Squeak-dev February 2013

squeak-dev@lists.squeakfoundation.org

61 participants
225 discussions

Mailing list message size restriction (Re: [squeak-dev] Fwd: Your message to Squeak-dev awaits moderator approval)
by H. Hirzel 25 Feb '13

25 Feb '13

Yes, for example a modest increase from 100kB to 130kB would do quite a lot. I often can do a screen shot for 70..90 kB but then the message as such needs some space as well. --Hannes On 2/25/13, Tobias Pape <Das.Linux(a)gmx.de> wrote: > Would anyone care to increase the message size restrictions a tad? > 100k seem somewhat low todays ;) > > Best > -Tobias > > Anfang der weitergeleiteten Nachricht: > >> Von: squeak-dev-bounces(a)lists.squeakfoundation.org >> Betreff: Your message to Squeak-dev awaits moderator approval >> Datum: 25. Februar 2013 09:20:16 MEZ >> […] > >> Your mail to 'Squeak-dev' with the subject >> >> Re: [squeak-dev] SqueakCore mentioned but not available. >> >> Is being held until the list moderator can review it for approval. >> >> The reason it is being held: >> >> Message body is too big: 106874 bytes with a limit of 100 KB >> >> […] > > > >

1 0

Fwd: Your message to Squeak-dev awaits moderator approval
by Tobias Pape 25 Feb '13

25 Feb '13

Would anyone care to increase the message size restrictions a tad? 100k seem somewhat low todays ;) Best -Tobias Anfang der weitergeleiteten Nachricht: > Von: squeak-dev-bounces(a)lists.squeakfoundation.org > Betreff: Your message to Squeak-dev awaits moderator approval > Datum: 25. Februar 2013 09:20:16 MEZ > […] > Your mail to 'Squeak-dev' with the subject > > Re: [squeak-dev] SqueakCore mentioned but not available. > > Is being held until the list moderator can review it for approval. > > The reason it is being held: > > Message body is too big: 106874 bytes with a limit of 100 KB > > […]

1 0

SqueakCore mentioned but not available.
by Tobias Pape 25 Feb '13

25 Feb '13

Dear all the Future Directions Workspace of 4.4 says “- A SqueakCore image is available at http://ftp.squeak.org/4.4” which, sadly, is not the case: Index of /4.4 Squeak4.4-12327.tgz 19-Jan-2013 21:32 8.7M Squeak4.4-12327.zip 31-Dec-2012 15:56 8.7M SqueakV41.sources.gz 17-Apr-2010 17:22 6.3M SqueakV41.sources.zip 16-Jan-2013 09:36 6.3M SqueakV44.sources.gz 17-Apr-2010 17:22 6.3M SqueakV44.sources.zip 16-Jan-2013 09:37 6.3M mac/ 11-Jan-2013 21:48 - unix-linux/ 11-Jan-2013 21:51 - win/ 11-Jan-2013 21:49 - Shouldn’t we fix that? Best -Tobias

2 1

Re: [squeak-dev] Getting rid of coloured code
by Chris Muller 25 Feb '13

25 Feb '13

Hi Ted! I've attached my style-table method (gz'd to avoid e-mail mangling the line-endings). It's one method on the class-side of SHTextStylerST80. After filing it in you can activate it with: SHTextStylerST80 styleTable: SHTextStylerST80 blackOnWhiteStyleTable The rationale are: reserved words :: bold and black. "Solidity" to indicate an immutable part of the syntax. Class names, symbols and structure (blocks, parens, etc) :: Black but not bold because the names can vary at least during "development". :) instVars :: Gold because they are the "resources" a class has available to send messages to. message selectors :: Dark Green because they invoke behaviors. comments :: Light-Gray to be non-instrusive, but still easily read. literal constants :: Dark Red to stand out and also to discourage heavy use of constants in code. arguments :: Rust because they are rigid, they cannot be assigned. temp-vars :: Dark Cyan because they're pliable like.. clouds or water.. And, finally, "bad" code :: magenta because its absolutely hideous, which strongly encourages me to take appropriate action to bring it down to a normal color. I typically never save a method with an unimplemented sender, so having the method name highlighted in majenta in a background window is a highly-visible reminder. It's brightness can sometimes be used to advantage to to "highlight" all occurrences of a variable in a method by temporarily removing one character from its declaration. I use the "Auto Enclose" preference to avoid majenta when using structure (parents, brackets, etc.). Cheers, Chris On Sat, Feb 23, 2013 at 7:04 PM, Ted Kaehler <Ted(a)vpri.org> wrote: > Chris, > I'd love to see your do-it to change the Shout colors to ones you > find satisfying. > A big problem is when you are in the middle of modifying a method. > As you type, it does not make sense to Shout, and the next 20 lines turn > bright red. That is annoying. I feel like saying, "I'm not done yet. > Don't get on my case!". > Perhaps if there are more than 2 lines in red after the insertion > point, just turn the rest of the method black. That would keep things much > calmer. The lack of any other colors would signal that something is not > complete. > > --Ted. > > > > > At 5:36 PM -0600 2/23/13, Chris Muller wrote: >> >> Wow, I'm surprised -- the human eye discerns color contrast much more >> easily than patterns. While I don't care for the standard Shout >> colors, by changing them to be more intuitive and constrasting, >> particularly for longer methods, the colors _really_ enhance the >> readability. For example, I'm able to instantly see references to >> instVars with just a glance because they're gold (dark yellow); or >> references to self and super because they're bolded, which quickly >> tells me whether this is a "utility" method. Further, by making >> comments a light-gray, they are just barely visible, so they're not >> cluttering the code, but still easily readable if I decide to focus on >> them for a clarification. Kind of like simply changing eye-focus >> "reveals" different aspects of the code -- what a powerful >> "gesture"!.. >> > > > -- > Ted Kaehler > "400 years of scientific enlightenment are being undone by TV shows on crop > circles, ancient astronauts, and haunted buildings." > http://www.vpri.org/html/team_bios/kaehler.html >

1 0

Daily Commit Log
by commits＠source.squeak.org 25 Feb '13

25 Feb '13

Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours: http://lists.squeakfoundation.org/pipermail/packages/2013-February/005720.h… Name: Kernel-eem.740 Ancestors: Kernel-fbs.739 Move the code to set the owning method of an AdditionalMethodState to the setter away from clients. Add an anti-sharing check. ============================================= http://lists.squeakfoundation.org/pipermail/packages/2013-February/005721.h… Name: Kernel-nice.741 Ancestors: Kernel-eem.740 Reduce the number of hardcoded constants used by #raisedTo:modulo: This will enable smooth transition to next generation of LargeIntegersPlugin. ============================================= http://lists.squeakfoundation.org/pipermail/packages/2013-February/005722.h… Name: KernelTests-nice.245 Ancestors: KernelTests-nice.244 Reduce the number of hardcoded constants used by #raisedTo:modulo: tests too. ============================================= http://lists.squeakfoundation.org/pipermail/packages/2013-February/005723.h… Name: CollectionsTests-nice.198 Ancestors: CollectionsTests-fbs.197 Diminish excessive usage of ReadWriteStream ============================================= http://lists.squeakfoundation.org/pipermail/packages/2013-February/005724.h… Name: Kernel-nice.742 Ancestors: Kernel-nice.741 Diminish excessive usage of ReadWriteStream ============================================= http://lists.squeakfoundation.org/pipermail/packages/2013-February/005725.h… Name: KernelTests-nice.246 Ancestors: KernelTests-nice.245 Diminish excessive usage of ReadWriteStream ============================================= http://lists.squeakfoundation.org/pipermail/packages/2013-February/005726.h… Name: System-nice.513 Ancestors: System-dtl.512 Diminish excessive usage of ReadWriteStream. Bonus: correct outdated code sample in comment ============================================= http://lists.squeakfoundation.org/pipermail/packages/2013-February/005727.h… Name: Tools-nice.447 Ancestors: Tools-fbs.446 Diminish excessive usage of ReadWriteStream ============================================= http://lists.squeakfoundation.org/pipermail/packages/2013-February/005728.h… Name: Morphic-nice.640 Ancestors: Morphic-tfel.639 Diminish excessive usage of ReadWriteStream ============================================= http://lists.squeakfoundation.org/pipermail/packages/2013-February/005729.h… Name: Morphic-nice.641 Ancestors: Morphic-nice.640 Correct blinkPrevParen: (only in absence of Shout). simplify an initialization. ============================================= http://lists.squeakfoundation.org/pipermail/packages/2013-February/005730.h… Name: Tools-nice.448 Ancestors: Tools-nice.447 FakeClassPool is obsolete in Squeak, remove recent reference to it. Note: we should deprecate this class ============================================= http://lists.squeakfoundation.org/pipermail/packages/2013-February/005731.h… Name: Compiler-nice.253 Ancestors: Compiler-eem.252 clean-up Parser from now unused requestor/category ============================================= http://lists.squeakfoundation.org/pipermail/packages/2013-February/005732.h… Name: Compiler-nice.254 Ancestors: Compiler-nice.253 revert Parser instance variable change, and think... ============================================= http://lists.squeakfoundation.org/pipermail/packages/2013-February/005733.h… Name: Monticello-nice.533 Ancestors: Monticello-fbs.532 Correct this bug: it was impossible to remove an instance variable from Compiler/Parser and probably a few other classes, because Compiler was invoked during its own modification. The fix will work as long as those class won't be Trait's composed... ============================================= http://lists.squeakfoundation.org/pipermail/packages/2013-February/005734.h… Name: Compiler-nice.255 Ancestors: Compiler-nice.254 Remove Parser ivars 'requestor' and 'category' again now that MC is patched. =============================================

1 0

The Trunk: Compiler-nice.255.mcz
by commits＠source.squeak.org 24 Feb '13

24 Feb '13

Nicolas Cellier uploaded a new version of Compiler to project The Trunk: http://source.squeak.org/trunk/Compiler-nice.255.mcz ==================== Summary ==================== Name: Compiler-nice.255 Author: nice Time: 24 February 2013, 11:55:43.511 pm UUID: f1c697bd-c388-4d60-94bc-56a1dc731591 Ancestors: Compiler-nice.254 Remove Parser ivars 'requestor' and 'category' again now that MC is patched. =============== Diff against Compiler-nice.254 =============== Item was changed: Scanner subclass: #Parser + instanceVariableNames: 'here hereType hereMark hereEnd prevMark prevEnd encoder parseNode failBlock requestorOffset tempsMark doitFlag properties queriedUnusedTemporaries cue' - instanceVariableNames: 'here hereType hereMark hereEnd prevMark prevEnd encoder requestor parseNode failBlock requestorOffset tempsMark doitFlag properties category queriedUnusedTemporaries cue' classVariableNames: '' poolDictionaries: '' category: 'Compiler-Kernel'! !Parser commentStamp: 'cwp 12/26/2012 23:34' prior: 0! I parse Smalltalk syntax and create a MethodNode that is the root of the parse tree. I look one token ahead.!

1 0

The Trunk: Monticello-nice.533.mcz
by commits＠source.squeak.org 24 Feb '13

24 Feb '13

Nicolas Cellier uploaded a new version of Monticello to project The Trunk: http://source.squeak.org/trunk/Monticello-nice.533.mcz ==================== Summary ==================== Name: Monticello-nice.533 Author: nice Time: 24 February 2013, 11:19:26.177 pm UUID: 082a31a2-8e19-4d30-9661-6fa84dfc436e Ancestors: Monticello-fbs.532 Correct this bug: it was impossible to remove an instance variable from Compiler/Parser and probably a few other classes, because Compiler was invoked during its own modification. The fix will work as long as those class won't be Trait's composed... =============== Diff against Monticello-fbs.532 =============== Item was changed: ----- Method: MCClassDefinition>>createClass (in category 'installing') ----- createClass | superClass class composition | superClass := superclassName == #nil ifFalse: [Smalltalk at: superclassName]. [class := (ClassBuilder new) name: name inEnvironment: superClass environment subclassOf: superClass type: type instanceVariableNames: self instanceVariablesString classVariableNames: self classVariablesString poolDictionaries: self sharedPoolsString category: category. ] on: Warning, DuplicateVariableError do:[:ex| ex resume]. "The following is written to support traits unloading" + composition := self traitComposition ifNil: [Array new] ifNotNil: [:traitComposition | Compiler evaluate: traitComposition]. - composition := Compiler evaluate: (self traitComposition ifNil:['{}']). (composition isCollection and:[composition isEmpty and:[class traitComposition isEmpty]]) ifFalse:[ class setTraitComposition: composition asTraitComposition. ]. + composition := self classTraitComposition ifNil: [Array new] ifNotNil: [:traitComposition | Compiler evaluate: traitComposition]. - composition := Compiler evaluate: (self classTraitComposition ifNil:['{}']). (composition isCollection and:[composition isEmpty and:[class class traitComposition isEmpty]]) ifFalse:[ class class setTraitComposition: composition asTraitComposition. ]. ^class!

1 0

The Trunk: Compiler-nice.254.mcz
by commits＠source.squeak.org 24 Feb '13

24 Feb '13

Nicolas Cellier uploaded a new version of Compiler to project The Trunk: http://source.squeak.org/trunk/Compiler-nice.254.mcz ==================== Summary ==================== Name: Compiler-nice.254 Author: nice Time: 24 February 2013, 11:04:25.634 pm UUID: 5a873a42-0c07-47db-98d3-efdb4e059d15 Ancestors: Compiler-nice.253 revert Parser instance variable change, and think... =============== Diff against Compiler-nice.253 =============== Item was changed: Scanner subclass: #Parser + instanceVariableNames: 'here hereType hereMark hereEnd prevMark prevEnd encoder requestor parseNode failBlock requestorOffset tempsMark doitFlag properties category queriedUnusedTemporaries cue' - instanceVariableNames: 'here hereType hereMark hereEnd prevMark prevEnd encoder parseNode failBlock requestorOffset tempsMark doitFlag properties queriedUnusedTemporaries cue' classVariableNames: '' poolDictionaries: '' category: 'Compiler-Kernel'! !Parser commentStamp: 'cwp 12/26/2012 23:34' prior: 0! I parse Smalltalk syntax and create a MethodNode that is the root of the parse tree. I look one token ahead.!

1 0

Re: [squeak-dev] The Trunk: Compiler-nice.253.mcz
by Nicolas Cellier 24 Feb '13

24 Feb '13

Oops, MC does not like this kind of intrusive change... I will try and delete this and rework... Nicolas 2013/2/24 <commits(a)source.squeak.org>: > Nicolas Cellier uploaded a new version of Compiler to project The Trunk: > http://source.squeak.org/trunk/Compiler-nice.253.mcz > > ==================== Summary ==================== > > Name: Compiler-nice.253 > Author: nice > Time: 24 February 2013, 10:43:13.671 pm > UUID: e6c74d32-dbba-41ef-9315-f333c8ce18de > Ancestors: Compiler-eem.252 > > clean-up Parser from now unused requestor/category > > =============== Diff against Compiler-eem.252 =============== > > Item was changed: > Scanner subclass: #Parser > + instanceVariableNames: 'here hereType hereMark hereEnd prevMark prevEnd encoder parseNode failBlock requestorOffset tempsMark doitFlag properties queriedUnusedTemporaries cue' > - instanceVariableNames: 'here hereType hereMark hereEnd prevMark prevEnd encoder requestor parseNode failBlock requestorOffset tempsMark doitFlag properties category queriedUnusedTemporaries cue' > classVariableNames: '' > poolDictionaries: '' > category: 'Compiler-Kernel'! > > !Parser commentStamp: 'cwp 12/26/2012 23:34' prior: 0! > I parse Smalltalk syntax and create a MethodNode that is the root of the parse tree. I look one token ahead.! > > Item was removed: > - ----- Method: Parser>>init:notifying:failBlock: (in category 'private') ----- > - init: sourceStream notifying: req failBlock: aBlock > - > - requestor := req. > - failBlock := aBlock. > - requestorOffset := 0. > - super scan: sourceStream. > - prevMark := hereMark := mark. > - self advance! > > Item was changed: > ----- Method: Parser>>setCue: (in category 'private') ----- > setCue: aCue > + cue := aCue! > - cue := aCue. > - > - "Also set legacy variables for methods that don't use cue yet." > - requestor := cue requestor. > - category := cue category.! > >

1 0

The Trunk: Compiler-nice.253.mcz
by commits＠source.squeak.org 24 Feb '13

24 Feb '13

Nicolas Cellier uploaded a new version of Compiler to project The Trunk: http://source.squeak.org/trunk/Compiler-nice.253.mcz ==================== Summary ==================== Name: Compiler-nice.253 Author: nice Time: 24 February 2013, 10:43:13.671 pm UUID: e6c74d32-dbba-41ef-9315-f333c8ce18de Ancestors: Compiler-eem.252 clean-up Parser from now unused requestor/category =============== Diff against Compiler-eem.252 =============== Item was changed: Scanner subclass: #Parser + instanceVariableNames: 'here hereType hereMark hereEnd prevMark prevEnd encoder parseNode failBlock requestorOffset tempsMark doitFlag properties queriedUnusedTemporaries cue' - instanceVariableNames: 'here hereType hereMark hereEnd prevMark prevEnd encoder requestor parseNode failBlock requestorOffset tempsMark doitFlag properties category queriedUnusedTemporaries cue' classVariableNames: '' poolDictionaries: '' category: 'Compiler-Kernel'! !Parser commentStamp: 'cwp 12/26/2012 23:34' prior: 0! I parse Smalltalk syntax and create a MethodNode that is the root of the parse tree. I look one token ahead.! Item was removed: - ----- Method: Parser>>init:notifying:failBlock: (in category 'private') ----- - init: sourceStream notifying: req failBlock: aBlock - - requestor := req. - failBlock := aBlock. - requestorOffset := 0. - super scan: sourceStream. - prevMark := hereMark := mark. - self advance! Item was changed: ----- Method: Parser>>setCue: (in category 'private') ----- setCue: aCue + cue := aCue! - cue := aCue. - - "Also set legacy variables for methods that don't use cue yet." - requestor := cue requestor. - category := cue category.!

1 0

← Newer
1
...
4
5
6
7
8
9
10
...
23
Older →

Jump to page: