February 2015 - Pharo-dev - lists.pharo.org

Re: [Pharo-dev] Easy way to write a proxy?

by Camille

> On 27 Feb 2015, at 03:43, Alexandre Bergel <alexandre.bergel(a)me.com> wrote: > > Hi! > > I have a near-to-be naive question. Maybe there is something obvious, but I cannot see it. Consider the following implementation of a proxy: > > -=-=-=-=-=-=-=-=-=-=-=-= > Object subclass: #MyProxy > instanceVariableNames: 'object'. > > MyProxy >> object: anObject > object := anObject > > MyProxy >> doesNotUnderstand: aMessage > ^ object perform: aMessage selector withArguments: aMessage arguments > > Object >> wrap > self becomeForward: (MyProxy new object: self; yourself) > -=-=-=-=-=-=-=-=-=-=-=-= > > The problem is with wrap. After the becomeForward:, MyProxy has itself in the object variable. Even self is apparently subject to the #become. > > It means that the following code loop indefinitely > > -=-=-=-=-=-=-=-=-=-=-=-= > d := Dictionary new. > d wrap. > d add: #foo -> 10 > -=-=-=-=-=-=-=-=-=-=-=-= > > I know there is Marianoâs library to write proxy. But I still wondering whether this could be done simply. > Any idea how to simply create a proxy? Hi Alexandre, You have to store the target and the proxy in different temps, do the become and (re)set the link between both after. Or in your case, something like: Object >> wrap | proxyThenTarget | proxyThenTarget := MyProxy new. self become: proxyThenTarget. âNow self is the proxy" self object: proxyThenTarget. ^ self Note that by subclassing Object many messages wonât be intercepted. With your dictionary example it means that you cannot intercept #at: . Subclassing ProtoObject reduces the number of such messages. Also one can bypass these methods by sending #doesNotUndestand: to the proxy. In the end all depends on your requirements: do you care about uniformity (all messages must be intercepted), composition of proxies, security (no target leaks, no interception bypass), what operations do you want to intercept (just message sends or iv accesses in addition), etc... For most cases Ghost is the way to go. Camille > > Cheers, > Alexandre > > -- > _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: > Alexandre Bergel http://www.bergel.eu > ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. > > > >

Feb. 27, 2015

Easy way to write a proxy?

by Alexandre Bergel

Hi! I have a near-to-be naive question. Maybe there is something obvious, but I cannot see it. Consider the following implementation of a proxy: -=-=-=-=-=-=-=-=-=-=-=-= Object subclass: #MyProxy instanceVariableNames: 'object'. MyProxy >> object: anObject object := anObject MyProxy >> doesNotUnderstand: aMessage ^ object perform: aMessage selector withArguments: aMessage arguments Object >> wrap self becomeForward: (MyProxy new object: self; yourself) -=-=-=-=-=-=-=-=-=-=-=-= The problem is with wrap. After the becomeForward:, MyProxy has itself in the object variable. Even self is apparently subject to the #become. It means that the following code loop indefinitely -=-=-=-=-=-=-=-=-=-=-=-= d := Dictionary new. d wrap. d add: #foo -> 10 -=-=-=-=-=-=-=-=-=-=-=-= I know there is Marianoâs library to write proxy. But I still wondering whether this could be done simply. Any idea how to simply create a proxy? Cheers, Alexandre -- _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: Alexandre Bergel http://www.bergel.eu ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.

Feb. 27, 2015

Re: [Pharo-dev] Bug? JPEG reading ignores EXIF Orientation

by Sean P. DeNigris

stepharo wrote > We should improve the Image support because it is dated. I've always found the "replace me with aTransformMorph on resize" thing baffling, so I started re-implementing a bare-bones SimpleImageMorph. Right now it holds onto the original form unaltered, but warps it to its inner bounds, and you can also specify to fix the aspect ratio. For the simple cases that cover 99% of my usage, it works perfectly. See https://github.com/seandenigris/SimpleImageMorph for more info. ----- Cheers, Sean -- View this message in context: http://forum.world.st/Bug-JPEG-reading-ignores-EXIF-Orientation-tp4807446p4… Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com.

Feb. 27, 2015

Re: [Pharo-dev] The evil twin in 40516

by Marcus Denker

> On 27 Feb 2015, at 00:26, Marcus Denker <marcus.denker(a)inria.fr> wrote: > > >> On 27 Feb 2015, at 00:15, Alexandre Bergel <alexandre.bergel(a)me.com> wrote: >> >> Wow!!! >> This means that one can call #createTwin, modify the AST, and simply call it! >> Impressive!!! > > You need to handle invalidation yourself, too. (for now). > I want to first concentrate on purely âannotationâ based MOP: instead of changing > the AST it is annotated with MetaLinks. When a link is set, the twin creation + invalidation > happens. When the last Link is removed, the twin disappears automatically. > >> Is it per default in Pharo 40516? > > Yes. > and it does not yet recompileâ¦ recompileAST self flag: #TBD. "compile the ASTâ :-)

Feb. 26, 2015

Re: [Pharo-dev] The evil twin in 40516

by Marcus Denker

> On 27 Feb 2015, at 00:24, Nicolai Hess <nicolaihess(a)web.de> wrote: > > Cool! > > (spotter does not work on this methods, do we have to change this > in the class ReactiveMethod or spotter?) we need to copy the methods overâ¦ Marcus

Feb. 26, 2015

Re: [Pharo-dev] The evil twin in 40516

by Nicolai Hess

Oh and inspecting the ReflectiveMethod ast with GTInspector has some strange effect 2015-02-27 0:24 GMT+01:00 Nicolai Hess <nicolaihess(a)web.de>: > Cool! > > (spotter does not work on this methods, do we have to change this > in the class ReactiveMethod or spotter?) >

Feb. 26, 2015

Re: [Pharo-dev] The evil twin in 40516

by Marcus Denker

> On 27 Feb 2015, at 00:15, Alexandre Bergel <alexandre.bergel(a)me.com> wrote: > > Wow!!! > This means that one can call #createTwin, modify the AST, and simply call it! > Impressive!!! You need to handle invalidation yourself, too. (for now). I want to first concentrate on purely âannotationâ based MOP: instead of changing the AST it is annotated with MetaLinks. When a link is set, the twin creation + invalidation happens. When the last Link is removed, the twin disappears automatically. > Is it per default in Pharo 40516? Yes. Marcus

Feb. 26, 2015

Re: [Pharo-dev] The evil twin in 40516

by Nicolai Hess

Cool! (spotter does not work on this methods, do we have to change this in the class ReactiveMethod or spotter?)

Feb. 26, 2015

Re: [Pharo-dev] zipped and unzipped methods for ByteArray

by Nicolas Cellier

2015-02-26 22:37 GMT+01:00 stepharo <stepharo(a)free.fr>: > Why do we have to put everything in string? > > why > Zip string: 'hello' > Zip bzip string: 'hello' > Zip gzip string: 'hello' > > would not be more modular? > Zip could be a factory and have subclasses. > > This is not because can extend classes that we should think about nicer > way. > > I like for example the DateFormatter strategy related to Date from other > languages because I can specify the way I want it. > > Stef > > Because somehow you can easily pipe unary messages. ^'hello' zipped base64Encoded versus: (Encoder base64 byteArray: (Zip gzip string: 'hello')). I much much prefer to read/maintain 1st version The 2nd one looks like overkill verbiage to me. > Le 25/2/15 16:52, Juraj Kubelka a Ã©crit : > > 25. 2. 2015 v 12:28, Sven Van Caekenberghe <sven(a)stfx.eu>: >>> >>> This seems related to adding 'easy of use' methods to String. >>> >>> There are many different compression schemes (zip, bzip, bzip2, gzip, >>> ...), are we going to add them all ? >>> >>> How much do we want to couple different parts of the system ? >>> >> I prefer to couple the compression schemes with String and ByteArray. >> Methods like #zipped, #unzipped are not that simple and it is not well >> apparent how to use that compression scheme. >> >> Cheers, >> Juraj >> >> On 25 Feb 2015, at 16:23, Juraj Kubelka <juraj.kubelka(a)gmail.com> wrote: >>>> >>>> Can I add ByteArray>>#zipped and #unzipped methods the way it is for >>>> String? I want to compress bytes before sending through internet, for >>>> example. >>>> >>>> Do you agree? >>>> Thanks, >>>> Juraj >>>> >>> >>> >> >> > >

Feb. 26, 2015

Re: [Pharo-dev] The evil twin in 40516

by Alexandre Bergel

Wow!!! This means that one can call #createTwin, modify the AST, and simply call it! Impressive!!! Is it per default in Pharo 40516? Alexandre > On Feb 26, 2015, at 8:07 PM, Marcus Denker <marcus.denker(a)inria.fr> wrote: > > Hi, > > In 40516 there is now a strange, but rather powerful mechanism for CompiledMethod: the Twin. > > What does that do? > > you can call on a method #createTwin > > (ReflectivityExamples>>#exampleMethod) createTwin. > > after this, the CompiledMethod has a high-level representation (the AST) attached that itself references the CompiledMethod > (they form a twin). > > (ReflectivityExamples>>#exampleMethod) reflectiveMethod > > The fun thing is now that one can install either in the class. Call #invalidate to make sure the reflective method is installed. > > ReflectiveMethod implements #run:with:in: which calls a compilation hook (too re-create from the AST) and then installs that in the > method dict and then executes the method: > > (ReflectivityExamples>>#exampleMethod) createTwin. > (ReflectivityExamples>>#exampleMethod) invalidate. > self assert: (ReflectivityExamples>>#exampleMethod) class = ReflectiveMethod. > self assert: ReflectivityExamples new exampleMethod = 5. > self assert: (ReflectivityExamples>>#exampleMethod) class = CompiledMethod. > > Which means that this gives us an in-image, on-demand JIT compiler AST->Bytecode. In 50 lines of code. > > e.g. try on Morph: > > Morph methods do: #createTwin. > Morph methods do: #invalidate. > > Counting which twin is installed shows us the working set of Morph: > > (Morph methods select: [ :each | each class = CompiledMethod ]) size > > some 330 method out of nearly 900â¦. > > So what can one do with this? In essence this turns the AST into a reflective representation for Methods > (if you care to set up twin creation + invalidation correctly). > > What will this allow us to do? stay tuned... > > Marcus > > > -- _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: Alexandre Bergel http://www.bergel.eu ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;.

Feb. 26, 2015