Squeak-dev November 2011

squeak-dev@lists.squeakfoundation.org

52 participants
168 discussions

Re: [squeak-dev] Re: 4.3 Release Schedule: Feature freeze today
by Edgar J. De Cleene 08 Dec '11

08 Dec '11

On 11/19/11 5:57 AM, "H. Hirzel" <hannes.hirzel(a)gmail.com> wrote: > From: Andreas Raab <andreas.raab(a)gmx.de> > Date: Thu, 17 Nov 2011 20:32:58 +0100 > Subject: [squeak-dev] Re: 4.3 Release Schedule: Feature freeze today > To: The general-purpose Squeak developers list > <squeak-dev(a)lists.squeakfoundation.org> > > Could someone update the alpha download? It would be useful to have a > current version for testing. > > Cheers, > - Andreas I have some ready, I interim nominate Squeak4.3gamma-11793 3081 run, 3050 passes, 7 expected failures, 20 failures, 2 errors, 2 unexpected passes Failures DependencyBrowserTest>>#testClassList ExceptionTests>>#testHandlerFromAction MCFileInTest>>#testStWriter MCMczInstallerTest>>#testInstallFromFile MCMczInstallerTest>>#testInstallFromStream MirrorPrimitiveTests>>#testMirrorAt MirrorPrimitiveTests>>#testMirrorInstVarAt PackageDependencyTest>>#testExceptions PackageDependencyTest>>#testMonticello PackageDependencyTest>>#testNetwork PackageDependencyTest>>#testST80 PackageDependencyTest>>#testSystem PackageDependencyTest>>#testToolBuilder PackageDependencyTest>>#testToolBuilderMVC PackageDependencyTest>>#testToolBuilderSUnit ProcessTest>>#testAtomicSuspend RWBinaryOrTextStreamTest>>#testExisiting ReleaseTest>>#testNoObsoleteClasses SocketTest>>#testSocketReuse SocketTest>>#testUDP SystemChangeFileTest>>#testClassAdded WeakSetInspectorTest>>#testSymbolTableM6812 Errors CompiledMethodTest>>#testPerformCanExecutelongMethodWithTemps CompiledMethodTest>>#testPerformInSuperclassCanExecutelongMethodWithTemps And when running the test , the debugger shows something related to primitive socket (sorry , no external log to attach) We need a Welcome Window showing what is new and About Squeak should say is 4.3 and not 4.2 I uploaded for all test Wait some minutes Edgar

4 7

[Bug][Fix] In ReferenceStream, code review request
by Juan Vuletich 08 Dec '11

08 Dec '11

Hi Folks, I found a bug in Cuis that made weak references to be improperly dumped in SmartRefStreams due to a bug in ReferenceStream. When I tried to reproduce the bug and fix in Squeak, I found out that in Squeak weak references are always dumped (and not just the more subtle bug in Cuis). The attach includes a test and fix. I'm confident about the fix in the context of Cuis, but I only played a few minutes with this in Squeak. Can anybody check this for Squeak and consider it for inclusion? Thanks, Juan Vuletich 'From Squeak4.2 of 20 November 2011 [latest update: #11796] on 27 November 2011 at 2:56:53 pm'! TestCase subclass: #ReferenceStreamTest instanceVariableNames: '' classVariableNames: '' poolDictionaries: '' category: 'System-Object Storage-Tests'! !Object methodsFor: 'objects from disk' stamp: 'jmv 11/27/2011 14:56'! 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)]]]! ! !DiskProxy methodsFor: 'i/o' stamp: 'jmv 11/25/2011 13:20'! storeDataOn: aReferenceStream "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! ! !ReferenceStream methodsFor: 'statistics' stamp: 'jmv 11/25/2011 13:21'! 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. n := 0. 'Finding Owners...' displayProgressFrom: 0 to: normalReferences size during: [:bar | normalReferences keysDo: [:parent | | kids | bar value: (n := n+1). kids := parent class isFixed 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]) do: [:child | parents at: child put: parent]]]. ownerBags := Dictionary new. tallies := Bag new. n := 0. 'Tallying Owners...' displayProgressFrom: 0 to: normalReferences size during: [:bar | normalReferences 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]]]! ! !ReferenceStream methodsFor: 'writing' stamp: 'jmv 11/25/2011 13:20'! addSpecialReference: aDiskProxy "See senders. Added to avoid breaking encapsulation (assuming that #references would answer the actual collection)" references at: aDiskProxy put: #none! ! !ReferenceStream methodsFor: 'writing' stamp: 'jmv 11/27/2011 14:56'! 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 ]! ! !ReferenceStreamTest methodsFor: 'testing' stamp: 'jmv 11/27/2011 14:56'! 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 ])! !

5 18

finalizing processes?
by Yoshiki Ohshima 01 Dec '11

01 Dec '11

Hello, I have trouble terminating some processes I make. More specifically, I'm trying to terminate some processes when the their associated "owner" object is garbage collected. The problem seems that these processes hold back pointers to my owner object via outerContext of the closures so the owner object does not get garbage collected when nobody else is pointing to it. Attached is my test case. After filing in this to a trunk image, evaluate: k := KEventStream new. KEventStream beTimer: k with: 100. in a workspace. You'll see a little counter going on the left edge of screen. What I want to do is to stop the counter by evaluating: k := nil. Smalltalk garbageCollect. but the processes hold onto the closure, which in turn holds onto the context for executing #beTimer:with:, and that context holds onto 'k', or 'anObject'. I thought that closure could just be detached from the outer context but it is not the case here. Why? Thank you! -- Yoshiki

2 2

New Cog VMs available at...
by Eliot Miranda 01 Dec '11

01 Dec '11

VM.r2518/ <http://www.mirandabanda.org/files/Cog/VM/VM.r2518/>. CogVM binaries as per VMMaker.oscog-eem.139/r2518. Add access to VM arguments to Mac. Make Windows, Mac and Unix consistent: Smalltalk getSystemAttribute: -1 => executable name -2 .. -n => VM arguments *including* image (if explicitly supplied). Replace win32 command line parsing with unixesque code using CommandLineToArgvW. linux launch script: Cope with old linuxes use of /lib/tls hack for an optional thread-local-storage version of libc/libpthread. The launch scripts add /lib/tls:/usr/lib/tls: to LD_LIBRARY_PATH instead of /lib:/usr/lib: if the VM is linked to /lib/tls/libc. -- best, Eliot

2 3

Daily Commit Log
by commits＠source.squeak.org 01 Dec '11

01 Dec '11

Changes to Trunk (http://source.squeak.org/trunk.html) in the last 24 hours: http://lists.squeakfoundation.org/pipermail/packages/2011-November/005104.h… Name: System-dtl.461 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. ============================================= http://lists.squeakfoundation.org/pipermail/packages/2011-November/005105.h… Name: Kernel-dtl.655 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. ============================================= http://lists.squeakfoundation.org/pipermail/packages/2011-November/005106.h… Name: Tests-dtl.136 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. =============================================

1 0

an interesting read ...
by Michael Haupt 30 Nov '11

30 Nov '11

Hi, for those interested in dynamic programming languages, here's one on Scala: https://gist.github.com/1406238 The background info is here: http://www.infoq.com/news/2011/11/yammer-scala Best, Michael

2 2

The Trunk: Tests-dtl.136.mcz
by commits＠source.squeak.org 30 Nov '11

30 Nov '11

David T. Lewis uploaded a new version of Tests to project The Trunk: http://source.squeak.org/trunk/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 Trunk: System-dtl.461.mcz
by commits＠source.squeak.org 30 Nov '11

30 Nov '11

David T. Lewis uploaded a new version of System to project The Trunk: http://source.squeak.org/trunk/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 Trunk: Kernel-dtl.655.mcz
by commits＠source.squeak.org 30 Nov '11

30 Nov '11

David T. Lewis uploaded a new version of Kernel to project The Trunk: http://source.squeak.org/trunk/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

SqueakSSL on Windows
by Jon Hylands 29 Nov '11

29 Nov '11

Hi everyone, I loaded SqueakSSL-Core-ar.22.mcz and SqueakSSL-Tests-ar-13.mcz from the Monticello repository http://www.squeaksource.com/SqueakSSL into a Squeak 4.2 All-in-One image directly from the squeak.org website (which has not been updated). The following method: SqueakSSLTest >> #expectedFailures is implemented thusly: expectedFailures "If we don't have a cert all the tests fail" SqueakSSL platformName = 'Mac OS' ifTrue:[ "The following tests all need certificate selection to work properly." ^#( testConnectAccept testEncryptDecrypt testMultiFrameDecrypt testSingleByteDecrypt testSplitTlsFrameRead testStreamAccept testStreamConnect testStreamTransfer ) ]. This clearly doesn't work on Windows at all, since TestCase >> #shouldPass sends #includes: to the result of that method, which (if you're not running Mac OS) is the SqueakSSLTest instance. I added: ^super expectedFailures at the bottom, and the tests at least don't choke before they start now. Now the tests all fail. I grabbed the latest DLL from http://code.google.com/p/squeakssl/downloads/list, replaced the one that came with 4.2, and now all the tests pass :-) I'm hoping to be able to send & receive email from a specific gmail account using this - any suggestions? - Jon

3 5

Jump to page: