Squeak-dev November 2011

squeak-dev@lists.squeakfoundation.org

52 participants
168 discussions

uncompacting classes
by Chris Muller 28 Nov '11

28 Nov '11

Compact classes cannot be made uncompact in a Cog+JIT VM. Can compact classes be made uncompact when running the StackInterpreter VM? Thanks.

2 1

Daily Commit Log
by commits＠source.squeak.org 28 Nov '11

28 Nov '11

Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours: http://lists.squeakfoundation.org/pipermail/packages/2011-November/005103.h… Name: Monticello-cmm.489 Ancestors: Monticello-cmm.485, Monticello-cmm.488 - Better guard in packageSelection: avoids debugger when viewing a repository with no packages. =============================================

1 0

The Inbox: Tests-dtl.136.mcz
by commits＠source.squeak.org 28 Nov '11

28 Nov '11

David T. Lewis uploaded a new version of Tests to project The Inbox: http://source.squeak.org/inbox/Tests-dtl.136.mcz ==================== Summary ==================== Name: Tests-dtl.136 Author: dtl Time: 27 November 2011, 6:51:19.271 pm UUID: 4b78cc43-10bc-4d80-a724-44c94ba846ac Ancestors: Tests-ul.135 Fix ReferenceStream handling of weak references. Test and patches by Juan Vuletich. <http://lists.squeakfoundation.org/pipermail/squeak-dev/2011-November/162285…> Packages affected: Kernel-Objects System-Object Storage Tests-Object Storage If we serialize a model with weak references to views, only the model should be serialized and not the views. The bug became apparent only when dumping a model to a SmartRefStream, that calls #references, and the serialized stream was later materialized in an image where the view classes had been deleted. In such rare cases, materialization would fail when trying to reference these absent classes. If serializing to a ReferenceStream, the bug didn't become apparent (views were never serialized). If serializing to a SmartRefStream, but view classes still existed, the bug didn't really become apparent (because views were not actually deserialized), the only effect was a larger file. =============== Diff against Tests-ul.135 =============== Item was changed: SystemOrganization addCategory: #'Tests-Exceptions'! SystemOrganization addCategory: #'Tests-Files'! SystemOrganization addCategory: #'Tests-Compiler'! SystemOrganization addCategory: #'Tests-Digital Signatures'! SystemOrganization addCategory: #'Tests-Object Events'! SystemOrganization addCategory: #'Tests-System-Support'! SystemOrganization addCategory: #'Tests-Bugs'! SystemOrganization addCategory: #'Tests-ObjectsAsMethods'! SystemOrganization addCategory: #'Tests-PrimCallController'! SystemOrganization addCategory: #'Tests-Release'! SystemOrganization addCategory: #'Tests-Utilities'! SystemOrganization addCategory: #'Tests-VM'! SystemOrganization addCategory: #'Tests-Hex'! SystemOrganization addCategory: #'Tests-Monticello'! SystemOrganization addCategory: #'Tests-Localization'! SystemOrganization addCategory: #'Tests-FilePackage'! SystemOrganization addCategory: #'Tests-Finalization'! SystemOrganization addCategory: #'Tests-Dependencies'! SystemOrganization addCategory: #'Tests-Monticello-Mocks'! + SystemOrganization addCategory: #'Tests-Object Storage'! + SystemOrganization addCategory: #'Tests-System-Object Storage'! Item was added: + TestCase subclass: #ReferenceStreamTest + instanceVariableNames: '' + classVariableNames: '' + poolDictionaries: '' + category: 'Tests-System-Object Storage'! Item was added: + ----- Method: ReferenceStreamTest>>testWeakDumps (in category 'testing') ----- + testWeakDumps + "Test that if we serialize a model with weak references to views, only the model is serialized and not the views. + + Note: The bug became apparent only when dumping a model to a SmartRefStream, that calls #references, and the serialized stream + was later materialized in an image where the view classes had been deleted. In such rare cases, materialization would fail when trying to reference these + absent classes. If serializing to a ReferenceStream, the bug didn't become apparent (views were never serialized). If serializing to a SmartRefStream, but + view classes still existed, the bug didn't really become apparent (because views were not actually deserialized), the only effect was a larger file. + + ReferenceStreamTest new testWeakDumps + " + | oldInstance refStream | + oldInstance :=StringHolder new contents: 'This is a text'. + oldInstance addDependent: Morph new. + refStream := ReferenceStream on: (DummyStream on: nil). + refStream nextPut: oldInstance. + self deny: (refStream references keys anySatisfy: [ :dumpedObject | dumpedObject isKindOf: Morph ])!

1 0

The Inbox: System-dtl.461.mcz
by commits＠source.squeak.org 28 Nov '11

28 Nov '11

David T. Lewis uploaded a new version of System to project The Inbox: http://source.squeak.org/inbox/System-dtl.461.mcz ==================== Summary ==================== Name: System-dtl.461 Author: dtl Time: 27 November 2011, 6:50:49.443 pm UUID: 11a6c0c8-eceb-43c1-bede-0c884b9323a6 Ancestors: System-eem.460 Fix ReferenceStream handling of weak references. Test and patches by Juan Vuletich. <http://lists.squeakfoundation.org/pipermail/squeak-dev/2011-November/162285…> Packages affected: Kernel-Objects System-Object Storage Tests-Object Storage If we serialize a model with weak references to views, only the model should be serialized and not the views. The bug became apparent only when dumping a model to a SmartRefStream, that calls #references, and the serialized stream was later materialized in an image where the view classes had been deleted. In such rare cases, materialization would fail when trying to reference these absent classes. If serializing to a ReferenceStream, the bug didn't become apparent (views were never serialized). If serializing to a SmartRefStream, but view classes still existed, the bug didn't really become apparent (because views were not actually deserialized), the only effect was a larger file. =============== Diff against System-eem.460 =============== Item was changed: ----- Method: DiskProxy>>storeDataOn: (in category 'i/o') ----- + storeDataOn: aReferenceStream - storeDataOn: aDataStream "Besides just storing, get me inserted into references, so structures will know about class DiskProxy." + super storeDataOn: aReferenceStream. + + "just so instVarInfo: will find it and put it into structures" + " aReferenceStream references at: self put: #none." + aReferenceStream addSpecialReference: self! - super storeDataOn: aDataStream. - aDataStream references at: self put: #none. - "just so instVarInfo: will find it and put it into structures"! Item was added: + ----- Method: ReferenceStream>>addSpecialReference: (in category 'writing') ----- + addSpecialReference: aDiskProxy + "See senders. Added to avoid breaking encapsulation (assuming that #references would answer the actual collection)" + references at: aDiskProxy put: #none! Item was changed: ----- Method: ReferenceStream>>references (in category 'writing') ----- references + "Do not include provisory references created in #nextPutWeak that never became normal references, + because the referenced object was never added from a call to #nextPut:" + ^ references select: [ :value | value isNumber ]! - ^ references! Item was changed: ----- Method: ReferenceStream>>statisticsOfRefs (in category 'statistics') ----- statisticsOfRefs "Analyze the information in references, the objects being written out" + | parents ownerBags tallies n nm owners normalReferences | + normalReferences := self references. "Exclude unrealized weaks" + parents := IdentityDictionary new: normalReferences size * 2. - | parents ownerBags tallies n nm owners | - parents := IdentityDictionary new: references size * 2. n := 0. 'Finding Owners...' + displayProgressFrom: 0 to: normalReferences size - displayProgressFrom: 0 to: references size during: [:bar | + normalReferences keysDo: - references keysDo: [:parent | | kids | bar value: (n := n+1). kids := parent class isFixed + ifTrue: [(1 to: parent class instSize) collect: [:i | parent instVarAt: i]] - ifTrue: [(1 to: parent class instSize) collect: [:i | parent - instVarAt: i]] ifFalse: [parent class isBits ifTrue: [Array new] + ifFalse: [(1 to: parent basicSize) collect: [:i | parent basicAt: i]]]. + (kids select: [:x | normalReferences includesKey: x]) - ifFalse: [(1 to: parent basicSize) collect: [:i | parent basicAt: - i]]]. - (kids select: [:x | references includesKey: x]) do: [:child | parents at: child put: parent]]]. ownerBags := Dictionary new. tallies := Bag new. n := 0. 'Tallying Owners...' + displayProgressFrom: 0 to: normalReferences size - displayProgressFrom: 0 to: references size during: [:bar | + normalReferences keysDo: "For each class of obj, tally a bag of owner classes" - references keysDo: "For each class of obj, tally a bag of owner - classes" [:obj | | objParent | bar value: (n := n+1). nm := obj class name. tallies add: nm. owners := ownerBags at: nm ifAbsent: [ownerBags at: nm put: Bag new]. (objParent := parents at: obj ifAbsent: [nil]) == nil ifFalse: [owners add: objParent class name]]]. ^ String streamContents: [:strm | tallies sortedCounts do: [:assn | n := assn key. nm := assn value. owners := ownerBags at: nm. strm cr; nextPutAll: nm; space; print: n. owners size > 0 ifTrue: [strm cr; tab; print: owners sortedCounts]]]!

1 0

The Inbox: Kernel-dtl.655.mcz
by commits＠source.squeak.org 28 Nov '11

28 Nov '11

David T. Lewis uploaded a new version of Kernel to project The Inbox: http://source.squeak.org/inbox/Kernel-dtl.655.mcz ==================== Summary ==================== Name: Kernel-dtl.655 Author: dtl Time: 27 November 2011, 6:50:19.012 pm UUID: a7fda999-52a1-42f1-92da-9cdae7a9614a Ancestors: Kernel-bf.654 Fix ReferenceStream handling of weak references. Test and patches by Juan Vuletich. <http://lists.squeakfoundation.org/pipermail/squeak-dev/2011-November/162285…> Packages affected: Kernel-Objects System-Object Storage Tests-Object Storage If we serialize a model with weak references to views, only the model should be serialized and not the views. The bug became apparent only when dumping a model to a SmartRefStream, that calls #references, and the serialized stream was later materialized in an image where the view classes had been deleted. In such rare cases, materialization would fail when trying to reference these absent classes. If serializing to a ReferenceStream, the bug didn't become apparent (views were never serialized). If serializing to a SmartRefStream, but view classes still existed, the bug didn't really become apparent (because views were not actually deserialized), the only effect was a larger file. =============== Diff against Kernel-bf.654 =============== Item was changed: ----- Method: Object>>storeDataOn: (in category 'objects from disk') ----- storeDataOn: aDataStream "Store myself on a DataStream. Answer self. This is a low-level DataStream/ReferenceStream method. See also objectToStoreOnDataStream. NOTE: This method must send 'aDataStream beginInstance:size:' and then (nextPut:/nextPutWeak:) its subobjects. readDataFrom:size: reads back what we write here." | cntInstVars cntIndexedVars | cntInstVars := self class instSize. cntIndexedVars := self basicSize. aDataStream beginInstance: self class size: cntInstVars + cntIndexedVars. 1 to: cntInstVars do: [:i | aDataStream nextPut: (self instVarAt: i)]. "Write fields of a variable length object. When writing to a dummy stream, don't bother to write the bytes" ((aDataStream byteStream class == DummyStream) and: [self class isBits]) ifFalse: [ + self class isWeak + ifTrue: [ + "For weak classes (for example DependentsArray) write the referenced object only + if referenced from elsewhere in the dumped object graph. + This means, for instance that if we only dump a model, no dependents are stored, + but if we store a view (i.e. a Morph), it is properly handled as a dependent after the object graph is revived." + 1 to: cntIndexedVars do: [ :i | + aDataStream nextPutWeak: (self basicAt: i)]] + ifFalse: [ + 1 to: cntIndexedVars do: [ :i | + aDataStream nextPut: (self basicAt: i)]]]! - 1 to: cntIndexedVars do: - [:i | aDataStream nextPut: (self basicAt: i)]]. - !

1 0

The Trunk: Monticello-cmm.489.mcz
by commits＠source.squeak.org 28 Nov '11

28 Nov '11

Chris Muller uploaded a new version of Monticello to project The Trunk: http://source.squeak.org/trunk/Monticello-cmm.489.mcz ==================== Summary ==================== Name: Monticello-cmm.489 Author: cmm Time: 27 November 2011, 5:34:34.624 pm UUID: bc247bba-c0c0-4614-ad4a-ed220b22ce5a Ancestors: Monticello-cmm.485, Monticello-cmm.488 - Better guard in packageSelection: avoids debugger when viewing a repository with no packages. =============== Diff against Monticello-cmm.488 =============== Item was changed: ----- Method: MCRepositoryInspector>>packageSelection: (in category 'morphic ui') ----- packageSelection: aNumber + selectedPackage := (aNumber between: 1 and: self packageList size) + ifTrue: [ (self packageList at: aNumber) asString ]. - selectedPackage := aNumber isZero - ifFalse: [ (self packageList at: aNumber) asString ]. self versionSelection: 0. versionNames := nil. self changed: #packageSelection; changed: #versionList!

1 0

[ANN] First try: Fuel for Squeak
by Mariano Martinez Peck 27 Nov '11

27 Nov '11

Hi guys. After a couple of efforts from Martin Dias and I, we have a "working" Fuel serializer for Squeak. I used the quotes because all I could do is to run tests. Fuel is a binary serializer and you can find all the information in http://rmod.lille.inria.fr/web/pier/software/Fuel This is a project sponsored by ESUG SummerTalk and mainly developed by Martin Dias. I only tried in a Squeak 4.2 image. To install it: (Installer monticello http: 'http://ss3.gemstone.com/ss') project: 'Fuel'; install: 'ConfigurationOfFuel'. ((Smalltalk at: #ConfigurationOfFuel) project version: '1.8-baseline') load. Most tests are green. There are only 2 categories of tests failing: 1) those under the category 'FuelTests-InternalMetalevel'. This is because Squeak has removed the traits there are in Pharo and hence most tests fails. However, this stuff of "InternalMetalevel" is only used if you want to serialize classes/traits completely. Most of the cases, the user just wants to serialize a class/trait as global (just serialize its name) and then during materialization it just searches the class. Of course, in the image where you materialize, such classes have to be present. 2) some tests that are long and it seems in squeak there is a timeout that make those tests to fail. As always, we appreciate any feedback. Cheers -- Mariano http://marianopeck.wordpress.com

2 5

squeakvm.org seems down
by Bernhard Pieber 26 Nov '11

26 Nov '11

According to http://www.downforeveryoneorjustme.com/ it is not just me. ;-) Who is maintaining that currently? Whom should I notify? Cheers, Bernhard

3 2

Re: [squeak-dev] Re: Testing Seaside in Squeak4.3alpha-11793
by H. Hirzel 25 Nov '11

25 Nov '11

Dale A retest some time later (Installer ss project: 'MetacelloRepository') install: 'ConfigurationOfSeaside30'. ((Smalltalk at: #ConfigurationOfSeaside30) project version: '3.0.5') load. (Smalltalk at: #WAPharoServerAdaptorBrowser) open. loaded fine into Squeak4.3gamma-11793 The 'start' button in the Seaside panel still does not work. However with the workaround of inspecting the adaptor and doing 'self start' it is fine. --Hannes On 6/30/11, Dale Henrichs <dhenrich(a)vmware.com> wrote: > Hannes, > > To give you additional information ... > > From the #record output, it appears that Seaside30 was already loaded in the > image when you ran #record ... otherwise I would expect to see a whole lot > of "Load ....mcz" records:) > > If you do something like the following: > > | version | > version := (Smalltalk at: #ConfigurationOfSeaside30) project > version: '3.0.5'. > version ignoreImage: true. > version record. > > The output should list _all_ of the packages that would be loaded into the > image if Seaside3.0 hadn't already been loaded ... sometimes that is useful > too. > > I haven't been trying to keep up with the evolution of the various > registration APIs that are used by the different versions of Pharo and > Squeak ... that's presumably why the control panel is missing from the > menu... > > Dale > ----- Original Message ----- > | From: "H. Hirzel" <hannes.hirzel(a)gmail.com> > | To: "Dale Henrichs" <dhenrich(a)vmware.com> > | Cc: seaside(a)lists.squeakfoundation.org, > squeak-dev(a)lists.squeakfoundation.org > | Sent: Wednesday, June 29, 2011 4:18:59 PM > | Subject: Re: [squeak-dev] Re: Testing Seaside in Squeak4.3alpha-11481 > | > | Thank you, Dale > | > | I did > | ((Smalltalk at: #ConfigurationOfSeaside30) project version: '3.0.5') > | record. > | in Squeak 4.3-11481 > | > | Unfortunately there is no Seaside control panel anymore in the apps > | menu. It was present in an ealier Seaside in Squeak 4.2. Perhaps > | Randal is referring to this.... > | > | Hannes > | > | > ---------------------------------------------------------------------------------------- > | > | Squeak4.3alpha-11481.image priorSource: 3256581 > | Loading 1.0-beta.29 of ConfigurationOfMetacello... > | Fetched -> ConfigurationOfGofer-dkh.34 --- > | http://seaside.gemstone.com/ss/metacello --- > | http://www.squeaksource.com/MetacelloRepository > | Loaded -> ConfigurationOfGofer-dkh.34 --- > | http://seaside.gemstone.com/ss/metacello --- > | http://www.squeaksource.com/MetacelloRepository > | Fetched -> ConfigurationOfHelpSystem-DaleHenrichs.13 --- > | http://www.squeaksource.com/MetacelloRepository --- > | http://www.squeaksource.com/MetacelloRepository > | Loaded -> ConfigurationOfHelpSystem-DaleHenrichs.13 --- > | http://www.squeaksource.com/MetacelloRepository --- > | http://www.squeaksource.com/MetacelloRepository > | Project: Gofer Core 1.0.5 > | Fetched -> Gofer-Core.squeak-dkh.135 --- > | http://seaside.gemstone.com/ss/metacello --- > | http://seaside.gemstone.com/ss/metacello > | Fetched -> Metacello-Core-dkh.465 --- > | http://seaside.gemstone.com/ss/metacello --- > | http://www.squeaksource.com/metacello > | Fetched -> Metacello-MC-dkh.522 --- > | http://www.squeaksource.com/metacello --- > | http://seaside.gemstone.com/ss/metacello > | Fetched -> Metacello-Platform.squeak-dkh.10 --- > | http://seaside.gemstone.com/ss/metacello --- > | http://www.squeaksource.com/metacello > | Fetched -> Metacello-ToolBox-dkh.105 --- > | http://www.squeaksource.com/metacello --- > | http://seaside.gemstone.com/ss/metacello > | Project: Help System 1.1 > | Fetched -> Metacello-Help-dkh.11 --- > | http://seaside.gemstone.com/ss/metacello --- > | http://www.squeaksource.com/metacello > | Loaded -> Gofer-Core.squeak-dkh.135 --- > | http://seaside.gemstone.com/ss/metacello --- cache > | Starting atomic load > | Loaded -> Metacello-Core-dkh.465 --- > | http://seaside.gemstone.com/ss/metacello --- cache > | Loaded -> Metacello-MC-dkh.522 --- > | http://www.squeaksource.com/metacello --- cache > | Loaded -> Metacello-Platform.squeak-dkh.10 --- > | http://seaside.gemstone.com/ss/metacello --- cache > | Loaded -> Metacello-ToolBox-dkh.105 --- > | http://www.squeaksource.com/metacello --- cache > | Loaded -> Metacello-Help-dkh.11 --- > | http://seaside.gemstone.com/ss/metacello --- cache > | Finished atomic load > | ...finished 1.0-beta.29 > | Loading 1.0-beta.29 of ConfigurationOfMetacello... > | Project: Gofer Core 1.0.5 > | Project: Help System 1.1 > | ...finished 1.0-beta.29 > | Recording 3.0.5 of ConfigurationOfSeaside30... > | Project: Grease Core 1.0.5 > | Fetched -> ConfigurationOfGrease-dkh.149 --- > | http://www.squeaksource.com/MetacelloRepository --- > | http://www.squeaksource.com/MetacelloRepository > | Loaded -> ConfigurationOfGrease-dkh.149 --- > | http://www.squeaksource.com/MetacelloRepository --- C:\Documents and > | Settings\user\My Documents\My Squeak\Squeak > | 4.3-alpha-11481\package-cache > | Project: Grease Core Tests 1.0.5 > | Project: Swazoo2 2.3.0.0 > | Fetched -> ConfigurationOfSwazoo2-janko.29 --- > | http://www.squeaksource.com/MetacelloRepository --- > | http://www.squeaksource.com/MetacelloRepository > | Loaded -> ConfigurationOfSwazoo2-janko.29 --- > | http://www.squeaksource.com/MetacelloRepository --- C:\Documents and > | Settings\user\My Documents\My Squeak\Squeak > | 4.3-alpha-11481\package-cache > | Project: Sport2 2.31 > | Fetched -> ConfigurationOfSPort2-DaleHenrichs.4 --- > | http://www.squeaksource.com/MetacelloRepository --- > | http://www.squeaksource.com/MetacelloRepository > | Loaded -> ConfigurationOfSPort2-DaleHenrichs.4 --- > | http://www.squeaksource.com/MetacelloRepository --- C:\Documents and > | Settings\user\My Documents\My Squeak\Squeak > | 4.3-alpha-11481\package-cache > | Project: Kom 1.0.9 > | Fetched -> ConfigurationOfKomHttpServer-dkh.25 --- > | http://www.squeaksource.com/MetacelloRepository --- > | http://www.squeaksource.com/MetacelloRepository > | Loaded -> ConfigurationOfKomHttpServer-dkh.25 --- > | http://www.squeaksource.com/MetacelloRepository --- C:\Documents and > | Settings\user\My Documents\My Squeak\Squeak > | 4.3-alpha-11481\package-cache > | Project: OB stable > | Fetched -> ConfigurationOfOmniBrowser-dkh.144 --- > | http://www.squeaksource.com/MetacelloRepository --- > | http://www.squeaksource.com/MetacelloRepository > | ConfigurationOfOmniBrowser>>postLoadOBStandardPharo12 (OBCodeBrowser > | is Undeclared) > | ConfigurationOfOmniBrowser>>postLoadOBStandardPharo13 (OBCodeBrowser > | is Undeclared) > | Loaded -> ConfigurationOfOmniBrowser-dkh.144 --- > | http://www.squeaksource.com/MetacelloRepository --- C:\Documents and > | Settings\user\My Documents\My Squeak\Squeak > | 4.3-alpha-11481\package-cache > | Project: SPort2 2.31 > | ...finished 3.0.5 > | > | > | > | On 6/29/11, Dale Henrichs <dhenrich(a)vmware.com> wrote: > | > Hannes, > | > > | > The #record message should be sent to a version so you need to do > | > something > | > like: > | > > | > ((Smalltalk at: #ConfigurationOfSeaside30) project version: > | > '3.0.5') > | > record. > | > > | > BTW, that is how load works as well: > | > > | > ((Smalltalk at: #ConfigurationOfSeaside30) project version: > | > '3.0.5') load. > | > > | > Dale > | > > | > ----- Original Message ----- > | > | From: "H. Hirzel" <hannes.hirzel(a)gmail.com> > | > | To: dhenrich(a)vmware.com > | > | Cc: seaside(a)lists.squeakfoundation.org > | > | Sent: Wednesday, June 29, 2011 3:16:14 PM > | > | Subject: Fwd: [squeak-dev] Re: Testing Seaside in > | > | Squeak4.3alpha-11481 > | > | > | > | Dale > | > | > | > | It seems that my answer was hidden under a pile of other emails. > | > | > | > | I actually tried to do what you recommend to use > | > | render > | > | > | > | (Smalltalk at: #ConfigurationOfSeaside30) record > | > | > | > | to get the load sequence of ConfigurationOfSeaside30-dkh.316 in > | > | Squeak 4.3-11481. But it did not work. > | > | > | > | What am I missing? > | > | > | > | Regards > | > | Hannes > | > | > | > | > | > | > | > | ---------- Forwarded message ---------- > | > | From: "H. Hirzel" <hannes.hirzel(a)gmail.com> > | > | Date: Tue, 21 Jun 2011 05:46:29 +0000 > | > | Subject: Re: [squeak-dev] Re: Testing Seaside in > | > | Squeak4.3alpha-11481 > | > | To: The general-purpose Squeak developers list > | > | <squeak-dev(a)lists.squeakfoundation.org> > | > | Cc: dhenrich(a)vmware.com > | > | > | > | On 6/6/11, Dale Henrichs <dhenrich(a)vmware.com> wrote: > | > | > Hannes, > | > | > > | > | > If you want to do a test load, you can use #record or #record: > | > | > instead of > | > | > #load or #load: and print the result ... you'll get a listing > | > | > of > | > | > the > | > | > packages and configurations that will be loaded into your image > | > | > ... > | > | > > | > | > I assume that the Button problem is related to an OB issue on > | > | > Squeak. I am > | > | > not aware of a fix, but if a fixed version appears I can update > | > | > the > | > | > ConfigurationOfOmniBrowser to point to the appropriate packages > | > | > .... > | > | > > | > | > Dale > | > | > | > | > | > | Dale, > | > | > | > | Thank you for pointing me to the use of the > | > | > | > | #record > | > | > | > | method instead of #load. > | > | I was not aware of this logging feature. However it did not work > | > | in > | > | ftp://ftp.squeak.org/4.3alpha/Squeak4.3alpha-11481.zip > | > | > | > | > | > | If I do > | > | > | > | "Seaside 3.0 http://www.seaside.st" > | > | (Installer ss project: 'MetacelloRepository') install: > | > | 'ConfigurationOfSeaside30'. > | > | (Smalltalk at: #ConfigurationOfSeaside30) load. > | > | (Smalltalk at: #WAPharoServerAdaptorBrowser) open. > | > | > | > | > | > | I get the error message that > | > | > | > | ConfigurationOfSeaside30 does not understand #record. > | > | > | > | What am I missing here? > | > | > | > | Regards > | > | Hannes > | > | > | > > | >

1 0

Squeak Smalltalk from the start videos
by Lawson English 24 Nov '11

24 Nov '11

Hi all, I'm creating a series of videos for programming Squeak that are intended for a reasonably beginning audience. I don't cover questions like "what is an ASCII character?" and the like, but I don't make too many assumptions about a viewer's background past the assumption that they know what "computer programming" means at some level. Please feel free to critique/make suggestions or otherwise offer feedback. Squeak from the very start Starting with squeak programming with as few assumptions as possible http://www.youtube.com/playlist?list=PL6601A198DF14788D&feature=viewall Lawson

2 2

← Newer
1
2
3
4
5
6
7
...
17
Older →

Jump to page: