Squeak-dev December 2000

squeak-dev@lists.squeakfoundation.org

96 participants
272 discussions

Re: [BUG?]New Event Architecture Queue Sequencing?
by Andrew C. Greenberg 04 Dec '00

04 Dec '00

I'll answer some of these questions now, and others later: >I have tried again, without success, to recreate this, so here are >some more questions: > >- if you do world menu/help/vm statistics, have there been any full GC's? >- how much free memory is Squeak showing? >- just to double check, your VM is 2.9alpha5, right? Yes. I also tried it with 4 and 3: identical results. >- when your test is running, it displays the event queue size on the >left. When you start to slow down, what do the numbers look like? the stack is non-empty when things start to run slow, and usually stops on a small, but non-zero and non-one number. If I type only a few events, it usually runs down to zero without the slowdown. After all the events have run down to completion, the next event shows a zero queue. This seems to result also when I strike several keys at once. I'm going to test this across several machines at my house.

1 0

Recent STP updates
by Stephen T. Pope 04 Dec '00

04 Dec '00

Hello! I'm not certain if these ever made it out; I sent them in August, but haven't been able to post to the list for several months. At any rate, if you tell me, I can post them separately as [ENH] so they make it into the automatic tracking system. The code is attached as a BinHexed StuffIt. Please advise... stp ----------------------------------------------------------- Recent STP updates -- 2000.08.24 (remote from Berlin) Browser-recent.st Make the size of the recent menu in the browser a variable. I use long ones. BrowserSel.cs Make the browser remember the mode selection (def/hier/comment) between class selections. There is also partially-working code (commented out) to remember the message category selection. ChangeList-recent.cs Make a short cut for the change list view to remove doits, clear old versions, and select current methods. Class-makePublic.st This is a safer version of the accessor creation method used by the "publish" goodie Collection-ifEmpty.st Short cut for control structures. Collection-isEmpty.st Made it more conservative. DateAndTime.st DateAndTime as merged class (subclass of Date with seconds) (easier for use, e.g., in databases) FileList-recent.cs Make the size of the recent directory list a variable size. FileStr-creation.st A couple of new file stream creation methods. Adds global FileSep with the path name delimiter FindSelAssign.cs Makes 2 changes to ParEd: Cmd-G inserts " := " and Cmd-H searches for the selection. Float class-twopi.st Accessor for the class variable two * pi Float-printPlaces.st Limited-width floating point number printing FormattedStr.cs C-style string formatting. Takes this: Transcript cr; show: ''Analyze '', sound name, '' D: '', (duration printPlaces: 3), '' R: '', sound rate printString, '' F: '', framesize printString, '' P: '', npoles printString; cr. and reduces it to this: {sound name. duration. sound rate. framesize. npoles} showFmt: '' Analyze %s D: %f R: %d F: %d P: %d''. Or this Transcript show: ''TSR: '', tsr printString, '' e: '', e truncated printString; cr. reduces to: {tsr . ext} showFmt: ''TSR: %s e: %d" Try this: { t0 / 1000 . nFrames * fftLen } showFmt: ''Took %5.2f sec, sz: %d.'' Interval-extent.st Get the extent of an interval. LauncherOption.cs Add the LauncherView to the system open menu. MessageTally-toFileNamed.st Convenience method for tallying to a file OrderedColl-size.st Added a nil check. Point-polar.st Answer a point that represents the receiver as r@theta RealEstate.cs Extent-from-user for new windows; assumes Mark's layouts. Do [Preferences enable: #userPlacement] SeqColl-first.st Added check for emptitude. String-conversions.cs A number of new string coersions for file I/O, e.g., asLocalFilename, asFileStream and sansLastPeriodSuffix. WinSizes.cs Make the default window sizes a bit larger and dependent on the screen size. -- stp Stephen Travis Pope http://www.create.ucsb.edu/~stp

1 0

RE: How to do a Sexy Squeak Demo?
by Torsten.Bergmann＠phaidros.com 04 Dec '00

04 Dec '00

Maybe you will look at my DHTML Squeak Tour at http://www.larrikin.org/accounts/astares It should help people getting a first impression of Squeaks possibilities. There you will find some nice screen shots. Actually it's a rough basic version, you have to wait a few seconds after clicking on "Squeak Tour" (otherwise you get a JavaScript error). Maybe I will find the time to incorporate some plugin demos. Torsten -----Original Message----- From: Stephane Ducasse [mailto:ducasse@iam.unibe.ch] Sent: Montag, 4. Dezember 2000 14:01 To: squeak(a)cs.uiuc.edu Subject: How to do a Sexy Squeak Demo? Hi all, over the last months I made a lot of Squeak demoes, basically everywhere I can ;) So I started to collect code snippets, ways of presenting, files..... I started to write everything in a set of wiki pages so that you can use it and add new material. I did not finish yet to upload all the files I used (like some cool flash). One of my goal is to have a kind of central repository where somebody willing to do a demo or have fun with Squeak can find enough information to do it. Look at if you want to know more and add things if you wnat to share your own presentation tricks and material http://titan.cnds.unibe.ch:8080/SmalltalkWiki/93 Stef Stephane 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 University of Bern, Institut fuer informatik and Mathematik IAM-SCG, 10 neubruckstrasse, CH-3012 Bern, Switzerland.

1 0

Re: 2.9a Squeak VM, apps, and themes
by Bruce ONeel 04 Dec '00

04 Dec '00

Hi, You probably get fairly far by first downloading the 2.8 package for your sysetm out of ftp://st.cs.uiuc.edu/Smalltalk/Squeak/2.8 and then getting the 2.9a images and changes. I think the 2.8 VMs still work fine with 2.9. cheers bruce J Scott Jaderholm <lispbliss(a)sage.cortland.com> wrote: > Thank you Bob for your help. > > > On 02 Dec 2000 20:00:19 -0700 J Scott Jaderholm <lispbliss(a)sage.cortland.com> wrote: > > >I'm interested in playing around with Nebraska and 2.9a of Squeak. I > > >downloaded Squeak2.9a-2988.zip which includes the .image and > > >..changes. The problem is I don't have a .sources or a VM for 2.9a and > > >I heard I need a new VM but all I can find is VMs built for MacOS. > > > > Looks like .sources is in > > > > ftp://st.cs.uiuc.edu/pub/Smalltalk/Squeak/2.0/SqueakV2.sources.sea.bin > > I can not extract this under linux. Are the .sources for 2.9a not > available in .zip or .tar.gz ? > > > I would try following the links on > > > > http://minnow.cc.gatech.edu/squeak/388 > > > appropriate to your OS for the VM > > <snip> > > ok, I downloaded the file Squeak-2.9a.pre1-src.tar.bz2. I extract, > create a build dir, cd build, ../src/unix/configure works fine, but > make gives me > > make: Nothing to be done for `all'. > > Here is the output of ../src/unix/configure > > creating cache ./config.cache > checking host system type... i686-pc-linux-gnu > checking whether make sets ${MAKE}... yes > checking for gcc... gcc > checking whether the C compiler (gcc ) works... yes > checking whether the C compiler (gcc ) is a cross-compiler... no > checking whether we are using GNU C... yes > checking whether gcc accepts -g... yes > checking for as... as > checking for ranlib... ranlib > checking for a BSD compatible install... /usr/bin/install -c > checking for BSD-compatible nm... /usr/bin/nm -B > checking build system type... i686-pc-linux-gnu > checking for ld used by GCC... /usr/bin/ld > checking if the linker (/usr/bin/ld) is GNU ld... yes > checking whether ln -s works... yes > updating cache ./config.cache > loading cache ./config.cache within ltconfig > checking for object suffix... o > checking for executable suffix... no > checking for gcc option to produce PIC... -fPIC > checking if gcc PIC flag -fPIC works... yes > checking if gcc supports -c -o file.o... yes > checking if gcc supports -c -o file.lo... yes > checking if gcc supports -fno-rtti -fno-exceptions ... yes > checking if gcc static flag -static works... -static > checking if the linker (/usr/bin/ld) is GNU ld... yes > checking whether the linker (/usr/bin/ld) supports shared libraries... yes > checking command to parse /usr/bin/nm -B output... ok > checking how to hardcode library paths into programs... immediate > checking for /usr/bin/ld option to reload object files... -r > checking dynamic linker characteristics... Linux ld.so > checking if libtool supports shared libraries... yes > checking whether to build shared libraries... yes > checking whether to build static libraries... no > checking for objdir... .libs > creating libtool > updating cache ./config.cache > loading cache ./config.cache > checking how to run the C preprocessor... gcc -E > checking for ANSI C header files... yes > checking for unistd.h... yes > checking for string.h... yes > checking for fcntl.h... yes > checking for sys/file.h... yes > checking for sys/param.h... yes > checking for sys/time.h... yes > checking for sys/filio.h... no > checking for sys/select.h... yes > checking for dlfcn.h... yes > checking for X... libraries /usr/X11R6/lib, headers > checking for dnet_ntoa in -ldnet... no > checking for dnet_ntoa in -ldnet_stub... no > checking for gethostbyname... yes > checking for connect... yes > checking for remove... yes > checking for shmat... yes > checking for IceConnectionNumber in -lICE... yes > checking whether time.h and sys/time.h may both be included... yes > checking for dirent.h that defines DIR... yes > checking for opendir in -ldir... no > checking whether struct tm is in sys/time.h or time.h... time.h > checking for tm_zone in struct tm... yes > checking for socklen_t in sys/socket.h... yes > checking for atexit or on_exit... atexit > checking for tzset... yes > checking for gmtoff in struct tm... yes > checking for timezone and daylight variables... yes > checking for gethostname in unistd.h... yes > checking for yp_bind in -lnsl... yes > checking for socket in -lsocket... no > checking for dlopen... no > checking for dlopen in -ldl... yes > checking for sin in -lm... yes > checking for XOpenDisplay in -lX11... yes > checking for XShmAttach in -lXext... yes > checking for mawk... mawk > checking whether we can gnuify interp.c... no > checking for Open Sound System... yes > checking for prefix to use for loadable modules... (none) > checking for compiler flags to force 32-bit addresses... > checking whether byte ordering is bigendian... no > checking whether misaligned access to doubles is ok... yes > checking whether doubles are stored in Squeak order... no > creating Makefile fragments... mh_frag mt_frag > updating cache ./config.cache > creating ./config.status > creating Makefile > creating sqUnixConfig.h > sqUnixConfig.h is unchanged > > -- > "Man's mind, once stretched by a new idea, never regains its original > dimensions." Oliver Wendell Holmes (a Lisp programmer at heart)

1 0

Re: [BUG?]New Event Architecture Queue Sequencing?
by Bob Arning 04 Dec '00

04 Dec '00

On Wed, 29 Nov 2000 11:38:43 -0500 "Andrew C. Greenberg" <werdna(a)mucow.com> wrote: >Running the following script in Morphic with a transcript open, I get >some unexpected behavior. Andrew, I couldn't duplicate the problem, but there have been some things happening in this area, so I would ask: - which VM are you using? - what is the latest change set in your image? Cheers, Bob

2 2

[ENH] Better albaphetizer again
by Andres Valloud 04 Dec '00

04 Dec '00

Hi. Now, it works with all class organizations. No, it's not nice. Yes, it works. Andres.

1 0

Re: Making an inverted ellipse morph?
by Mark van Gulik 04 Dec '00

04 Dec '00

on 12/2/00 4:37 PM, Dino at deanmao(a)cc.gatech.edu wrote: > Actually, the Form>>replaceColor:withColor: thing doesn't work. The > stuff in the drawOn: method uses canvas functions that completely avoid > drawing the object if the color is transparent. Simply replacing the > color with transparency would seem to work, but it doesn't. If you > replace the color with transparency, you get a white background. > > It would be cool if somehow we could have boolean operations with shapes > to get "holes" in shapes... like circle xor: square... > > Seems like the only way to draw a circle cut out of a square is to do > the brute force method which I don't really like doing. In VisualWorks I would do this building a Mask that was the inverted ellipse shape (worst case, build a regular ellipse on the Mask and invert either its bits or its palette). Then I would draw the mask with the appropriate colored paint onto the final output device. In Squeak, the following seems to work: | dot | dot := Form dotOfSize: 400. dot offset: 0@0. dot copy: (0@0 extent: dot extent) from: 0@0 in: nil rule: Form reverse. dot displayOn: Display at: 0@0 clippingBox: (0@0 extent: 400@400) rule: Form oldPaint fillColor: Color green

1 0

RE: QKS and .NET (was: RE: Pentium 4 tinyBenchmark results)
by Alan Kay 04 Dec '00

04 Dec '00

Now, come on George, tell us a little more where the .NET VM did come from ..... Cheers, Alan ------ At 12:38 PM -0800 12/4/00, George Bosworth wrote: > >Isn't the .NET technology based on QKS' Smalltalk Agents? >no. QKS was one of a number of companies and institutions that ported their >stuff on top of an early pre-beta version of .NET > >-----Original Message----- >From: Jeff Szuhay [mailto:jeff@szuhay.org] >Sent: Monday, December 04, 2000 10:53 AM >To: Squeak Public Maliing List >Subject: QKS and .NET (was: RE: Pentium 4 tinyBenchmark results) > > >>P.S. Just got the betas of Micro$oft's Visual Studio.NET. >>Give me Smalltalk over C# any day! > >Isn't the .NET technology based on QKS' Smalltalk Agents? > >imagine the "innovation"... taking a now 20 year old technology >and basing a product strategy on it? > >BTW, I was reminded this weekend of the value of longevity of >certain technologies, in this case film. The remark was that any >100 year old negative could be used to make a picture. Can we say >that about computer media? > > > >-- > ************************************************** > Jeff Szuhay A randomly-directed > www.szuhay.org chaotical wetware pattern > jeff(a)szuhay.org recognizer/generator. > > > Make things as simple as possible, > but not simpler. > -- Albert Einstein

1 0

RE: StandardFileStream>>readOnlyFileNamed:
by Peter Crowther 04 Dec '00

04 Dec '00

> From: Tim Rowledge [mailto:tim@sumeru.stanford.edu] > Time to redo it all... separate the name from the stream, add some more > capable prims, clean it up. I think the key here is the more capable primitives. Most modern OSs can tell you whether a file exists. Some can even tell you whether you can read it or not (Windows NT, sadly, does not appear to be one of them --- there's a separate COM component to tell you). The question is, as always, whether the effort required to maintain N different implementations of the primitive is worth the gain. In this case, I'd argue that it is... but I don't maintain a VM :-). > System going down at 1:45 for disk crashing. (Chuckle). Next time you see Ian Piumarta, ask him about the number of disks mushroom went through at the University --- and what happens when you do a 23-hour coding stint and lose the disk! - Peter

2 1

[updates] 14 more
by Dan Ingalls 03 Dec '00

03 Dec '00

Oops; I meant these to be included with the last batch. These were culled by SMA (again thanks!). Tim (Exceptions-tpr.cs) and Duane (AlphaBlendOpt-dsm.cs) please don't feel slighted -- I need to check with Ian and Andreas respectively on these. - Dan Here are the summaries... ----------------------------------- 3063FindBtwSubstrs-dew -- Doug Way -- 15 November 2000 Adds a better comment to String>>findBetweenSubStrs: to make the behavior more clear." 3064FasterChangeSet-sqr -- Andres Valloud -- 14 November 2000 Greatly improves class order determination for changeset fileout" 3065BetterAlbaphetizer-sqr -- Andres Valloud -- 16 November 2000 A better category alphabetizer, with a convenient method so you can do Smalltalk sortAllCategories" 3066GCCPPCPatchDispatTab-jmm -- johnmci(a)smalltalkconsulting.com -- 22 November 2000 Patch the mpw gcc generated Squeak interpreter to remove a check for 256, produces faster byte code numbers, patch already exists for CodeWarrior" 3067ReadOnlyFile-mir -- Mike Rutenberg -- 4 September 2000 Files are usually opened using oldFileNamed: (which allows writing) when in fact they will only be read. This causes two problems: (1) file opening primitve will fail (to open the file for writing!) if the file is marked by the OS as readonly (including all files on readonly CDROMs) and (2) the underlying OS will believe the file needs to be locked to facilitate the writing until it is explicitly closed (rarely done!) or garbage collected. This is a conservative conversion of many uses of oldFileNamed: to readOnlyFileNamed: where it is clear the file is not written. No other modifications of the methods were done. 3068BrowserFix-ak -- Andreas Kuckartz -- 24 November 2000 Selecting 'class' button while '?' was selected resulted in both buttons being active." 3069ChngSorterKeyFix-sma -- Stefan Matthias Aust -- 26 November 2000 I reported a bug that the change sorter's changeset menu keyboard shortcuts were not working in MVC. As nobody fixed this, I had to do it myself..." 3070IRCConnectionFix-gwc -- Geoff Corey -- 21 November 2000 Attached is a fix to IRCConnection. The problem was the local variables being used in openConnectionDialogue were being garbage collected under NT and Macinstosh because of a missing #fixTemps call. My thanks go to Lex Spoon, Bob Arning, and Tim Rowledge for teaching an old dog new tricks." 3071SketchRepaintFix-jcg -- Joshua Gargus -- 19 November 2000 I experienced a problem where SketchMorphs that I wanted to paint were being shown clipped by the SketchEditor bounds. I traced this to the offset of the Sketch's form being added to the origin when the form is first painted on paintedForm. This appears to fix the problem." 3072CelesteAddons-dvf -- Daniel Vainsencher -- 18 November 2000 The addition of undo has broken the use of Celeste current by mail out changeset and the new mail out bug report. This change set replaces that mechanism." 3073BugReports-dvf -- Daniel Vainsencher -- 16 November 2000 Inspired by Chris Nortons recent idea, this CS adds a new option to the debugger, mail out bug report, which creates a good basis for such: the reciever, its state, the method vars, the stack, and all the image and vm info I could find. Comments, suggestions, feedback all welcome." 3074PasswordRequest-rca -- Russell Allen -- 13 November 2000 This short changeset changes the Password class to request passwords from the user with FillInTheBlankMorph>>requestPassword:, not FillInTheBlankMorph>>request:" 3075UndoWorks-sqr -- Andres Valloud -- 14 November 2000 Fixes some problems with Undo" 3076CelesteFixes-beo -- Bruce O'Neel -- 3 December 2000 - updateToc has left the building. Replace the call with cacheTOC. - mail messages weren't parsing correctly for TOC entries. Changed comparisions with field names such as 'date' to case insensitive comparisons. - attachment separaters were looking for content-type rather than Content-Type."

1 0

← Newer
1
...
21
22
23
24
25
26
27
28
Older →

Jump to page: