January 2012 - Pharo-users - lists.pharo.org

Re: [Pharo-users] Pharo Morphic View Application Framework

by Benjamin

If you have around the pdf you are talking about on your website, I will be glad to read it. Thanks in advance ;) Ben On Jan 31, 2012, at 5:58 AM, S Krish wrote: > http://skrishnamachari.wordpress.com/2012/01/31/pharo-application-framework… > > Obliged for feedback ... > > For details and links.. > Sharing this little framework that should go some way to ease creating large applications ( currently morphic oriented). > > AbstractMorphicView subclass: #MyNewView â¦ > > #createMorphsSpecs > > ^{ > > âpanel1â² -> #SomeOtherView. > âlabel1â² -> #LabelMorph. > âtext1â² -> #PluggableTextMorph. > âbutton1â² -> #PluggableButtonMorph. > } > > layoutSpecsArray > > ^{ > âwidgetNameInMorphsDictionaryâ -> { > #fractions -> (0 @ 0 corner: 1 @ 0.5). â fractional value 0 to 1 range â > #offsets -> (10 @ 10 corner: 0 @ 0). â offsets relative to the fractional position â > }. > > ââ¦ add other widgetspecsâ¦â > } > > Thats it.. you should have a prototype UI neatly laid out.. and then specify: #morphsPrimaryPropertiesSpecs and #morphsSecondaryPropertiesSpecs for full functionality that composes views inside views and scales well. > > AbstractSimpleMorphicView can make it simpler for TableLayout stuff without requiring #layoutSpecsArray. > > ..... the main post has other details and links to package... >

Jan. 31, 2012

Re: [Pharo-users] focus setting

by Werner Kassens

i noticed that although i dont get a keystroke event for shift-tab, i do get keyUp and keyDown events. obviously they all come from the vm and the new vm in Pharo 1.3 probably cant keep up with my very old linux version. otoh i cant open the 1.3 image with the vm from Pharo 1.2 (This interpreter (vers. 6502) cannot read image file (vers. 6505) ). i guess, one day or another i have to update to a newer linux version. for the moment i just changed the shift-tab keyDown event to a keystroke event, since i need the keystroke event but not the keyDown one. or is there an older vm more similar to the one in Pharo 1.2, that can open the 1.3 image? werner

Jan. 31, 2012

Re: [Pharo-users] [Pharo-project] Pharo Morphic View Application Framework

by Denis Kudriashov

2012/1/31 StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> > > Denis > we do not want to rely on saving a Morph to a method (serialization). > Now it does not mean that we cannot use the prototype idea. > We could have a spec that creates an object used as a prototype for > building UI. > > Saving a prototypes and building interface are orthogonal. > > Stef > > Yes, I understand this

Jan. 31, 2012

Re: [Pharo-users] [Pharo-project] Pharo Morphic View Application Framework

by StÃ©phane Ducasse

On Jan 31, 2012, at 3:04 PM, Denis Kudriashov wrote: > 2012/1/31 S Krish <krishnamachari.sudhakar(a)gmail.com> > Sure I will review this pdf.. and come back on this..how well should > integrate the specs part.. > > The framework though is based on the spec and extensibly on direct > composition and does a little more towards building a scalable/ > flexible framework that incorporates in one simpler hierarchy all that > goes into a standard large application creation.. > > How is the Presenty work supposed to work along with this..? > > > For Presenty Specs can be used for code presentation of concrete morphs (view prototypes). > If Specs can save arbitraty morph to method I will use it. > And Presenty will got monticello support. Denis we do not want to rely on saving a Morph to a method (serialization). Now it does not mean that we cannot use the prototype idea. We could have a spec that creates an object used as a prototype for building UI. Saving a prototypes and building interface are orthogonal. Stef

Jan. 31, 2012

Re: [Pharo-users] [Pharo-project] Pharo Morphic View Application Framework

by StÃ©phane Ducasse

On Jan 31, 2012, at 1:23 PM, S Krish wrote: > Sure I will review this pdf.. and come back on this..how well should > integrate the specs part.. but Spec is a spec based UI builder. > The framework though is based on the spec and extensibly on direct > composition and does a little more towards building a scalable/ > flexible framework that incorporates in one simpler hierarchy all that > goes into a standard large application creation.. Ok but did you ***read*** what benjamin said: he iterated on it and now it is a good basis to build support for a lot of widgets. Have a look at his document. It would be good to avoid to have several frameworks doing the same. For example I would like to have more tests. The prototype part is open. We did not think much about it. > > How is the Presenty work supposed to work along with this..? > > > On 1/31/12, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> wrote: >> did you see what benjamin is doing with Spec. >> Because you should join effort >> >> >

Jan. 31, 2012

Re: [Pharo-users] focus setting

by Werner Kassens

> bug in your copy of the image.. possibly.. i just downloaded another Pharo 1.3, used one of the examples in the unchanged image and got the same result. i then made a descendant of StandardWindow with handlesKeyboard: evt ^true and: keyStroke: anEvent anEvent wasHandled: true. self inform: anEvent asString . while this reacts eg on tab or shift-up, it does not react on shift-tab, somehow strange. and it shoudnt be the OS, since Pharo 1.2 gets the shift-tab. werner

Jan. 31, 2012

Re: [Pharo-users] focus setting

by S Krish

bug in your copy of the image.. possibly.. shift-tab seems to work fine for me over most variants I have worked.. just hacked to also include tab windows through ctrl+tab.. which is actually missing.. capability.. when you want to cycle through windows within pharo Morph>>tabKey: event "Check for tab key activity and change focus as appropriate." event controlKeyPressed ifFalse: [*event keyCharacter = Character tab* * ifTrue: [event commandKeyPressed* * ifTrue: [World tabOpenWindows: event].* *event shiftPressed* ifTrue: [self navigateFocusBackward] ifFalse: [self navigateFocusForward]. ^ true]]. ^ false PasteUpMorph>> keyStroke: anEvent ... anEvent keyCharacter == Character tab ifTrue: [self tabAmongFields ifTrue: [ anEvent commandKeyPressed ifTrue: [^ self tabOpenWindows: anEvent]. ... tabOpenWindows: aKeyboardEvent | currentFocus index | currentFocus := aKeyboardEvent hand keyboardFocus. SystemWindow noteTopWindowIn: self. SystemWindow topWindow sendToBack. " could actually raise the next window to top rather" On Tue, Jan 31, 2012 at 6:53 PM, Werner Kassens <wkassens(a)libello.com>wrote: > Hi, > in Pharo 1.2 i could tab through the submorphs using tab and shift-tab. in > Pharo 1.3 shift-tab seems not to work anymore. what can i use to move > backwards through the morphs to set the focus? > werner > >

Jan. 31, 2012

Re: [Pharo-users] [Pharo-project] Pharo Morphic View Application Framework

by Denis Kudriashov

2012/1/31 S Krish <krishnamachari.sudhakar(a)gmail.com> > Sure I will review this pdf.. and come back on this..how well should > integrate the specs part.. > > The framework though is based on the spec and extensibly on direct > composition and does a little more towards building a scalable/ > flexible framework that incorporates in one simpler hierarchy all that > goes into a standard large application creation.. > > How is the Presenty work supposed to work along with this..? > > For Presenty Specs can be used for code presentation of concrete morphs (view prototypes). If Specs can save arbitraty morph to method I will use it. And Presenty will got monticello support.

Jan. 31, 2012

focus setting

by Werner Kassens

Hi, in Pharo 1.2 i could tab through the submorphs using tab and shift-tab. in Pharo 1.3 shift-tab seems not to work anymore. what can i use to move backwards through the morphs to set the focus? werner

Jan. 31, 2012

Re: [Pharo-users] [Pharo-project] Pharo Morphic View Application Framework

by S Krish

Sure I will review this pdf.. and come back on this..how well should integrate the specs part.. The framework though is based on the spec and extensibly on direct composition and does a little more towards building a scalable/ flexible framework that incorporates in one simpler hierarchy all that goes into a standard large application creation.. How is the Presenty work supposed to work along with this..? On 1/31/12, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> wrote: > did you see what benjamin is doing with Spec. > Because you should join effort > >

Jan. 31, 2012

Re: [Pharo-users] Pharo Morphic View Application Framework

by StÃ©phane Ducasse

did you see what benjamin is doing with Spec. Because you should join effort the idea is that we can define object with spec and compose them and their logic using value holder. Stef On Jan 31, 2012, at 5:58 AM, S Krish wrote: > http://skrishnamachari.wordpress.com/2012/01/31/pharo-application-framework… > > Obliged for feedback ... > > For details and links.. > Sharing this little framework that should go some way to ease creating large applications ( currently morphic oriented). > > AbstractMorphicView subclass: #MyNewView â¦ > > #createMorphsSpecs > > ^{ > > âpanel1â² -> #SomeOtherView. > âlabel1â² -> #LabelMorph. > âtext1â² -> #PluggableTextMorph. > âbutton1â² -> #PluggableButtonMorph. > } > > layoutSpecsArray > > ^{ > âwidgetNameInMorphsDictionaryâ -> { > #fractions -> (0 @ 0 corner: 1 @ 0.5). â fractional value 0 to 1 range â > #offsets -> (10 @ 10 corner: 0 @ 0). â offsets relative to the fractional position â > }. > > ââ¦ add other widgetspecsâ¦â > } > > Thats it.. you should have a prototype UI neatly laid out.. and then specify: #morphsPrimaryPropertiesSpecs and #morphsSecondaryPropertiesSpecs for full functionality that composes views inside views and scales well. > > AbstractSimpleMorphicView can make it simpler for TableLayout stuff without requiring #layoutSpecsArray. > > ..... the main post has other details and links to package... >

Jan. 31, 2012

Pharo Morphic View Application Framework

by S Krish

http://skrishnamachari.wordpress.com/2012/01/31/pharo-application-framework… Obliged for feedback ... For details and links.. Sharing this little framework that should go some way to ease creating large applications ( currently morphic oriented). AbstractMorphicView subclass: #MyNewView â¦ #createMorphsSpecs ^{ âpanel1â² -> #SomeOtherView. âlabel1â² -> #LabelMorph. âtext1â² -> #PluggableTextMorph. âbutton1â² -> #PluggableButtonMorph. } layoutSpecsArray ^{ âwidgetNameInMorphsDictionaryâ -> { #fractions -> (0 @ 0 corner: 1 @ 0.5). â fractional value 0 to 1 range â #offsets -> (10 @ 10 corner: 0 @ 0). â offsets relative to the fractional position â }. ââ¦ add other widgetspecsâ¦â } Thats it.. you should have a prototype UI neatly laid out.. and then specify: #morphsPrimaryPropertiesSpecs and #morphsSecondaryPropertiesSpecs for full functionality that composes views inside views and scales well. AbstractSimpleMorphicView can make it simpler for TableLayout stuff without requiring #layoutSpecsArray. ..... the main post has other details and links to package...

Jan. 31, 2012

Re: [Pharo-users] Command Line Interface using FFI

by Guillermo Polito

Hmm, I've tried doing it from a non-headless image. Maybe it has something to do with it? Guille On Mon, Jan 30, 2012 at 8:34 AM, Sven Van Caekenberghe <sven(a)beta9.be>wrote: > > On 30 Jan 2012, at 11:44, Guillermo Polito wrote: > > > FileStream stdout nextPutAll: 'something';crlf. > > > > Works great, but > > > > FileStream stdin next. > > > > Never returns :/. Whatever I try. > > > > Any idea? > > That is not good, I must admit I only tried #stout, so I did: > > $ cat run.sh > > #!/bin/bash > script_home=$(dirname $0) > script_home=$(cd $script_home && pwd) > image=$script_home/experimental.image > script=$script_home/run.st > args=$* > vm=$script_home/bin/CogVM > options="-vm-display-null -vm-sound-null" > #echo $vm $options $image $script $args > $vm $options $image $script $args > > $ cat run.st > [ FileStream stdin atEnd ] whileFalse: [ > FileStream stdout nextPut: FileStream stdin next ]. > ! > Smalltalk snapshot: false andQuit: true. > > $ cat /tmp/foo.txt > Hello there! > Welcome to Smalltalk. > > > > $ cat /tmp/foo.txt | ./run.sh > Hello there! > Welcome to Smalltalk. > > > > So that seems to work. > > HTH, > > Sven > > >

Jan. 30, 2012

Re: [Pharo-users] Command Line Interface using FFI

by Guillermo Polito

On Mon, Jan 30, 2012 at 9:17 AM, David T. Lewis <lewis(a)mail.msen.com> wrote: > On Mon, Jan 30, 2012 at 07:44:40AM -0300, Guillermo Polito wrote: > > FileStream stdout nextPutAll: 'something';crlf. > > > > Works great, but > > > > FileStream stdin next. > > > > Never returns :/. Whatever I try. > > > > Any idea? > > > > This is because no data was received on the standard input stream. Try > typing something followed by <enter> so that the operating system terminal > actually sends some data. > I did :/. I always open my images from the terminal. And the stdout prints in the terminal where I've opened the image. But when if I use stdin and write to that console, it does not work :/. However, if I do this with fgets+FFI it works... (It blocks but when I enter data into the terminal the control is returned to the image). > By default, the stdin, stdout, and stderr streams are all in blocking mode > unless you set them otherwise. If you want to set the stdin to nonblocking > mode, see #setNonBlocking in package OSProcess. Note that there is no > "correct" setting for this, it depends on what you are trying to do with > the stream. > > HTH, > Dave > > Thanks!

Jan. 30, 2012

**Important** Pharo Conference Call for Date Feedback

by StÃ©phane Ducasse

Hi guys We want to organize a Pharo Conference @ lille around business and companies in May/June 2012. Now we need you to understand what is the best to organize because we have some constraints (like not good access to rooms on saturday). Could you reply to the following questions: Are you planning to attend: Yes No Would you attend two days: Yes No Would you like to present something: Yes No Would you prefer to have (remove the one you do not like) - one friday and one saturday or - two days during the week would be not a problem. Would you like to have Handons: Yes No Business/Experience returns: Yes No What are the following dates that are ok for you? Frid 11- Satur 12 May 2012: Yes No Thur 10- Frid 11 May 2012: Yes No Mond 14- Tuesd 15 May 2012: Yes No Tuesd 15- Wedn 16 May 2012: Yes No Thur 24 - Frid 25 May 2012: Yes No Frid 25- Satur 26 May 2012: Yes No Stef

Jan. 30, 2012

Re: [Pharo-users] Command Line Interface using FFI

by David T. Lewis

On Mon, Jan 30, 2012 at 07:44:40AM -0300, Guillermo Polito wrote: > FileStream stdout nextPutAll: 'something';crlf. > > Works great, but > > FileStream stdin next. > > Never returns :/. Whatever I try. > > Any idea? > This is because no data was received on the standard input stream. Try typing something followed by <enter> so that the operating system terminal actually sends some data. By default, the stdin, stdout, and stderr streams are all in blocking mode unless you set them otherwise. If you want to set the stdin to nonblocking mode, see #setNonBlocking in package OSProcess. Note that there is no "correct" setting for this, it depends on what you are trying to do with the stream. HTH, Dave

Jan. 30, 2012

Re: [Pharo-users] Command Line Interface using FFI

by Sven Van Caekenberghe

On 30 Jan 2012, at 11:44, Guillermo Polito wrote: > FileStream stdout nextPutAll: 'something';crlf. > > Works great, but > > FileStream stdin next. > > Never returns :/. Whatever I try. > > Any idea? That is not good, I must admit I only tried #stout, so I did: $ cat run.sh #!/bin/bash script_home=$(dirname $0) script_home=$(cd $script_home && pwd) image=$script_home/experimental.image script=$script_home/run.st args=$* vm=$script_home/bin/CogVM options="-vm-display-null -vm-sound-null" #echo $vm $options $image $script $args $vm $options $image $script $args $ cat run.st [ FileStream stdin atEnd ] whileFalse: [ FileStream stdout nextPut: FileStream stdin next ]. ! Smalltalk snapshot: false andQuit: true. $ cat /tmp/foo.txt Hello there! Welcome to Smalltalk. $ cat /tmp/foo.txt | ./run.sh Hello there! Welcome to Smalltalk. So that seems to work. HTH, Sven

Jan. 30, 2012

Re: [Pharo-users] Command Line Interface using FFI

by Guillermo Polito

FileStream stdout nextPutAll: 'something';crlf. Works great, but FileStream stdin next. Never returns :/. Whatever I try. Any idea? On Mon, Jan 30, 2012 at 1:29 AM, Guillermo Polito <guillermopolito(a)gmail.com > wrote: > > > On Sun, Jan 29, 2012 at 6:58 PM, Sven Van Caekenberghe <sven(a)beta9.be>wrote: > >> Hi Guillermo, >> >> On 29 Jan 2012, at 21:11, Guillermo Polito wrote: >> >> > I've spent a few hours playing with Alien+FFI and I did want to have a >> Command line interface to play and use as a quick browser. >> > >> > Gofer it >> > squeaksource: 'CLInterface'; >> > package: 'ConfigurationOfCLInterface'; >> > load. >> > >> > ConfigurationOfCLInterface load. >> > >> > >> > To play with it write a script, for example startCLI.st, with the >> following: >> > >> > CLInterface start >> > >> > >> > >> > And then, from the command line >> > >> > CogVM -vm-display-null <image> startCLI.st >> > >> > >> > What it does support? >> > - executes smalltalk expressions >> > - handles the errors and shows them into the stdout (so it does not >> explode :P) >> > - stores variables as a workspace >> > Since I used standard C functions (fgets and printf only) this should >> work on every platform from scratch (having a VM with FFI working, of >> course)... >> > >> > The problems I've ran so far: >> > - blocking FFI blocks the vm. I've not tried yet to use threads >> through FFI (and I don't know if it works. Does it?). But It should be >> great to start a headless seaside server with a command line :/. >> > >> > Ok, it's a very simple thing, but it allows to execute a expression >> from a remote server via ssh without doing much :). >> > >> > Guille >> >> Cool, but why do you need to access those C functions ? >> You do know that you can access std* from Smalltalk, do you ? >> > > No! :D > > >> Have a look at the FileStream class 'stdio' category. >> > > will do :). Will that solve the vm-block issue? > > >> >> Apart from that, nice small, readable and useful hack. >> >> Thanks :P. > > >> Regards, >> >> Sven >> >> >> >

Jan. 30, 2012

Re: [Pharo-users] Command Line Interface using FFI

by Guillermo Polito

On Sun, Jan 29, 2012 at 6:58 PM, Sven Van Caekenberghe <sven(a)beta9.be>wrote: > Hi Guillermo, > > On 29 Jan 2012, at 21:11, Guillermo Polito wrote: > > > I've spent a few hours playing with Alien+FFI and I did want to have a > Command line interface to play and use as a quick browser. > > > > Gofer it > > squeaksource: 'CLInterface'; > > package: 'ConfigurationOfCLInterface'; > > load. > > > > ConfigurationOfCLInterface load. > > > > > > To play with it write a script, for example startCLI.st, with the > following: > > > > CLInterface start > > > > > > > > And then, from the command line > > > > CogVM -vm-display-null <image> startCLI.st > > > > > > What it does support? > > - executes smalltalk expressions > > - handles the errors and shows them into the stdout (so it does not > explode :P) > > - stores variables as a workspace > > Since I used standard C functions (fgets and printf only) this should > work on every platform from scratch (having a VM with FFI working, of > course)... > > > > The problems I've ran so far: > > - blocking FFI blocks the vm. I've not tried yet to use threads through > FFI (and I don't know if it works. Does it?). But It should be great to > start a headless seaside server with a command line :/. > > > > Ok, it's a very simple thing, but it allows to execute a expression from > a remote server via ssh without doing much :). > > > > Guille > > Cool, but why do you need to access those C functions ? > You do know that you can access std* from Smalltalk, do you ? > No! :D > Have a look at the FileStream class 'stdio' category. > will do :). Will that solve the vm-block issue? > > Apart from that, nice small, readable and useful hack. > > Thanks :P. > Regards, > > Sven > > >

Jan. 30, 2012

Re: [Pharo-users] Command Line Interface using FFI

by Sven Van Caekenberghe

Hi Guillermo, On 29 Jan 2012, at 21:11, Guillermo Polito wrote: > I've spent a few hours playing with Alien+FFI and I did want to have a Command line interface to play and use as a quick browser. > > Gofer it > squeaksource: 'CLInterface'; > package: 'ConfigurationOfCLInterface'; > load. > > ConfigurationOfCLInterface load. > > > To play with it write a script, for example startCLI.st, with the following: > > CLInterface start > > > > And then, from the command line > > CogVM -vm-display-null <image> startCLI.st > > > What it does support? > - executes smalltalk expressions > - handles the errors and shows them into the stdout (so it does not explode :P) > - stores variables as a workspace > Since I used standard C functions (fgets and printf only) this should work on every platform from scratch (having a VM with FFI working, of course)... > > The problems I've ran so far: > - blocking FFI blocks the vm. I've not tried yet to use threads through FFI (and I don't know if it works. Does it?). But It should be great to start a headless seaside server with a command line :/. > > Ok, it's a very simple thing, but it allows to execute a expression from a remote server via ssh without doing much :). > > Guille Cool, but why do you need to access those C functions ? You do know that you can access std* from Smalltalk, do you ? Have a look at the FileStream class 'stdio' category. Apart from that, nice small, readable and useful hack. Regards, Sven

Jan. 29, 2012

Command Line Interface using FFI

by Guillermo Polito

I've spent a few hours playing with Alien+FFI and I did want to have a Command line interface to play and use as a quick browser. Gofer it squeaksource: 'CLInterface'; package: 'ConfigurationOfCLInterface'; load. ConfigurationOfCLInterface load. To play with it write a script, for example startCLI.st, with the following: *CLInterface start* And then, from the command line CogVM -vm-display-null <image> startCLI.st What it does support? - executes smalltalk expressions - handles the errors and shows them into the stdout (so it does not explode :P) - stores variables as a workspace Since I used standard C functions (fgets and printf only) this should work on every platform from scratch (having a VM with FFI working, of course)... The problems I've ran so far: - blocking FFI blocks the vm. I've not tried yet to use threads through FFI (and I don't know if it works. Does it?). But It should be great to start a headless seaside server with a command line :/. Ok, it's a very simple thing, but it allows to execute a expression from a remote server via ssh without doing much :). Guille

Jan. 29, 2012

Re: [Pharo-users] Enhanced Scamper or Real Browser

by StÃ©phane Ducasse

Hi john long time ago in the microsoft paul allen interval lab, jon hyland developed a browser all in squeak. Now some people enhanced scamper but I do not have the idea of its state. Let us know if you learn something. Stef > Hello Everyone, > > Just wondering if someone had either enhanced Scamper or developed a real web browser that compares with Firefox in browsing and rendering of web pages... don't really care about Javascript support. > > Please let me know. > > Thanks, > > John. > > -- > John Gardner, CMC > Certified Management Consultant and Certified Enterprise Architect > > JVG Consulting Inc. > Management Consulting and Enterprise Architecture > 905-933-1372

Jan. 28, 2012

Enhanced Scamper or Real Browser

by John Gardner

Hello Everyone, Just wondering if someone had either enhanced Scamper or developed a real web browser that compares with Firefox in browsing and rendering of web pages... don't really care about Javascript support. Please let me know. Thanks, John. -- *John Gardner, CMC* Certified Management Consultant and Certified Enterprise Architect *JVG Consulting Inc.* Management Consulting and Enterprise Architecture 905-933-1372

Jan. 28, 2012

Fwd: [Pharo-project] Pharo sprint in Brussels on Feb 4th

by Mariano Martinez Peck

---------- Forwarded message ---------- From: Andy Kellens <akellens(a)vub.ac.be> Date: Fri, Jan 27, 2012 at 11:35 AM Subject: [Pharo-project] Pharo sprint in Brussels on Feb 4th To: Pharo-project(a)lists.gforge.inria.fr Dear Pharoers, In an earlier mail Stephan Eggermont already mentioned that there will be Pharo sprint collocated with the FOSDEM conference next week. This sprint will take place on Saturday February 4th from 10h until 18h in the rooms of the Software Languages Lab of the Vrije Universiteit Brussel. More precisely: Room 10F720 (10th floor of building F) Vrije Universiteit Brussel Pleinlaan 2 1050 Brussels Due to renovations to our building, you will have to enter via the entrance between buildings F and G. As the university is normally closed on a Saturday, you might have to pass via security in order to be allowed entrance to the building (they have been notified about the sprint). If you are coming by car, please send me your license plate so that I can make sure that you are allowed to enter the campus. To get some idea of how many people to expect, could you please let us know if you are planning to attend. Kind regards, Andy -- Mariano http://marianopeck.wordpress.com

Jan. 27, 2012

Primitive failed when compiling Vm on Windows

by Guillermo Polito

Hi! I've downloaded version 3.8 of ConfigurationOfCog on Pharo 1.4, and when sending #topDir: to a CogWindowsConfig instance, I get a primitive failed error: PrimitiveFailed: primitive #primCreateDirectory: in DosFileDirectory failed 25 January 2012 7:24:30 pm VM: Win32 - IX86 - 5.1 - CoInterpreter VMMaker.oscog-eem.140 uuid: 2487009c-2c13-4609-b89e-77f0e195f433 Dec 12 2011, StackToRegisterMappingCogit VMMaker.oscog-eem.139 uuid: c2849383-9768-4948-b9b2-a5c22d482b07 Dec 12 2011, r2522 http://www.squeakvm.org/svn/squeak/branches/Cog Image: Pharo1.4a [Latest update: #14296] DosFileDirectory(Object)>>primitiveFailed: Receiver: DosFileDirectory on '' Arguments and temporary variables: selector: #primCreateDirectory: Receiver's instance variables: pathName: FilePath('') DosFileDirectory(Object)>>primitiveFailed Receiver: DosFileDirectory on '' Arguments and temporary variables: Receiver's instance variables: pathName: FilePath('') DosFileDirectory(FileDirectory)>>primCreateDirectory: Receiver: DosFileDirectory on '' Arguments and temporary variables: fullPath: '\\MinGW' Receiver's instance variables: pathName: FilePath('') DosFileDirectory(FileDirectory)>>createDirectory: Receiver: DosFileDirectory on '' Arguments and temporary variables: localFileName: '\\MinGW' Receiver's instance variables: pathName: FilePath('') DosFileDirectory(FileDirectory)>>assureExistenceOfPath: Receiver: DosFileDirectory on '' Arguments and temporary variables: lPath: '\\MinGW' localPath: '\\MinGW' Receiver's instance variables: pathName: FilePath('') DosFileDirectory(FileDirectory)>>assureExistenceOfPath: Receiver: DosFileDirectory on '\\MinGW' Arguments and temporary variables: lPath: 'msys' localPath: 'msys' Receiver's instance variables: pathName: FilePath('\\MinGW') DosFileDirectory(FileDirectory)>>assureExistenceOfPath: Receiver: DosFileDirectory on '\\MinGW\msys' Arguments and temporary variables: lPath: '1.0' localPath: '1.0' Receiver's instance variables: pathName: FilePath('\\MinGW\msys') DosFileDirectory(FileDirectory)>>assureExistenceOfPath: Receiver: DosFileDirectory on '\\MinGW\msys\1.0' Arguments and temporary variables: lPath: 'home' localPath: 'home' Receiver's instance variables: pathName: FilePath('\\MinGW\msys\1.0') DosFileDirectory(FileDirectory)>>assureExistenceOfPath: Receiver: DosFileDirectory on '\\MinGW\msys\1.0\home' Arguments and temporary variables: lPath: 'Guille' localPath: 'Guille' Receiver's instance variables: pathName: FilePath('\\MinGW\msys\1.0\home') DosFileDirectory(FileDirectory)>>assureExistenceOfPath: Receiver: DosFileDirectory on '\\MinGW\msys\1.0\home\Guille' Arguments and temporary variables: lPath: 'cog' localPath: 'cog' Receiver's instance variables: pathName: FilePath('\\MinGW\msys\1.0\home\Guille') DosFileDirectory(FileDirectory)>>assureExistenceOfPath: Receiver: DosFileDirectory on '\\MinGW\msys\1.0\home\Guille\cog' I followed the error up to: DosFileDirectory class>>privateFullPathForURI: aURI | path | path := aURI path unescapePercents. "Check for drive notation (a: etc)" path size > 1 ifTrue: [ ((path at: 3) = $:) ifTrue: [path := path copyFrom: 2 to: path size] ifFalse: [ "All other cases should be network path names (\\xxx\sdsd etc)" path := '/' , path]]. ^path copyReplaceAll: '/' with: self slash Which seems not being resolving well a full path like 'c:\something\another\directory' and is answering something like '\\MinGW\msys\1.0\home\Guille\cog\gpolito-cogvm' which then is treated as a relative path... I've done a temporary fix like this in my image so I can continue, but maybe someone has a little more knowledge on this class: DosFileDirectory class>>privateFullPathForURI: aURI | path | path := aURI asString unescapePercents. "Check for drive notation (a: etc)" path size > 1 ifTrue: [ ((path at: 2) = $:) ifTrue: [path := path copyFrom: 1 to: path size] ifFalse: [ "All other cases should be network path names (\\xxx\sdsd etc)" path := '/' , path]]. ^path copyReplaceAll: '/' with: self slash Guille

Jan. 25, 2012