- Pharo-users - lists.pharo.org

Re: [Pharo-users] Threads safety in Pharo

by Alidra Abdelghani

Thanks Guille for your answer, >From what I understand, it is mainly a synchronisation problem. Why cant' existing synchronisation mechanisms (semaphores and mutex) be used? Abdelghani > On 31 Jul 2017, at 18:44, Guillermo Polito <guillermopolito(a)gmail.com> wrote: > > I believe there is no such a document. It would be however interesting to investigate it a bit deeper. In general, the problem we talk about when we talk about thread safety is the following: Can we run a workspace in a separate thread than a browser and provide correct results? Can we run two browsers concurrently and change code from both of them? > > The thread safety problem has several levels I'd say: > > - the kernel of the language (classes, methods, compilation, and so on) cannot safely be modified while other threads are running. Thus, if we compile a new version of a method in a thread, that could break a separate thread that was using that method. > > - there are several libraries using global state. We should investigate if they work correctly when using them concurrently. > E.g., source code management is not thread safe. This could cause a source code corruption if several methods are modified concurrently. > Like that, we sh n ould investigate all libraries and see what should be adapted and if there is a need at all. > > - in a third level, general core libraries (collections, networking, files, etc) are not designed to be thread safe and that is natural. For example, most of the time you don't want that your collection is accessed concurrently. For such cases, some libraries could provide some extensions/wrappers/whatsoever to provide some synchronization mechanism. Otherwise it is the responsibility of the user to synchronize usages with semaphores/mutexes. > > > On Mon, Jul 31, 2017 at 1:38 AM, Alidra Abdelghani via Pharo-users <pharo-users(a)lists.pharo.org <mailto:pharo-users@lists.pharo.org>> wrote: > > > ---------- Forwarded message ---------- > From: Alidra Abdelghani <alidrandco(a)yahoo.fr <mailto:alidrandco@yahoo.fr>> > To: pharo-users(a)lists.pharo.org <mailto:pharo-users@lists.pharo.org> > Cc: "StÃ©phane Ducasse" <stephane.ducasse(a)inria.fr <mailto:stephane.ducasse@inria.fr>>, farid arfi <arfi_f(a)hotmail.com <mailto:arfi_f@hotmail.com>> > Bcc: > Date: Mon, 31 Jul 2017 01:38:58 +0200 > Subject: Threads safety in Pharo > Hi, > > Somebody once evoked the problem of threads safety in Pharo. With a friend of mine who is expert in formal methods and process scheduling, we would like to have a look on it. > Does anyone knows a good document describing the problem of Pharo with threads safety or at least any document that we can start with? > > Thanks in advance, > Abdelghani > > > > > > > -- > > Guille Polito > > Research Engineer > French National Center for Scientific Research - http://www.cnrs.fr <http://www.cnrs.fr/> > > > Web: http://guillep.github.io <http://guillep.github.io/> > Phone: +33 06 52 70 66 13

Aug. 4, 2017

How to remove packages and reclaim image size?

by Tim Mackinnon

Hi - Iâm wondering how you best remove packages in the image to reclaim space? I am a bit confused because I see 2 package systems MCPackage and RPackage? And I see 2 different ways of removing things: (MCPackage named: name) unload (RPackage organizer packageNamed: name) unregister So which is the correct one to use (or are they the same?). Iâve tried both, but then when I save my image (headless in a script) - it doesnât seem to make any difference in image size. So Iâm also wondering if there is an extra step to forcing size reduction. I have tried: 3 timesRepeat: [ Smalltalk garbageCollect. Smalltalk cleanOutUndeclared. Smalltalk fixObsoleteReferences]. Which is what I saw in some ImageCleaner tool, but again I donât see any image size difference, which makes me wonder if Iâm doing the wrong thing or missing a step? Tim

Aug. 4, 2017

Re: [Pharo-users] Glorp reconnection management (was: Re: Glorp: how to reconnect after image shutdown / load?)

by Pierce Ng

On Thu, Aug 03, 2017 at 03:27:16PM +0200, Herby VojÄÃk wrote: > Looking at state of the art code: > GlorpSession >> loginIfError: aBlock > | result | > result := self accessor loginIfError: aBlock. > system platform characterEncoding: accessor encoding. > ^result In the line "result := self accessor loginIfError: aBlock", "self accessor" is not an instance of DatabaseAccessor, because DatabaseAccessor>>loginIfError: aBlock self subclassResponsibility "self accessor" is an instance of PharoDatabaseAccessor, and PharoDatabaseAccessor>>loginIfError: aBlock self log: 'Login'. self databaseDriver: self connectionClass new. self execute: [ self databaseDriver connect: currentLogin. currentLogin secure ifTrue: [ currentLogin discardPassword ] ] ifError: aBlock. self log: 'Login finished' PharoDatabaseAccessor>>isLoggedIn self databaseDriver ifNil: [^ false]. ^ self databaseDriver isConnected Connecting to my instance of Reddit (actually RedditSt20 using SQLite not PostgreSQL), adding a link and then clicking refresh, Transcript shows: Login Login finished SELECT t1.id, t1.url, t1.title, t1.created, t1.points FROM REDDIT_LINKS t1 ORDER BY t1.points DESC LIMIT 20 an OrderedCollection() SELECT t1.id, t1.url, t1.title, t1.created, t1.points FROM REDDIT_LINKS t1 ORDER BY t1.created DESC LIMIT 20 an OrderedCollection() Begin Transaction INSERT INTO REDDIT_LINKS (url,title,created,points) VALUES (?,?,?,?) an Array('http://planet.smalltalk.org' 'Planet Smalltalk' 2017-08-04T22:25:14.558667+08:00 0) SELECT last_insert_rowid() (0.0 s) Commit Transaction SELECT t1.id, t1.url, t1.title, t1.created, t1.points FROM REDDIT_LINKS t1 ORDER BY t1.points DESC LIMIT 20 an OrderedCollection() SELECT t1.id, t1.url, t1.title, t1.created, t1.points FROM REDDIT_LINKS t1 ORDER BY t1.created DESC LIMIT 20 an OrderedCollection() Login happened once. Clicking "New Session": Logout Logout finished Login Login finished SELECT t1.id, t1.url, t1.title, t1.created, t1.points FROM REDDIT_LINKS t1 ORDER BY t1.points DESC LIMIT 20 an OrderedCollection() SELECT t1.id, t1.url, t1.title, t1.created, t1.points FROM REDDIT_LINKS t1 ORDER BY t1.created DESC LIMIT 20 an OrderedCollection() Clicking "Refresh" in the application: SELECT t1.id, t1.url, t1.title, t1.created, t1.points FROM REDDIT_LINKS t1 ORDER BY t1.points DESC LIMIT 20 an OrderedCollection() SELECT t1.id, t1.url, t1.title, t1.created, t1.points FROM REDDIT_LINKS t1 ORDER BY t1.created DESC LIMIT 20 an OrderedCollection() Pierce

Aug. 4, 2017

Re: [Pharo-users] Threads safety in Pharo

by Alidra Abdelghani

> On 01 Aug 2017, at 09:36, StÃ©phane Ducasse <stephane.ducasse(a)inria.fr> wrote: > > To complement. > > What would be fun is to investigate the implementation of ProcessScheduler, Process and check the implementation. > > Abdel there is a nice presentation of the processor and process in the Squeak french book. Great, I taught some of theses concepts to my students some times ago and we did some fun lab sessions with the Linux scheduler > This was on my todo to turn it into a book chapter for Pharo. > > So may be we could work together to write a little booklet on this aspect to get started. I would love to :D > > Stef

Aug. 4, 2017

Re: [Pharo-users] Threads safety in Pharo

by Alidra Abdelghani

Hi Ben, > On 01 Aug 2017, at 13:31, Ben Coman <btc(a)openinworld.com> wrote: > > Not sure I'll have what you're looking for, but to start, do you mean Pharo's green threads or vm native threads? I believe this is about Pharo threads. I am not quite sure of what the vm native threads are or if there is a safety problem with them. But still, it could be interesting to investigate :) Thanks Abdelghani > cheers -ben > > On Mon, Jul 31, 2017 at 7:38 AM, Alidra Abdelghani via Pharo-users <pharo-users(a)lists.pharo.org <mailto:pharo-users@lists.pharo.org>> wrote: > > > ---------- Forwarded message ---------- > From: Alidra Abdelghani <alidrandco(a)yahoo.fr <mailto:alidrandco@yahoo.fr>> > To: pharo-users(a)lists.pharo.org <mailto:pharo-users@lists.pharo.org> > Cc: "StÃ©phane Ducasse" <stephane.ducasse(a)inria.fr <mailto:stephane.ducasse@inria.fr>>, farid arfi <arfi_f(a)hotmail.com <mailto:arfi_f@hotmail.com>> > Bcc: > Date: Mon, 31 Jul 2017 01:38:58 +0200 > Subject: Threads safety in Pharo > Hi, > > Somebody once evoked the problem of threads safety in Pharo. With a friend of mine who is expert in formal methods and process scheduling, we would like to have a look on it. > Does anyone knows a good document describing the problem of Pharo with threads safety or at least any document that we can start with? > > Thanks in advance, > Abdelghani > > > >

Aug. 4, 2017

Re: [Pharo-users] Compiling lots of method source without debugger

by Henrik Sperre Johansen

Patrick Scherer wrote > Hello everyone, > > This is my first time here so I hope this is the right place to ask > questions. > > In an attempt to move lots of code from VA Smalltalk to Pharo, I need to > compile a lot of methods whose code may or may not work due to missing > classes in Pharo etc. > I want to be able to compile these methods programatically without > interruptions (using method source strings only) and gather all compiler > errors that occur. What would be the best way to go about this? > > Also: How do I control in which package a compiled method ends up if the > target class is in a different package (Method extension)? > > Any help would be greatly appreciated! > > Kind regards, > Patrick Scherer IIRC, Stephanne recently listed preservation of unloaded supers / extensions of unloaded classes as an area the INRIA group was working on, with a paper scheduled for ESUG, he might be able to help. I also recently went through roughly the same to import VSE code (for inspection in Moose, not execution). Can't say it was the best way, but I ended up creating a custom CodeImporter with several error/notification handlers around the "normal" loading code, and modifying the base system to raise said exceptions at appropriate places (rather than as currently, invoking dialogs directly) For instance, a MissingClass notification in MethodChunk >> handleMissingBehavior, and a MissingPool from TClass >> #sharing:, (both with defaultAction implemented as per current dialog-raising behavior, so they work the same unless explicitly handled). For my case, it was sufficient to have the handlers put classes/pools auto-generated by handlers in predetermined packages. The biggest pain / time drain was dealing with creating instvars in autogenerated classes when referenced from imported code (ie in methods on classes with missing superclasses) so the analysis wouldn't treat them as unknown globals, so if you don't initially need that, I'd skip it. I also did a compiler plugin to be able to rewrite certain selectors on import (mostly base class extensions from VSE clashing with existing implementations in Pharo), if you go that route, beware that the compiler doesn't really like you actually redefining source from outside an editor, and there will be some pains to have it store the correct(ed) source strings. To create a method as an extension in Package MyX, you need to give it a categoriy of *MyX (with an optional -originalCategory suffix if you want that) Good luck! Cheers, Henry -- View this message in context: http://forum.world.st/Compiling-lots-of-method-source-without-debugger-tp49… Sent from the Pharo Smalltalk Users mailing list archive at Nabble.com.

Aug. 4, 2017

Compiling lots of method source without debugger

by Patrick Scherer

Hello everyone, This is my first time here so I hope this is the right place to ask questions. In an attempt to move lots of code from VA Smalltalk to Pharo, I need to compile a lot of methods whose code may or may not work due to missing classes in Pharo etc. I want to be able to compile these methods programatically without interruptions (using method source strings only) and gather all compiler errors that occur. What would be the best way to go about this? Also: How do I control in which package a compiled method ends up if the target class is in a different package (Method extension)? Any help would be greatly appreciated! Kind regards, Patrick Scherer

Aug. 4, 2017

Re: [Pharo-users] Iceberg and removing packages

by Esteban Lorenzano

> On 3 Aug 2017, at 19:13, Esteban A. Maringolo <emaringolo(a)gmail.com> wrote: > > How can I know which version it is loaded? > > My system version is: "Pharo6.0 of 13 May 2016 update 60508â this is old 6.0, you need 6.1(which is 60510) but then you need the new VM :S I checked: 0.5 fixes the problem of removing, but it fixes it bad: it removes correctly, but sync window shows no-changes (and commit will do the remove anyway). I will fix this once I come backâ¦ Esteban > > > Regards. > > > Esteban A. Maringolo > > 2017-08-03 6:27 GMT-03:00 Esteban Lorenzano <estebanlm(a)gmail.com <mailto:estebanlm@gmail.com>>: > so, this is probably a bug. > are you using latest iceberg version? (it should be 0.5.5â¦ the one present on P6.1) > > Esteban > > > > On 3 Aug 2017, at 00:32, Tim Mackinnon <tim(a)testit.works> wrote: > > > > I just hit the same problem - I wanted to remove a package that I no longer wanted - I went to the iceberg packages tab for the project and picked remove from disk and then I did a synchronise and commit. At the time I did notice that it didnât show a package removal change (but I assumed that as it was gone, maybe you donât show that?). > > > > Tim > > > >> On 31 Jul 2017, at 13:30, Esteban Lorenzano <estebanlm(a)gmail.com <mailto:estebanlm@gmail.com>> wrote: > >> > >> Did you commit? > >> > >>> On 31 Jul 2017, at 14:06, Esteban A. Maringolo <emaringolo(a)gmail.com <mailto:emaringolo@gmail.com>> wrote: > >>> > >>> That's exactly what I did. But after synchronizing the repo, the > >>> package wasn't listed on Iceberg but it still was in in the repo. > >>> It happened with another package in another repo as well. > >>> > >>> If it happens again I'll let you know. > >>> > >>> Regards! > >>> > >>> Esteban A. Maringolo > >>> > >>> > >>> 2017-07-31 7:03 GMT-03:00 Esteban Lorenzano <estebanlm(a)gmail.com <mailto:estebanlm@gmail.com>>: > >>>> Hi, > >>>> > >>>> In iceberg, you have to go to package tab and say "remove package from disk" > >>>> > >>>> Esteban > >>>> > >>>>> On 30 Jul 2017, at 17:01, Esteban A. Maringolo <emaringolo(a)gmail.com <mailto:emaringolo@gmail.com>> wrote: > >>>>> > >>>>> I got into the pharo-local/iceberg/... and git rm'ed the directory, > >>>>> commited and synchronized the project in Iceberg. > >>>>> > >>>>> I hope it doesn't break anything since I don't know how much "magic" > >>>>> does Iceberg behind the scenes other than automating the git commands > >>>>> and providing a centralized UI. > >>>>> > >>>>> What is the "This is all you need to read to understand Iceberg?" > >>>>> document I should read? I read its wiki, but it seems there is more to > >>>>> go. > >>>>> > >>>>> Regards! > >>>>> > >>>>> Esteban A. Maringolo > >>>>> > >>>>> > >>>>> 2017-07-30 11:28 GMT-03:00 Peter Uhnak <i.uhnak(a)gmail.com <mailto:i.uhnak@gmail.com>>: > >>>>>> This was supposedly fixed in April https://github.com/pharo-vcs/iceberg/issues/317 <https://github.com/pharo-vcs/iceberg/issues/317> > >>>>>> > >>>>>> however I had the same issue ~two months ago, so I had to delete the packages by hand. > >>>>>> > >>>>>> P > >>>>>> > >>>>>> > >>>>>>> On Sun, Jul 30, 2017 at 11:04:20AM -0300, Esteban A. Maringolo wrote: > >>>>>>> Hi, > >>>>>>> > >>>>>>> I'm playing around with Iceberg in Pharo 6, and even when I find the > >>>>>>> workflow streamlined, but since it doesn't map 1:1 with git workflow > >>>>>>> from other IDEs or command line, I find myself not knowing how to do > >>>>>>> certain tasks. > >>>>>>> > >>>>>>> One thing that happened is that I published a few packages to one of > >>>>>>> my repos in Github, then I decided to remove one of the packages from > >>>>>>> the repo, then I synchronized it but in the repo there is still is the > >>>>>>> package folder for the package I removed. > >>>>>>> > >>>>>>> What should I do to definitely remove those files from the commit? > >>>>>>> > >>>>>>> Regards! > >>>>>>> > >>>>>>> Esteban A. Maringolo > >>>>>>> > >>>>>> > >>>>> > >>>> > >>> > >> > > > > > > >

Aug. 4, 2017

Re: [Pharo-users] Settings - Server Mode - should I use it?

by Tim Mackinnon

Hey Ben - thanks for sharing this with me - you know, I never knew you could right click on the settings and browseâ¦ Iâve always (only partially successfully) used haloâs on ui items and tried to dig in that wayâ¦ It turns out that in the minimal images that Pavel has been churning out - WorldState is not there (which makes sense, as you would expect all of the UI pieces to be stripped out). Tim > On 2 Aug 2017, at 18:09, Ben Coman <btc(a)openInWorld.com> wrote: > > hi Tim, > > This is Pharo! You can dig in to understand it yourself. ;) > 1. On the setting, right-click and "Browse". > 2. Scroll down to... label: 'Server mode' > then on #serverMode, right-click > Code Search > Implementors of it > 3. Looking at WorldState>>serverMode, click "Senders" > and you'll come to WorldState>>interCyclePause: > Look for senders of this and inspect MinCycleLapse. > > Basically that setting makes the UI loop delay for a longer time ==> less CPU > but I don't know whether that loop still runs in headless mode. > I guess if you are bootstrapping your Lambda image, you can choose. > > But since your Lambda is meant to finish quite quickly, alternatively maybe during image shutdown you could set a flag such that next startup #interCyclePause: blocks for 500ms and if the image exits cleanly and quickly, #doOneCycleNowFor: is never executed at all. > > cheers -ben > > On Wed, Aug 2, 2017 at 10:39 PM, Tim Mackinnon <tim(a)testit.works <mailto:tim@testit.works>> wrote: > Hi - I hadnât looked in the settings browser much before, but I was looking at what settings there were and noticed âserver modeâ. > > Should I be enabling this when building a headless server image to run on Lambda? I saw a 2016 post by John Brant that indicated it made a difference when doing some benchmarking in the image (presumably headful) - but Iâm wondering if when running headless on a server whether it really makes a difference. > > Has anyone played with this much before? > > Tim >

Aug. 4, 2017

Re: [Pharo-users] Syntax Error: Message pattern expected

by Stephane Ducasse

It may happen that you are trying to load an old package with a syntax error. We should improve the compiler infrastructure because I hate such error not You can bring the menu on the syntax error window and bring the debugger. (yes it is totally hidden and I hate it too). Stef On Wed, Aug 2, 2017 at 9:22 PM, jb <brauer(a)nordakademie.de> wrote: > Hi everyone! > > I get this error when I try to load a package from my Monticello rep. I have > never seen this before and have not a clue where to look for the mistake. > > The error message window show: > Object <none> > > Any hints what to do? > > Johannes > > > > -- > View this message in context: http://forum.world.st/Syntax-Error-Message-pattern-expected-tp4958351.html > Sent from the Pharo Smalltalk Users mailing list archive at Nabble.com. >

Aug. 4, 2017