Squeak-dev December 2002

squeak-dev@lists.squeakfoundation.org

173 participants
456 discussions

Re: Uploading files to the minnow swiki?
by Anthony Hannan 01 Dec '02

01 Dec '02

Doug Way <dway(a)riskmetrics.com> wrote: > > I was just trying to upload a file to the Squeak Swiki on minnow, and I > notice that all uploads are now password-protected. (I don't remember > seeing any announcement on the list about this particular password > protection.) There was an announcement a while ago by Mark Guzdial. The username is 'squeak' and the password is 'viewpoints'. Cheers, Anthony

1 0

Uploading files to the minnow swiki?
by Doug Way 01 Dec '02

01 Dec '02

I was just trying to upload a file to the Squeak Swiki on minnow, and I notice that all uploads are now password-protected. (I don't remember seeing any announcement on the list about this particular password protection.) This is probably a reasonable scheme to prevent abuse, but could a login/password be posted to the list, so that listmembers can at least upload files? I think this should be perfectly safe, since the password ("squeak") for the now-locked top page was posted on the list awhile ago, and there has been no abuse whatsoever on that page since then. - Doug Way

1 0

[SM][ANN] IOHandle, CommandShell, and OSProcess on SM with SAR archives
by David T. Lewis 01 Dec '02

01 Dec '02

I've put together SAR archives for OSProcess, CommandShell, and IOHandle and updated the SqueakMap registrations. Kudos to all who are putting SM together, this is great stuff! The SAR archive and loader was particularly gratifying. I was able to get the IOHandle changes to load automatically from a single SAR file. Don't look at how I did it, it's ugly. But at least it was possible, and SqueakMap and the SAR loader make it look easy. Dave

2 2

70 packages on SqueakMap!
by goran.hultgren＠bluefish.se 01 Dec '02

01 Dec '02

Yiha! Amazing. Really cool guys, I feel the pressure to make SM even better... regards, Göran

4 10

Bug in teh program
by Jaril 01 Dec '02

01 Dec '02

Sirs, As a trial project I worked on the Path through Net problem, where I use a Point and Vector model. At some point I encountered a strange situation: the following code did not work correctly for one of vectors in the helpSet, though I did test the situation thoroughly and endPoint is postively in pointsOnPath : helpSet do: [:vector | ( pointsOnPath includes: (vector endPoint )) ifTrue: [ helpSet remove: vector ifAbsent: [^'Error 2, class NetClass'] ]. ]. But once I applied the following code on the same data: helpSet := helpSet reject: [:vector | pointsOnPath includes: (vector endPoint )]. the result was correct. I am new to ST and can't find my mistake. Any help is appreciated. LP Jaril Lavrenčič

3 2

Re: Bug in teh program
by Bob Arning 01 Dec '02

01 Dec '02

Jaril, Your problem is a basic one: "Do not alter collections while iterating over them". The messages, such as #do:, that iterate over collections are unaware that that the collection has changed, so they keep chugging from 1 to n. As a result, they can skip valid elements, process the same one twice and/or try to access beyond the end of the collection. Your solution is a good one. If the situation were more complicated, then replacing helpSet do: [:vector | with helpSet copy do: [:vector | might be useful. Cheers, Bob On Sun, 1 Dec 2002 18:16:27 +0100 "Jaril" <diris(a)volja.net> wrote: >As a trial project I worked on the Path through Net problem, where I use a >Point and Vector model. At some point I encountered a strange situation: the >following code did not work correctly for one of vectors in the helpSet, >though I did test the situation thoroughly and endPoint is postively in >pointsOnPath : > > > > > helpSet do: [:vector | > ( pointsOnPath includes: (vector endPoint )) > ifTrue: [ > helpSet remove: vector ifAbsent: [^'Error 2, class >NetClass'] > ]. > ]. > > > > >But once I applied the following code on the same data: > > helpSet := helpSet reject: [:vector | pointsOnPath includes: (vector >endPoint )]. > >the result was correct. > > >I am new to ST and can't find my mistake. Any help is appreciated.

1 0

About method comment evaluation
by Stephane Ducasse 01 Dec '02

01 Dec '02

Hi I would like to find the person who did the following: you can select a comment and evaluate it. In VW and Squeak now you can write "self ..." in the comment and in such a case it refers to the ***class*** I was so used to this browser trick that I never realized how it could be dangerous for newbie. Imagine self in comment and in the code represent different object: instance and its class. So I would like to know if we could change or introduce another browser trick class selfClass this way we could have "selfClass new xxx" and not "self new xxx" Is the author of the previous trick around? What the other are thinking about that? Stef Dr. Stéphane DUCASSE (ducasse(a)iam.unibe.ch) http://www.iam.unibe.ch/~ducasse/ "if you knew today was your last day on earth, what would you do different? ... especially if, by doing something different, today might not be your last day on earth" Calvin&Hobbes

2 3

Re: [ANN] v3 of StarBrowser registered on SqueakMap: try it!
by Daniel Vainsencher 01 Dec '02

01 Dec '02

For example, running SmallLint on all the code in a classification (convert classification to BrowserEnvironment, then run lint). Yup, that'd be pretty cool. Hmm... now somewhat busy turning my iPaq into a linux box, but I like the SB, and it's looking like it could already be a nice place to live... It also gives a good reason to refactor various GUI away from all sort of logic, to make available to SB, which is a good thing... Daniel Stephane Ducasse <ducasse(a)iam.unibe.ch> wrote: > > - Refactoring Browser taking advantages of classification ;) > > > > > -- > > Ned Konz > > http://bike-nomad.com > > GPG key ID: BEEA7EFE > > > > > > > Dr. StÈphane DUCASSE (ducasse(a)iam.unibe.ch) > http://www.iam.unibe.ch/~ducasse/ > "if you knew today was your last day on earth, what would you do > different? ... especially if, by doing something different, today > might not be your last day on earth" Calvin&Hobbes

1 0

Squeak Map question
by David T. Lewis 01 Dec '02

01 Dec '02

Packages on SqueakMap are categorized by "Squeak Version". Is this intended to mean "the oldest version of Squeak on which this package is known to work" or "the most recent version of Squeak on which this packing is known to work"? Dave

2 2

Re: 70 packages on SqueakMap!
by Daniel Vainsencher 01 Dec '02

01 Dec '02

Not really, this isn't the same situation as the Connectors/SAR one, because the base installer is SM itself. The script can just say "SqueakMap: load SARInstaller and then load my package using it". No need to find the stream, because the load script can be external to the package. This is a benefit of holding dependencies externally to the packages - it's should be easy to simply compose packaging work. Daniel Ned Konz <ned(a)bike-nomad.com> wrote: > On Saturday 30 November 2002 09:26 am, Doug Way wrote: > > I think that would be simpler than requiring all package owners who > > use the SAR format to have a special load script to install > > SARInstaller > > The problem with this is back to the "self-installing archive" > problem: we'd then need to put that script in the front of the SAR, > and it would have to crawl up the stack contexts to find the stream, > etc. > > I think a better strategy would be just to notify the user when no > installer is available. Which would go along with the "better error > reporting" that we want for SM loading in general. > > -- > Ned Konz > http://bike-nomad.com > GPG key ID: BEEA7EFE

1 0

Jump to page: