On Fri, Nov 7, 2014 at 2:06 PM, Alain Rastoul <alf.mmm.cat@gmail.com> wrote:

Well, it looks like this thread have somewhat derailed :)

My initial�� question was about a problem I now think (but have to
confirm) is a vm system call signals side effects with socket polling
in zeroMQ code (remember this is the vmdev mailing list not the smalltalk
user list).

I don't like zealots of any kind (object, relational, functional, religious ...),
my pun was essentially a joke about trying to send a stateless electron�� :o)
over a wire�� as a computation�� to someone's answer I suspect beeing
a troll (plug a chakra socket here for a zealot effect).

But electrons are not stateless; they have spin and energy ;-)

��

To me at some point, we have to go into some kind of materialization and state,
(even with pivotal who have to stores it's code as bits I guess ?)
and certainly no judgement of any kind about other realms or langages,
I like new ideas , new langages, and�� new experiences.
Of course, functional programming, monads, lambda calculus
interests everybody here (me too), I also would like to try functional programming too
but have a project I want to take at some interesting level first, and this is my priority.

If the intent was not a troll, and he wants to discuss
about smalltalk object oriented realm why doesn't he *please* start a new thread
and lot of people would give him valuable answers (some did here)
about the small*talk* spirit which is (IMHO) essentially about object perception
and message sending (hence behaviour or functional) but certainly not "only state"

Regards,

Alain

Le 07/11/2014 22:04, kilon alios a ��crit :

Its definetly an interesting concept. I am curious to give it a try
because it would take me outside my comfort zone and there is nothing
more that I love than getting outside my comfort zone. But I blame Pharo
and the Pharo people who dont let me to try another language, damn them
!!!! DAMN THEM I SAY!!!!

Side effects certainly can be a source of trouble but alas they are not
the holy grail of trouble. Coding is a complex subject so introducing
restrictions of purity will produce some side effects. OH YES PUN WAS
INTENDED !!!!

Its easy to dismiss such an approach however without trying it in
practice with real projects. But then I do also recognise that mixing
types can be a problem too at some point that does not convince me into
going to�� static type language.

I think for Pharo state is much less of a problem because Pharo has very
powerful inspector , debuger and IDE to track down state. So its easy
for a pharo developer to be aware of state compared to some developer
using another language and not using an IDE at all. If state becomes too
complex then its a sign to refactor code and make it simpler. I do think
however with powerful IDEs you can get around this problem easily.

So I cant say I am a big believer of Haskell.

Also I real hate the terminology "side effect" ... sounds too.....
elitist to me.

On Fri, Nov 7, 2014 at 5:51 PM, Kjell Godo
<squeaklist@gmail.com
<mailto:squeaklist@gmail.com>> wrote:

�� This is off topic.

�� I tried to post it as a top level thread but I have become unknown.

�� I don't know if you want this crap in here but I have decided not to
�� wait for the

�� postmaster to get back to me on the subject of becoming known.�� Feel
�� free.

�� ( Original-SUBJECT:�� "( picoVerse-:( what about state , is state
�� really evil? ) )"�� )

�� I am a Smalltalker.

�� But in the past few months i have been running with the Haskellers.

�� The Haskellers hate state.

�� This seemed strange at first because as a Smalltalker i love(d)
�� state.�� State iswas my friend.

�� 90% of my life as a Smalltalker is state wrangling.�� I am a state
�� herder.

�� The debugger is my staff I use to whack the state.�� And TestCase is
�� my sheep dog.

�� But to the Haskellers

�� state is

�� the evil trinity

�� of

�� satan the anti christ and the false prophet

�� all rolled into one.

�� State is the true dev incarnation of the total catastrophe of
�� development Armageddon.

�� Blood up to the bridles for hundreds of miles.�� Dogs and cats living
�� together.�� Mass hysteria.

�� They say.

�� I'm not sure i quite get it yet but they keep preaching on this one
�� point most of all.

�� State is evil.

�� You must keep all state in a Monad.�� As many methods/functions m as
�� possible

�� must be 100% dependent on the input parameters ONLY.

�� No hidden instance variables affecting the return value of m are
�� allowed.

�� The only effect m can have is to return a value.

�� If all this is true then m is pure.

�� And pure is good.�� Pure is very good.�� And the wind says

�� very.

�� So i wonder if any of you fellow

�� Smalltalkers

�� have thought about this at all.

�� Thanks

�� Kjell E God��

�� (((((((((( Maybe Smalltalk should be called Statewalk

�� as in yak it up fuzz ball. ))))))))))

�� On Tuesday, November 4, 2014, Alain Rastoul
�� <alf.mmm.cat@gmail.com
�� <mailto:alf.mmm.cat@gmail.com>> wrote:

�� Stef,
�� As I said to Igor, the main problem about NativeBoost was between
�� the chair and the keyboard... :)
�� It is my first use of NativeBoost, I simply overlooked the very good
�� existing chapter of EnterprisePharo on NativeBoost
�� (NativeBoost recipes, the X11 journey) and misused the
�� NativeBoost marshalling of data types.
�� NAtiveBoost is great.

�� If I achieve my experiments with zeromq and ends up with
�� something clean (not the case actually, and not my first goal),
�� I will be glad to add�� a chapter about that.

�� My current problem is about a different socket behaviour between
�� windows and linux.
�� I think this is not a problem with ZeroMQ, the C program works
�� as expected,
�� I have to look for an explanation.
�� No stress about that

�� Alain

�� Le 04/11/2014 10:39, stepharo a ��crit :

�� Alain

�� which nativeboost chapter :)?
�� Could you propose a paragraph so that we cover the problem
�� you faced?

�� Stef

�� On 4/11/14 00:44, Alain Rastoul wrote:

�� Hi Igor,

�� Thank you for your answer, it worked perfectly
�� Looks like I overlooked the nativeboost chapter
�� .. 10 timesRepeatAfterMe: [self rtfm ] .
�� And my suggestion about testing nil was stupid, much
�� better to fail soon.
�� ...�� I am an ass on this one...

�� However, I am struggling on another point you already
�� answered in the
�� past
�� in the mailing list to a guy who wanted to use socket.io
�� <http://socket.io> :
�� (http://forum.world.st/socket-__io-td3891592.html#a3893031
�� <http://forum.world.st/socket-io-td3891592.html#a3893031>)
�� "Sockets in Pharo are not blocking the whole VM.
�� What they block is the process which working with
�� concrete socket. But
�� other processes can still run, while
�� one waiting data / even from single socket. "
�� on windows, zmq socket receive is blocking, on linux,
�� not blocking
�� (hence not working).
�� As zmq is doing is IO on another thread, I guess there
�� is some side
�� effect of
�� socket receive timeout settings somewhere (in the plugin
�� ?) - just a
�� guess...
�� Getting socket options shows no difference, but I don't
�� know how it is
�� done on the vm
�� side with regards to threads and if the socket is the
�� same (zmq socket
�� is not the tcp socket)...
�� And on linux, the equivalent C program of to the
�� smalltalk version
�� blocks as expected.

�� I a mperplexified ...
�� may be I should look at vm and plugin code (VMMaker
�� package IIRC) ?
�� Do you have another advice ?

�� Thanks you

�� Alain
�� Le 03/11/2014 02:12, Igor Stasenko a ��crit :

�� NBExternalArray instances cannot be passed directly
�� to functions
�� expecting pointers.

�� use 'myarray address' for arguments.

�� On 3 November 2014 00:20, Alain Rastoul
�� <alf.mmm.cat@gmail.com
�� <mailto:alf.mmm.cat@gmail.com>>
�� wrote:

�� Hi,

�� I have a problem with a nativeboost call, but
�� I don't see what I do
�� wrong.

�� library function prototype is:
�� int zmq_getsockopt (void *socket, int
�� option_name, void
�� *option_value, size_t *option_len);

�� my calling method in pharo is:
�� zmq_getsockopt: socket option_name: option_name
�� option_value:
�� option_value option_len: option_len
�� <primitive: #primitiveNativeCall
�� module: #NativeBoostPlugin
�� error: errorCode>
�� ^self nbCall: #(int zmq_getsockopt
�� (void *socket, int
�� option_name, void * option_value, size_t*
�� option_len)�� )

�� when I call it with
�� ...
�� optionValue := (NBExternalArray ofType: 'int')
�� externalNew: 1.
�� optionLen := (NBExternalArray ofType: 'size_t'
�� ) externalNew: 1.
�� [�� optionValue at: 1 put: 0 .
�� optionLen at: 1 put: (NBExternalType
�� sizeOf: 'int')�� .
�� rc := self zmq_getsockopt: socket
�� option_name: option_name
�� option_value:�� optionValue
�� option_len: optionLen�� .
�� value := optionValue at: 1 .
�� ] ensure: [�� optionValue free.
�� optionLen free�� ].
�� ...
�� I allways get an exception: "error during FFI
�� call : nil"

�� After stepping in NBFFICallout generation, I
�� found something
�� strange, the code
�� generation seems not to be called because
�� lastError stays at nil ?

�� handleFailureIn: aContext nativeCode: aBlock
�� lastError := self getErrorFrom:
�� aContext lastError:
�� NativeBoost lastError.

�� >>getErrorFrom: aContext lastError:
�� errorCode
�� ...�� checks�� pragmas etc
�� >>getErrorFrom: aContext lastError:
�� errorCode
�� ...�� lastError := aContext
�� tempAt: method
�� numTemps.
�� => lastError = nil ???
�� shouldn't be
�� ErrNoNativeCodeInMethod ?
�� "install native code and retry the send"
�� lastError = ErrNoNativeCodeInMethod
�� ifTrue: [ ^ self generateCode:
�� aBlock andRetry:
�� aContext ].
�� never gets called ...

�� "ok, we're out of options, signal an
�� error here"
�� ^ self signalError: lastError

�� Could it be because I use this image on windows
�� and unix ?
�� Or because I had an exception at prototype
�� parsing the first time
�� because I forgot a ; at the end of the prototype ?

�� Is my prototype correct or is it the origin of
�� the error ?
�� Is there a way to reset the lastError
�� (aContext�� tempAt: method
�� numTemps) ?
�� I will try to reset it in debugger but may be
�� there is a cleaner
�� way ?
�� would it be ok to change�� the test in
�� handleFailure to
�� (lastError = ErrNoNativeCodeInMethod) or:[
�� lastError isNil ] ?
�� (I can open a bug in this case )

�� Any idea or comment is welcome
�� Thanks in advance

�� Alain

�� --
�� Best regards,
�� Igor Stasenko.

best,

Eliot