Squeak-dev September 2001

squeak-dev@lists.squeakfoundation.org

168 participants
408 discussions

Semaphore>>critical: ?
by G�ran Hultgren 03 Sep '01

03 Sep '01

Hi guys! I am hacking away on a Flow based client/server thingy and needed to use a Semaphore as a mutex. So I did something like this: myMutex critical: [ ...code code... ^self ...code code...] ...and oops, got bitten by the fact that the Semaphore does not get signalled when I have a return in my critical code block! Of course, it was simple to rewrite the code without returns but... Is there any particular reason (performance etc) that prevents us from changing the implementation of #critical: to this (thus not missing to signal the Semaphore)?: critical: mutuallyExcludedBlock "Evaluate mutuallyExcludedBlock only if the receiver is not currently in the process of running the critical: message. If the receiver is, evaluate mutuallyExcludedBlock after the other critical: message is finished." self wait. ^mutuallyExcludedBlock ensure: [self signal] The funny thing is that this seems to work EXCEPT for one Unit test I have which seems to "hang" on the "self signal" part above. Funny enough, if I alt-. it and then just "proceed" it moves on! Obviously I am missing some piece of the puzzle, right? In any case this should perhaps be mentioned in the comment to #critical: regards, G�ran PS. This is in a rather old image pdated to 2915 with JMM's Flow and SUnit 2.7.1. DS ===== G�ran Hultgren, goran.hultgren(a)bluefish.se GSM: +46 70 3933950, http://www.bluefish.se "Department of Redundancy department." -- ThinkGeek __________________________________________________ Do You Yahoo!? Get email alerts & NEW webcam video instant messaging with Yahoo! Messenger http://im.yahoo.com

3 4

[Modules] a summary of joseph's work
by Stephane Ducasse 03 Sep '01

03 Sep '01

Hi I discussed and saw a lot of demoes of joseph's work at ESUG and I would like to summarize it. Joseph's idea is to find a minimal structure that support the modularisation and on top of which other can build - Based on a minimal semantic model: a layer on top of the smalltalk meta-model (implemented lazzily) that represents declaratively Smalltalk elements. Around 15 classes - a Module/Cluster structure influenced by Team/V in which classes and classes extensions can go. - a set of errors that can be captured to resolve differently conflicts: the base to build other tools such as Colleage. - a modularizer: this class allows one to define the basic structure of the modules and automatically move classes in those modules. This is an extremely important aspect to let people explore different ways of organizing the code. - The possibility to have envy like prerequisites validation or to let it lose so that we can hack fast something. - a set of Browsers for the module ....: really cool to already have them! I hope I did not forget anything. The point for me is that being able to compile Squeak from VA or VW is just a proof of concept that we could get a really small image, and even less minimal if we would stream in byte-code and remove the compiler ;) Stef I hope this helps

1 0

About Celeste
by raymondasselin＠sympatico.ca 03 Sep '01

03 Sep '01

Is there a way to not display, in the bottom pane of Celeste email browser, all the stuff like 'delivered-to: squeak-dev(a)lists.squeakfoundation.org reply-to: squeak-dev(a)lists.squeakfoundation.org list-help: <mailto:squeak-dev-request@lists.squeakfoundation.org?subject=help> list-post: <mailto:squeak-dev@lists.squeakfoundation.org> list-subscribe: <http://lists.squeakfoundation.org/listinfo/squeak-dev>, <mailto:squeak-dev-request@lists.squeakfoundation.org?subject=subscribe> list-id: The general-purpose Squeak developers list <squeak-dev.lists.squeakfoundation.org> list-unsubscribe: <http://lists.squeakfoundation.org/listinfo/squeak-dev>, <mailto:squeak-dev-request@lists.squeakfoundation.org?subject=unsubscrib e> list-archive: <http://lists.squeakfoundation.org/pipermail/squeak-dev/>' for each message and keeping this optional? I thought somebody did something about it but can't find it on 'http://macos.tuwien.ac.at:9009/Server.home' ...may be this was a dream?

1 0

About Celeste
by raymondasselin＠sympatico.ca 03 Sep '01

03 Sep '01

1 0

SPSBrowser and RefactoringBrowser as Squeak image segments on BSS
by Torsten.Bergmann＠phaidros.com 03 Sep '01

03 Sep '01

Hi, I converted the code for the SPSBrowser and the RefactoringBrowser into Squeak Image segments for Bobs Super swiki (BSS). So it's much easier to load them into Squeak. SPSBrowser http://209.143.91.36/super/425 Refactoring Browser http://209.143.91.36/super/426 Note these projects require an updated 3.1alpha image. To open the project either browse the pages with scamper and click on the .pr links or use the orange "find" buttom at the bottom of your squeak image to locate them on BSS. Torsten

1 0

[ANN] Fixes/enhancements are now being reviewed for inclusion into the base image
by Doug Way 03 Sep '01

03 Sep '01

This is an announcement to let everyone know that a small group of people is now working on culling and reviewing fixes & enhancements, from the Squeak list, for inclusion into the base Squeak image. The process is similar to what has been done before with individual volunteers culling fixes from the list, except now a group of volunteers is in place. This should help keep the culling/inclusion process going on a regular basis, since it won't be dependent on the availability of a single overworked volunteer. (A culling/inclusion of submissions from the list hasn't happened in about 8 months, partly as a result of this problem.) Also, in theory, the decisions as to what fixes & enhancements to include should be more balanced, coming from a group. :-) Squeak Central will still have the final say on what is ultimately incorporated into the image, of course, but the bulk of the culling and incorporating work will be done by the volunteer group. (By the way, I'm a member of the group, along with Stefan Matthias Aust, Bob Arning, Dan Ingalls (as the SqC rep) and possibly a couple of others. Hm, I guess I still need to add this group as a possible Squeak Foundation project to the SqF swiki.) Anyway, it's been quite awhile since the last official culling/inclusion sweep, December 3rd of last year to be exact. We'll be taking a look at all of the submissions on the SQFIXES archive since then, although probably only a brief look at many of the older ones. (The SQFIXES archive is at http://swiki.gsug.org:8080/SQFIXES/) So, in order to help us do this: 1. If there are any fixes or enhancements, especially older ones, that are really important to you and you are afraid might be overlooked, please let us know about them. (You can reply in this thread.) 2. If you authored a fix/enhancement that is older and no longer working in 3.1alpha-4261, it would help us to include it if you brought it up to date to work with 3.1alpha-4261 and reposted it to the list. (under the same subject as before, if possible) 3. If you know of any important fixes out there which are not included in the SQFIXES archive for any reason (e.g. they weren't posted with the right subject tags, or they were never posted to the list at all), please post them to the list with the appropriate [FIX] or [ENH] or [GOODIE] subject tag. These extra steps are hopefully a one-time thing. In the future, the culling/inclusion process should be frequent enough that these steps won't be necessary. Thanks! - Doug Way dway(a)riskmetrics.com

2 1

[ANN] Refactoring Browser port to Squeak goes beta
by danielv＠netvision.net.il 03 Sep '01

03 Sep '01

So, the Squeak RB port team reports: Our goal was to advance the RB port as done by Bob Hartwig and the people at the previous CS from alpha state to beta. At a high level, this means that we expect just about anyone should be able to load the Refactoring Browser and do all their daily work on it. Since the port creates a subclass of the regular browser, this should be a definite improvement on what's available now, even if not everything is perfect. Specifics - the Refactoring Browser as we release it, loads cleanly on an updated image (4164 and up), lots of refactorings work (almost everything done over methods and variables), and the critical UI support code is there. So - we urge everyone to take a look, especially people familiar with the refactoring browser (to find problems) and also people that aren't (because everyone have the tools to make code better). Ok, so where's the code? The code is on the squeak foundation ftp site, at - ftp://ftp.squeakfoundation.org/pub/packages/RefactoringBrowser/ What you need are 4 files, to be loaded in this order - ClassBuilder.1.cs - a tiny patch to the ClassBuilder to avoid a problem loading one of the next files. I'm not sure this is a good permanent solution, but it's very meek and harmless. RBModel_26.cs - slightly modified version of the refactoring engine as appears on BobH's site. Note this file is pretty big and takes a while to load. RBUI_18.cs - exact copy of BobH's file RBPortBeta.6.cs - Our modifications. and then, RefactoringBrowser openBrowser Other files - Bob Hartwig's site includes a UI for the Lint tool (find bugs, code smells, and oddities automatically) which is uses the RB model. We haven't tested or modified it. You might also be interested in downloading the latest SUnit framework and tests for the model classes (especially if you want to help improve the browser), this is also described on Bob Hartwig's site. What isn't there (and why the code is beta, ie, not ready to go into the base image) - * Lots of obsolete support classes and changes that are deeper than needed to the base image are still there. * No rewrite rule editor UI yet. * Operations on classes not tested - some definitely don't work. * Extract Method To Component doesn't work yet (requires more UI work, soon) * The UI code is written mostly in the tradition Browser big ball of mud style. This should definitely be refactored over time... * Lots of features we consider secondary (or didn't remember that existed) - the class hierarchy view for example - are simply not there. You want 'em, say so. we got all kinds of support, technical and moral, from other Campers: Joseph Pelrine Markus Denker Hans Martin Mosner Roger Whitney The Squeak RB at CS3 team - Markus Gaelli Michael Pruemm Frank (I didn't get his last name) and Yours Squeakily, Daniel Vainsencher PS - For those of you who have never been to CampSmalltalk (like until this time) - you should. It's great to meet the wider Smalltalk family, and generally work together.

2 2

Re: [Fix] the #mailOut feature of the ChangeSorter still refers to the old address. This fixes it
by danielv＠netvision.net.il 03 Sep '01

03 Sep '01

Third fix to the same thing... We also have had the updated header cutters for Celeste for a while, and the mail out bug reports also. Could the update culling guys please put this matter out of our misery? please? "Rob Withers" <rwithers12(a)mediaone.net> wrote: > This is a multi-part message in MIME format. > > ------=_NextPart_000_008F_01C12E9D.042D37E0 > Content-Type: text/plain; > charset="iso-8859-1" > Content-Transfer-Encoding: 7bit > > > > ------=_NextPart_000_008F_01C12E9D.042D37E0 > Content-Type: application/octet-stream; > name="ChangeSet-mailOut.1.cs" > Content-Transfer-Encoding: quoted-printable > Content-Disposition: attachment; > filename="ChangeSet-mailOut.1.cs" > > 'From Squeak3.1alpha of 7 March 2001 [latest update: #4282] on 26 August = > 2001 at 5:29:16 pm'!=0D=0D!ChangeSet methodsFor: 'fileIn/Out' stamp: = > 'rww 8/26/2001 17:28'!=0DmailOut=0D "File out the receiver, to a file = > whose name is a function of the =0D =0A= > change-set name and either of the date & time or chosen to have a =0D =0A= > unique numeric tag, depending on the preference =0D =0A= > 'sequentialChangeSetRevertableFileNames'."=0D | subjectPrefix slips = > message compressBuffer compressStream data compressedStream = > compressTarget |=0D (Smalltalk includesKey: #Celeste)=0D ifFalse: [^ = > self notify: 'no mail reader present'].=0D subjectPrefix _ self = > chooseSubjectPrefixForEmail.=0D self checkForConversionMethods.=0D = > Cursor write=0D showWhile: ["prepare the message"=0D message _ = > MailMessage empty.=0D message setField: 'from' toString: Celeste = > userName.=0D message setField: 'to' toString: = > 'squeak-dev(a)lists.squeakfoundation.org'.=0D message setField: = > 'subject' toString: subjectPrefix , name.=0D message=0D body: = > (MIMEDocument=0D contentType: 'text/plain'=0D content: = > (String=0D streamContents: [:str | =0D str nextPutAll: = > 'from preamble:';=0D cr;=0D cr.=0D self = > fileOutPreambleOn: str])).=0D "Prepare the gzipped data"=0D data _ = > data _ WriteStream on: String new.=0D data header.=0D self = > fileOutPreambleOn: data.=0D self fileOutOn: data.=0D self = > fileOutPostscriptOn: data.=0D data trailer.=0D data _ ReadStream on: = > data contents.=0D compressBuffer _ ByteArray new: 1000.=0D = > compressStream _ GZipWriteStream on: (compressTarget _ WriteStream=0D = > on: (ByteArray new: 1000)).=0D [data atEnd]=0D whileFalse: = > [compressStream=0D nextPutAll: (data nextInto: compressBuffer)].=0D = > compressStream close.=0D compressedStream _ ReadStream on: = > compressTarget contents asString.=0D message addAttachmentFrom: = > compressedStream withName: name , '.cs.gz'.=0D CelesteComposition = > openForCeleste: Celeste current initialText: message text].=0D = > Preferences suppressCheckForSlips=0D ifTrue: [^ self].=0D slips _ self = > checkForSlips.=0D (slips size > 0=0D and: [self confirm: 'Methods in = > this fileOut have halts > or references to the Transcript > or other ''slips'' in them. > Would you like to browse them?'])=0D ifTrue: [Smalltalk = > browseMessageList: slips name: 'Possible slips in ' , name]! !=0D > ------=_NextPart_000_008F_01C12E9D.042D37E0--

1 0

Re: Semaphore>>critical: ?
by Leandro Caniglia 02 Sep '01

02 Sep '01

=?ISO-8859-1?Q?"G=F6ran"?= Hultgren <gohu(a)rocketmail.com> wrote: > Is there any particular reason (performance etc) that prevents > us from changing the implementation > of #critical: to this (thus not missing to signal the Semaphore)?: > > critical: mutuallyExcludedBlock > "Evaluate mutuallyExcludedBlock only if the receiver is not currently in > the process of running the critical: message. If the receiver is, evaluate > mutuallyExcludedBlock after the other critical: message is finished." > > self wait. > ^mutuallyExcludedBlock ensure: [self signal] In GemSqueak we implemented this method just as you. We didnt find any problem with the ensure: (and many without it). > The funny thing is that this seems to work EXCEPT for one Unit test > I have which seems to "hang" on the "self signal" part above. Funny > enough, if I alt-. it and then just "proceed" it moves on! We've also observed the same. /Leandro

1 0

[ENH] BraveEmptyTrash
by danielv＠netvision.net.il 02 Sep '01

02 Sep '01

from preamble: "Change Set: BraveEmptyTrash Date: 2 September 2001 Author: Daniel Vainsencher When emptying your Celeste trash, allow the user to choose whether to delete only messages that appear only in the trash, or to delete all messages in it, even if filed in other categories. This allows one to find duplicates, delete them all, and empty the trash, without caring whether I filed any of them already by mistake."!

1 0

Jump to page: