Squeak-dev April 2015

squeak-dev@lists.squeakfoundation.org

38 participants
642 discussions

Error when saving class comments
by Fabian Windheuser 28 May '15

28 May '15

Hello, Whenever I try to modify and save a class comment I get the following error: "Error: RemoteString past end of file" (see the attachment for the error trace). My changes are not getting saved and the file seems to get corrupted. I am no longer able to reopen this class comment because it always throws the same error. I tried restarting my image and using new unmodified .image and .changes files however the error still occurred. I am running Squeak 4.5 with update number #13702 on a 64bit Arch Linux distribution. Best regards, Fabian

5 5

Bug whith contextOn:do:
by Tobias Pape 28 May '15

28 May '15

Hey fellow Squeakers [Attention: low level lengthy stuff] I today encountered a strange behavior. When running the Startup code, somewhen ContextPart>>#complete is called, which, in the process issues #contextOn:do:, which subsequently somewhere interanlly does a jump: contextOn: exceptionClass do: block "Create an #on:do: context that is ready to return from executing its receiver" | ctxt chain | ctxt := thisContext. [chain := thisContext sender cut: ctxt. ctxt jump] on: exceptionClass do: block. "jump above will resume here without unwinding chain" ^ chain The idea is that we end up right before '^ chain' as the comment indicates. so what happens is that ctxt is the context of #contextOn:do: itself and it gets send #jump in the closure. Jump now does the following: jump "Abandon thisContext and resume self instead (using the same current process). You may want to save thisContext's sender before calling this so you can jump back to it. Self MUST BE a top context (ie. a suspended context or a abandoned context that was jumped out of). A top context already has its return value on its stack (see Interpreter>>primitiveSuspend and other suspending primitives). thisContext's sender is converted to a top context (by pushing a nil return value on its stack) so it can be jump back to." | top | "Make abandoned context a top context (has return value (nil)) so it can be jumped back to" thisContext sender push: nil. "Pop self return value then return it to self (since we jump to self by returning to it)" stackp = 0 ifTrue: [self stepToSendOrReturn]. stackp = 0 ifTrue: [self push: nil]. "must be quick return self/constant" top := self pop. thisContext privSender: self. ^ top So. bytecode for #contextOn:do: is: 29 <8A 01> push: (Array new: 1) 31 <6B> popIntoTemp: 3 32 <89> pushThisContext: 33 <6A> popIntoTemp: 2 34 <12> pushTemp: 2 35 <13> pushTemp: 3 36 <8F 20 00 0A> closureNumCopied: 2 numArgs: 0 bytes 40 to 49 40 <89> pushThisContext: 41 <D2> send: sender 42 <10> pushTemp: 0 43 <E1> send: cut: 44 <8E 00 01> popIntoTemp: 0 inVectorAt: 1 47 <10> pushTemp: 0 48 <D3> send: jump 49 <7D> blockReturn 50 <10> pushTemp: 0 51 <11> pushTemp: 1 52 <F0> send: on:do: 53 <87> pop 54 <8C 00 03> pushTemp: 0 inVectorAt: 3 57 <7C> returnTop The jump lands right at 53 and does a pop. HOWEVER, at this point the stack of this context is empty and the pop actually pops the 3rd temp from the temps that 'just happens' to be right under the stack. This should be fatal. HOWEVER again, Squeak actually does not pop but only decrement the SP so the temp access still works(this _could_ be fine but some implementations (Eg, RSqueak) tried to separate temps and stack; which is not possible currently). What could be the problem here? - are the 'stackp = 0'-checks in #jump wrong and they actually should check for the actual stack depth _after_ temps? - should we put in a "sacrificial anode" in #contextOn:do: so that the pop does not pop the empty stack? (like this: contextOn: exceptionClass do: block "Create an #on:do: context that is ready to return from executing its receiver" | ctxt chain | ctxt := thisContext. [chain := thisContext sender cut: ctxt. ctxt push: nil. "sacrifical anode" ctxt jump ] on: exceptionClass do: block. "jump above will resume here without unwinding chain" ^ chain - Or is there an even better way? Best -Tobias

4 9

Font handling
by karl ramberg 02 May '15

02 May '15

Hi, the current font handling could need a refactoring/ cleanup. Emphasis of a font are stored in different places and that makes it inconsistent and very hard to interface. As far as I can tell should the emphasis info be stored as a text attribute. But it seems that many tools set the emphasis directly in the font. Here is a couple of examples from text in Workspaces: {a TextEmphasis code: 1 . aTextFontReference(TTCFont(BitstreamVeraSansMono 24 Bold Oblique))} This font has bold and italics set in the font (corresponds to value 3), but the emphasis attribute is 1 which corresponds to bold. {a TextFontReference(a StrikeFont(Accumon10 12))} Here is also emphasis set in the font and no emphasis attribute is present at all. Karl

1 1

Re: [squeak-dev] The Trunk: Morphic-mt.895.mcz
by Nicolas Cellier 01 May '15

01 May '15

2015-04-17 15:58 GMT+02:00 <commits(a)source.squeak.org>: > Marcel Taeumel uploaded a new version of Morphic to project The Trunk: > http://source.squeak.org/trunk/Morphic-mt.895.mcz > > ==================== Summary ==================== > > Name: Morphic-mt.895 > Author: mt > Time: 17 April 2015, 3:58:36.602 pm > UUID: 955fb45c-fbb5-d545-8e18-2f22524bcd24 > Ancestors: Morphic-mt.894 > > Added the missing hooks so that text models can not only handle printing > but also debugging, exploring, inspecting, and even evaluation. > > This avoids having to subclass PluggableTextMorph, TextMorph, and > TextEditor for achieving the same variation point. > > =============== Diff against Morphic-mt.894 =============== > > Item was changed: > ----- Method: TextEditor>>debugIt (in category 'do-its') ----- > debugIt > > + | receiver context | > - | method receiver context | > - (model respondsTo: #doItReceiver) > - ifTrue: > - [receiver := model doItReceiver. > - context := model doItContext] > - ifFalse: > - [receiver := context := nil]. > self lineSelectAndEmptyCheck: [^self]. > + > + (model respondsTo: #debugIt:) > + ifTrue: [^ model perform: #debugIt: with: self selection]. > + > + receiver := (model respondsTo: #doItReceiver) > + ifTrue: [model doItReceiver] > + ifFalse: [nil]. > + > + context := (model respondsTo: #doItContext) > + ifTrue: [model doItContext] > + ifFalse: [nil]. > + > + (self compileSelectionFor: receiver in: context) ifNotNil: > [:method | > + self debug: method receiver: receiver in: context].! > - method := self compileSelectionFor: receiver in: context. > - method notNil ifTrue: > - [self debug: method receiver: receiver in: context].! > > Item was changed: > ----- Method: TextEditor>>evaluateSelectionAndDo: (in category 'do-its') > ----- > evaluateSelectionAndDo: aBlock > "Treat the current selection as an expression; evaluate it and > invoke aBlock with the result." > | result rcvr ctxt | > self lineSelectAndEmptyCheck: [^ nil]. > > + (model respondsTo: #evaluate:) ifTrue: [ > + ^ aBlock value: (model perform: #evaluate: with: self > selection)]. > + > In one image with Tweak loaded (http://squeaksource.com/TweakCore) this change had the effect loosing the ability to do it/print it/inspect it For example, typing one of most basic Smalltalk expression 3+4 -> a Text for '3+4' This is because the Scripting package installs Object>>evaluate: aBlock ^aBlock value What are these hooks usefull for? They do not seem in usage currently in the image. Nicolas > (model respondsTo: #doItReceiver) > ifTrue: [ rcvr := model doItReceiver. > ctxt := model doItContext] > ifFalse: [rcvr := ctxt := nil]. > result := [ > rcvr class evaluatorClass new > evaluate: self selectionAsStream > in: ctxt > to: rcvr > notifying: self > ifFail: [morph flash. ^ nil] > logged: true. > ] > on: OutOfScopeNotification > do: [ :ex | ex resume: true]. > > (model respondsTo: #evaluated:result:) ifTrue: [ > model perform: #evaluated:result: with: self selection > with: result]. > > ^aBlock value: result! > > Item was changed: > ----- Method: TextEditor>>exploreIt (in category 'do-its') ----- > exploreIt > + > + self evaluateSelectionAndDo: [:result | > + (model respondsTo: #exploreIt:result:) > + ifTrue: [model > + perform: #exploreIt:result: > + with: self selection > + with: result] > + ifFalse: [result explore]].! > - self evaluateSelectionAndDo: [:result | result explore]! > > Item was changed: > ----- Method: TextEditor>>inspectIt (in category 'do-its') ----- > inspectIt > + > + self evaluateSelectionAndDo: [:result | > + (model respondsTo: #inspectIt:result:) > + ifTrue: [model > + perform: #inspectIt:result: > + with: self selection > + with: result] > + ifFalse: [result inspect]].! > - self evaluateSelectionAndDo: [:result | result inspect]! > > >

2 2

The Trunk: GetText-nice.36.mcz
by commits＠source.squeak.org 01 May '15

01 May '15

Nicolas Cellier uploaded a new version of GetText to project The Trunk: http://source.squeak.org/trunk/GetText-nice.36.mcz ==================== Summary ==================== Name: GetText-nice.36 Author: nice Time: 1 May 2015, 1:40:18.364 am UUID: 4b76e586-b5fb-4611-a82a-56ac5461f57d Ancestors: GetText-kfr.35 Though it can't really be tested in current trunk image state (InternalTranslator is not initialized) , modernize the way to utf8+gzip+base64 encode... This how how simpler the world could be without those Multi RW Binary or Text things, no need to send those reset/ascii/binary incantations. =============== Diff against GetText-kfr.35 =============== Item was changed: ----- Method: LanguageEditor>>codeSelectedTranslationAsMimeString (in category 'gui methods') ----- codeSelectedTranslationAsMimeString + | keys code cont | - | keys code tmpStream s2 gzs cont | keys := selectedTranslations collect: [:key | self translations at: key]. code := String streamContents: [:aStream | self translator fileOutOn: aStream keys: keys withBOM: false]. - tmpStream := MultiByteBinaryOrTextStream on: ''. - tmpStream converter: UTF8TextConverter new. - tmpStream nextPutAll: code. - s2 := RWBinaryOrTextStream on: ''. - gzs := GZipWriteStream on: s2. - tmpStream reset. - gzs nextPutAll: (tmpStream binary contentsOfEntireFile asString) contents. - gzs close. - s2 reset. - cont := String streamContents: [:strm | + strm nextPutAll: '"UTF8+Gzip+Base64 encoded translation for;'; cr. - strm nextPutAll: '"Gzip+Base64 encoded translation for;'; cr. strm nextPutAll: '#('. keys do: [:each | strm nextPutAll: '''', each, ''' '.]. strm nextPutAll: ')"'; cr; cr. strm nextPutAll: 'NaturalLanguageTranslator loadForLocaleIsoString: '. strm nextPut: $'. strm nextPutAll: translator localeID isoString. strm nextPut: $'. strm nextPutAll: ' fromGzippedMimeLiteral: '. strm nextPut: $'. + strm nextPutAll: code squeakToUtf8 asByteArray zipped base64Encoded. - strm nextPutAll: (Base64MimeConverter mimeEncode: s2) contents. strm nextPutAll: '''.'. strm cr. ]. (StringHolder new contents: cont) + openLabel: 'exported codes in UTF8+Gzip+Base64 encoding' translated! - openLabel: 'exported codes in Gzip+Base64 encoding' translated!

1 0

The Trunk: Collections-nice.622.mcz
by commits＠source.squeak.org 01 May '15

01 May '15

Nicolas Cellier uploaded a new version of Collections to project The Trunk: http://source.squeak.org/trunk/Collections-nice.622.mcz ==================== Summary ==================== Name: Collections-nice.622 Author: nice Time: 1 May 2015, 12:40:25.475 am UUID: 93c764da-c5b6-4fa6-aa98-a9ad54611907 Ancestors: Collections-mt.621 #toBraceStack: is not used for compiling { } for so long that it's really time to get rid of it. Symbol>>numArgs: does not need to copy self into a temp var. =============== Diff against Collections-mt.621 =============== Item was removed: - ----- Method: Collection>>toBraceStack: (in category 'private') ----- - toBraceStack: itsSize - "Push receiver's elements onto the stack of thisContext sender. Error if receiver does - not have itsSize elements or if receiver is unordered. - Do not call directly: this is called by {a. b} := ... constructs." - - self size ~= itsSize ifTrue: - [self error: 'Trying to store ', self size printString, - ' values into ', itsSize printString, ' variables.']. - thisContext sender push: itsSize fromIndexable: self! Item was changed: ----- Method: Symbol>>numArgs: (in category 'system primitives') ----- numArgs: n "Answer a string that can be used as a selector with n arguments. TODO: need to be extended to support shrinking and for selectors like #+ " + | numArgs aStream offs |. + (numArgs := self numArgs) >= n ifTrue: [^self]. - | selector numArgs aStream offs | - - selector := self. - (numArgs := selector numArgs) >= n ifTrue: [^self]. aStream := WriteStream on: (String new: 16). aStream nextPutAll: self. (numArgs = 0) ifTrue: [aStream nextPutAll: ':'. offs := 0] ifFalse: [offs := 1]. 2 to: n - numArgs + offs do: [:i | aStream nextPutAll: 'with:']. + ^aStream contents asSymbol! - ^aStream contents asSymbol - - !

1 0

Daily Commit Log
by commits＠source.squeak.org 30 Apr '15

30 Apr '15

Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours: http://lists.squeakfoundation.org/pipermail/packages/2015-April/008182.html Name: Files-cmm.144 Ancestors: Files-dtl.143 - Fix up asBytesDescription. - Remove a redundant override. ============================================= http://lists.squeakfoundation.org/pipermail/packages/2015-April/008183.html Name: Morphic-mt.951 Ancestors: Morphic-mt.950 Some layout issues fixed in Font importer tool. Code clean-up, too. ============================================= http://lists.squeakfoundation.org/pipermail/packages/2015-April/008184.html Name: Tools-mt.611 Ancestors: Tools-mt.610 Minor cosmetic, pixel-based changes in browsers, package browsers, and debuggers. ============================================= http://lists.squeakfoundation.org/pipermail/packages/2015-April/008185.html Name: Tools-mt.612 Ancestors: Tools-mt.611 Minor cosmetic fix regarding space between code pane and annotation pane in code holders. It now correctly corresponds to the (surrounding) panel's #cellInset. ============================================= http://lists.squeakfoundation.org/pipermail/packages/2015-April/008186.html Name: Monticello-mt.613 Ancestors: Monticello-mt.612 Some hard-coded layout properties refactored to adapt spacing according to fonts. ============================================= http://lists.squeakfoundation.org/pipermail/packages/2015-April/008187.html Name: MonticelloConfigurations-mt.130 Ancestors: MonticelloConfigurations-dtl.129 Some hard-coded layout properties refactored to adapt spacing according to fonts. ============================================= http://lists.squeakfoundation.org/pipermail/packages/2015-April/008188.html Name: Morphic-mt.952 Ancestors: Morphic-mt.951 Faster check for scroll bar visibility in scroll panes. ============================================= http://lists.squeakfoundation.org/pipermail/packages/2015-April/008189.html Name: System-nice.731 Ancestors: System-kfr.730 A DummyStream may do nothing, but shall mimic the answer of a regular stream. =============================================

1 0

The Trunk: System-nice.731.mcz
by commits＠source.squeak.org 30 Apr '15

30 Apr '15

Nicolas Cellier uploaded a new version of System to project The Trunk: http://source.squeak.org/trunk/System-nice.731.mcz ==================== Summary ==================== Name: System-nice.731 Author: nice Time: 30 April 2015, 11:21:11.155 pm UUID: bef33999-82b0-49a5-a17a-536c57394b83 Ancestors: System-kfr.730 A DummyStream may do nothing, but shall mimic the answer of a regular stream. =============== Diff against System-kfr.730 =============== Item was changed: ----- Method: DummyStream>>nextStringPut: (in category 'accessing') ----- nextStringPut: aString + "do nothing" + ^aString! - "do nothing"!

1 0

Wrong error message from BlockClosure ?
by karl ramberg 30 Apr '15

30 Apr '15

#(1 2) do:[:a :b| b] "do it" Error: 'This block accepts 2 argument, but was called with 1 arguments.' Should not the numArgs and the numArgsForInvocation switch place here ? BlockClosure>>numArgsError: numArgsForInvocation | printNArgs | printNArgs := [:n| n printString, ' argument', (n = 1 ifTrue: [''] ifFalse:['s'])]. self error: 'This block accepts ', (printNArgs value: numArgs), ', but was called with ', (printNArgs value: numArgsForInvocation), '.'

3 3

The Trunk: Morphic-mt.952.mcz
by commits＠source.squeak.org 30 Apr '15

30 Apr '15

Marcel Taeumel uploaded a new version of Morphic to project The Trunk: http://source.squeak.org/trunk/Morphic-mt.952.mcz ==================== Summary ==================== Name: Morphic-mt.952 Author: mt Time: 30 April 2015, 2:07:32.024 pm UUID: 993bb899-d198-9245-8d5a-69a3acb7f9a9 Ancestors: Morphic-mt.951 Faster check for scroll bar visibility in scroll panes. =============== Diff against Morphic-mt.951 =============== Item was changed: ----- Method: ScrollPane>>hIsScrollbarShowing (in category 'geometry testing') ----- hIsScrollbarShowing "Return true if a horz scroll bar is currently showing" + ^ hScrollBar owner notNil! - ^submorphs includes: hScrollBar - ! Item was changed: ----- Method: ScrollPane>>vIsScrollbarShowing (in category 'geometry testing') ----- vIsScrollbarShowing "Return true if a retractable scroll bar is currently showing" + ^ scrollBar owner notNil - ^submorphs includes: scrollBar !

1 0

Jump to page: