- Pharo-dev - lists.pharo.org

Re: [Pharo-project] Debugger fixes was Re: Getting Pharo 1.0 released

by Michael Roberts

Stef, all, a process question... so I am merging closure fixes from squeak trunk into pharo-core.... Prior to doing this in MC I had filed out Eliot's first exceptions change set from a squeak trunk image giving me the diff that was applied to the base squeak-trunk image during the system update. Comparing this to Pharo-core I saw we already had all the changes his specific cs/MC made so i marked that initially on the tracker as complete, separately you did the same in a different issue. However, this is not the complete difference that could be applied as you commented on the tracker. In principle unless we have access to the base image from which the fixes came, assuming they did not come from squeak trunk, we don't know if squeak trunk already had fixes required by Eliot and also in his base (presume a Qwaq image), therefore not needing to be merged, or if the patch applied to Squeak trunk contains everything to fix the job at hand (closure integration). So, in principle we have to merge in everything and I guess this could require quite a bit of review and testing. this could apply to any fixes we want to apply from Squeak trunk so i thought I would ask. obviously it's up to the merger to make these choices, but now squeak trunk is moving at a good pace I guess we should think about it. what would be really helpful in the merge tool would be to see the ancestry of the source so you could see which named MC package it came from and where it was in the ancestry chain. i.e if it is not the head package. Gary, any spare time to add features to the merge tool? thanks, Mike

Sept. 7, 2009

Re: [Pharo-project] request for the new merge tool :)

by Michael Roberts

the merge tool is driving me a little mad... How hard would it be to make this change? Also is there any documentation available on the polymorph merge tool? thanks, Mike On Mon, Jul 6, 2009 at 4:03 PM, Damien Cassou<damien.cassou(a)gmail.com> wrote: > 2009/7/6 Gary Chambers <gazzaguru2(a)btinternet.com>: >> For reference, when merging (rather than comparing) the left hand side is >> always what's currently in the image and the right hand side that which is >> incoming from the package selected to be merged. > > That note would already be appreciated in the tool as a comment. > > -- > Damien Cassou > http://damiencassou.seasidehosting.st > > "Lambdas are relegated to relative obscurity until Java makes them > popular by not having them." James Iry > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >

Sept. 7, 2009

Re: [Pharo-project] problem with running Pharo under Solaris 10 9/05

by Bryce Kampjes

On Mon, 2009-09-07 at 15:58 +0200, ivaylo.ganchev(a)univ-paris8.fr wrote: > Hello. > > I can't manage to run the virtual machine under solaris 10. I have > permanently the following output: > > -bash: ./squeak: Invalid argument > > Here is the output when I try to run the image with the vm of squeak: > > iconv_open: Invalid argument > This interpreter (vers. 6502) cannot read image file (vers. 1746468864). > Press CR to quit... > > Could someone provide me with a solution or with instructions to compile a > vm for Solaris 10? The makefile is somewhat obscure with plenty of > includes at /home/bryce/squeak/exupery without knowing whatś in > there. > > Help would be much appreciated. First you are probably better off using Ian's squeak VMs which now have closure support. The makefiles are generated, regenerate them using: ../platforms/unix/config/configure Which will re-create all the makefiles without the references to my home directories. Bryce

Sept. 7, 2009

Re: [Pharo-project] OSProcess and SqueakMap

by Ross Boylan

On Mon, 2009-09-07 at 13:50 -0700, Ross Boylan wrote: > OSProcess loaded fine, but CommandShell says PluggableTextController > and > PluggableTextView. ... are required classes. Ross

Sept. 7, 2009

[Pharo-project] Patterns for Extensible Code Models

by Andrew P. Black

I was talking today about ways of extending a code model (like, the one inside Pharo that represents my packages, classes, and methods), and realized that I need to publicize some patterns that Daniel Vanisencher and I worked out some years ago when we were trying to separate the computation of the required-methods property (needed for traits) from the particular browser that one might use to look at it. The key idea behind these patterns is to take the work of recomputing some property of the model out of the browser, so that we do not get complaints that "Pharo is too slow". The paper is at http://web.cecs.pdx.edu/~black/publications/ModelExtensions.pdf It's quite long, but I think that it is easy to read. A version has been accepted by the new Springer Journal "LNCS Transactions on Pattern Languages of Programming", if it ever appears. ____________________________ Prof. Andrew P. Black Department of Computer Science Portland State University Oregon, USA http://www.cs.pdx.edu/~black Telephone: +1 503 725 2411

Sept. 7, 2009

[Pharo-project] VNC on mac?

by Lawson English

I've been trying to test the RFBServer on a Mac using the Seaside 2.8.4 image. I'm using Chicken of the VNC as my VNC client, but all I get is a static iamge of the Squeak desktop. Can't referesh and certainly can't interact. The connection info says it IS an interactive session though. Any ideas for what I'm doing wrong? THanks Lawson

Sept. 7, 2009

Re: [Pharo-project] OSProcess and SqueakMap

by Ross Boylan

On Mon, 2009-09-07 at 16:00 -0400, David T. Lewis wrote: > On Mon, Sep 07, 2009 at 12:15:28PM -0700, Ross Boylan wrote: > > I want to run the magma tester, which is much easier if OS Process is > > around. It doesn't seem to be in the image or Monticello, and I think > > it is only available from SqueakMap. > > OSProcess and CommandShell (formerly part of OSProcess until I split it) > are both on SqueakSource. For Pharo, I recommend that you load the latest > versions from SqueakSource rather than from SqueakMap (I made a fix just > yesterday that is important for Pharo users). Be sure to get the unit tests > also, they are a good source of examples. OSProcess loaded fine, but CommandShell says PluggableTextController and PluggableTextView. > > I have not tested on the latest Pharo, but I did notice one test failure > for OSProcess the last time I checked. It was something involving coordinating > lots of (Smalltalk) processes. One of the CommandShell tests was not running > to completion either, probably for the same reason. I'm not sure if it is > still a problem on the latest Pharo or not (sorry, I have not had time to > track it down yet). > > Also, CommandShell will not show up in the menus on Pharo, so you will > need to evaluate "CommandShell open" to start a shell. > > > What's the best way to get OSProcess into the image? Is it already > > there? Not compatible? > > It should not be necessary to add it to the image. I meant into my image, not the baseline image. > I always intended for > OSProcess to be an external package, and it should be easy enough to load > for anyone who wants to use it. Do I need the OSProcess plugin? Thanks. Ross

Sept. 7, 2009

Re: [Pharo-project] CLFramework-UIBuilder

by nullPointer

I donÂ´t understand. The last dev image (10440) works for me in that point :( . Regards. Torsten Bergmann wrote: > >>i published the CLFramework package > > Nice! > > > First bug report: > In CLBuilderMainForm>>createNewView you send > #alert:text:title: to the current UI theme > > but this message is not in the latest core image. > > Bye > Torsten > > > -- > Neu: GMX Doppel-FLAT mit Internet-Flatrate + Telefon-Flatrate > fÃ¼r nur 19,99 Euro/mtl.!* http://portal.gmx.net/de/go/dsl02 > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > -- View this message in context: http://n2.nabble.com/CLFramework-UIBuilder-tp3596854p3599379.html Sent from the Pharo Smalltalk mailing list archive at Nabble.com.

Sept. 7, 2009

Re: [Pharo-project] OSProcess and SqueakMap

by David T. Lewis

On Mon, Sep 07, 2009 at 12:15:28PM -0700, Ross Boylan wrote: > I want to run the magma tester, which is much easier if OS Process is > around. It doesn't seem to be in the image or Monticello, and I think > it is only available from SqueakMap. OSProcess and CommandShell (formerly part of OSProcess until I split it) are both on SqueakSource. For Pharo, I recommend that you load the latest versions from SqueakSource rather than from SqueakMap (I made a fix just yesterday that is important for Pharo users). Be sure to get the unit tests also, they are a good source of examples. I have not tested on the latest Pharo, but I did notice one test failure for OSProcess the last time I checked. It was something involving coordinating lots of (Smalltalk) processes. One of the CommandShell tests was not running to completion either, probably for the same reason. I'm not sure if it is still a problem on the latest Pharo or not (sorry, I have not had time to track it down yet). Also, CommandShell will not show up in the menus on Pharo, so you will need to evaluate "CommandShell open" to start a shell. > What's the best way to get OSProcess into the image? Is it already > there? Not compatible? It should not be necessary to add it to the image. I always intended for OSProcess to be an external package, and it should be easy enough to load for anyone who wants to use it. Dave

Sept. 7, 2009

[Pharo-project] OSProcess and SqueakMap

by Ross Boylan

I want to run the magma tester, which is much easier if OS Process is around. It doesn't seem to be in the image or Monticello, and I think it is only available from SqueakMap. What's the best way to get OSProcess into the image? Is it already there? Not compatible? A few months ago the recommendation for SM seemed to be (Adrian Leinhard) "you can download the MCZ file using a web browser from http://map.squeak.org/ ". There were also indications of other alternatives present or on the way. Thanks. Ross Boylan

Sept. 7, 2009