July 2016 - Pharo-users - lists.pharo.org

Re: [Pharo-users] GTSpotter froze my image again

by Offray Vladimir Luna CÃ¡rdenas

Hi, My image becomes slow when I change my Internet connection and use spotter, for example when I suspend my laptop leaving my image open and change from home to the hackerspace or viceversa. In that case, the best way to get responsiveness back is to reopen the image. May be you can try to check if there is some network issue involved. Cheers, Offray On 02/07/16 17:50, Bernardo Ezequiel Contreras wrote: > Hi, > > it's the second time that my image freeze when i open gt spotter, > pressing Alt + . doesn't interrupt it. is there something that you > can do to interrupt it? > > im on debian wheezy using > Pharo5.0 > Latest update: #50760 > > thanks. > > > -- > Bernardo E.C. > > Sent from a cheap desktop computer in South America.

July 3, 2016

GTSpotter froze my image again

by Bernardo Ezequiel Contreras

Hi, it's the second time that my image freeze when i open gt spotter, pressing Alt + . doesn't interrupt it. is there something that you can do to interrupt it? im on debian wheezy using Pharo5.0 Latest update: #50760 thanks. -- Bernardo E.C. Sent from a cheap desktop computer in South America.

July 2, 2016

Recovering PNG/JPG images that were automatically serialized as FUEL when storing into external database

by Offray Vladimir Luna CÃ¡rdenas

Hi, I was storing Pharo objects into an external SQLite3 database. It went fine, but now all my png/jpg images are serialized with FUEL, when I try to see them inside SQLite and when I import them back to the image, they're long byte streams. There is any way to recover the images as png/jpg morphs that visible inside the image again? How can I disable this automatic serialization for making the database portable without using FUEL to recover back the PNG/JPG images? Thanks, Offray

July 2, 2016

Re: [Pharo-users] Styling a Pharo UI application

by Johan Fabry

Hi Tommaso, thanks for the explanation! I did not mean to point fingers at you for complaining, I was just trying to figure out what the problem was. So thanks again for sharing. This is indeed not yet possible, and it would be cool to have it working. However there is a workaround since you can take any Morph and embed it in a Spec UI. For example: initializeWidgets | â¦ textMorph â¦ | [â¦ do a lot of other stuff â¦] textMorph := LabelMorph newLabel:âasdfâ. textMorph color: Color red. text := textMorph asSpecAdapter. Voila, a label with red text. -- Does this mail seem too brief? Sorry for that, I donât mean to be rude! Please see http://emailcharter.org <http://emailcharter.org/> . Johan Fabry - http://pleiad.cl/~jfabry <http://pleiad.cl/~jfabry> PLEIAD and RyCh labs - Computer Science Department (DCC) - University of Chile > On Jun 30, 2016, at 07:56, Tommaso Dal Sasso <tommaso.dalsasso(a)gmail.com <mailto:tommaso.dalsasso@gmail.com>> wrote: > > I was not complaining about Spec, just reporting my experience with the various UI toolkits in Pharo as a relatively newbie to this area of Pharo. > > Anyway, as I wrote in my first email, I had difficulties with Spec to go beyond building a standard interface, especially in performing text manipulation (font size) or coloring the background of a string, or encapsulating pieces of text in frames. I tried to understand how to do that, but there is very little documentation on the matter, and the main answer I got was "be prepared to use Morphic". > > Trying to understand how to access the properties of the elements I found several comments like: > > [talking about font size in Spec] > On 03/10/15 09:24, stepharo wrote: > > I think that this part has been completely forgotten by spec. > > Normally the layout should take into account the size of the label and font > > and I do not see why we could not specify the color and other attributes we want to have but we did not work on this point. > > Now we will have to see that in the context of Brick skin. > > > > Stef > > Don't get me wrong, I like Spec, it is just unclear to me if it wants to be the default toolkit to build widgets in Pharo (in this case it should probably access more properties of Morphic), or if wants to be completely engine-agnostic, therefore preferring abstraction over expressiveness. > > Tommaso > >

July 2, 2016

Re: [Pharo-users] Provide a mapping for a collection of <Association>s in NeoJSON

by Sven Van Caekenberghe

> On 01 Jul 2016, at 00:56, Sven Van Caekenberghe <sven(a)stfx.eu> wrote: > > However, I also think that it might be possible to solve the ordering requirement with a regular mapping, provided I make a little internal change - I will investigate that and report back. I committed the following (#bleedingEdge): === Name: Neo-JSON-Core-SvenVanCaekenberghe.34 Author: SvenVanCaekenberghe Time: 2 July 2016, 12:03:08.919979 am UUID: ded70581-c06f-4687-936c-11a029a39ba4 Ancestors: Neo-JSON-Core-SvenVanCaekenberghe.33 Use an OrderedDictionary instead of a regular Dictionary to store the properties inside a NeoJSONObjectMapping - this allows the user to control the order of properties while writing, if needed Add NeoJSONWriterTests>>#testPreservePropertyOrder === Name: Neo-JSON-Tests-SvenVanCaekenberghe.34 Author: SvenVanCaekenberghe Time: 2 July 2016, 12:03:30.417248 am UUID: 883b6ef1-e5fc-4d85-b86c-b21c1b5a3d6c Ancestors: Neo-JSON-Tests-SvenVanCaekenberghe.33 Use an OrderedDictionary instead of a regular Dictionary to store the properties inside a NeoJSONObjectMapping - this allows the user to control the order of properties while writing, if needed Add NeoJSONWriterTests>>#testPreservePropertyOrder === This allows for the following: NeoJSONWriterTests>>#testPreservePropertyOrder | writer testObject | (testObject := NeoJSONTestObject2 new) id: 123; width: 100; height: 50; data: 'test'. writer := [ :object | String streamContents: [ :stream | (NeoJSONWriter on: stream) for: NeoJSONTestObject2 do: [ :mapping | mapping mapInstVars: #(id width height data) ]; nextPut: object ] ]. self assert: (writer value: testObject) equals: '{"id":123,"width":100,"height":50,"data":"test"}'. For any permutation of the list of instance variables, each resulting in differently ordered JSON. This applies to all mapping techniques. Sven

July 1, 2016

Re: [Pharo-users] [ANN]Windows previewer / switcher for Pharo

by Merwan Ouddane

On 29/06/2016 23:42, Nicolai Hess wrote: > Hi Julien, > > yes, finding a shortcut satisfying all users (for all different > platforms) is not easy, see > 15546 > <https://pharo.fogbugz.com/f/cases/15546/alt-tab-does-not-work-correctly> > alt+tab does not work correctly > and > 17924 > <https://pharo.fogbugz.com/f/cases/17924/Switch-Windows-shortcut> > Switch Windows shortcut > > > 2016-06-29 0:31 GMT+02:00 Merwan Ouddane <merwanouddane(a)gmail.com > <mailto:merwanouddane@gmail.com>>: > > Argh i was so thrilled by this ! But it doesn't seem to work on > windows :/ > > Ctrl+tab gives the focus to the front window or makes an "inspect > it" on the current editor line... > > > For windows, to make ctrl+tab working, you need to disable the > "hasSpecialCTRLKeyValue"-check (I think this check is not needed > anymore, since we changed the > windows vm for ctrl+number/tab/space... keystrokes) Yes it works, thanks Nicolai. Is there a case where this method is still relevant ? I noticed some other weird behaviours in the events from the keyboard. ctrl + 'arrow up' and down are sending two keydown events instead of a keydown and a keystroke like the left and right arrow are sending. > > Thank you anyway ! > > > On 28/06/2016 14:51, Julien Delplanque wrote: > > Also, here is a screenshot (I forgot to attach it to the > preceding mail). > > Julien > > > On 28/06/16 11:06, Julien Delplanque wrote: > > Hello, > > This mail to announce you that I am working on a windows > previewer / switcher for > Pharo [1]. > > It is still in early stage of development but it is usable. > > To install it: > > |Metacello new repository: > 'github://JulienDelplanque/WindowsPreviewer/repository'; > baseline: 'WindowsPreviewer'; load| > > See the README for details on activation/shortcuts. > > Please, if you have some ideas or critics let me know! > > Also, a better name than "Windows Previewer" would be cool > but I have > no other idea :-)... > > Regards, > > Julien > > Links: > [1]: https://github.com/juliendelplanque/WindowsPreviewer > > > > >

July 1, 2016

Re: [Pharo-users] Provide a mapping for a collection of <Association>s in NeoJSON

by Offray Vladimir Luna CÃ¡rdenas

Hi, I use both, NeoJSON and STON and I agreed with Sven here. Having both separated give different entry points for different developers. Data oriented ones would start probably with NeoJSON while devs looking for more general data serialization language for graphs, documents and so on will start with STON. Cheers, Offray On 01/07/16 04:54, Sven Van Caekenberghe wrote: >> On 01 Jul 2016, at 11:27, Esteban Lorenzano <estebanlm(a)gmail.com> wrote: >> >> btw, why NeoJSON is not deprecated in benefit of STON? > Although there is obviously a relation between the two (STON is inspired by JSON, STON has some basic backward compatibility with JSON, same project style, same author), there are also differences. > > JSON is by definition much simpler and much more limited (no type info for objects, no complex structures like shared and circular references) while STON was designed specifically to cover those. > > JSON needs much more explicit mapping support to deal with anything except for simple maps and lists. > > In terms of documentation, the API, it would confuse the hell out of regular JSON users to see things mixed up with STON which they are probably not interested in at all. I think this last point might even be the most important one. > >> Esteban >> >>> On 01 Jul 2016, at 08:52, stepharo <stepharo(a)free.fr> wrote: >>> >>> Thanks sven >>> >>> I was thinking to add a section in the chapter but I will wait for your solution. >>> >>> >>> stef >>> >>> >>> Le 1/7/16 Ã 00:56, Sven Van Caekenberghe a Ã©crit : >>>> Hi, >>>> >>>> Yes, object properties in JSON are not ordered, by definition. But for humans it can make sense. >>>> Converting your domain objects to OrderedDictionaries or overwriting #neoJSONOn: are valid ways to control NeoJSON. >>>> >>>> However, I also think that it might be possible to solve the ordering requirement with a regular mapping, provided I make a little internal change - I will investigate that and report back. >>>> >>>> Sven >>>> >>>>> On 30 Jun 2016, at 17:29, PBKResearch <peter(a)pbkresearch.co.uk> wrote: >>>>> >>>>> Bernardo >>>>> Thanks for this. After sending my previous mail, I did what I should have done first and looked at list of collection objects in the system browser. So I discovered OrderedDictionary for the first time. I am pleased to see that this is the key to the solution. >>>>> Peter >>>>> From: Pharo-users [mailto:pharo-users-bounces@lists.pharo.org] On Behalf Of Bernardo Ezequiel Contreras >>>>> Sent: 30 June 2016 16:17 >>>>> To: Any question about pharo is welcome <pharo-users(a)lists.pharo.org> >>>>> Subject: Re: [Pharo-users] Provide a mapping for a collection of <Association>s in NeoJSON >>>>> Peter, >>>>> i had the same issue, i couldn't find the way to do it, but i hack my own writing like this >>>>> NDBFood>>neoJsonOn: neoJSONWriter >>>>> "custom writing" >>>>> neoJSONWriter >>>>> writeMap: >>>>> (OrderedDictionary new >>>>> add: 'ndb_no' -> ndbno; >>>>> add: 'full_name' -> name; >>>>> add: 'food_group' -> fg; >>>>> add: 'factors' -> self factors; >>>>> add: 'nutrients' -> self nutrients; >>>>> yourself) >>>>> i just implement my own writing for my objects. >>>>> HTH >>>>> On Thu, Jun 30, 2016 at 12:06 PM, PBKResearch <peter(a)pbkresearch.co.uk> wrote: >>>>>> Hello >>>>>> I have recently used NeoJSON for the first time. All went pretty well, but I would like to make a small tweak to the output, which probably involves defining a mapping. I have read through the information in the NeoJSON pdf, and also looked at the class comments, but I am a bit lost. Could Sven or anyone please give me a hint? >>>>>> What I want is to construct a JSON object as a Pharo dictionary, but then control the order in which the fields are output to the JSON file. Clearly the order is irrelevant for a computer, but it could be helpful to a human reader to put them in an order which reflects the way humans think about the objects. The keys of a dictionary are of course unordered, so this is not possible, but if we step back a bit we see that a dictionary is often constructed from a collection of associations; if we could stop at this stage, and have a mapping which outputs the associations in the order in which they appear in the collection, this would give the control I am looking for. >>>>>> Am I talking nonsense, or is there any way of implementing this using NeoJSON mappings? >>>>>> Best wishes >>>>>> Peter Kenny >>>>> -- >>>>> Bernardo E.C. >>>>> Sent from a cheap desktop computer in South America. >>>> >>> >> > >

July 1, 2016

Re: [Pharo-users] Styling a Pharo UI application

by Dimitris Chloupis

No Spec is a helper API to help you make GUIs quickly, its not a Morphic replacement. Bloc is the Morphic replacement and it has something similar to Spec called Brick. The GUI situation in pharo is not good, Pharo forked Morphic from Squeak, a well designed API that its design was not followed as it extended and it turned ugly in the process and none actually bother documenting it other than small fragments here and there. Bloc tries to clean up that design at least the Bloc I tried a year ago, I hear now that a lot of thing have changed. Personally I make my own GUI API that fits my needs, I take what Morphic offers and I provide a design that is far closer to the way I think. Unfortunately right now GUI is in a state of flux in Pharo but that is actually a good thing since it definitely needs improvement. Morphic is very useful to learn because right now is the only GUI API that works well for Pharo. Bloc is not ready for public consumption and Spec was never meant for heavily customisable and customised GUIs. Having to read Morphic code to understand it is definetly painful. For example for my customised ChronosManager I decided to provide Taskbar functionality. So I try to copy the taskbar functionality that is implemented in SystemWindow and adjusted to my needs, it was a nightmare. I manage to freeze my image several time and it took a ton of trial and error to figure out how it worked and it still does not behave exactly as I wanted. I am still considering to move my entire GUI to Html/JS/CSS , its ugly and fragmented to work with these technologies but at least they are well documented and very powerful. This is a reason why I designed a way to use Python libraries from Pharo. Python libraries are beautifully designed, well documented and very mature. So you have plenty of options and there is no reason to limit yourself on a pure pharo solution. Modern applications are written in multiple languages and frameworks. You just use the right tool for the job. On Thu, Jun 30, 2016 at 2:57 PM Tommaso Dal Sasso < tommaso.dalsasso(a)gmail.com> wrote: > I was not complaining about Spec, just reporting my experience with the > various UI toolkits in Pharo as a relatively newbie to this area of Pharo. > > Anyway, as I wrote in my first email, I had difficulties with Spec to go > beyond building a standard interface, especially in performing text > manipulation (font size) or coloring the background of a string, or > encapsulating pieces of text in frames. I tried to understand how to do > that, but there is very little documentation on the matter, and the main > answer I got was "be prepared to use Morphic". > > Trying to understand how to access the properties of the elements I found > several comments like: > > [talking about font size in Spec] > On 03/10/15 09:24, stepharo wrote: > > I think that this part has been completely forgotten by spec. > > Normally the layout should take into account the size of the label and > font > > and I do not see why we could not specify the color and other attributes > we want to have but we did not work on this point. > > Now we will have to see that in the context of Brick skin. > > > > Stef > > Don't get me wrong, I like Spec, it is just unclear to me if it wants to > be the default toolkit to build widgets in Pharo (in this case it should > probably access more properties of Morphic), or if wants to be completely > engine-agnostic, therefore preferring abstraction over expressiveness. > > > > Tommaso > > > > On 26/06/16 23:39, Offray Vladimir Luna CÃ¡rdenas wrote: > > +1 :) > > Offray > > On 26/06/16 02:56, stepharo wrote: > > + 1 > > Stef > > Tommaso, > > can you give some concrete examples of what it is that you wanted to do to > refine your application and that was really hard (or impossible) to do with > Spec? > > As long as we do not know what is wrong, we cannot fix it. > > As long as we do not know what important things are wrong, we cannot > prioritize. > > -- > Does this mail seem too brief? Sorry for that, I donât mean to be rude! > Please see http://emailcharter.org . > > Johan Fabry - http://pleiad.cl/~jfabry > PLEIAD and RyCh labs - Computer Science Department (DCC) - University > of Chile > > On Jun 25, 2016, at 15:54, Tommaso Dal Sasso <tommaso.dalsasso(a)gmail.com> > wrote: > > The beauty that I am looking for as a developer for has to do with the > code and the API. I consider important how fast I am able to prototype a > working example --in this Spec is excellent-- and how much I can extend the > code to refine my application --in this Spec is really hard to use--. > > > > > >

July 1, 2016

Re: [Pharo-users] [UFFI] Getting an array of opaque structures filled

by Julien Delplanque

And it works! Thanks a lot! Yes I should join slack, do I need an invitation or something? Julien On 01/07/16 14:03, Esteban Lorenzano wrote: > ok, can you reload development version and try again? > > ps: this is easier on slack ;) > >> On 01 Jul 2016, at 14:00, Esteban Lorenzano <estebanlm(a)gmail.com> wrote: >> >> yes, I made a mistakeâ¦ the problem of not being capable of test here :( >> >> 1 minute. >> >> Esteban >> >>> On 01 Jul 2016, at 13:56, Julien Delplanque <julien(a)tamere.eu> wrote: >>> >>> Ok, now I get a different exception! :-) >>> >>> "MessageNotUnderstood: LUDevice class>>fromHandle:" >>> >>> when executing: "array first." >>> >>> Where LUDevice is defined like this: >>> >>> FFIOpaqueObject subclass: #LUDevice >>> instanceVariableNames: '' >>> classVariableNames: '' >>> package: 'LibUsb-Kernel' >>> >>> Julien >>> >>> >>> On 01/07/16 13:51, Esteban Lorenzano wrote: >>>> I need you to test :) >>>> >>>> can you execute this: >>>> >>>> (ConfigurationOfUnifiedFFI project version: #development) load. >>>> >>>> and then retry your example? >>>> >>>> thanks, >>>> Esteban >>>> >>>> ps: this became pharo-dev, but wellâ¦ still here :) >>>> >>>>> On 01 Jul 2016, at 13:48, Julien Delplanque <julien(a)tamere.eu> wrote: >>>>> >>>>> Oh, ok :p >>>>> >>>>> Can you notify me know when the bug will be resolved? >>>>> >>>>> Thanks, >>>>> >>>>> Julien >>>>> >>>>> >>>>> On 01/07/16 13:36, Esteban Lorenzano wrote: >>>>>> â¦ and now you hit a bug. >>>>>> bah, an ânon implemented featureâ :P >>>>>> >>>>>> I never tested arrays with complex types. >>>>>> it should not be hardâ¦ I will provide a fix. >>>>>> >>>>>> Esteban >>>>>> >>>>>> >>>>>>> On 01 Jul 2016, at 13:30, Julien Delplanque <julien(a)tamere.eu> wrote: >>>>>>> >>>>>>> Ok, the ffi call is done without problem with your method thanks. :) >>>>>>> >>>>>>> But now that I have the array, when I try to access the first element using: >>>>>>> >>>>>>> array first. >>>>>>> >>>>>>> I get a exception saying I need to override a method: >>>>>>> >>>>>>> "SubclassResponsibility: FFIOpaqueObjectType had the subclass responsibility to implement #basicHandle:at:" >>>>>>> >>>>>>> I looked at others implementors but I am not sure of how I should override it, nor if I should. >>>>>>> >>>>>>> Thanks for the time your taking to answer me. :-) >>>>>>> >>>>>>> Julien >>>>>>> >>>>>>> On 01/07/16 12:29, Esteban Lorenzano wrote: >>>>>>>>> On 01 Jul 2016, at 12:27, Esteban Lorenzano <estebanlm(a)gmail.com> wrote: >>>>>>>>> >>>>>>>>> hi, >>>>>>>>> >>>>>>>>> this is because you want to get an array with opaque objects. Do not let the *** confuses youâ¦ cwhat you actually want, conceptually is something like this: >>>>>>>>> >>>>>>>>> *((FFIOpaqueObject*)[]) >>>>>>>>> >>>>>>>>> means: you pass the address of an array of opaque types (who are always pointers). >>>>>>>>> >>>>>>>>> So, what you need to pass is the address of an arrayâ¦ you will need to work a bit here, since translation is not automatic. >>>>>>>>> >>>>>>>>> 1) you need to declare your type. Letâ say it will be: >>>>>>>>> >>>>>>>>> FFIOpaqueObject subclass: USBDevice. >>>>>>>>> >>>>>>>>> 2) then you need to get the pointer to this array. You need to rewrite you code as this: >>>>>>>>> >>>>>>>>> arrayPtr := ExternalAddress new. >>>>>>>>> size := self getDevice: ctx list: arrayPtr. >>>>>>>>> >>>>>>>>> getDevice: ctx list: list >>>>>>>>> ^ self ffiCall: #(size_t libusb_get_device_list (libusb_context *ctx, void **list)) >>>>>>>>> >>>>>>>>> NOTE that we change the type to "void **â. This is because this is what you actually want: the array of devices >>>>>>>>> >>>>>>>>> 3) now you will have size and arrayPtr. Then you declare the array: >>>>>>>>> >>>>>>>>> array := FFIExternalArray fromPointer: arrayPtr type: USBDevice >>>>>>>> sorry, this expression (while possible) does not stops in âsizeââ¦ it has to be like this, instead: >>>>>>>> >>>>>>>> array := FFIExternalArray fromPointer: arrayPtr type: USBDevice size: size. âyou already has the size for the result" >>>>>>>> >>>>>>>>> â¦ and you should be able to iterate this array normally :) >>>>>>>>> >>>>>>>>> let me know if this works for youâ¦ Iâm âcoding in mail clientâ, so it can fail :P >>>>>>>>> >>>>>>>>> cheers, >>>>>>>>> Esteban >>>>>>>>> >>>>>>>>> >>>>>>>>>> On 01 Jul 2016, at 12:02, Julien Delplanque <julien(a)tamere.eu> wrote: >>>>>>>>>> >>>>>>>>>> Thanks, it works I get an instance of FFIOpaqueObject. >>>>>>>>>> >>>>>>>>>> But how do I get a specific libusb_device object from the list? >>>>>>>>>> >>>>>>>>>> Thanks again for you quick answer :-). >>>>>>>>>> >>>>>>>>>> Julien >>>>>>>>>> >>>>>>>>>> On 01/07/16 11:44, Esteban Lorenzano wrote: >>>>>>>>>>> Hi, >>>>>>>>>>> >>>>>>>>>>> an opaque structure is not an FFIExternalObject but an FFIOpaqueObject. >>>>>>>>>>> in that case, something like (simplified): >>>>>>>>>>> >>>>>>>>>>> ctx := (your context). >>>>>>>>>>> list := FFIOpaqueObject new. >>>>>>>>>>> size := self getDevice: ctx list: list. >>>>>>>>>>> >>>>>>>>>>> getDevice: ctx list: list >>>>>>>>>>> ^ self ffiCall: #(size_t libusb_get_device_list (libusb_context *ctx, FFIOpaqueObject ***list)) >>>>>>>>>>> >>>>>>>>>>> should work (note that of course you can make libusb_device an alias for your typeâ¦ that does not matters much, this is just a simplification). >>>>>>>>>>> >>>>>>>>>>> if this does not works, if you can provide me code to reproduce it, I can give it a shot and see what happens :) >>>>>>>>>>> >>>>>>>>>>> Esteban >>>>>>>>>>> >>>>>>>>>>>> On 01 Jul 2016, at 11:19, Julien Delplanque <julien(a)tamere.eu> wrote: >>>>>>>>>>>> >>>>>>>>>>>> Hello everyone, >>>>>>>>>>>> >>>>>>>>>>>> I have a question about the UFFI API. I have the following function: >>>>>>>>>>>> >>>>>>>>>>>> ssize_t libusb_get_device_list (libusb_context *ctx, libusb_device ***list) >>>>>>>>>>>> >>>>>>>>>>>> where libusb_device is an opaque structure. >>>>>>>>>>>> >>>>>>>>>>>> I made a type mapping "ssize_t" -> "int" and an object inheriting >>>>>>>>>>>> from FFIExternalObject for the "libusb_context" (others methods >>>>>>>>>>>> using it are working). >>>>>>>>>>>> >>>>>>>>>>>> I can not find how to get an array filled with libusb_device using UFFI from Pharo. :-( >>>>>>>>>>>> >>>>>>>>>>>> Thanks in advance, >>>>>>>>>>>> >>>>>>>>>>>> Julien >>>>>>>>>>>> >>> >

July 1, 2016

Re: [Pharo-users] [UFFI] Getting an array of opaque structures filled

by Julien Delplanque

No problem, give me as much version(s) to test as you want. :p Julien On 01/07/16 14:00, Esteban Lorenzano wrote: > yes, I made a mistakeâ¦ the problem of not being capable of test here :( > > 1 minute. > > Esteban > >> On 01 Jul 2016, at 13:56, Julien Delplanque <julien(a)tamere.eu> wrote: >> >> Ok, now I get a different exception! :-) >> >> "MessageNotUnderstood: LUDevice class>>fromHandle:" >> >> when executing: "array first." >> >> Where LUDevice is defined like this: >> >> FFIOpaqueObject subclass: #LUDevice >> instanceVariableNames: '' >> classVariableNames: '' >> package: 'LibUsb-Kernel' >> >> Julien >> >> >> On 01/07/16 13:51, Esteban Lorenzano wrote: >>> I need you to test :) >>> >>> can you execute this: >>> >>> (ConfigurationOfUnifiedFFI project version: #development) load. >>> >>> and then retry your example? >>> >>> thanks, >>> Esteban >>> >>> ps: this became pharo-dev, but wellâ¦ still here :) >>> >>>> On 01 Jul 2016, at 13:48, Julien Delplanque <julien(a)tamere.eu> wrote: >>>> >>>> Oh, ok :p >>>> >>>> Can you notify me know when the bug will be resolved? >>>> >>>> Thanks, >>>> >>>> Julien >>>> >>>> >>>> On 01/07/16 13:36, Esteban Lorenzano wrote: >>>>> â¦ and now you hit a bug. >>>>> bah, an ânon implemented featureâ :P >>>>> >>>>> I never tested arrays with complex types. >>>>> it should not be hardâ¦ I will provide a fix. >>>>> >>>>> Esteban >>>>> >>>>> >>>>>> On 01 Jul 2016, at 13:30, Julien Delplanque <julien(a)tamere.eu> wrote: >>>>>> >>>>>> Ok, the ffi call is done without problem with your method thanks. :) >>>>>> >>>>>> But now that I have the array, when I try to access the first element using: >>>>>> >>>>>> array first. >>>>>> >>>>>> I get a exception saying I need to override a method: >>>>>> >>>>>> "SubclassResponsibility: FFIOpaqueObjectType had the subclass responsibility to implement #basicHandle:at:" >>>>>> >>>>>> I looked at others implementors but I am not sure of how I should override it, nor if I should. >>>>>> >>>>>> Thanks for the time your taking to answer me. :-) >>>>>> >>>>>> Julien >>>>>> >>>>>> On 01/07/16 12:29, Esteban Lorenzano wrote: >>>>>>>> On 01 Jul 2016, at 12:27, Esteban Lorenzano <estebanlm(a)gmail.com> wrote: >>>>>>>> >>>>>>>> hi, >>>>>>>> >>>>>>>> this is because you want to get an array with opaque objects. Do not let the *** confuses youâ¦ cwhat you actually want, conceptually is something like this: >>>>>>>> >>>>>>>> *((FFIOpaqueObject*)[]) >>>>>>>> >>>>>>>> means: you pass the address of an array of opaque types (who are always pointers). >>>>>>>> >>>>>>>> So, what you need to pass is the address of an arrayâ¦ you will need to work a bit here, since translation is not automatic. >>>>>>>> >>>>>>>> 1) you need to declare your type. Letâ say it will be: >>>>>>>> >>>>>>>> FFIOpaqueObject subclass: USBDevice. >>>>>>>> >>>>>>>> 2) then you need to get the pointer to this array. You need to rewrite you code as this: >>>>>>>> >>>>>>>> arrayPtr := ExternalAddress new. >>>>>>>> size := self getDevice: ctx list: arrayPtr. >>>>>>>> >>>>>>>> getDevice: ctx list: list >>>>>>>> ^ self ffiCall: #(size_t libusb_get_device_list (libusb_context *ctx, void **list)) >>>>>>>> >>>>>>>> NOTE that we change the type to "void **â. This is because this is what you actually want: the array of devices >>>>>>>> >>>>>>>> 3) now you will have size and arrayPtr. Then you declare the array: >>>>>>>> >>>>>>>> array := FFIExternalArray fromPointer: arrayPtr type: USBDevice >>>>>>> sorry, this expression (while possible) does not stops in âsizeââ¦ it has to be like this, instead: >>>>>>> >>>>>>> array := FFIExternalArray fromPointer: arrayPtr type: USBDevice size: size. âyou already has the size for the result" >>>>>>> >>>>>>>> â¦ and you should be able to iterate this array normally :) >>>>>>>> >>>>>>>> let me know if this works for youâ¦ Iâm âcoding in mail clientâ, so it can fail :P >>>>>>>> >>>>>>>> cheers, >>>>>>>> Esteban >>>>>>>> >>>>>>>> >>>>>>>>> On 01 Jul 2016, at 12:02, Julien Delplanque <julien(a)tamere.eu> wrote: >>>>>>>>> >>>>>>>>> Thanks, it works I get an instance of FFIOpaqueObject. >>>>>>>>> >>>>>>>>> But how do I get a specific libusb_device object from the list? >>>>>>>>> >>>>>>>>> Thanks again for you quick answer :-). >>>>>>>>> >>>>>>>>> Julien >>>>>>>>> >>>>>>>>> On 01/07/16 11:44, Esteban Lorenzano wrote: >>>>>>>>>> Hi, >>>>>>>>>> >>>>>>>>>> an opaque structure is not an FFIExternalObject but an FFIOpaqueObject. >>>>>>>>>> in that case, something like (simplified): >>>>>>>>>> >>>>>>>>>> ctx := (your context). >>>>>>>>>> list := FFIOpaqueObject new. >>>>>>>>>> size := self getDevice: ctx list: list. >>>>>>>>>> >>>>>>>>>> getDevice: ctx list: list >>>>>>>>>> ^ self ffiCall: #(size_t libusb_get_device_list (libusb_context *ctx, FFIOpaqueObject ***list)) >>>>>>>>>> >>>>>>>>>> should work (note that of course you can make libusb_device an alias for your typeâ¦ that does not matters much, this is just a simplification). >>>>>>>>>> >>>>>>>>>> if this does not works, if you can provide me code to reproduce it, I can give it a shot and see what happens :) >>>>>>>>>> >>>>>>>>>> Esteban >>>>>>>>>> >>>>>>>>>>> On 01 Jul 2016, at 11:19, Julien Delplanque <julien(a)tamere.eu> wrote: >>>>>>>>>>> >>>>>>>>>>> Hello everyone, >>>>>>>>>>> >>>>>>>>>>> I have a question about the UFFI API. I have the following function: >>>>>>>>>>> >>>>>>>>>>> ssize_t libusb_get_device_list (libusb_context *ctx, libusb_device ***list) >>>>>>>>>>> >>>>>>>>>>> where libusb_device is an opaque structure. >>>>>>>>>>> >>>>>>>>>>> I made a type mapping "ssize_t" -> "int" and an object inheriting >>>>>>>>>>> from FFIExternalObject for the "libusb_context" (others methods >>>>>>>>>>> using it are working). >>>>>>>>>>> >>>>>>>>>>> I can not find how to get an array filled with libusb_device using UFFI from Pharo. :-( >>>>>>>>>>> >>>>>>>>>>> Thanks in advance, >>>>>>>>>>> >>>>>>>>>>> Julien >>>>>>>>>>> >> >

July 1, 2016