Squeak-dev December 2008

squeak-dev@lists.squeakfoundation.org

91 participants
114 discussions

Magritte: Validation deprecated?
by Udo Schneider 05 Dec '08

05 Dec '08

All, I'm currently playing a bit with Magritte and I can !"§"&%! myself for not take a look at it earlier. One thing I'm not sure about at the moment is validation. It seems that validation code is not called when saving values. I'm having a description like this: descriptionEmail ^ MAStringDescription new autoAccessor: 'name'; group: 'Internet'; beEditable; beSearchable; addCondition: [:value | self halt. (value includes: $@) and: [(value subStrings: $@) size = 2]] labelled: 'Invalid E-Mail'; comment: 'Please enter the name of the object'; default: 'Unnamed'; label: 'E-Mail' If I call the model using "model asMorph addButtons; addWindow ; callInWorld" and save a bogus E-Mail the breakpoint is not reached. Are Conditions still supported? CU, Udo

3 4

MultiByteFileStream>>skipSeparatorsAndPeekNext doesNotUnderstand: #isSeparator
by Greg A. Woods; Planix, Inc. 04 Dec '08

04 Dec '08

Grrrr.. I've been messing about trying to put together a little change set for a minor little bug fix in SMTPClient #initiateSession, and I've been hacking about perfecting it, viewing versions and making new ones and so on, and just as I'm trying to view the final diffs using the "versions" button in a method browser window I get a doesNotUnderstand: #isSeparator error, apparently from MultiByteFileStream #skipSeparatorsAndPeekNext. From poking around in the debugger window I can't see that any package I've filed in has modified that method, and also I'm sure I've used the versions buffer in a method browser window before too without any errors. It looks like the reason for the error is that the object being sent #isSeparator is actually nil! I say that because the instance variable "peek" is nil according to the debugger and I'm assuming the assignment returns the value being sent #isSeparator after it has assigned it to the variable. But how did "self next" return nil when "self atEnd" is false? This is with Squeak3.10.2-7179-basic.image plus whatever packages I've filed in using the Package Universe browser. (i'm still having a _lot_ of trouble filing in some packages, or their dependencies, too!) Has anyone seen this before? Am I missing something obvious in the changes from packages I've filed in that might be causing the bug to show up? -- Greg A. Woods; Planix, Inc. <woods(a)planix.ca>

4 4

[FIX] SMTPClient literal helo fix
by woods＠planix.ca 04 Dec '08

04 Dec '08

from preamble: "Change Set: SMTPClient literal helo fix Date: 4 December 2008 Author: Greg A. Woods If Squeak is connecting to an SMTP server from a host with an IP address that has no reverse DNS set up, such as is often the case for hosts on private networks, especially wireless networks, then NetNameResolver will return a string representation of the client's source IP address. SMTPClient #initiateSession must send a HELO greeting with this IP address in proper literal form, i.e. enclosed in square brackets, instead of in the unadorned form it previously used. "! (I think it would be really nice if the diff between original and final versions was also put in the text of the message in something like normal unidiff form so that readers of the list don't have to directly examine the change set itself inside a smalltalk image just to get an overall view of what's being changed and how)

1 0

Re: (Mantis) #07212 flaps render *properly* during StandardSystemFontsTest
by Jerome Peace 04 Dec '08

04 Dec '08

[squeak-dev] #07212 flaps render *properly* during StandardSystemFontsTest Hi Timothy, :-) I am really happy you posted this. It means someone else notices the small symptoms of big bugs, too. If you run some of the font tests. The fonts will change and then be restored the defaults. Changing the flap fonts causes the flaps to be laid out again. This may either accidentally and temporarily cure the off-by-one/roundoff error that affects the flaps. There are some other bugs in flaps as well. I made repairs for them but I don't know and rather doubt whether they actually got into 3.10.2. If you view sub issues of #7212 on mantis you can see what else has been discoved. (http://bugs.squeak.org/view.php?id=6486 is one of the main reports.) Yours in curiosity and service, --Jeorme Peace *** >Mr. Timothy C. Johnson tjohnson at iwu.edu >Sat Nov 29 01:32:39 UTC 2008 > > >Hi, > >It is known that flaps render improperly in 3.10.x -- there is a ~1 >pixel dead zone on some of the flaps. But I just ran >System-Support-Tests>>StandardSystemFontsTest and something it did made >the flaps render properly, all of a sudden. Any idea why? > >- timj ***

2 1

Botsinc free
by stephane ducasse 04 Dec '08

04 Dec '08

Hi all I'm happy to announce that my book BotsInc http://smallwiki.unibe.ch/botsinc will be free soon (in fact it is since 23 september). I would like to thank ESUG http://www.esug.org for supporting financially the buying of the book rights and giving the book to the kid world. A catalan version will be available really soon thanks to jordi delgado! (Yes the french were not the first this time... We were the first for the etoy book and squeak by example. :)) If you want to participate to a translation please contact me. I will be setting up soon an infrastructure to support any language translation. Stef ------------------------ Stéphane Ducasse USTL - LILF - CNRS UMR 8022 - INRIA stephane.ducasse(a)inria.fr http://stephane.ducasse.free.fr "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 Tel 33 3 59 57 78 66 - Fax 33 3 59 57 78 50 40, avenue Halley, Parc Scientifique de la Haute Borne, Bât.A, Park Plaza, Villeneuve d'Ascq, FR-59650, France.

5 7

leadership team meeting notes for 2008-12-03
by Craig Latta 04 Dec '08

04 Dec '08

Hi-- - Attending: Bert, Craig, Edgar, Igor, Randal, and Yoshiki. Absent: Giovanni. We meet on first and third Wednesdays. - Yoshiki reports progress from Matthew Fulmer on Squeak 4 (see [1]). - Randal says there's been no change on the "captcha" issue[2], because he's been too busy himself. - Craig received a recent invoice for Squeak's servers, hasn't finished taking over the payments. - Randal suggests changing the licensing page on the Squeak website to mention the unacceptability of the GPL for Squeak contributions. The consensus was to make it clear that we accept only MIT-licensed contributions, leaving discussion of all other licenses to a Frequently-Aaked Questions page (featuring heavy use of the word "no" :). We expect to formalize the license situation via future membership with the Software Freedom Conservancy, after we can deliver to them our impending attempt at a license-clean release, Squeak 4. thanks again, -C [1] Matthew, via Yoshiki: "I worked on this three days last week. I started with your 3.10 history image, made a file containing all the dirty methods, and now I'm annotating it with every change you made to etoys, to see what would still need relicensing after I applied your etoys 3.2 changes, and whether they would apply cleanly. So far, I have gone through about 4 of your 10 etoys changesets. My plan is something like this: - Check to see what yoshiki's changesets don't cover (40% done) - Apply yoshiki's changesets to 3.10, carefully - Confirm and fix the licensing status of new modules in 3.10: - Traits (almost certainly clean) - Universes (probably clean) - Monticello (yikes) - Do whatever I missed above [2] From the 2008-11-19 meeting: Randal says there's been no progress mentioned by the web team, since the last meeting, in setting up a "captcha" system for community communication with the leadership team via web mail. He says he'll take the lead on implementing it. In the meantime, please send agenda item suggestions to Craig.

5 6

Loading my monticello package - very slow at end; how to diagnose?
by Sophie (itsme213) 04 Dec '08

04 Dec '08

When I load my monticello package into Damien's 3.10 web image, the progress bar moves really quickly to about the 99% mark, then pauses there for a good 15 seconds. Is there a way to figure out (or guess) what it is doing, and if something in my code can be changed to improve it? Thanks!

4 7

Kernel Extensions Missing in Web Dev Image 3.10
by Thomas Kowark 04 Dec '08

04 Dec '08

Hi list, I have downloaded a squeak 3.10 based web-dev image from Damien Cassou's website and stumbled upon the following problem: Some of the tests in my current project are setting #shouldInheritSelectors to true. This leads to a method #withInheritedSelectors being called. This method now tries to call #allSelectorsBelow, which unfortunately is not implemented by any class. A quick Google lookup revealed that this method comes with the KernelExtensions package. After installing the package, the tests worked just fine. Damien, could you include this package into the next release of the images, or are there any problems with it, I am not aware of? Oh, and before I forget. The SqueakMapPackageLoader crashes when I try to start it from the world menu. If it doesn't work anymore, it would also be a good idea to remove it from the world menu, wouldn't it? kind regards, Thomas Kowark

4 8

Rio: exceptions documentation
by Casimiro de Almeida Barreto 03 Dec '08

03 Dec '08

Hello, I had a code like: [ aStream := CrLfFileStream new open: 'aGivenFileName' forWrite: false ] on: FileDoesNotExistException do: [ :exception | <code ....> ]. It worked ok. But with: [ aStream := 'aGivenFileName' asRio reader ] on: FileDoesNotExistException do: [ :exception | <code ...> ]. things just don't work... Where I can find exceptions documentation (in particular for Rio) ? Besides, it seems that the atEnd message is delayed by one entry (in the first case 1025 lines are correctly read and in the second case it goes one line ahead in the file).

3 6

[BUG] signedLongAt: 1 put:−2,147,483,648
by John M McIntosh 02 Dec '08

02 Dec '08

I noted the the current VMs don't handle the case of storing −2,147,483,648 (primitive failure) via signed32BitValueOf: usually called via FFI signedLongAt:put: For plugin writers who used signed64BitValueOf: (primitive failure) and signed64BitIntegerFor: returned zero if passed −9,223,372,036,854,775,808. signed64BitIntegerFor: is the one that is nasty since your value of 0x8000000000000000 silently turns in 0. I reopened problem http://bugs.squeak.org/view.php?id=6987 and wrote some SUnits. I also provided a suggested fix, which allows me to run the SUnits. Lastly for Alien I added long long (signed/unsigned) support which is still missing in the FFI plugin so I could do some testing. -- = = = ======================================================================== John M. McIntosh <johnmci(a)smalltalkconsulting.com> Corporate Smalltalk Consulting Ltd. http://www.smalltalkconsulting.com = = = ========================================================================

2 5