Squeak-dev October 2004

squeak-dev@lists.squeakfoundation.org

156 participants
313 discussions

[FIX][VM][UNIX] utf8 support for key events
by Martin Kuball 24 Oct '04

24 Oct '04

Hi! Here are some changes for the unix vm to improve the handling of key events in environments with utf-8 locales. Basically I did the following: - change recode in sqUnixX11.c to use all information from the input event. The result is still a single 8bit byte, though. - add code in x2sqKeyInput to check the XIM locale. If it is utf8 change the uxXWinEncoding to utf-8. - add a new commandline arg and environment variable to explicitly set the uxXWinEncoding. I added a patch for Makefile.in to make it work with newer versions of the gnu auto* tools. I used automake 1.7, autoconf 2.59 and libtool 1.5.6. Should be compatible with older versions. Martin

3 3

[FIX] TimeComment-dew
by dway＠mailcan.com 24 Oct '04

24 Oct '04

from preamble: "Change Set: TimeComment-dew Date: 23 October 2004 Author: Doug Way Fixes the comment for Time. The comment previously read 'This represents a period of time.' which implies a duration or interval of time, which is *not* what Time represents. It really represents a particular point in time, during an unknown day. (Although it can be converted to a Duration.) The new comment reads: 'This represents a particular point in time during any given day. For example, '5:19:45 pm'. If you need a point in time on a particular day, use DateAndTime. If you need a duration of time, use Duration.' "!

1 0

[FIX] Unnamed
by tomkoenig＠mindspring.com 23 Oct '04

23 Oct '04

1 0

[BUG]
by tomkoenig＠mindspring.com 23 Oct '04

23 Oct '04

The error report: 23 October 2004 1:08:16 pm VM: Win32 - a SmalltalkImage Image: Squeak3.8alpha [latest update: #6326] SecurityManager state: Restricted: false FileAccess: true SocketAccess: true Working Dir C:\Documents and Settings\Thomas Koenig\Desktop\Squeak Trusted Dir C:\Documents and Settings\Thomas Koenig\Desktop\Squeak\Thomas Koenig Untrusted Dir C:\My Squeak\Thomas Koenig Rectangle>>rightCenter Receiver: 100@100 corner: 200@200 Arguments and temporary variables: Receiver's instance variables: origin: 100@100 corner: 200@200 DecentReviewerNPBugReportTest>>setUp Receiver: DecentReviewerNPBugReportTest>>#testSend Arguments and temporary variables: Receiver's instance variables: testSelector: #testSend aDecentReviewerNotePad: a DecentReviewerNotePad(2962) aCrashDmp: AttachmentResource aMethodContext: Rectangle>>rightCenter [] in DecentReviewerNPBugReportTest(TestCase)>>runCase {[self setUp. self performTest]} Arguments and temporary variables: BlockContext>>ensure: Receiver: [] in DecentReviewerNPBugReportTest(TestCase)>>runCase {[self setUp. self performTest]} Arguments and temporary variables: aBlock: [] in DecentReviewerNPBugReportTest(TestCase)>>runCase {[self tearDown]...etc... returnValue: nil b: nil Receiver's instance variables: sender: BlockContext>>ensure: pc: 28 stackp: 0 nargs: 0 startpc: 26 home: DecentReviewerNPBugReportTest(TestCase)>>runCase --- The full stack --- Rectangle>>rightCenter DecentReviewerNPBugReportTest>>setUp [] in DecentReviewerNPBugReportTest(TestCase)>>runCase {[self setUp. self performTest]} BlockContext>>ensure: - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - BlockContext>>sunitEnsure: DecentReviewerNPBugReportTest(TestCase)>>runCase [] in TestResult>>runCase: {[aTestCase runCase. true]} BlockContext>>on:do: BlockContext>>sunitOn:do: [] in TestResult>>runCase: {[[aTestCase runCase. true] sunitOn: self class failure do: [:signal | ...]} BlockContext>>on:do: BlockContext>>sunitOn:do: TestResult>>runCase: DecentReviewerNPBugReportTest(TestCase)>>run: [] in TestSuite>>run: {[:each | self sunitChanged: each. each run: aResult]} OrderedCollection>>do: TestSuite>>run: [] in TestSuite>>run {[self run: result]} BlockContext>>ensure: BlockContext>>sunitEnsure: TestSuite>>run [] in TestRunner>>runSuite: {[result := suite run]} BlockContext>>ensure: [] in TestRunner>>runSuite: {[[result := suite run] ensure: [running := nil. suite removeDependent: ...]} [] in BlockContext>>newProcess {[self value. Processor terminateActive]}

1 0

Croquet on SqueakMap?
by David Faught 23 Oct '04

23 Oct '04

Fri Oct 22 15:20:02 CEST 2004, David Faught dave_faught at yahoo.com wrote: >Are there any thoughts about adding a Croquet Jasmine release to the >categories for SqueakMap? I'm a little unsure just how far the >revolutionary versus evolutionary thinking of the Croquet team extends. Maybe I should put this another way. As a member of both the Squeak community and the Croquet community, I would very much like to be able to use SqueakMap for Croquet Jasmine packages. If there are no serious objections, can a category for the Jasmine release of Croquet please be added to SqueakMap? I guess that a reference to the Croquet license would probably also be in order, but that's not my main concern. _______________________________ Do you Yahoo!? Declare Yourself - Register online to vote today! http://vote.yahoo.com

1 0

RE: Storing floats, etc in GOODS
by Jarvis, Robert P. (Bob) (Contingent) 23 Oct '04

23 Oct '04

3.2 is one of the many numbers which cannot be exactly represented using IEEE floating-point values. Changing to use IEEE 64-bit format won't fix the problem either because 3.2, when stored as an IEEE 64-bit floating point value, is stored as 3.2000000000000002. This is an example of why putting monetary values in floating point variables is not a good idea. Perhaps you could use Fraction or ScaledDecimal if they are appropriate? Very respectfully, Bob Jarvis > -----Original Message----- > From: Yar Hwee Boon [mailto:hboon@motionobj.com] > Sent: Friday, October 22, 2004 9:09 AM > To: squeak-dev(a)lists.squeakfoundation.org > Subject: Storing floats, etc in GOODS > > > Hi > > Storing floats, etc in GOODS, using Avi Bryant's client, > numbers such as > 3.2 are returned as 3.199999809265137. Seems that GOODS > stores floats > using the IEEE 32 bit format. How do you work around it? I'm > storing money > one in currency only. Create a currency class, maybe? Thanks. > > -- > Regards > HweeBoon > MotionObj > ----------------------------------------- ********************************************************************** PLEASE NOTE: The above email address has recently changed from a previous naming standard -- if this does not match your records, please update them to use this new name in future email addressed to this individual. This message and any attachments are intended for the individual or entity named above. If you are not the intended recipient, please do not forward, copy, print, use or disclose this communication to others; also please notify the sender by replying to this message, and then delete it from your system. The Timken Company **********************************************************************

3 2

REMINDER: Harvesting Party Sunday October 24th 09:00 GMT
by Ken Causey 23 Oct '04

23 Oct '04

Just a reminder that the next Harvesting Party will take place between 09:00 (9 AM) and 13:00 (1 PM) GMT Sunday October 24th, 2004. As usual we will meet in #squeak on irc.freenode.net. More information about the #squeak IRC channel can be found at http://people.squeakfoundation.org/article/7.html Information about the Harvesting Party idea can be found at http://people.squeakfoundation.org/article/29.html If you have the time please attend and work toward the future of Squeak. The next seven parties are scheduled for 13:00 (01:00 PM) GMT Thursday October 28, 2004 16:00 (04:00 PM) GMT Monday November 01, 2004 20:00 (08:00 PM) GMT Friday November 05, 2004 00:00 (12:00 AM) GMT Wednesday November 10, 2004 04:00 (04:00 AM) GMT Sunday November 14, 2004 08:00 (08:00 AM) GMT Thursday November 18, 2004 12:00 (12:00 PM) GMT Monday November 22, 2004

1 0

[BUG] BlockContext storeString hangs
by Stéphane Rollandin 23 Oct '04

23 Oct '04

try this (printIt) in a workspace (beware, it will not return and may hang the system for good): [1 + 1] storeString Stef

2 2

Re: MySql connection
by Colin Putney 23 Oct '04

23 Oct '04

On Oct 22, 2004, at 8:58 AM, squeak(a)ajr.e4ward.com wrote: > On Friday 22 October 2004 2:18 am, Hilaire Fernandes wrote: >> Hello, >> >> Exploring the use of the MySQL driver, I have found that each time I >> want to do a SQL query I need to create a new JdmConnection. > > I also ran into problems trying to reuse a connection for more than > one query. > It looks like the driver was designed that way although I have heard > from > others (Colin?) that it does work, maybe if the resultset is > completely > read in between. But my experience and observation is that connections > are > not serially reusable much less reentrant. But multiple concurrent > connections are ok (if not elegant). You know, I never ran into this problem. I did find that not reading all of the data after a query would hose the connection, but it would remain open. Concurrent queries over the same connection won't work either, but that's normal, for this driver at least. I don't know if the protocol supports that. I wish I could be of more help with this but I don't have access to a MySQL server anymore. Colin

1 0

RE: getting rid of anObsoleteOBNotifier
by Jeff Sparkes 22 Oct '04

22 Oct '04

> No, OBNotifier was my attempt at a simple way to handle updating the > browser when the system is changed by something other than that browser. > It had nothing to do with OBNotification. I said it was a guess :-) >> The best solution I found was to explore "Smalltalk obsoleteClasses". >> It turns out that anObsoleteNotifier was at: 29, so I did "(Smalltalk obsoleteClasses >> at: 29) becomeForward: OBNotification class", which seems to have done the trick. > I'm surprised this hasn't caused you any problems. If you find yourself > in this situation again, the best way to fix it is to send #unregister > to the instance of OBNotifier. For this image, I think you want > something like the following. I don't have a Squeak image handy, so I > can't check this code... > SystemChangeNotifier uniqueInstance noMoreNotificationsFor: > theObsoleteOBNotifer. (Smalltalk obsoleteClasses at: 29) allInstances do: [:i | SystemChangeNotifier uniqueInstance noMoreNotificationsFor: i]. did the job. I don't think you can access obsolete classes by name. >> Is this a general technique to use after renaming classes? > In general, you shouldn't need to do this kind of thing. Renaming a > class should "just work." (Although this was a case of removing a > class.) This was a pretty ugly hack that I didn't clean up properly when > I took it out. I'll add the clean up code to the OB installer. Thanks for the heads up, Colin

1 0

← Newer
1
...
6
7
8
9
10
11
12
...
32
Older →

Jump to page: