjustments.
Brian T. Rice <water@...>**20060321031705]
[Added a Makefile rollback target to undo the effects of blessnew should something go wrong.
Brian T. Rice <water@...>**20060330085023]
[Bug-fix for slotNames to avoid #slotCount call problems.
Brian T. Rice <water@...>**20060330080148]
[Added #slotNames and #delegateNames to Map code, to replace the neglected pre-bootstrap.slate versions.
Brian T. Rice <water@...>**20060330005422]
[Moved interp/map.slate (Map code) up in the prelude load order so that other non-essential libraries can use immutability.
Brian T. Rice <water@...>**20060329211748]
[Clean-ups of derive-related code.
Brian T. Rice <water@...>**20060329221633]
[Fixed #derive to work with immutable prototypes.
Brian T. Rice <water@...>**20060329214038]
[Changed addPrototype: to be the basis of addPrototype:derivedFrom: and added an &immutable option to make immutable objects by default.
Brian T. Rice <water@...>**20060329202053]
[Bug-fixes for Fraction code.
Brian T. Rice <water@...>**20060329034500]
[Allowed for Integer division to provisionally work if Fraction code is not loaded.
Brian T. Rice <water@...>**20060329024738]
[Moved fraction.slate from the core bootstrap to the prelude, since it's not needed.
Brian T. Rice <water@...>**20060329024005]
[Moved Fraction code from numeric.slate to fraction.slate.
Brian T. Rice <water@...>**20060328223610]
[Update code which exercises Semaphore and SharedQueue
Tony Garnock-Jones <tonyg@...>**20060325160851]
[Use signedLongInt_asObject instead of longInt_asObject
Tony Garnock-Jones <tonyg@...>**20060325121803]
[Only add -m32 on i386 CPUs
Tony Garnock-Jones <tonyg@...>**20060325121319]
[Added #parseConcatenatively which enables the old-style statement-separator-less syntax. It is enabled by default now.
Brian T. Rice <water@...>**20060322091155]
[Moved #expected:butFound: and #check:is: from token.slate to parser.slate since it is Parser-centric code.
Brian T. Rice <water@...>**20060322033227]
[Parser code cleanups.
Brian T. Rice <water@...>**20060317201622]
[Added a hook protocol for objects to override lookup in a post-facto manner using #didNotUnderstand:at:.
Brian T. Rice <water@...>**20060321071259]
[Further bug-fixes of silly uses of #[] (they don't make compile-time blocks, they evaluate at compile-time to Nil!).
Brian T. Rice <water@...>**20060321055200]
[Cleaner Slate-quitting from the Makefile targets.
Brian T. Rice <water@...>**20060321032757]
[Added #die and #quit to Image for convenient exit signalling.
Brian T. Rice <water@...>**20060321032434]
[Added the platform plugin as a "newvm" Makefile target prerequisite.
Brian T. Rice <water@...>**20060321031902]
[Fixed #unlessCompletes:.
Brian T. Rice <water@...>**20060320205119]
[Renamed ifCompletes:otherwise: to ifCompletes:ifFails:.
Brian T. Rice <water@...>**20060210195948]
[Bug-fix for SortedArray #copy method.
Brian T. Rice <water@...>**20060320202439]
[More SortedArray and SortedSet bug-fixes.
Brian T. Rice <water@...>**20060320201302]
[SortedArray refactorings, some bug-fixes for safety.
Brian T. Rice <water@...>**20060320184113]
[Moved Perl-style idioms for conditional execution (#if: #if:then: #if:then:else: #while: #until: #unless:) into the normal method.slate control-flow library.
Brian T. Rice <water@...>**20060319200907]
[More pidginPrimitive cleanups and added checks for SmallInteger tags to arguments.
Brian T. Rice <water@...>**20060319013838]
[Refactored #indicesOf: to use a new #indicesOfAllSatisfying: for Sequences.
Brian T. Rice <water@...>**20060318201047]
[Refactored the Makefile newboot target into newvm and newimages, and added a run-check to installcheck target.
Brian T. Rice <water@...>**20060317221547]
[Added the url for the monads in Ruby blog post that inspired this little excursion.
Brian T. Rice <water@...>**20060317204838]
[Syntax Writer cleanups.
Brian T. Rice <water@...>**20060317012134]
[Syntax refactorings to use #surroundings and Block #new.
Brian T. Rice <water@...>**20060316212818]
[Moved SmallInteger primitives from prims.slate to smallint.slate.
Brian T. Rice <water@...>**20060301234509]
[Added #immutable to the VM-Image linkage array SpecialObjectsArray/specialOops to signal trying to change an immutable object.
Brian T. Rice <water@...>**20060111234215]
[Added an Immutable map flag bit, and #isImmutable and #restrictsDelegation methods for Object pointers.
Brian T. Rice <water@...>**20060111232439]
[Mobius IR Optimizer code cleanups.
Brian T Rice <water@...>**20060315025118]
[Syntax Mode code cleanups.
Brian T Rice <water@...>**20060315025043]
[IR code cleanups.
Brian T Rice <water@...>**20060308053854]
[Syntax code cleanups and bug-fixes for creation methods.
Brian T. Rice <water@...>**20060316173930]
[Added support for returning a given number of lexical scopes up the chain.
Brian T. Rice <water@...>**20060219023317]
[Cleanups of pidginPrimitive macro code.
Brian T. Rice <water@...>**20060315032227]
[Cleaned up TypeMismatch signalling for local variable stores.
Brian T. Rice <water@...>**20060315005006]
[Created a sketch version of a conditions: inliner based on the caseOf: inliner, with TODO notes where extra work is needed. This is for the ByteCompiler, of course.
Brian T. Rice <water@...>**20060316054207]
[Created a conditions:(otherwise:) method to complement caseOf:(otherwise:); basically a COND vs. case-switch.
Brian T. Rice <water@...>**20060316050735]
[Fixed printing of the faux-prototype String.
Brian T. Rice <water@...>**20060316041105]
[Fixed the FFI VM module to work with the asObject type change in pidgin.
Brian T. Rice <water@...>**20060314221253]
[Changed type declarations in some Pidgin code from LongInt to UnsignedLongInt for clarity (and reduced warnings).
Brian T. Rice <water@...>**20060313213832]
[Added a PathFinder type which is a stream of possible Path objects to a given target, using pluggable criteria.
Brian T. Rice <water@...>**20060309213352]
[Further split up the Makefile newboot target and allowed a separate imagemaker.image target and vmmaker.image target for fine-grained bootstrap setups.
Brian T. Rice <water@...>**20060309072927]
[Included basic INSTALL notes for the XCode project.
Brian T. Rice <water@...>**20060309072847]
[More pseudo-code detail on an Objective-C bridge.
Brian T. Rice <water@...>**20060309064537]
[Moved RoleEntry-specific code from map.slate to role.slate.
Brian T. Rice <water@...>**20060309013801]
[Updates to the BSD Portfile.
Brian T. Rice <water@...>**20060309010202]
[Renamed File handle-generating primitives from open:/openForInput:/openForOutput:/openForNew: to handleFor*.
Brian T. Rice <water@...>**20060313040654]
[Added a #freeze/#isFrozen protocol for recursive immutable-object creation.
Brian T. Rice <water@...>**20060309010253]
[Refactored the fullCleanBootstrap method so its parts may be called separately.
Brian T. Rice <water@...>**20060307223528]
[Silenced C compiler warning in boot.c.
Brian T. Rice <water@...>**20060307205826]
[Bug-fix for C Generator cleanups.
Brian T. Rice <water@...>**20060307203530]
[Used some method-generation to clean up Types C setup code.
Brian T. Rice <water@...>**20060307095026]
[More C Generator cleanups.
Brian T. Rice <water@...>**20060307093905]
[Defined log: and log10.
Brian T. Rice <water@...>**20060307081842]
[Added nested indentation levels to the printouts for load: calls that recurse.
Brian T. Rice <water@...>**20060307071002]
[Removed the notice from the VM bootstrap sequence for each major line-number since we can now get that in the Debugger easily or from the other notices.
Brian T. Rice <water@...>**20060307064333]
[Cleaned up C-generation of for-loops.
Brian T. Rice <water@...>**20060307062552]
[Added user-land code for immutability.
Brian T. Rice <water@...>**20060302000803]
[Added support to the setTo:/:= macro for selectors that have variants with put: appended.
Brian T. Rice <water@...>**20060305220728]
[Added #arg:put: to the Debugger's added-commands list to set indexed arguments.
Brian T. Rice <water@...>**20060305225946]
[Fixed printing of the SmallInteger "prototype".
Brian T. Rice <water@...>**20060304212703]
[Used cloneSettingSlots: in UI code.
Brian T. Rice <water@...>**20060302092459]
[Used cloneSettingSlots:to: in bootstrap code.
Brian T. Rice <water@...>**20060302085231]
[Non-bootstrap uses of cloneSettingSlots:to:.
Brian T. Rice <water@...>**20060302074222]
[Added cloneSettingSlots:to: to the primitives and out of root.slate to make it faster and useful for using immutable objects.
Brian T. Rice <water@...>**20060302030307]
[Added bitAt: to SmallInteger.
Brian T. Rice <water@...>**20060301224242]
[Fixed the previous load: error-handler.
Brian T. Rice <water@...>**20060301202231]
[Added a generic load: implementation which just throws a warning and tells the user what's not-quite-right about what they're doing.
Brian T. Rice <water@...>**20060301184453]
[Small cleanups to i18n encoding code.
Brian T. Rice <water@...>**20060301004505]
[Fixed the use of combined VM+image in the case where the executable is invoked via use of the PATH.
Brian T. Rice <water@...>**20060227220055]
[Added comments and quick bug-fix for the image-locating code where the VM is called without a fully-qualified path, and FIXME note for later.
Brian T. Rice <water@...>**20060227210153]
[Removed quadtree from the AutoLoad list.
Brian T. Rice <water@...>**20060226225026]
[Added a newautoloaddb target to the Makefile to re-generate the AutoLoad database without undo fuss.
Brian T. Rice <water@...>**20060226212729]
[Renamed the XCode project C file group to "VM base sources".
Brian T. Rice <water@...>**20060227031707]
[Added an XCode project for MacOS X build support (which mostly just wraps the Makefile).
Brian T. Rice <water@...>**20060227030626]
[Added a sketch version of a StateMachine library and FSM subtype that would wrap a Digraph.
Brian T. Rice <water@...>**20060227011613]
[Added a #removeAll and #clear protocol for ExtensibleCollections.
Brian T. Rice <water@...>**20060226211754]
[Added graph/digraph to the AutoLoad scan sequence.
Brian T. Rice <water@...>**20060226210732]
[Defined Association as: Sequence to return a two-element array of key and value.
Brian T. Rice <water@...>**20060226205952]
[More ByteCompiler code cleanups.
Brian T. Rice <water@...>**20060226000801]
[Replaced the #method accessor of the bytecode compiler with #currentMethod and added #currentContext to clarify code.
Brian T. Rice <water@...>**20060225234633]
[Unicode / i18n code cleanups.
Brian T. Rice <water@...>**20060226202507]
[Fixed #from:to:do: to perform proper safety checks, fixing allButFirst/allButLast calls for single-element Sequences.
Brian T. Rice <water@...>**20060226002555]
[Driving addPrototype: to extinction.
Brian T. Rice <water@...>**20060223233238]
[Moved QuadTree code to unfinished/.
Brian T. Rice <water@...>**20060223221936]
[SmartConsole code cleanups.
Brian T. Rice <water@...>**20060222222819]
[Added a #refreshWorkspace method to the REPL.
Brian T. Rice <water@...>**20060224231758]
[Bug-fix for the last REPL changes.
Brian T. Rice <water@...>**20060224222634]
[Modified the REPL to use scratch namespaces delegating to the lobby by default and an #in: method to change that delegation.
Brian T. Rice <water@...>**20060224220257]
[Added #newDelegatingTo: Namespace convenience method.
Brian T. Rice <water@...>**20060224215043]
[Added a skeletal slot.slate to unfinished with a basic ValueSlot type.
Brian T. Rice <water@...>**20060224200104]
[Cleanups of define: to use &slots:.
Brian T. Rice <water@...>**20060223224259]
[Removed redundant line in slate-startup.el.
Brian T. Rice <water@...>**20060224000948]
[Added #reset to Syntax Parser.
Brian T. Rice <water@...>**20060219020426]
[Improved Pidgin code generator handling of #do:/#doWithIndex: by making a more basic #keysDo:.
Brian T. Rice <water@...>**20060218223147]
[Added #above:by:do: support to the Pidgin code generator.
Brian T. Rice <water@...>**20060218203116]
[Bug-fix for #below:by:do: in the Pidgin generator.
Brian T. Rice <water@...>**20060218202946]
[Added support for #do: to the Pidgin generator, and a stub for #doWithIndex:.
Brian T. Rice <water@...>**20060218202650]
[Pidgin code-generator cleanups.
Brian T. Rice <water@...>**20060218201536]
[Syntax code cleanups.
Brian T. Rice <water@...>**20060218180400]
[ByteCompiler random code cleanups.
Brian T. Rice <water@...>**20060218174812]
[Cleaned up the ByteCompiler code for return statements.
Brian T. Rice <water@...>**20060218174801]
[Separated &optionals: functionality from applyTo: into applyTo:optionals:.
Brian T. Rice <water@...>*-20060218010716]
[Separated SmallInteger math primitives into ext/smallint.slate.
Brian T. Rice <water@...>**20060218013533]
[Fixed the bootstrap process "make" call to only build a VM executable and not try a full bootstrap when the new images aren't made.
Brian T. Rice <water@...>**20060218011928]
[Separated &optionals: functionality from applyTo: into applyTo:optionals:.
Brian T. Rice <water@...>**20060218010716]
[Added a blessnew target for the Makefile which backs up bootstrap sources and copies the newboot results to the regular filenames.
Brian T. Rice <water@...>**20060217223211]
[Fixed the parsing of optional argument declarations in literal block headers.
Brian T. Rice <water@...>**20060217221739]
[More Interval code, and some fixes.
Brian T. Rice <water@...>**20060217015906]
[Made Syntax Parser Error an attribute of the traits object, not every Parser object.
Brian T. Rice <water@...>**20060214164105]
[Bug-fix for newLiteralFor: code (commented out an incomplete override).
Brian T. Rice <water@...>**20060214165825]
[Separated the newboot target into one that creates a bootstrapper image for re-use.
Brian T. Rice <water@...>**20060214024819]
[Range code cleanups.
Brian T. Rice <water@...>**20060213231325]
[Cleanup of extprim.c code.
Brian T. Rice <water@...>**20060213231232]
[ExternalLibrary error primitive cleanup.
Brian T. Rice <water@...>**20060213231154]
[Made use of newLiteralFor: in the Lexer to clean up code.
Brian T. Rice <water@...>**20060213191502]
[Cleanups for primitive Types code.
Brian T. Rice <water@...>**20060213191346]
[More Interval code.
Brian T. Rice <water@...>**20060213210129]
[Added #BareDefinition to the Image-building code as a quick illustration of how to template a custom build.
Brian T. Rice <water@...>**20060213190258]
[Added the skeleton of an Interval type, in unfinished/.
Brian T. Rice <water@...>**20060213185607]
[Added #TODO: message to complement #TODO and add a note.
Brian T. Rice <water@...>**20060213184914]
[Added #TODO as a selector to raise a debugger for methods which are incomplete.
Brian T. Rice <water@...>**20060213184517]
[Various BUGS updates.
Brian T. Rice <water@...>**20060210194747]
[Formatting fixes for the manual.
Brian T. Rice <water@...>**20060210194643]
[Extracted Console/DebugConsole code into console.slate, and fixed the use of Console I/O for stderr.
Brian T. Rice <water@...>**20060210205823]
[Re-factored ifCompletes:/unlessCompletes: in terms of an ifTrue:ifFalse:-like ifCompletes:otherwise: setup. This also fixes ifCompletes:.
Brian T. Rice <water@...>**20060210174159]
[Small Makefile fixes.
Brian T. Rice <water@...>**20060210004018]
[Bug-fix for ExternalLibrary loading error-handling.
Brian T. Rice <water@...>**20060210173240]
[Adjusted the slateMain local declarations for portability.
Brian T. Rice <water@...>**20060209234919]
[Updated INSTALL instructions for the new kernel image naming conventions.
Brian T. Rice <water@...>**20060209222830]
[Changed the bootstrap kernel image output name to kernel.endianness.wordsize.image.
Brian T. Rice <water@...>**20060209205817]
[Added a basic NullSoft installer configuration for Windows, already tested.
Brian T. Rice <water@...>**20060209154257]
[Fixes for starting Slate in emacs.
Brian T. Rice <water@...>**20060204004047]
[Moved the Portfile to pkg/.
Brian T. Rice <water@...>**20060202232000]
[Added support for embedding the image into the VM executable in boot.c.
Brian T. Rice <water@...>**20060203042915]
[Added a Portfile for DarwinPorts.
Brian Templeton <bpt@...>**20060201080316]
[Further fixed PositionableStream #upToAll: to use whileFalse: and peek: vs. loop and next:, fixing a further bug and cleaning it up.
Brian T. Rice <water@...>**20060202051304]
[Bug-fix for PositionableStream upToAll:.
Brian T. Rice <water@...>**20060202045710]
[TAG StablePoint-0.3.5.3
The Slate Team <slate@...>**20060201213801]
Patch bundle hash:
3b584a78dea15a904fb5c3ae8c3b3eafc02b637b