January 2017 - Pharo-dev - lists.pharo.org

enumerated return values from FFI callback

by Ben Coman

Following on from my "Parting words" at [1] I'd like to add the following... FFIExternalEnumerationType >> callbackReturnOn: callbackContext for: anObject ^ super callbackReturnOn: callbackContext for: anObject value but I'm seeking some expert feedback before logging an issue. For discussion background, here is a reference C program that sets up the framework for a callback that returns an enumeration... ----------example.c #include <stdio.h> typedef enum { A = 1, B = 2, C = 3 } CallBackReturnValue; CallBackReturnValue callbackFn() { return B; } CallBackReturnValue callout( CallBackReturnValue (*callback)() ) { return callback(); } int main() { printf("%d\n", callout(callbackFn) ); return 0; } ---------- And validation... $ gcc -Wall -Werror example.c $ ./a.out 2 ---------- Now after deleting main() and callback(), then compiling to a shared library per [2]... $ gcc -c -Wall -Werror -fpic example.c $ gcc -shared -o passthru.so example.o $ cp example.so /tmp the Pharo equivalent is.... FFIExternalEnumeration subclass: #CallBackReturnValue instanceVariableNames: '' classVariableNames: '' package: 'AAAExample' CallBackReturnValue class >> enumDecl ^ #( A 1 B 2 C 3 ) CallBackReturnValue class >> initialize self initializeEnumeration And evaluate "CallBackReturnValue initialize." FFILibrary subclass: #Example instanceVariableNames: '' classVariableNames: '' poolDictionaries: 'CallBackReturnValue' package: 'AAAExample' Example >> unixModuleName ^ '/tmp/example.so' Example class >> callout: callbackBlock ^ self ffiCall: #( int callout ( FFICallback callbackBlock ) ) module: Example Example class >> doCallback | callbackFn | callbackFn := FFICallback signature: #( CallBackReturnValue ( void ) ) block: [ B ]. ^ Example callout: callbackFn. ----- But "Example doCallback" produces error in VMCallbackContext32>>wordResult: anInteger where anInteger is aCallBackReturnValue(#B) rather than 2 I can resolve this using... callbackFn := FFICallback signature: #( CallBackReturnValue ( void ) ) block: [ B value ]. "<===" but the original direct use of the enumeration "B" directly is nicer. This can be facilitated by moving #value as follows... FFIExternalEnumerationType >> callbackReturnOn: callbackContext for: anObject ^ super callbackReturnOn: callbackContext for: anObject value "<==" where the existing method for super is... FFIExternalType >> callbackReturnOn: callbackContext for: anObject ^ callbackContext wordResult: anObject Anyone see a problem with that addition? btw, Would that example be useful for the book? cheers -ben [1] http://blog.openinworld.com/2016/09/pharo-libclang-ffi-part-4-ast-walking-w… [2] http://www.cprogramming.com/tutorial/shared-libraries-linux-gcc.html

Jan. 20, 2017

Re: [Pharo-dev] [Ann] Calypso update: method browser and better UI

by Denis Kudriashov

2017-01-12 15:44 GMT+01:00 Hilaire <hilaire(a)drgeo.eu>: > > And of course it should open in new tab too. > > > > > > Yes. Problem that I not hook into text editor yet. So any navigation > > from source code is standard Pharo "global commands". But I need to use > > Calypso commands instead. And it is not trivial task to hack this place. > > I guess so, but you will be real winner. And this is done. I just upload new stable version: - source code shortcuts are managed by Calypso. I found simple hook how to achieve this - senders/implementors by source code shortcuts in method browser are now opened in new tabs instead of separate window. - no Spec text model anymore. Rubric widgets are used directly. It improves speed of building text editors

Jan. 20, 2017

[pharo-project/pharo-core]

by GitHub

Branch: refs/tags/60346 Home: https://github.com/pharo-project/pharo-core

Jan. 20, 2017

[pharo-project/pharo-core] a5f125: 60346

by GitHub

Branch: refs/heads/6.0 Home: https://github.com/pharo-project/pharo-core Commit: a5f125365fef092f200dca731327759aae6265ce https://github.com/pharo-project/pharo-core/commit/a5f125365fef092f200dca73… Author: Jenkins Build Server <board(a)pharo-project.org> Date: 2017-01-20 (Fri, 20 Jan 2017) Changed paths: M ConfigurationOfFlatQA.package/ConfigurationOfFlatQA.class/instance/symbolic versions/stable_.st A ConfigurationOfFlatQA.package/ConfigurationOfFlatQA.class/instance/versions/versionv3%5F2%5F11_.st R Renraku-Test.package/RenrakuTest.class/instance/tests/testUnusedVar.st A Renraku.package/ReVarSearchSourceAnchor.class/instance/accessing/interval.st A Renraku.package/ReVarSearchSourceAnchor.class/instance/accessing/searchString.st R Renraku.package/ReVarSearchSourceAnchor.class/instance/as yet unclassified/interval.st R Renraku.package/ReVarSearchSourceAnchor.class/instance/as yet unclassified/searchString.st R Renraku.package/extension/RBOnlyReadOrWrittenVariableRule/instance/check_forCritiquesDo_.st R Renraku.package/extension/RBUnreferencedVariablesRule/instance/check_forCritiquesDo_.st R ScriptLoader60.package/ScriptLoader.class/instance/pharo - scripts/script60345.st A ScriptLoader60.package/ScriptLoader.class/instance/pharo - scripts/script60346.st R ScriptLoader60.package/ScriptLoader.class/instance/pharo - updates/update60345.st A ScriptLoader60.package/ScriptLoader.class/instance/pharo - updates/update60346.st M ScriptLoader60.package/ScriptLoader.class/instance/public/commentForCurrentUpdate.st Log Message: ----------- 60346 19565 QA v3.2.11 https://pharo.fogbugz.com/f/cases/19565 http://files.pharo.org/image/60/60346.zip

Jan. 20, 2017

[pharo-project/pharo-core]

by GitHub

Branch: refs/tags/60345 Home: https://github.com/pharo-project/pharo-core

Jan. 20, 2017

[pharo-project/pharo-core] eb4c24: 60345

by GitHub

Branch: refs/heads/6.0 Home: https://github.com/pharo-project/pharo-core Commit: eb4c24b44409bbde4df845f6558c98564300ad48 https://github.com/pharo-project/pharo-core/commit/eb4c24b44409bbde4df845f6… Author: Jenkins Build Server <board(a)pharo-project.org> Date: 2017-01-20 (Fri, 20 Jan 2017) Changed paths: M GT-Inspector.package/extension/GlamourPresentationModel/class/inspectorOn_.st M Glamour-SpecIntegration.package/GlamourPresentationModel.class/README.md M Growl.package/GrowlMorph.class/class/settings/growlSettingsOn_.st M Morphic-Base.package/ShortcutReminder.class/instance/settings/customSettingsOn_.st A Nautilus-Tests.package/NautilusUITest.class/instance/tests/testMethodSortBlock.st A Nautilus.package/NautilusUI.class/instance/private/methodSortBlock.st M Nautilus.package/NautilusUI.class/instance/widget method/methodsForCategory_.st M OpalCompiler-Core.package/OCASTTranslator.class/README.md R ScriptLoader60.package/ScriptLoader.class/instance/pharo - scripts/script60344.st A ScriptLoader60.package/ScriptLoader.class/instance/pharo - scripts/script60345.st R ScriptLoader60.package/ScriptLoader.class/instance/pharo - updates/update60344.st A ScriptLoader60.package/ScriptLoader.class/instance/pharo - updates/update60345.st M ScriptLoader60.package/ScriptLoader.class/instance/public/commentForCurrentUpdate.st Log Message: ----------- 60345 19548 Fix Spec Integration class comments and examples https://pharo.fogbugz.com/f/cases/19548 19545 Add comment about OC translator subclasses (for effect / for value) https://pharo.fogbugz.com/f/cases/19545 19555 Nautilus should sort all binary methods before non-binary methods https://pharo.fogbugz.com/f/cases/19555 19550 Typo in settings: Popup notifaction --> Popup notification https://pharo.fogbugz.com/f/cases/19550 http://files.pharo.org/image/60/60345.zip

Jan. 20, 2017

Re: [Pharo-dev] In the need of Pharo magic

by Peter Uhnak

On Thu, Jan 19, 2017 at 06:46:22PM +0100, Stephane Ducasse wrote: > I would love to get a functionality that dynamically removes from my code > all the self halt that are encountered during test execution. > This way I will not need to run the test, walk the stack, edit the code and > retart the process. > > It would be really cool You mean to _permanently_ remove them after the test passed, or just temporarily disable them during the execution? The latter can be done with MetaLinks, e.g.: Something>>add: aNumber to: anotherNumber self halt. ^ anotherNumber + aNumber SomethingTest>>testAddTo self assert: (Something new add: 4 to: 10) equals: 14 SomethingTest>>setUp | halts | super setUp. halts := (Something >> #add:to:) ast sendNodes select: [ :each | each selector = #halt ]. link := MetaLink new control: #instead. halts do: [ :each | each link: link ]. SomethingTest>>tearDown super tearDown. link uninstall Now if I call the method by hand it will halt, but the test will skip the halts. Peter

Jan. 20, 2017

Re: [Pharo-dev] In the need of Pharo magic

by Henrik Nergaard

Here is an example of using the ast to remove -all- halts from the code if one or more is encountered during execution. ------------------------------------------------------------ | haltTypes haltingMethods testCode | haltTypes := #( halt halt: haltIf: haltIfNil haltOnCount: haltOnce). haltingMethods := IdentitySet new. "example code" Object compile: 'aMethod | a | self halt. a := 42. self haltIf: a = 42; haltOnCount: 2; yourself. self haltOnce; halt: ''Halting!''. self yourself; haltIfNil; yourself. [ 1 halt ] value. ^ a'. testCode := (Object >> #aMethod) ast formattedCode. [ "do stuff here" Object new perform: #aMethod ] on: Halt do: [ :ex | haltingMethods add: ex signalerContext method. "ex debug" "<- show debugger on a halt" ex resume. ]. haltingMethods do: [ :method | | ast cascades | ast := method ast copy. cascades := IdentitySet new. ast allChildren select: [ :child | child isMessage and: [ haltTypes includes: child selector ] ] thenDo: [ :child | | parent | parent := child parent. parent isCascade ifTrue: [ parent messages remove: child. cascades add: parent ] ifFalse: [ parent removeNode: child ]. ]. cascades select: [ :cascadeNode | cascadeNode messages isEmpty] thenDo: [ :cascadeNode | cascadeNode parent removeNode: cascadeNode ]. method origin compile: ast formattedCode classified: method protocol. ]. ( DiffMorph from: testCode to: (Object >>#aMethod) sourceCode contextClass: Object ) openInWindow ------------------------------------------------------------ Best regards, Henrik Fra: Pharo-dev [mailto:pharo-dev-bounces@lists.pharo.org] PÃ¥ vegne av Stephane Ducasse Sendt: 19 January 2017 18:46 Til: Pharo Development List <pharo-dev(a)lists.pharo.org> Emne: [Pharo-dev] In the need of Pharo magic I would love to get a functionality that dynamically removes from my code all the self halt that are encountered during test execution. This way I will not need to run the test, walk the stack, edit the code and retart the process. It would be really cool

Jan. 19, 2017

Re: [Pharo-dev] [squeak-dev] Support for LP64 & LLP64 in FFI Kernel

by David T. Lewis

+1 This sounds like a good approach to me. Dave > Hi All, > > currently the FFI has support for 8, 16, 32 & 64 bit integers, but no > support for C's long type. In almost all C implementations byte is 8 > bits, > short is 16 bits, int is 32 bits and long long is 64 bits. But long may > be > either 32 or 64 bits depending on programming model. > > In LP32 long is 32 bits, which is used in 32-bit unix, mac os and windows. > In LP64 long is 64 bits, which is used in 64-bit unix and mac os. > In LLP64 long is 32 bits, which is used in 64-bit windows. > > So having a type which is either 32 or 64 bits depending on platform is > useful in modeling C's long type across platforms. Esteban and I propose > adding such a type to FFI kernel, actually two types, one unsigned, one > signed. So we propose extending the below (these constants are used to > define signatures for FFI calls and the contents of structs) > > FFIConstants class>>#initializeTypeConstants > "type void" > FFITypeVoid := 0. > > "type bool" > FFITypeBool := 1. > > "basic integer types. > note: (integerType anyMask: 1) = integerType isSigned" > > FFITypeUnsignedByte := 2. > FFITypeSignedByte := 3. > FFITypeUnsignedShort := 4. > FFITypeSignedShort := 5. > FFITypeUnsignedInt := 6. > FFITypeSignedInt := 7. > > "64bit types" > FFITypeUnsignedLongLong := 8. > FFITypeSignedLongLong := 9. > > "special integer types" > FFITypeUnsignedChar := 10. > FFITypeSignedChar := 11. > > "float types" > FFITypeSingleFloat := 12. > FFITypeDoubleFloat := 13. > > "type flags" > FFIFlagAtomic := 16r40000. "type is atomic" > FFIFlagPointer := 16r20000. "type is pointer to base type" > FFIFlagStructure := 16r10000. "baseType is structure of 64k length" > FFIStructSizeMask := 16rFFFF. "mask for max size of structure" > FFIAtomicTypeMask := 16r0F000000. "mask for atomic type spec" > FFIAtomicTypeShift := 24. "shift for atomic type" > > to add to > "special integer types" > FFITypeUnsignedChar := 10. > FFITypeSignedChar := 11. > FFITypePlatformUnsignedLong := 14. > FFITypePlatformSignedLong := 15. > > This is backward compatible, even if it complicates the marshaling a > little > since it used to be that all integer types were less than > FFITypeSingleFloat. But breaking backward compatibility would be a bad > idea. > > Then in ByteArray et al we could add accessors for platformLongAt: > platformUnsignedLongAt:. > > Comments, thoughts, objections? > _,,,^..^,,,_ > best, Eliot > >

Jan. 19, 2017

Re: [Pharo-dev] In the need of Pharo magic

by Graham McLeod

Hi Stef, see an idea below.. Stephane Ducasse wrote: > > I would love to get a functionality that dynamically removes from my > code all the self halt that are encountered during test execution. > This way I will not need to run the test, walk the stack, edit the > code and retart the process. > > It would be really cool One approach I have used is to have a global called "DebugMode" which is s boolean. The, usie two types of debugging statements, like: DebugMode ifTrue: [self halt]. and DebugMode: ifTrue: [Transcrip show: somethingUseful ] Then you can just turn DebugMode on or off. This could be made simpler if the self halt of the system at object level checked such a flag and if the Transcript had a subclass which wouldn't log if DebugMode was false. KR Graham

Jan. 19, 2017