September 2009 - Pharo-dev - lists.pharo.org

[Pharo-project] Building a VM and/or plugins

by Ken.Dickey

> "Schwab,Wilhelm K" <bschwab(a)anest.ufl.edu> ... > The question then becomes whether to connect to OpenSSL via FFI/Alien or to > make a plugin. Â One absolute requirement is that the image not turn to > stone when an address is wrong, a peer crashes, or a router gets unplugged > w/o warning. Â The most obvious solution to me (please speak up if you > disagree) is to connect/bind, read/write and disconnect on a > connection-specific OS thread. I would think a separate thread (or process) is required to handle and recover from bad pointer references. For general client/server multiplexing however, you might look at liboop for ideas: http://liboop.ofb.net/ "Liboop is a low-level event loop management library for POSIX-based operating systems. It supports the development of modular, multiplexed applications which may respond to events from several sources. It replaces the "select() loop" and allows the registration of event handlers for file and network I/O, timers and signals. Since processes use these mechanisms for almost all external communication, liboop can be used as the basis for almost any application." There are refs to other event dispatchers on the liboop web page. FYI, -KenD

Sept. 6, 2009

[Pharo-project] [ann] PharoCore-1.0-10439-BETA

by Marcus Denker

http://gforge.inria.fr/frs/download.php/23084/PharoCore-1.0-10439-BETA.zip -- Marcus Denker - http://marcusdenker.de PLEIAD Lab - Computer Science Department (DCC) - University of Chile

Sept. 6, 2009

Re: [Pharo-project] [Esug-list] ESUG Pictures

by Tudor Girba

Where exactly are they? :) Doru On 6 Sep 2009, at 20:53, Hernan Wilkinson wrote: > And here are the ones I took... > > On Sun, Sep 6, 2009 at 5:49 AM, James Foster > <Smalltalk(a)jgfoster.net> wrote: > http://picasaweb.google.com/WeybridgeWay/ESUG2009InBrestFrance# > > (Sorry for the cross-posting). > > Feel free to reuse, comment, or contact me as needed :) > > James Foster > _______________________________________________ > Esug-list mailing list > Esug-list(a)lists.esug.org > http://lists.esug.org/listinfo/esug-list > > _______________________________________________ > Esug-list mailing list > Esug-list(a)lists.esug.org > http://lists.esug.org/listinfo/esug-list -- www.tudorgirba.com "The coherence of a trip is given by the clearness of the goal."

Sept. 6, 2009

[Pharo-project] [update] #10439

by Marcus Denker

10439 ----- Issue 606: Problem with FreeType set to HintingNormal Issue 1068: Mac: Quit from host menu not working anymore Issue 1167: Failing test: TraitTest>>testPrinting reports tagged Milestone-1.0 is down to 19! http://code.google.com/p/pharo/issues/list?q=milestone%3D1.0 -- Marcus Denker - http://marcusdenker.de PLEIAD Lab - Computer Science Department (DCC) - University of Chile

Sept. 6, 2009

Re: [Pharo-project] [squeak-dev] some thoughts on coding practices.

by StÃ©phane Ducasse

> (2) I wanted to make a subclass, but then discovered >>>> RFBServer which >>>> I need actually hard codes a reference to InputEventSensor versus >>>> saying >>>> InputEventHandler defaultClass (or something) >> >> -> yes not hardcoding class is a good design and it should be >> reinforced. > > So is there a Lint rule for this or someone thinking of doing that > cleanup? Or a > bug? report? I have to check for the smallLint rule. For now open a ticket > >>> I fixed this in Refactoring-Spelling-lr.7, so that the spell checker >>> is only initialized when it is actually used. >>> >>> It looks to me that you are using a Pharo-Dev image, but I guess you >>> would rather use a Pharo-Core image? Maybe the Pharo-Core image >>> should >>> be placed more prominent on the download page, I've seen numerous >>> experienced people at ESUG downloading the "wrong" image. >> >> Yes which tells me that we should also iron pharo-dev. > > > Er well I thought the iPhone users would *like* to access the dev > image? Is that wrong? No but we got some glitches with the different browsers related to OB because code diverged and we would love to get it on track but david is working on papers :) Stef > > > -- > = > = > = > = > = > ====================================================================== > John M. McIntosh <johnmci(a)smalltalkconsulting.com> Twitter: > squeaker68882 > Corporate Smalltalk Consulting Ltd. http:// > www.smalltalkconsulting.com > = > = > = > = > = > ====================================================================== > > > > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

Sept. 6, 2009

Re: [Pharo-project] question about Issue 207: ImageSegment discoverActiveClasses strange error

by StÃ©phane Ducasse

Sheridan if you want you can commit in the pharoTaskForces project. Like that your work is not mixed with the incoming flow of the inbox that I try to keep as clean as possible. > > >> On Sep 5, 2009, at 3:10 AM, sheridan(a)sheridan-mahoney.com wrote: > > [...] > >> >>> I have found a report that seems quite interesting, and have done >>> some digging into the issue described here: >>> http://code.google.com/p/pharo/issues/detail?id=207 >>> ( ImageSegment discoverActiveClasses strange error ) >>> and now I have a few questions I was hoping to get feedback on. >> >> >> We would love to have a couple of tests for imagesegment!!!! >> Adrian did you publish the ones you did? > > I committed my image segment tests to the inbox (Tests-AdrianLienhard. > 19). There are some initial tests in ImageSegmentTest (one fails, > which documents a problem of loading symbols). I also implemented > support code that allows one to analyze a segment binary (i.e., to > iterate over the instances in the segment without instantiating the > objects). This should simplify writing more tests. > > Cheers, > Adrian > > [...] > > > ___________________ > http://www.adrian-lienhard.ch/ > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

Sept. 6, 2009

Re: [Pharo-project] [Esug-list] ESUG Pictures

by Hernan Wilkinson

And here are the ones I took... On Sun, Sep 6, 2009 at 5:49 AM, James Foster <Smalltalk(a)jgfoster.net> wrote: > http://picasaweb.google.com/WeybridgeWay/ESUG2009InBrestFrance# > > (Sorry for the cross-posting). > > Feel free to reuse, comment, or contact me as needed :) > > James Foster > _______________________________________________ > Esug-list mailing list > Esug-list(a)lists.esug.org > http://lists.esug.org/listinfo/esug-list >

Sept. 6, 2009

Re: [Pharo-project] [squeak-dev] some thoughts on coding practices.

by John M McIntosh

On 6-Sep-09, at 8:54 AM, StÃ©phane Ducasse wrote: >>> (2) I wanted to make a subclass, but then discovered >>> RFBServer which >>> I need actually hard codes a reference to InputEventSensor versus >>> saying >>> InputEventHandler defaultClass (or something) > > -> yes not hardcoding class is a good design and it should be > reinforced. So is there a Lint rule for this or someone thinking of doing that cleanup? Or a bug? report? >> I fixed this in Refactoring-Spelling-lr.7, so that the spell checker >> is only initialized when it is actually used. >> >> It looks to me that you are using a Pharo-Dev image, but I guess you >> would rather use a Pharo-Core image? Maybe the Pharo-Core image >> should >> be placed more prominent on the download page, I've seen numerous >> experienced people at ESUG downloading the "wrong" image. > > Yes which tells me that we should also iron pharo-dev. Er well I thought the iPhone users would *like* to access the dev image? Is that wrong? -- = = = ======================================================================== John M. McIntosh <johnmci(a)smalltalkconsulting.com> Twitter: squeaker68882 Corporate Smalltalk Consulting Ltd. http://www.smalltalkconsulting.com = = = ========================================================================

Sept. 6, 2009

[Pharo-project] Finding unpackaged methods

by Schwab,Wilhelm K

Hello all, Are there any existing tools that will, for example, find all methods marked with a specific author and not in a given list of packages? At least I _think_ that's what I want :) I have been extending an install script that I used to initially create the image I am using, and it might be best converted to a list of packages known to a builder that could also help look for methods that might be about to be left behind. I am thinking ahead to moving to a later beta and fearing leaving important code behind. Clearly I will keep my current image and be able to return to it, but some good tools could help avoid a lot of frustration. Some of my concern is over a few strange things that I've noticed about SIF's response to Dolphin's virtual categories. Bill

Sept. 6, 2009

Re: [Pharo-project] question about Issue 207: ImageSegment discoverActiveClasses strange error

by Adrian Lienhard

On Sep 5, 2009, at 10:46 , StÃ©phane Ducasse wrote: > On Sep 5, 2009, at 3:10 AM, sheridan(a)sheridan-mahoney.com wrote: [...] > >> I have found a report that seems quite interesting, and have done >> some digging into the issue described here: >> http://code.google.com/p/pharo/issues/detail?id=207 >> ( ImageSegment discoverActiveClasses strange error ) >> and now I have a few questions I was hoping to get feedback on. > > > We would love to have a couple of tests for imagesegment!!!! > Adrian did you publish the ones you did? I committed my image segment tests to the inbox (Tests-AdrianLienhard. 19). There are some initial tests in ImageSegmentTest (one fails, which documents a problem of loading symbols). I also implemented support code that allows one to analyze a segment binary (i.e., to iterate over the instances in the segment without instantiating the objects). This should simplify writing more tests. Cheers, Adrian [...] ___________________ http://www.adrian-lienhard.ch/

Sept. 6, 2009