- Pharo-dev - lists.pharo.org

Re: [Pharo-dev] [Vm-dev] Methods with more than 15 args sketch

by Tudor Girba

Hi, > On Jun 20, 2016, at 9:03 AM, Marcus Denker <marcus.denker(a)inria.fr> wrote: > >> >> On 20 Jun 2016, at 08:58, Tudor Girba <tudor(a)tudorgirba.com> wrote: >> >> Hi, >> >>> On Jun 20, 2016, at 8:30 AM, Marcus Denker <marcus.denker(a)inria.fr> wrote: >>> >>>> >>>> On 19 Jun 2016, at 11:35, Tudor Girba <tudor(a)tudorgirba.com> wrote: >>>> >>>> Hi, >>>> >>>> Does anyone want to pick this up on the Pharo image side? >>>> >>>> I took a brief look at this and while playing I noticed that the errors when we send more than 15 arguments seem a bit strange. Take a look at this snippet: >>>> http://ws.stfx.eu/958BXP5GR136 >>>> (you can paste it in Spotter to get the playground) >>>> >>>> In summary: >>>> - If we compile a method with a signature containing more than 15 arguments we get: >>>> "SyntaxErrorNotification: Too many argumentsââ. >>>> This is good. >>>> >>>> - If we compile a method with a message send containing between 16-31 arguments we get: >>>> "'InMidstOfFileinNotificationââ. >>>> Ok-ish. >>>> >>>> - If we compile a method with a message send containing more than 31 arguments we get: >>>> "'Error: genSend:numArgs: numArgs index 32 is out of range 0 to 31â" >>>> This is wrong. >>>> >>>> >>>> Did I misunderstand something? >>>> >>> >>> The VM does not support it, and as nobody is using it, the error messages are not that perfect. >>> (to have them nicer, someone would have seen it and be bothered to fix itâ¦) >> >> Sure. No problem. I asked the question because I am new in this area and I just wanted to see if I maybe I misunderstood something. >> >>> What Clement proposes is to implement support for more than 15 arguments in the compiler. >>> (which means that instead of error messages, it would compile, with the idea as he described >>> in the mail). >> >> Yes, I know :). So, I come back to my original question: Is anyone interested in following this up on the image level in Pharo? >> > Isnâ t that what the mail of clement was all about? To change the compiler to emit code according to the proposed scheme? Yes, it is. But, as that is only a mail, the question is who is going to implement this concretely :) Doru > Marcus -- www.tudorgirba.com www.feenk.com "One cannot do more than one can do."

June 20, 2016

Re: [Pharo-dev] [Vm-dev] Methods with more than 15 args sketch

by ClÃ©ment Bera

Hi, I think there was a misunderstanding Tim. The idea is not necessarily to let developers write code with this many arguments, the IDE could actually enforces the programmer not to write methods with too many arguments. It's really about non programmers, like code generating tools such as SmaCC. In Java for example, the "Clean code" code convention limits the number of arguments per method to 3, the "Code complete" code convention limits it to 7, but Java can compile methods up to 255 arguments for code generators. If you want to limit the number of arguments the programmer can write, enforce it in the IDE according to your code convention. *Please talk about the maximum number of arguments a developer can write in another thread, this is non related to the implementation discussed here.* As Jan said, there are already too many limitations in Smalltalk to write proper code generators. We've worked on those limitations already: - the SistaV1 bytecode set will go to production in Pharo 6 (likely Squeak at some point too) and it removes an important part of the limitations (jump sizes up to 65k instructions instead of 1024, ...). - Spur removed other limitations (methods with up to 32k literals instead of 255, behaviors with up to 65k inst vars instead of 255). The important limitations remaining are the max frame size (56), which in practice limits the number of temps to around 55, whereas we could have 64 temps or more, and the number of arguments of methods and blocks. This scheme is about solving those problems. There are 0.5% of methods using a large frame so we could grow from 56 to 128 or 256 without any significant problems. For arguments bytecode calling convention looks so much simpler than VM support. On Mon, Jun 20, 2016 at 8:30 AM, Marcus Denker <marcus.denker(a)inria.fr> wrote: > > > On 19 Jun 2016, at 11:35, Tudor Girba <tudor(a)tudorgirba.com> wrote: > > > > Hi, > > > > Does anyone want to pick this up on the Pharo image side? > > > > I took a brief look at this and while playing I noticed that the errors > when we send more than 15 arguments seem a bit strange. Take a look at this > snippet: > > http://ws.stfx.eu/958BXP5GR136 > > (you can paste it in Spotter to get the playground) > > > > In summary: > > - If we compile a method with a signature containing more than 15 > arguments we get: > > "SyntaxErrorNotification: Too many argumentsââ. > > This is good. > > > > - If we compile a method with a message send containing between 16-31 > arguments we get: > > "'InMidstOfFileinNotificationââ. > > Ok-ish. > > > > - If we compile a method with a message send containing more than 31 > arguments we get: > > "'Error: genSend:numArgs: numArgs index 32 is out of range 0 to > 31â" > > This is wrong. > > > > > > Did I misunderstand something? > > > > The VM does not support it, and as nobody is using it, the error messages > are not that perfect. > (to have them nicer, someone would have seen it and be bothered to fix itâ¦) > > What Clement proposes is to implement support for more than 15 arguments > in the compiler. > (which means that instead of error messages, it would compile, with the > idea as he described > in the mail). > > Marcus > > >

June 20, 2016

Re: [Pharo-dev] Code-Icons in the "message" browser

by Nicolas Passerini

That is great! On Mon, Jun 20, 2016 at 8:16 AM, Marcus Denker <marcus.denker(a)inria.fr> wrote: > Hi, > > The âmessageâ (should be âmethodâ) browser (which shows methods on > âsenders ofâ or âimplementors ofâ > now shows editor-icons: > > > BreakPoints are not yet fully working, but that should be easy to fix. > (this is in 60 098) > > Marcus >

June 20, 2016

Re: [Pharo-dev] [Vm-dev] Methods with more than 15 args sketch

by Marcus Denker

> On 20 Jun 2016, at 08:58, Tudor Girba <tudor(a)tudorgirba.com> wrote: > > Hi, > >> On Jun 20, 2016, at 8:30 AM, Marcus Denker <marcus.denker(a)inria.fr> wrote: >> >>> >>> On 19 Jun 2016, at 11:35, Tudor Girba <tudor(a)tudorgirba.com> wrote: >>> >>> Hi, >>> >>> Does anyone want to pick this up on the Pharo image side? >>> >>> I took a brief look at this and while playing I noticed that the errors when we send more than 15 arguments seem a bit strange. Take a look at this snippet: >>> http://ws.stfx.eu/958BXP5GR136 >>> (you can paste it in Spotter to get the playground) >>> >>> In summary: >>> - If we compile a method with a signature containing more than 15 arguments we get: >>> "SyntaxErrorNotification: Too many argumentsââ. >>> This is good. >>> >>> - If we compile a method with a message send containing between 16-31 arguments we get: >>> "'InMidstOfFileinNotificationââ. >>> Ok-ish. >>> >>> - If we compile a method with a message send containing more than 31 arguments we get: >>> "'Error: genSend:numArgs: numArgs index 32 is out of range 0 to 31â" >>> This is wrong. >>> >>> >>> Did I misunderstand something? >>> >> >> The VM does not support it, and as nobody is using it, the error messages are not that perfect. >> (to have them nicer, someone would have seen it and be bothered to fix itâ¦) > > Sure. No problem. I asked the question because I am new in this area and I just wanted to see if I maybe I misunderstood something. > >> What Clement proposes is to implement support for more than 15 arguments in the compiler. >> (which means that instead of error messages, it would compile, with the idea as he described >> in the mail). > > Yes, I know :). So, I come back to my original question: Is anyone interested in following this up on the image level in Pharo? > Isnâ t that what the mail of clement was all about? To change the compiler to emit code according to the proposed scheme? Marcus

June 20, 2016

Re: [Pharo-dev] [Vm-dev] Methods with more than 15 args sketch

by Tudor Girba

Hi, > On Jun 20, 2016, at 8:30 AM, Marcus Denker <marcus.denker(a)inria.fr> wrote: > >> >> On 19 Jun 2016, at 11:35, Tudor Girba <tudor(a)tudorgirba.com> wrote: >> >> Hi, >> >> Does anyone want to pick this up on the Pharo image side? >> >> I took a brief look at this and while playing I noticed that the errors when we send more than 15 arguments seem a bit strange. Take a look at this snippet: >> http://ws.stfx.eu/958BXP5GR136 >> (you can paste it in Spotter to get the playground) >> >> In summary: >> - If we compile a method with a signature containing more than 15 arguments we get: >> "SyntaxErrorNotification: Too many argumentsââ. >> This is good. >> >> - If we compile a method with a message send containing between 16-31 arguments we get: >> "'InMidstOfFileinNotificationââ. >> Ok-ish. >> >> - If we compile a method with a message send containing more than 31 arguments we get: >> "'Error: genSend:numArgs: numArgs index 32 is out of range 0 to 31â" >> This is wrong. >> >> >> Did I misunderstand something? >> > > The VM does not support it, and as nobody is using it, the error messages are not that perfect. > (to have them nicer, someone would have seen it and be bothered to fix itâ¦) Sure. No problem. I asked the question because I am new in this area and I just wanted to see if I maybe I misunderstood something. > What Clement proposes is to implement support for more than 15 arguments in the compiler. > (which means that instead of error messages, it would compile, with the idea as he described > in the mail). Yes, I know :). So, I come back to my original question: Is anyone interested in following this up on the image level in Pharo? Cheers, Doru > > Marcus -- www.tudorgirba.com www.feenk.com "Problem solving efficiency grows with the abstractness level of problem understanding."

June 20, 2016

Re: [Pharo-dev] [Vm-dev] Methods with more than 15 args sketch

by Marcus Denker

> On 19 Jun 2016, at 11:35, Tudor Girba <tudor(a)tudorgirba.com> wrote: > > Hi, > > Does anyone want to pick this up on the Pharo image side? > > I took a brief look at this and while playing I noticed that the errors when we send more than 15 arguments seem a bit strange. Take a look at this snippet: > http://ws.stfx.eu/958BXP5GR136 > (you can paste it in Spotter to get the playground) > > In summary: > - If we compile a method with a signature containing more than 15 arguments we get: > "SyntaxErrorNotification: Too many argumentsââ. > This is good. > > - If we compile a method with a message send containing between 16-31 arguments we get: > "'InMidstOfFileinNotificationââ. > Ok-ish. > > - If we compile a method with a message send containing more than 31 arguments we get: > "'Error: genSend:numArgs: numArgs index 32 is out of range 0 to 31â" > This is wrong. > > > Did I misunderstand something? > The VM does not support it, and as nobody is using it, the error messages are not that perfect. (to have them nicer, someone would have seen it and be bothered to fix itâ¦) What Clement proposes is to implement support for more than 15 arguments in the compiler. (which means that instead of error messages, it would compile, with the idea as he described in the mail). Marcus

June 20, 2016

Code-Icons in the "message" browser

by Marcus Denker

Hi, The âmessageâ (should be âmethodâ) browser (which shows methods on âsenders ofâ or âimplementors ofâ now shows editor-icons: BreakPoints are not yet fully working, but that should be easy to fix. (this is in 60 098) Marcus

June 20, 2016

[pharo-project/pharo-core]

by GitHub

Branch: refs/tags/60098 Home: https://github.com/pharo-project/pharo-core

June 19, 2016

[pharo-project/pharo-core] 13e616: 60098

by GitHub

Branch: refs/heads/6.0 Home: https://github.com/pharo-project/pharo-core Commit: 13e61670de0fe81cbfa08498bf7ea898b773b1f6 https://github.com/pharo-project/pharo-core/commit/13e61670de0fe81cbfa08498… Author: Jenkins Build Server <board(a)pharo-project.org> Date: 2016-06-19 (Sun, 19 Jun 2016) Changed paths: M Kernel-Tests.package/CompiledMethodTest.class/instance/tests - testing/testContainsHalt.st A Kernel.package/CompiledMethod.class/instance/literals/symbolLiterals.st M Kernel.package/CompiledMethod.class/instance/scanning/messages.st M Kernel.package/CompiledMethod.class/instance/testing/containsHalt.st R Kernel.package/Halt.class/class/halting/halt.st R Kernel.package/Halt.class/class/halting/halt_.st M Kernel.package/Halt.class/class/halting/now.st M Reflectivity-Tools.package/HaltOnCountIconStyler.class/instance/testing/shouldStyleNode_.st M Reflectivity-Tools.package/HaltOnceIconStyler.class/instance/testing/shouldStyleNode_.st M Ring-Core-Kernel.package/RGMethodDefinition.class/instance/testing/containsHalt.st R ScriptLoader60.package/ScriptLoader.class/instance/pharo - scripts/script60097.st A ScriptLoader60.package/ScriptLoader.class/instance/pharo - scripts/script60098.st R ScriptLoader60.package/ScriptLoader.class/instance/pharo - updates/update60097.st A ScriptLoader60.package/ScriptLoader.class/instance/pharo - updates/update60098.st M ScriptLoader60.package/ScriptLoader.class/instance/public/commentForCurrentUpdate.st M Spec-Tools.package/MessageBrowser.class/instance/initialization/initializePresenter.st Log Message: ----------- 60098 18534 add Style icons so messageBrowser https://pharo.fogbugz.com/f/cases/18534 18526 better #containsHalt https://pharo.fogbugz.com/f/cases/18526 http://files.pharo.org/image/60/60098.zip

June 19, 2016

Re: [Pharo-dev] Pharo 5.0 Tiny Feature Request

by Peter Uhnak

On Sun, Jun 19, 2016 at 10:05:45AM +0200, Max Leske wrote: > > > On 19 Jun 2016, at 01:44, Sean P. DeNigris <sean(a)clipperadams.com> wrote: > > > > Max Leske wrote > >> Why donât you use OSSubprocess? > > > > My scripts require this functionality to be pre-loaded in a vanilla image. > > Ok, just wanted to check. > > > Also, on a minor note, some projects use OSProcess and I thought OSSP and > > OSP are incompatible. > > Not as far as I know. OSSubProcess uses the OSProcess VM plugin but is otherwise independent. They should be incompatible at the moment. There was a discussion about this some time ago in regards to Moose where you couldn't load & run OSSubprocess because OSProcess was already installed. It was solved by replacing OSProcess with this system call Sean is proposing to have in Pharo. Peter

June 19, 2016