Squeak-dev May 1998

squeak-dev@lists.squeakfoundation.org

82 participants
182 discussions

Mouse Buttons and Modifier Keys (was Yellow Button and Morphic)
by Maloney 29 May '98

29 May '98

Hi, Bill. Re: At 11:33 AM -0400 5/29/98, William O. Dargel wrote: >Where should I be looking to get a perspective on how input events, >particularly the mouse, are handled with Morphic? Look at the event-handling code in HandMorph. Re: >One thing that I noticed: it looks like any mouse button will act like >the "yellow" one if the "Option" key is pressed. Trouble is, I don't >have an option key. As I recall, that's something that's on the Mac? >Does the Mac still sport a single button mouse? Sorry, I'm culturally >deprived ;-) The Smalltalk image from which Squeak is derived was designed to run with a three-button mouse on an Alto or Dorado computer. These buttons were known as "red", "blue", and "yellow", and some mice really had buttons of these colors. Win32 machines have only two mouse buttons, so the mapping is: left-mouse -> "red" alt-left-mouse -> "blue" right-mouse -> "yellow" The Mac has only a single button, so the mapping is: mouse -> "red" cmd-mouse -> "blue" option-mouse -> "yellow" I believe the mapping of modifier keys is: Mac Win32 Unix shift shift shift ctrl ctrl ctrl cmd alt meta(?) option --- --- I don't believe there is any Win32 or Unix equivalent of the option key. Andreas and Ian, please correct me if I'm wrong. If that's the case, the option key should be used only to simulate the yellow button on the Mac, and should not be tested for directly, since functions that depend on the option key would not be accessible on other platforms. Unfortunately, this is not the case right now-- we do have some code that tests for the option key. Re: >On a more general note: Are there any efforts going on that are related >to making the user input interface more easily configured or customized, >without the need to change methods in the base system? Not that I know of. -- John

2 1

Spawn this file missing in 2.0
by Mark Guzdial 29 May '98

29 May '98

I just noticed that "spawn this file" is missing from the FileList menu of 2.0. The below change seems to bring it back: 'From Squeak 2.0 of May 22, 1998 on 29 May 1998 at 1:14:37 pm'! !FileList methodsFor: 'file list menu' stamp: 'mjg 5/29/1998 13:14'! fileSelectedMenu: aMenu ^ aMenu labels: 'fileIn file into new change set browse changes spawn this file copy name to clipboard open image in a window read image into GIFImports play midi file sort by name sort by size sort by date rename delete add new file broadcast as update' lines: # (5 7 8 11) selections: #(fileInSelection fileIntoNewChangeSet browseChanges editFile copyName openImageInWindow importImage playMidiFile sortByName sortBySize sortByDate renameFile deleteFile addNewFile putUpdate) ! ! -------------------------- Mark Guzdial : Georgia Tech : College of Computing : Atlanta, GA 30332-0280 (404) 894-5618 : Fax (404) 894-0673 : guzdial(a)cc.gatech.edu http://www.cc.gatech.edu/gvu/people/Faculty/Mark.Guzdial.html

1 0

Deploying "shrink-wrapped" Squeak Apps?
by Paul Fernhout 29 May '98

29 May '98

What's the current thinking on deploying "shrink-wrapped" Squeak apps? Does anyone think it is doable at this point (in terms of stability)? Are any VMs in good enough shape to let non-programmer end customers use (especially without extensive further in-house testing on the relevant hardware)? If Morphic is used in an app, how does one prevent end users from rearranging/deleting the GUI by accident? Any general comments on this issue? -Paul Fernhout Kurtz-Fernhout Software ========================================================= Developers of custom software and educational simulations Creators of PlantStudio(TM) Botanical Illustration Software http://www.kurtz-fernhout.com

2 1

reading objects from DataStream
by Bolot Kerimbaev 29 May '98

29 May '98

If somebody knows what happens, please advise. I implemented a classroom management (turnin/grading) system for Mark Guzdial's class, and recently we discovered the following problem (I am skipping hours of raiding for the lost shark): Large turnins (i.e., for which text instance variable is >16384 bytes) could not be recovered from a DataStream. Running a debugger revealed that the error occurred after the text (that was supposed to be over 16K long) was recovered into an empty String. However, the error was raised only when the next instance variable was being read: DataStream raised 'Expected start of object but found 0' error. It is my understanding that the text instance variable (containing a String at the time of save) was written as ByteArray (because it was longer than 16384 bytes). However, when it was read from DataStream, it did not recover correctly (tried to read as String?). Bolot Kerimbaev PS. I am using Mark Guzdial's FileDictionary, which provides a Dictionary like access to data in a PositionableDataStream (which extends DataStream with position, position:, and setToEnd methods). This is a message I sent to Mark and his TAs. ---------- Forwarded message ---------- Date: Thu, 28 May 1998 17:16:15 -0400 (EDT) From: Bolot Kerimbaev <bolot(a)cc.gatech.edu> To: CS 2390 Admin <cs2390-ta(a)cc.gatech.edu> Subject: saga: root of problems New theory: When a Turnin is saved to disk, it iterates through its instance variables and tries to save each of them. Since code is-a String, it is subject to the 16384 byte limitation (DataStream>>writeString), i.e., instead of trying to save it as a String, DataStream saves that large chunk of code as a ByteArray. But ByteArray and String use 4 and 2 bytes to save length respectively. In other words, they are different. Now, when that Turnin is read from the disk, Turnin tries to restore code (which is in Turnin's text instVar). But for some reason (have not figured out how exactly), it stores '' (at least on my test runs) in text, and then chokes on Timestamp. Apparently, it does not correctly process text instVar, which messes up its DataStream position counter, and then tries to read Timestamp from an incorrect position! I reproduced the problem using 'bare' FileDictionary: | sqFD data | data := Turnin new initialize; text: (String new: 17000 withAll: $a). sqFD := FileDictionary fileNamed: 'test'. sqFD open. sqFD at: 'fred' put: data. sqFD close. and then to read (and crash): | sqFD data | sqFD := FileDictionary fileNamed: 'test'. sqFD open. [data := sqFD at: 'fred' ] ifError: [:msg :rec | Transcript show: 'This is the error! Trying to read Timestamp, but Turnin instVar>>text was not properly handled'; cr. sqFD close]. sqFD close. " you won't get here " data inspect. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Bolot Kerimbaev (bolot(a)cc.gatech.edu) ------------------------------------- The difficult -- we'll do immediately. The impossible will take a bit longer.

2 1

Rich Text Processing in Squeak
by Dan Shafer 29 May '98

29 May '98

I've delved somewhat into the classes in Squeak 2.0 Beta that support rich text editing and formatting and I find them quite intriguing. Comparing them to the Swing Set classes in Java for the same general purpose, they seem quite robust, though they fall short at the document level (unless I'm missing something in Squeak, which is certainly possible). I'm trying to compare the two environments in terms of their suitability for writing a net-based word processor. Are there any example applications out there that point the way and perhaps do a better job of explaining the classes that would support such an effort in Squeak? -- |=================================================| | Dan Shafer, Executive Technical Producer and | | Editor-at-Large, CNET's Site for Site Builders | | dshafer(a)cnet.com - Preferred Method of Contact | | http://www.builder.com - Main Builder Site | |http://buzz.builder.com - Online Discussion Area | |http://web.builder.com - Semi-Annual Conferences| | Save the Dates - Dec. 7-9, 1998, Web.Builder | | New Orleans Strikes Again! | |=================================================|

1 0

Re: More general problem for .sqo files (Re: reading objects fromDataStream)
by Bolot Kerimbaev 29 May '98

29 May '98

I've made a 'quick-n-dirty' fix to the DataStream's behavior. DataStream>>readString method uses PositionableStream>>nextString to do the job. The latter only handles Strings that are shorter than 16384 bytes. Further, it appears that when a String is saved, DataStream saves Strings signature, and only then checks if it is longer than 16K. If so, it saves it as a ByteArray. I changed PositionableStream>>nextString method, and my program immediately recovered seemingly lost data. The change is as follows: I read a byte that designates length. If it is zero, I peek the next byte, if that one is zero as well, I assume that this String is longer than 16K, thus 4 bytes comprise the length byte. It has not been tested extensively with other applications or Squeak itself (e.g., this could break if you legally had two consecutive zeros). Please let me know if I am playing with fire. Bolot ---DO NOT FILE IT IN YET! LET'S HEAR WHAT SQUEAK CENTRAL SAYS--- PositionableStream>>nextString "Read a string from the receiver. The first byte is the length of the string, unless it is greater than 192, in which case the first two bytes encode the length." | aString char length| length _ self next. "first byte." length >= 192 ifTrue: [length _ (length - 192) * 256 + self next]. " I added these lines " (length = 0 and: [self peek = 0]) ifTrue: [length _ (length bitShift: 24) + (self next bitShift: 16) + (self next bitShift: 8) + self next]. aString _ String new: length. 1 to: length do: [:i | aString at: i put: self next asCharacter]. ^aString

1 0

Re: Squeak for UnixWare
by Ian Piumarta 29 May '98

29 May '98

> During linking an error occurred: some references could not be resolved: > .XShmDetach [...] > Is it possible that I need X11 Release 6 Either that, or I need to add something that allows XSHM to be disabled from the compile command. Yes, that's the solution -- I'll add a test for NO_XSHM (along with the current test for HP-UX that also disables shared memory ;) and then you just need to add -DNO_XSHM to the CFLAGS. Look near the beginning od sqXWindow.c and you'll see where you need to change #ifdef HPUX into #if defined(HPUX) || defined(NO_XSHM) (or somesuch). Bye, Ian

1 0

Re: Squeak for UnixWare
by Ian Piumarta 29 May '98

29 May '98

> Oh my. I actually managed to get it working with just: > cc -D__GNUC__ -o Squeak *.c -lX11 -lXext -lm -lsocket -lnsl > I still like makefiles better, though. Adding "-D__GNUC__" to the CFLAGS definition in the UnixWare section of GNUmakefile.conf should do the trick... Ian

2 1

Re: Squeak for UnixWare
by Sabine van Loon 29 May '98

29 May '98

Chris, > 2) Edit interp.c and comment out the line: > #include "sqMachDep.h" /* needed only by the JIT virtual machine */ Alternatively, you could define __GNUC__ (like in cc -D__GNUC___ -oSqueak *.c). That's how I do it for the AIX. You probably won't be needing steps 2 to 4 then ;-). Cheers, Reinier.

2 1

Re: Squeak API support
by David Anderson 29 May '98

29 May '98

>>Hi Ale, >Hi! Hi all... :^) >>I am using your sqAPI2.0.exe file, and when I run your test examples, >> some of them crash the VM, giving me a crash dump as below. >Me too! :^( >But it work ok during debug ! >(I am using Microsoft Developer Studio for C++ version 4.0) Same here, except I am using version 5.0. Works during debug, but not release. >Can any squeaker help us? >Are there any special consideration to compile the VM under Win32 platform? >Ale. I'm won't consider myself a true squeaker yet, but I did make some progress toward a solution and it was Andreas' response to an eariler post of mine that led me to it. Basically, I set the compiler to use __stdcall functions instead of __cdecl functions. Now everything works except your call back tests. I haven't played with your callback routines yet, so I figured with this info it might be easier for you to track down what the problem might be then for me. Just in case you don't know... The compiler option is something like /gx (or in the code generation portion of the compiler options). When I went to compile, it gave me one error (atexit() requires the passed function to be of type __cdecl, so I just declared void Cleanup() as void __cdecl Cleanup()). When I recompiled everything seemed happy. No more crash during the sleep call, or any of the other examples for the api, just during the callback examples. Dave Anderson deandrsn(a)uiuc.edu

1 0

Jump to page: