Squeak-dev August 2008

squeak-dev@lists.squeakfoundation.org

103 participants
139 discussions

[ANN] Chalten 2.0
by Torsten Bergmann 28 Aug '08

28 Aug '08

Hi Maximiliano, "Four slightly different intercalary schemes have been described in the literature..." see http://www.phys.uu.nl/~vgent/islam/islam_tabcal.htm >explained in the book Calendrical Calculations Since (August twentieth, 2008) asIslamic returns the 17th in your example it seems like the book describes either type Ic (‛15’, civil]) or type IIc (‛16’, civil) from the above web page. Microsoft is using the "Kuwaiti Algorithm", see http://en.wikipedia.org/wiki/Kuwaiti_algorithm >The model is provided AS-IS, under the MIT license. Thanks. Found it deep in the FAQ section of the SqueakSource page. (Expected it in the projects Overview as in http://www.squeaksource.com/Seaside29.html which states "Code commited to this repository will be automatically under MIT license.") Bye Torsten -- Psssst! Schon das coole Video vom GMX MultiMessenger gesehen? Der Eine für Alle: http://www.gmx.net/de/go/messenger03

1 0

Searching for WinCE 1.0 Squeak VM port
by Vladimir Machulsky 27 Aug '08

27 Aug '08

Hello, friends! Please don't laugh. :-) I'm searching for Squeak VM port to Windows CE 1.0. Accordingly to archived web page at http://web.archive.org/web/19990128074440/sugarweb.com/Miniatures/files/Win… there was a one in existence. Presumably, it was alive some time ago at ftp://ftp.squeak.org/SmallSqueaksForPDAs/MiniSqueak1.2/ directory. And long before seems it was available at ftp://ftp.squeak.org/wince/ . So, please, if any of You have this retro, please write me a line. -- BR, Vladimir Machulsky

3 3

Is there an up to date tutorial for making plugins?
by Ang Beepeng 27 Aug '08

27 Aug '08

I have search the web and found several links that sounds useful. But they seem out of date and do not work as instructed. Is there an up to date one for a simple example plugin? Thanks, -- View this message in context: http://www.nabble.com/Is-there-an-up-to-date-tutorial-for-making-plugins--t… Sent from the Squeak - Dev mailing list archive at Nabble.com.

5 7

[ANN] Chalten 2.0
by Torsten Bergmann 27 Aug '08

27 Aug '08

Maximiliano Taborda wrote: >(August twentieth, 2008) asIslamic --> Return Sha'ban 17, 1429 Depending on the underlying arithmetical calendar type this should return either Sha'ban 17, 1429 or Sha'ban 18, 1429 or Sha'ban 19, 1429 See http://www.phys.uu.nl/~vgent/islam/islam_tabcal.htm Is the library able to handle different intercalary schemes? Whats the license of the code? Found no info on Squeaksource about that. Thx Torsten -- GMX Kostenlose Spiele: Einfach online spielen und Spaß haben mit Pastry Passion! http://games.entertainment.gmx.net/de/entertainment/games/free/puzzle/61691…

2 1

running test suites
by Esteban Lorenzano 27 Aug '08

27 Aug '08

Hi, I'm looking for a simple way to run test suites and have results like test runner... and test runner does not understand suites (just cases). Is there a "suite runner" somewhere, or maybe I'm using test runner wrong? Thanks, Esteban

2 1

how to tell when Squeak's GUI is active?
by Craig Latta 27 Aug '08

27 Aug '08

Hi-- I'm using Growl/Snarl[1] to post notifications from Squeak. Of course, they're somewhat annoying when Squeak is already the active host application. Is there some way to tell whether Squeak is in the foreground or background, as far as the host window system is concerned? I only care about getting the user's attention when some other host application's GUI is active (and thus possibly obscuring Squeak's GUI). thanks! -C [1] http://en.wikipedia.org/wiki/Growl_(software)

2 1

SandstoneDb question
by Esteban Lorenzano 27 Aug '08

27 Aug '08

Hi, I'm using SandstoneDb for a project... and I have this (one-to-many) scenario: a) Product as Parts b) a Part can be owner by a Product, or can be orphan so, I have two roots: Product and Part well, by design, Sandstone does not allow me to make a recursive commit, and I have to make that commit by hand, something like this: self product partsDo: [ :each | each isNew ifTrue: [ each commit ]. self product commit. but... I need this behaivor in many parts of my environment, so I rewrited #commit, this way: Product>>commit self partsDo: [ :each | each isNew ifTrue: [ each commit ]. super commit. My question is: is this design right? Is there another -cleaner- way to do this? Thanks in advance, Esteban pd: maybe #commitWithRelations is better than overwrite commit?

3 4

[ANN] Chalten 2.0
by Maximiliano Taborda 26 Aug '08

26 Aug '08

Hi. We uploaded to SqueakSource (for Squeak) and to the Cincom Public Repository (for Visualworks) a new release of the project called "Chalten" (2.0) that is our implementation of the Time Domain (not only of the Gregorian Calendar now). This model was born as a representation of the Gregorian Calendar, and, in this last version, the model support the Gregorian, Julian, Islamic an Hebrew Calendars. Many refactorings was done from the original model to get this last version wich reifyes many concepts of the Time Domain that Smalltalk-80 does not; for example: day, day of month, month of year, month, year, calendars, time zones, filters over the timeline, etc.. The main objective of this model is to easily, and in a "SDL way", solve problems of the Time Domain that are not easy (or impossible) to do with the Smalltalk-80 abstractions. t provides also some concepts that are useful for financial application, among others, like relative dates and time line filters. This model also uses the units models (Aconcagua). Because it uses the units model, there are some concepts that are not useful anymore, like Duration, because a Duration is a Measure of time. The model covers many concepts of the Time Domain, but lacks of some abstractions; for example, the representation of the leap second. It comes with more than 1600 tests (all green). Here are some examples: "Create some time entities" August twentieth, 2008 --> Creates an instance of the gregorian Date for 20/08/2008 August twentieth --> Creates an instance of the gregorian DayOfMonth for 20/08 August, 2008 --> Creates an instance of the gregorian MonthOfYear for August of 2008 "But, not only gregorian entities" Shaban seventeenth, 1429 --> Creates an instance of the islamic Date for 17/08/1429 Shaban seventeenth --> Creates an instance of the islamic DayOfMonth for 17/08 Shaban, 1429 --> Creates an instance of the islamic MonthOfYear for Shaban of 1429 JulianAugust seventh, 2008 --> Creates an instance of the julian Date for 07/08/2008 JulianAugust seventh --> Creates an instance of the julian DayOfMonth for 07/08 JulianAugust, 2008 --> Creates an instance of the julian MonthOfYear for August of 2008 Av nineteenth, 5768 --> Creates an instance of the hebrew Date for 19/05/2008 Av nineteenth --> Creates an instance of the hebrew DayOfMonth for 19/05 Av, 5768 --> Creates an instance of the hebrew MonthOfYear for Av of 5768 "Converting dates between diferents calendars" (August twentieth, 2008) asIslamic --> Return Sha'ban 17, 1429 (August twentieth, 2008) asHebrew --> Return Av 19, 5768 (August twentieth, 2008) asJulian --> August 7, 2008 (julian) (Shaban seventeenth, 1429) asHebrew --> Return Av 19, 5768 (JulianAugust seventh, 2008) asIslamic --> Return Sha'ban 17, 1429 (Av nineteenth, 5768) asGregorian --> August 20, 2008 (gregorian) "Measuring distances" August twentieth, 2008 distanceTo: December thirtieth, 2008 --> Return an instance of Measure <132 days> Shaban seventeenth, 1429 distanceTo: Muharram second, 1430 --> Return an instance of Measure <132 days> Av nineteenth, 5768 distanceTo: Tevet third, 5769 --> Return an instance of Measure <132 days> "Also, measuring distance expressed in diferents way (calendars)" August twentieth, 2008 distanceTo: Tevet third, 5769 --> Return an instance of Measure <132 days> Shaban seventeenth, 1429 distanceTo: December thirtieth, 2008 --> Return an instance of Measure <132 days> Av nineteenth, 5768 distanceTo: Muharram second, 1430 --> Return an instance of Measure <132 days> "Collect some entities" (ChaltenYear number: 2008 calendar: GregorianCalendar) months collect: [:monthOfYear | monthOfYear lastDate] --> Returns all the last dates of the 2008 months. (ChaltenYear number: 2008 calendar: GregorianCalendar) dates select: [:date | date is: Monday] --> Returns all Mondays of 2008 (ChaltenYear number: 5768 calendar: HebrewCalendar) dates select: [:date | date is: YomShabbat] --> Returns all yom shabbats of 5768 "Let's create a filter for all dates..." nonWorkingDays := TimeLineFilter named: 'Non Working Days' "Now, we want Saturdays to be on that filter" nonWorkingDays addDayRule: Saturday "Now we want Sundays from January 1st of year 1000 to the end of time..." nonWorkingDays addRule: (nonWorkingDays dayRule: Sunday) from: (January first, 1000) to: TheEndOfTime "Now we want all July 9th since 1816 because is the Independence Day in Argentina". nonWorkingDays addRule: (nonWorkingDays dayOfMonthRule: July ninth) from: (July ninth, 1816) to: TheEndOfTime "Testing some dates..." nonWorkingDays includes: (July ninth, 2008) --> Returns true nonWorkingDays includes: (July eighth, 2008) --> Returns false nonWorkingDays includes: (July twelfth, 2008) --> Returns true, it is Saturday "But, how about to filter some like the hebrew new year day" nonWorkingDays addDayOfMonthRule: Tishri first nonWorkingDays includes: (Tishri first, 5769) --> Return true, it is the next hebrew new year nonWorkingDays includes: (September thirtieth, 2008) --> Return true, it is the next hebrew new year (in gregorian) "21/08/2008 is a Thursday" timespan := TimeSpan from: (August twentyfirst, 2008) duration: (48 * TimeUnits hour) settleDate := RelativeDate timespan: timespan using: nonWorkingDays negated nonWorkingDays includes: (August twentyfifth, 2008) --> Returns false because 25/08/2008, a Monday, is a working day settleDate absoluteDate --> Returns 25/08/2008 "Now a new non working day is added to the filter" nonWorkingDays addDateRule: (August twentyfifth, 2008) nonWorkingDays includes: (August twentyfifth, 2008) --> Return true. "Now 25/08/2008, is a not working day" settleDate absoluteDate --> Now it returns 26/08/2008 because the filter has changed "Working with time zones" buenosAiresDateTime := TimeZonedDateTime dateTime: (DateTime date: August twentieth, 2008 timeOfDay: (TimeOfDay hours: 19 minutes: 35)) zone: TimeZones buenosAires greenwichDateTime := TimeZonedDateTime dateTime: (DateTime date: August twentieth, 2008 timeOfDay: (TimeOfDay hours: 22 minutes: 35)) zone: TimeZones greenwich buenosAiresDateTime = greenwichDateTime --> Return true, it is the same instant but measure in diferent zone buenosAiresDateTime distanceTo: greenwichDateTime --> Return a measure <0 days>, because it is the same instant buenosAiresDateTime := buenosAiresDateTime next: (TimeUnits hour * 3) buenosAiresDateTime = greenwichDateTime --> Return false, the hour is the same but the zone is different (buenosAiresDateTime distanceTo: greenwichDateTime) convertTo: TimeUnits hour --> Return a measure <-3 hours>, just the offset between zones Regards.- Maximiliano

3 3

AGI (Re: [squeak-dev] How to find...?)
by Yoshiki Ohshima 26 Aug '08

26 Aug '08

At Mon, 25 Aug 2008 22:10:23 +0000, dmurphy(a)infratecture.com wrote: > > no problem! > > btw...is anyone at ViewPoints Research doing anything with AGI? > (Artificial General Intelligence). i saw a google summer of code > project to port NARS to Squeak. is anyone pursuing? Some of folks are pursuing knowledge and heuristics representations and processing in relation to the programming languages, but it is not for the general, grand AI vision direction. -- Yoshiki

2 1

Re: [squeak-dev] How to find...?
by jecel＠merlintec.com 26 Aug '08

26 Aug '08

>From early 1999 to at least 2004 Dan Ingalls sent many reports to this list about what changes had been included in the official image. The first line had the name of the change set (starting with a number), the name of the author and the date. Almost always the last part of the change set's name was the author initials, so I generated the following report: http://paste.lisp.org/display/65680 I started by extracting the first line of each change from the emails and then cleaned that up. Based on the numbering not all changes made it to my list. This file itself might be interesting to some people so it might be a good idea to upload it to the wiki. The next step was extracting the second and fourth element (separated by "-") in each line. Some change sets have no second part while others have multiple initials. I manually removed most of the resulting noise but left duplication where the same name was spelled a bit differently or the initials appeared as both lower case and upper case. This makes me think that 'md\' is probably Marcus Denker. My list didn't help with any of the other initials. -- Jecel

7 25

← Newer
1
2
3
4
5
6
7
8
...
14
Older →

Jump to page: