On Sat, May 26, 2018 at 8:07 AM, Denis Kudriashov
wrote:
> Hi
>
> 2018-05-26 8:46 GMT+03:00 Clément Bera :
>
>> Just mentioning another use-case:
>>
>> getDatabaseInstance
>> ^ (Production CifTrue: [Database] CifFalse: [MockDatabase]) new
>
n the image for this case right now but it's probably a couple
lines of code with Reflectivity.
On Sat, May 26, 2018 at 7:28 AM, Clément Bera
wrote:
>
>
> On Fri, May 25, 2018 at 10:44 PM, Esteban Lorenzano
> wrote:
>
>>
>>
>> On 25 May 2018, at 17:
On Fri, May 25, 2018 at 10:44 PM, Esteban Lorenzano
wrote:
>
>
> On 25 May 2018, at 17:30, Clément Bera wrote:
>
> What about a preprocessor like the Java preprocessors ? The Truffle
> project relies heavily on that for high performance Java and it's quite
> nice. It
What about a preprocessor like the Java preprocessors ? The Truffle project
relies heavily on that for high performance Java and it's quite nice. It's
difficult to do that in Smalltalk right now.
I think if you want to do what are asking for you just need to write a
bytecode compiler extension.
I
That is the answer I would give:
Biggest Smalltalk community in Silicon valley is likely the Lam Research
one (dozens of devs). On my blog post talking about Smalltalk, I have 2k
views per Smalltalk community-wide audience post, so the community is at
least that big. Smalltalk conferences (Smallta
The guy who asked the question said: "...when you want to shorten some
object initialization"
Using ClassVariable is an alternative way to shorten object initialization,
reading a ClassVariable is almost the same performance as reading a
literal, so that looked like a good alternative to me. Both
Not really.
You can use ClassVariables though. Here's an example for Sets:
Object subclass: #MyClass
instanceVariableNames: ''
classVariableNames: 'MyClassVar'
package: 'MyPackage'
MyClass class>>initialize
super initialize.
MyClassVar := Set new.
MyClass>>foo
self bar: MyClassVar
"Alpha software can be unstable and could cause crashes or data loss" [1]
Experiencing instability, crashes and data loss in the alpha version of a
software, including Pharo alpha, is to be expected and nothing to be
surprised about. The support of the Pharo stable version (currently 6.1)
has sign
Hi Hilaire,
Do you have to use the world or Morphic at all ?
I wrote 2 different real-time 2D games at 50fps in Pharo [1,2]. I tried
using Morphic but that was crazy slow I could not even get 20fps. Instead I
use direct bindings to Cairo surface and SDL2 window and I don't open the
world morph at
On Fri, Jun 2, 2017, 17:05 Steffen Märcker wrote:
> Dear all,
>
> thanks for the many suggestions. I didn't had time to test all
> import/export ways yet. But for now, I can report on two:
>
> 1) NGFileOuter
> Unfortunately It raised several MNUs in my image. I'll investigate them
> later.
>
> 2)
p.3
simpliciyty -> simplicity
p.4
we will have the define their semantics.
-> to define ? I am not sure I understand the sentence
p.10
and may converted on the fly
-> and maybe ? I am not sure I understand the sentence
At the beginning I started to read and understood nothing, then I read a 24
s
I think the roles got changed I am no longer admin (Note that it does not
really matter since I did not do much admin work but still)
On Sat, Mar 10, 2018 at 11:48 AM, Dimitris Chloupis
wrote:
> Hey guys after my mistake we got "hacked" and someone banned all our
> members in Discord
>
> I have
lve.
> Cheers,
>
> Hernán
>
> 2018-02-10 9:47 GMT-03:00 Clément Bera :
> > Hi,
> >
> > In short, everything that is not namespace/module compatible will be
> > deprecated/changed/removed in the future, so it is not recommended to use
> > it.
> >
o look up MyProject.Array, no?
>
> So if the argument for asClass is based on future addition of modules,
> then I will need to manually change it anyway regardless of what approach I
> used, because I don't know in which namespace the class will end up.
>
> Am I missing something?
Hi,
In short, everything that is not namespace/module compatible will be
deprecated/changed/removed in the future, so it is not recommended to use
it.
a.2) #Array asClassInEnvironment: Smalltalk globals
b.1) Smalltalk globals at: #Array
=> Ok-ish, note that you may need to change 'Smalltalk globa
Hi,
I am currently parsing lua and JSON-like file in Pharo. They contain both
Simplified and Traditional Chinese characters for comments and for strings
displayed in the UI. Lua files are parsed correctly. However the JSON-like
files aren't.
In attachment I put one of the problematic file with Si
I don't think we do. Do you need it on SequenceableCollection or
HashedCollection too ?
Recently I was trying to iterate over the first N elements of a collection
and since there was no #first:do: I used #from:to:do:. I guess you could
use that too:
aCollection from: 1 to: (aCollection size min:
There is a REPL Image we use for VM debugging. It's a REPL in the command
line and in the VM Simulator.
File-in in attachement to load and start the REPL.
On Tue, Nov 28, 2017 at 8:18 AM, Stephane Ducasse
wrote:
> Sorry I wanted to have it in pharo :)
>
> On Mon, Nov 27, 2017 at 3:56 AM, Holge
Hi,
You can start the VM with this parameter:
--memory [mk]use fixed heap size (added to image size)
When the memory is reached you have an out of memory error.
Check --help on the VM.
On Tue, Oct 24, 2017 at 10:15 PM, PAUL DEBRUICKER
wrote:
> I'm trying to get some understanding about
On Wed, Oct 18, 2017 at 7:07 AM, James Ladd wrote:
> Hey David,
>
> If you could only access instance variables in the class that defined them
> and therefore only in a subclass via an accessor / mutator method w that be
> a big problem in your view?
>
Yes.
I don't want to have to define access
Hi Phil,
I did that using Ronie's OpenAL binding and Merwann's wav parser back in
2014-2015:
MCSmalltalkhubRepository
owner: 'ronsaldo'
project: 'OpenAL'
MCSmalltalkhubRepository
owner: 'MerwanOuddane'
project: 'WAVParser'
I did things such as (build yourself something derived from this code I
Or write a script that obfuscate the code. Not so hard to do.
On Tue, Sep 26, 2017 at 5:13 PM, Christophe Demarey <
christophe.dema...@inria.fr> wrote:
> Hi
>
> What about deploying a minimal image without UI ?
> But, as you probably know, headless mode does not prevent to dump the code.
>
> Chr
der in the future making it work on the latest Pharo version on
multiple OS if it makes sense to show people how to do it. I'm busy with
other concerns right now.
Anyway, it's just one approach, you can use another one.
> Cheers,
>
> Jeff
>
>
>
>
>
> On Sun
Hi,
I describe in the next paragraphs what I do to build application with Pharo
at 50 fps. Now I am using a SDL2 window with OSWindow and not the default
Pharo window for rendering. There is a fullscreen option with OSWindow, I
don't use it but I've just tried it works fine.
To use Athens/Cairo I
Hi,
If your tool works in Pharo 6, you can use the other bytecode set which
supports up to 32k literals. To do so, go to:
World Menu > Settings > Compiler > Encoder
and pick SistaV1 instead of V3PlusClosures
Try to load your code. The default Pharo 6 VM supports both bytecode sets.
Alternatively
Hi,
About the best way to help...
A reproducible case that we (VM developers) can reproduce easily is the
best way to help. If we do not have a way to reproduce your bug, we cannot
do anything about it. The best is an image which crashes immediately after
start-up or a do-it that reliably crash t
Hi,
You need a VM with the SqueakV3 memory manager and compiled in the past few
months for Mac OS X Sierra support.
All recent Pharo VM are compiled with the Spur memory manager leading to
this error “This interpreter (vers. X) cannot read image file (vers. 6505)”.
I would try the latest VM from
Hi,
All symbols are interned in the Symbol table. If one does (#foo , #bar ,
#baz) and each returned value would be a symbol, then both #foobar and
#foobarbaz would be registered in the symbol table.
I would guess that's why the concatenated value is a string and not a
symbol, to avoid registerin
Hi Raffaello,
Reportedly, the VM without the JIT (pure interpreter, also called PharoS or
StackVM) works on windows 64 and FFI works with it if the VM compiled with
clang but not with gcc. There is no configuration for Pharo right now. It
should not be hard to add a pharo configuration to have the
If I am correct, when you do:
double x=0;
while(x<10)
{
x = x+1;
}
C++ compiles the + to double operation.
In Pharo, this code: |x| x := 0. 1 to: 10 do:[:each| x := x +1]
is using SmallInteger operations. The code you wrote is fully inlined in
Cog's JIT (no message s
The VM you use is the fastest one available.
It's slower mostly due to non Smalltalk concerns (processor architecture,
energy consumption, frequency, overall hardware).
There is an abstraction over the back-ends in the JIT and it could be that
overall the JIT is a little bit more optimised for in
>
> Regards,
> Vitor
>
> On Wed, Oct 12, 2016 at 9:38 AM, Clément Bera
> wrote:
>
>> The simulation of primitives is done in Context>>#doPrimitive:method:r
>> eceiver:args:
>>
>> Basically, specific numbers are simulated in the image while othe
The simulation of primitives is done in
Context>>#doPrimitive:method:receiver:args:
Basically, specific numbers are simulated in the image while other numbers
are run using the VM code.
Quick methods (what you call inlined methods) are encoded with primitive
numbers between 256 and 512. If you lo
eas Sunardi
>> wrote:
>>
>>> Thank you for the impressive quick response. Unfortunately, I have older
>>> glibc. So now I'm struggling with compiling glibc 2.15 for 32 bit on my 64
>>> bit CentOS 6.5 machine. Not an easy thing to do.
>>>
>&
Thanks for reporting the problem.
The error means the VM is incompatible with the image. There was a change
of image format in Pharo 5, so the package has likely an old VM while the
image has the new format, or the new VM while the image has the old format.
Someone will look into that problem in
General documentation on all the base frameworks is present in the *Pharo
by example* book. You can download it for free here:
http://pharobyexample.org/versions/PBE1-2009-10-28.pdf
Or here for the updated version:
https://ci.inria.fr/pharo-contribution/view/Books/job/UpdatedPharoByExample/lastSucc
19, 2016 at 3:51 PM, Clément Bera
> wrote:
> > I am not sure this is the right mailing list.
> >
> > Spur32BitPreen could be used as part of the Pharo release process as
> long as
> > the compactor is not working well. This way released images can be more
> >
Hello Ben,
The limit you see for instance variables is due to:
- Class format encoding (Memory manager dependent)
- Bytecode set encoding (Bytecode set dependent)
The 255 inst var limit is enforced both by:
- the V3 Memory manager
- the SqueakV3PlusClosures bytecode set.
Now we have Spur instead
On Thu, Sep 8, 2016 at 3:44 AM, Vitor Medina Cruz
wrote:
> Hello,
>
> While profiling some I/O code that takes ~20 seconds to execute under my
> local image, the report says that about ~13 seconds is waste on
> OtherProcesses -> ProcessorScheduler class>>idleProcess. I could not
> understand what
What ruby runtime exactly ? The standard ruby interpreter, rubinius, JRuby ?
What do you mean by performance ? Smallest time to run long computation ?
Latency for web servers ? Pauses in real time applications ?
The standard ruby interpreter is really slow (likely ~100 times slower than
the Pharo
Hi,
I've never seen that error before. That looks interesting.
You need to provide more information. For example the backtrace or
something written in PharoDebug.log.
Else, as suggested, set a breakpoint in malloc_error_break and try to
figure out what is going on.
On Wed, Sep 7, 2016 at 4:56 P
The 64 bits linux and mac VM have been working for more than 6 months.
They've been in production in a company since April or something like that.
Latest product are built by Travis and available here:
https://bintray.com/opensmalltalk/vm/cog#files. Pharo provides a
Pharo-flavored VM with a differe
On old versions of Pharo you need to use:
#subclass:instanceVariableNames:classVariableNames:poolDictionaries:package:
That might be it.
On Fri, Jul 29, 2016 at 6:14 PM, Peter Uhnak wrote:
> Can you show the full trace? Because I am regularly doing something
> similar without any problems.
> In
We measured Spur vs pre-Spur, but I am not sure we can measure VW without
Cincom's approval.
On Tue, Jun 7, 2016 at 7:56 AM, Gerry Weaver wrote:
> Hi,
>
> I was reading about the VM stuff and assumed (wrongly) that the post
> applied to that. Sorry for the noise.
>
> Thanks,
> -G
>
>
> -Orig
;
>
> you should add that to a simple blog post :)
>
>
> Stef
>
> Le 24/5/16 à 19:35, Clément Bera a écrit :
>
> Hi,
>
> Sorry the mail is quite long... I CC'd Nevena for section II, she used the
> VM caches for type inference.
>
> *Section I. VM param
Hi,
Sorry the mail is quite long... I CC'd Nevena for section II, she used the
VM caches for type inference.
*Section I. VM parameters*
* 46 size of machine code zone, in bytes*
Well, the size of the machine code zone :-). To speed-up execution, the cog
uses internally a JIT compiler whic
I usually get it from http://files.pharo.org/
In media, pharoCheatSheet or flyer-cheat-sheet.
On Fri, May 13, 2016 at 10:32 AM, Dimitris Chloupis
wrote:
> thanks any idea where i can get the card that contains the pharo syntax ?
>
> On Fri, May 13, 2016 at 10:33 AM stepharo wrote:
>
>> good lu
I think in addition of the parser/compilation chain, a language is about
having a good IDE.
The Pharo IDE depends more and more on the AST and not on the source code,
so if your new language uses an AST polymorphic with the Smalltalk AST I
guess you can have IDE tools for free (at least partly).
Alternatively, I do that using a SDL window. This way, the production game
can be launched using SDL and pharo headless, and not have the overhead of
the morphic rendering loop.
2016-03-13 15:22 GMT+01:00 Henrik Nergaard :
> http://ws.stfx.eu/1RCNV18R5J7E
>
> Best regards,
> Henrik
>
> -Origi
sourceCode is the best way to do it. It answers the original sources if
available, and the decompiled sources if not available.
2016-03-10 11:22 GMT+01:00 Cyril Ferlicot :
> Hi,
>
> Do you really need to decompile the method? A CompiledMethod know his
> source code.
>
> See: Object methods first
Hello,
As Stephan Eggermont said, you need to make some measurements.
There are a few big problems.
In short, if the FFI call is going to do quite some computation (at least a
few millisecond), assuming tinyC has performance similar to gcc or llvm,
you may get faster, if the FFI call is going to
Alternatively, convertor from pdf to epub are available and free. Pillar
can generate pdf.
I know converting does not produce a file as good as if there were a pillar
exporter, but that's a good enough temporary solution to me.
2016-02-25 22:54 GMT+01:00 stepharo :
> Hi martin
>
> Why don't yo
Ok.
Then I guess you can do something like Slang, which is used for the VM.
Slang is a restrictive Smalltalk compiling to C.
The slang compiler parses the code using the Smalltalk Compiler parser,
then translate the Smalltalk AST to its own AST, do some manipulation based
on pragmas available in
Are you sure C++ compilation is the best way ? There are a lot of flaws:
Smalltalk needs to be restrictive or will be compiled to very slow code,
you will loose debugging features, etc.
I would try to integrate Pharo the same way lua is integrated in engines
like Source2 from valve: you run your U
Hello Sven,
You can use the VM with the JIT on the Pie now (Spur Cog VM). It is
deployed by default on the recent raspberry pis and has millions of users
for Squeak and Scratch since August 2015 (the raspberry pie fondation paid
Tim for the ARM JIT on the Cog because of the size of their user base
The latest FFI uses the old VM FFI which works on x86 and ARM and a new
front-end which is seemingly x86 dependent.
So OSSubProcess will work if you use FFI, but the old syntax not the new
one.
2016-02-13 0:23 GMT+01:00 Mariano Martinez Peck :
> Hi Sven,
>
> Since OSSubprocess uses the latest FF
Hello,
Both classes (SmallFloat64 and BoxedFloat64) are present on 32 bits and 64
bits by default right now. The code base is common between 32 and 64 bit
images. It is possible to run a 32 bit image without the class
SmallFloat64, but it is not the default behavior. There are no instances of
Smal
Yeah I did in in early version of Pharo 5 (50155).
2016-01-24 10:10 GMT+01:00 stepharo :
> the support for SDL2.0 should be better in Pharo50 (even if due to ffi
> changes and new objcte format we may have some
> glitches).
>
>
> Le 23/1/16 22:45, Edwin Ancaer a écrit :
>
>> I'm not a smalltalk e
2016-01-12 0:40 GMT+01:00 Vincent BLONDEAU <
vincent.blond...@polytech-lille.net>:
> Hi,
>
>
>
> It is expected.
>
> When you evaluate a block with a return, the method containing the block
> definition (here testBlock) will return.
>
> So the test will pass.
>
>
>
> BTW, you don’t need to put a r
There is no way of creating such blocks currently.
In fact only non local returns and debugging are problems as remote
variables are accessed through an indirection. A lightweight block (to
reuse your terms) with no outer context but accesses to remote temporary
variables would work fine.
The com
2015-12-20 8:15 GMT+01:00 Marcus Denker :
>
>
> On Sat, Dec 19, 2015 at 9:24 PM, Tudor Girba wrote:
>
>> Hi,
>>
>> Strings are supposed to be immutable, and that is why we do not support
>> editing by default :).
>>
>>
> I do not think this is good... yes, it would be nice if *literal* strings
>
Have you tried this (from this page:
http://pharo.org/gnu-linux-installation#ubuntu-ppa-headless-vm ):
Ubuntu (ppa)
Ubuntu users can use the dedicated ppa to install Pharo in various ways:
Install the headless Pharo VM only
sudo add-apt-repository ppa:pharo/stable
sudo dpkg --add-architecture i38
PharoSound was not working on Mac on my machine a few month ago. It was
reported to be working on the Raspberry Pie and windows.
2015-11-18 12:23 GMT-03:00 Merwan Ouddane :
> This summer I made a wav parser, there is an example to play wav files
> with OpenAL.
>
> https://pharoweekly.wordpress.co
Hello,
You can try to reload ConfigurationOfOSWindow or try to get the latest
ConfigurationOfOSWindow then load the bleeding edge, that might be it.
Else for the #doSemanticAnalysisIn: you can try to flush the ASTCache (I
think it's "ASTCache rest")
Have fun
2015-07-18 17:54 GMT+02:00 Hilaire :
2015-07-15 12:34 GMT+02:00 Sean P. DeNigris :
> Noury Bouraqadi-2 wrote
> > Regarding the naming PharoJS is more than a bridge. It allows:
> > -Develop and test apps in Pharo
> > -Generate a javascript file for a standalone app
>
> How does it compare to Amber? Similar use case?
>
>
The resulting
What is PharoJS ? Are you talking about Pharo on top of the Bert's SqueakJS
VM ?
If so, it starts, but some primitives fail making it freeze after a short
while as you describe.
It would be fun to have it running even though most of the Pharo folks went
into the Amber direction for Pharo on Javas
What about "1 tinyBenchmarks" ?
Just to know if the VM is slower as a whole or only compilation / source
access ?
2015-06-30 0:36 GMT+02:00 Jan Blizničenko :
> And one another benchmark of linux in VM on that desktop PC:
> Roassal loading - 58 s
> compilations per second - avg: 262.4682, min: 25
Silent compilation is when the compilation is done without telling any
other objects.
In the regular compilation of a method, for example, you tell all the class
browsers opened that the method was compiled so that if a class browser
shows an old version of the same method it is refreshed to show
Code critics relies on simple rules, such as the use of the selector
#isKindOf: .
It does not detect the second case because the rule is simple, though both
are questionable.
Best,
2015-04-24 18:24 GMT+02:00 Peter Uhnák :
> Why does CodeCritic (QualityAssistant) considers this questionable
>
>
Seemingly, UI-wise, it was not good to have a button for the bytecode.
As I use it al the time, I reintroduced it in the menu and you can use the
shortcut Cmd+b+b, which reportedly was better.
In other IDE tools that nautilus you can't see anymore the bytecodes though.
2015-04-01 8:06 GMT-07:00
Hey Sean,
Now to create a proxy you need to subclass ProtoObject and not nil.
This was changed years ago to avoid some issues. An example of issue that
existed: when you subclass nil, as
#doesNotUnderstand: is not implemented by default on your proxy, any proxy
created receiving a message would c
Hello,
I believe that topics related to robotics are supervised by Noury Bouraqadi
and / or Luc Fabresse.
I think you should ask them for more information.
You can find their mail on their web page (just type their names on google)
Good luck,
Clement
2015-03-27 9:51 GMT+01:00 Damien Cassou
Well, this year google it looks like accepted less organizations.
2012: 180 organizations accepted.
2013: 177 organizations accepted.
2014: 190 organizations accepted.
2015: 137 organizations accepted.
A few projects with ESUG could be nice for sure. But there we will have in
the same issues agai
It depends on your OrderedCollection implementation.
OrderedCollection has a variable size in memory. When instantiated, it has
for example 10 slots, and if you have more than 10 objects, it needs to
change its size to hold more slots.
The current implementation relies on an indirection to an arr
Hey,
Inspectors are different between Moose and Pharo 3. Can you precise, are
you using the Glamour inspector, the EyeInspector or the old Morphic
inspector ?
I have never seen that in Pharo. Are your titles dynamic (I mean, when the
object changes, can your title change ?) ?
2015-02-18 11:35 GM
h the updated result once Pharo4 is out.
> >
> > Of course, you can add the benchmark and tweak it for your needs.
> >
> > Thank you for all the good work you are doing! Reaching a speed near
> > pypy would be a real game changer!
> >
> > 2015-02-17 11:24
Hello Andrea,
The way you wrote you algorithm is nice but makes extensive use of closures
and iterates a lot over collections. Those are two aspects where the
performance of Pharo have issues. Eliot Miranda and myself are working
especially on those 2 cases to improve Pharo performance. If you don
Hello,
There could be many reasons why you would have such a big image.
The most common reason is the monticello metadata, as Phil said.
Another reason could be because some caches and objects that are flushed on
snapshot, so if you looked at the image size at runtime and not the file
size (for
2015-01-28 16:41 GMT+01:00 Laura Risani :
> Hi all,
>
> What is the explanation why Smalltalk designers preferred single
> inheritance+traits to multiple inheritance? Why is the former better than
> the latter?
> Do traits let you share state also or only behavior?
>
Traits only share behaviors.
Hello,
Or simpler:
1 second wait.
In your example:
Transcript open.
[
10 timesRepeat:[Transcript crShow:(DateAndTime now).
1 second wait]
] fork.
2015-01-27 20:21 GMT+01:00 Nicolai Hess :
>
> 2015-01-27 19:46 GMT+01:00 Laura Risani :
>
>> Hi all,
>>
>> I need an object/method lik
2015-01-21 10:31 GMT+01:00 nacho <0800na...@gmail.com>:
> philippeback wrote
> > why running pharo with a squeakvm?
> >
> > I do it, in my Raspberry Pi :p and its slo
>
Actually I think only the UI is slow on the Pi because on the squeak VM
made by Tim he reimplemented differently BitBlt.
Is
aside.image \
> st ./startServer.st \
> --no-quit
>
> Apart from running it under a limited user/group, it is also possible to
> chroot it for an extra security measure.
>
>
> Michal
>
>
>
>
> On 21.1.2015 09:16, Clément Bera wrote:
>
> Just a deta
Just a detail. When you run headless with the latest pharo-vm (that you can
download here: http://files.pharo.org/vm/pharo/ in your OS directory, file
stable.zip), add --no-quit after --no-display in the VM start-up options.
2015-01-21 7:21 GMT+01:00 p...@highoctane.be :
> why running pharo with
Typically, if the image can still access the change file but not the source
file (or the other way around), some methods still show temporary variable
names, whereas other methods show t1, t2, etc , depending on which files
their sources were in.
2015-01-20 10:21 GMT+01:00 Sven Van Caekenber
Yeah Object>>#finalize is supposed to do what you said but unfortunately it
does not work...
Maybe it's a VM bug or maybe it's not supported any more.
We'll see if Eliot answer the thread.
2015-01-13 15:55 GMT+01:00 Johan Fabry :
>
> No, just putting the cleanup code in a finalize method does n
There's a direct train from Paris CDG airport to Lille. It's usually more
complex to go from brussels airport to Lille.
There's an airport in Lille but there's very few flights from there (only
to Marseilles and Porto AFAIK).
2014-12-27 12:21 GMT+01:00 Udo Schneider :
> All,
>
> what's the best
2014-12-17 7:42 GMT+01:00 Damien Cassou :
>
>
> Le 16 déc. 2014 22:48, "dboeren" a écrit :.
> > Off the top of my head, I would be interested in support for "slices"
> > meaning taking being able to treat a subset of a larger
> OrderedCollection as
> > its own OrderedCollection rather than using c
Hello,
I don't really know how to answer.
Are your examples really from smalltalk-80 ? Were they running on the
Xerox-D microcoded machine ? Because I don't think there are anymore
smalltalk-80 available right now, nor machine that can run it. There's a
smalltalk-78 running on javascript on the w
Hello,
What you want to do is difficult because the vm internals rely on the word
size.
There are two solutions:
1) editing the VM (JIT native code back end + memory manager) to support 64
bits. That's a work in progress. You can discuss about it on the Squeak
vm-dev mailing list.
2) compiling th
Hello,
this is possible but non trivial.
I think the easiest way is to bind the C functions to manage C libraries:
dlopen, dlsym, dlclose, dlerror. Then you can manually open and close the
dynamic libs (dlopen, dlclose) and call the function dynamically loaded
with dlsym. dlerror is used for erro
Depending on the VM you use, different objects are passed by values.
Currently -> only smallintegers
on Spur 32 bits VM -> smallintegers and characters
on Spur 64 bits VM -> smallintegers, characters and smallfloat
But as Stef said it does not really matter in smalltalk.
2014-09-25 21:55 GMT+02:
Let's say no.
It will be possible in the future but it requires quite some work in the
VM. If you want to do this work, ask the question on VM-dev mailing list
they will tell you what you need to do to make it possible.
2014-09-18 22:05 GMT+02:00 Annick Fron :
>
> Is it possible to call pharo fr
Hello,
The additional assignment is not needed.
You get a warning because the warning detection is approximate and
sometimes raises warning whereas it should not, but if you proceed, it
compiles fine.
2014-09-10 8:54 GMT+02:00 Hilaire :
> Hello,
>
> I noticed this change described bellow and I
I got unmatched string quote.
If it's French, could it be a ' ?
It looks like a parser bug to me...
I hope not :-)
>
> Alain
>
> On Tue, Sep 9, 2014 at 5:43 AM, Clément Bera
> wrote:
>
>> What kind of bug do you have with accent ? Do you have a stack trace
What kind of bug do you have with accent ? Do you have a stack trace ?
More importantly, can you give us a reproducible case ? I will investigate
tomorrow if you have a reproducible case to check if this is due to Opal.
It may also be recent changes related to fonts.
2014-09-08 21:07 GMT+02:00 S
I don't understand, do you want to compile your library using FFI or do you
want to bind a library compiled your way with FFI ?
If you want to compile your library using FFI, then use OSProcess to run
the compilation line you showed.
If you want to bind a library compiled your way with FFI, the e
2014-08-07 10:47 GMT+02:00 aria2end :
> Yes Exactly. I want to see all the potential methods called from from this
> method which is the second case. but I receive error using that script.
>
> MessageNotUnderstood: ByteSymbol>>implementors
>
> Ok this method is implemented in the image I use (Phar
2014-08-07 5:11 GMT+02:00 aria2end :
> Hi, I know that I can see senders and implementers of a method but is there
> any way to see all methods that are used in a method ?
>
This is impossible. The method called for each message send can only been
known at runtime, because it depends on the recei
2014-08-07 6:38 GMT+02:00 aria2end :
> Hi james,
>
> Your post not only resolved the issue but also introduced me a new approach
> which I was looking for it.
> Thank you
>
> do you know the reason why assigning to a passed-in argument is illegal by
> any chance ?
>
> This is illegal because then
Well the method with the code you give was compiled without any problem in
my image (Pharo #30851 with FFI loaded with configuration browser as
explained in the other thread).
Perhaps this is related to your bug in the other thread where you were
(probably) missing a package ?
2014-07-10 18:15
1 - 100 of 162 matches
Mail list logo
