Squeak-dev November 2002

squeak-dev@lists.squeakfoundation.org

189 participants
695 discussions

[BUG] Install SqueakMap in 3.2.1
by Jim Benson 04 Dec '02

04 Dec '02

Hi, I attempted to install SqueakMap in a freshly created 3.2.1 image. I created the 3.2.1 image by: 1. Downloading Squeak 3.2-4956.zip from: ftp://st.cs.uiuc.edu/pub/Smalltalk/Squeak/3.2/ 2. Performing 'update code from server' 3. Selecting 'advance to 3.2.1' when prompted 4. Save image to 'Squeak 3.2.1-4957' after update. 5. Close image. 6. Reopen image 5. Execute: ChangeSorter newChangesFromStream: (('http://map2.squeakfoundation.org/sm/packagebyname/squeakmap/downloadurl' asUrl retrieveContents content) asUrl retrieveContents content unzipped readStream) named: 'SqueakMap'. 6. The attached screenshot shows the error message: 'SqueakMap can only be installed in 3.2.1 or 3.4 alpha (update 5104 or later).Please load updates to your image to reach one of those versions and try again! I will now halt this filein so you can abort it. (3.2.1 is the maintenance branch of 3.2 so there is no real point in staying at 3.2)Windows XP boxJim Benson

4 5

Re: [BUG?] objects tool has no pictures
by Ned Konz 03 Dec '02

03 Dec '02

On Friday 29 November 2002 01:35 pm, Ross Boylan wrote: > The objects browser now mostly doesn't show pictures of the objects > it makes available (o off the main menu). It used to; I have 3.4a > #5125. Is this by design? Is there a way to change the behavior? Yes, by resetting the thumbnails <g>... > AudioChat under Collaborative and the StarSqueak items show images > in the tool; nothing else does. > > Ah, I found reset thumbnails off the Morphic main menu, and that > fixed things up. But shouldn't they be on by default? They should be. I don't know why this happened (if it helps, they're OK at 16 bit depth; it's just when you go to 32 bits that you can't see them). -- Ned Konz http://bike-nomad.com GPG key ID: BEEA7EFE

2 1

[BUG]3.4-5125 FileDirectory>exists faulty?
by Tim Rowledge 02 Dec '02

02 Dec '02

I've just discovered a nasty side effect of the recent change to Filedirectory>exists (update 5114). The immediate effect is that is stops VMMakerTool from working on Acorn, but I can't see how it works properly at all. Given the latest CVS platform sources I have (a few days old) it looks to me that unix, win32 & Acorn all carefully handle _not_ returning an entry for the directory entry in prim dir_Lookup(). I can't tell about Mac since the way it works is so foreign to me. The new update seems to rely on the prim actually returning an entry with quite specific contents. I'm puzzled; obviously I could change the Acorn VM but a) it would break backwards compatability (big deal eh? nobody uses Acorn Squeak you think? Wrongo, dude, wrongo) b) since it looks like it is wrong on win32/unix, why change to match wrongness? I wonder when we will actually get a filename system that works cleanly... sigh. tim -- Tim Rowledge, tim(a)sumeru.stanford.edu, http://sumeru.stanford.edu/tim Useful Latin Phrases:- Fac ut vivas. = Get a life.

3 9

"Pattern Hatching"
by Richard A. O'Keefe 02 Dec '02

02 Dec '02

I've just been reading the book "Pattern Hatching: Design Patterns Applied" by John Vlissides, Addison-Wesley 1998, ISBN 0-201-43293-5. I have come away from this with two overwhelming impressions: (1) Here is a better than competent OO programmer, and C++ is fighting him every inch of the way. Example after example is basically "how do I design-hack a way around such-and-such a limitation of C++." I've always been fairly positive about patterns, and I've taught them to students. This book, by one of the GoF, made me wonder. The examples strike me as Heath Robinson-ish; is this because of pattern-think, or is it because of C++? It would be really illuminating to see some of his examples revisited by a better than competent Smalltalker, because I think you'd end up with a completely different set of patterns being applied, or the patterns would look different. (2) One of the chapters describes the "GENERATION GAP" pattern. The example is a visual interface builder, where you draw your interface and the tool then generates code. "Normally, you can't create a finished, nontrivial application automatically with one of these tools; you have to implement some functionality the old-fashioned way, by hand in a programming language." So GENERATION GAP helps you by doubling the number of classes. (Er, really.) But WHY do you have to develop part of the interface in one environment (the interface builder) and the rest in another environment? Why shouldn't "what to do when you are pressed" be something you can plug in *and test* in the interface builder? The answer appears to be "because the interface builder is generating C++" (see point 1 above) "and can't itself run the code it generates". POOF! An entire pattern disappears up its own fundament if your environment can represent code as data and can run dynamically generated code. I just wish I understood Morphic better. Smalltalk: back to the future????

9 11

70 packages on SqueakMap!
by goran.hultgren＠bluefish.se 01 Dec '02

01 Dec '02

Yiha! Amazing. Really cool guys, I feel the pressure to make SM even better... regards, Göran

4 10

[ANN] v3 of StarBrowser registered on SqueakMap: try it!
by Ned Konz 01 Dec '02

01 Dec '02

The Squeak port of the StarBrowser has been considerably improved, and is now ready for day-to-day use (that is: I'm using it myself). It is registered on SqueakMap, and can also be found using the SqueakMap web interface at: http://map2.squeakfoundation.org/sm/package/9187f67b-a437-4e52-93f9-eecb9b4… It now also includes a basic SqueakMap browser/loader and a DVS UI. Try the window frame menu and the context menus on individual items, the toolbar buttons, and drag/drop to/from other browsers and within or between StarBrowsers. It's not complete yet; I've noted some of the missing parts are in the install/changes.txt entry in the SAR. Most notably, don't bother trying to use the history arrows (<- and -> buttons), as they may confuse the browser. However, the left-most button (hide/show tree) works fine. This is the third version of the port of the StarBrowser from VisualWorks to Squeak. Most of the functionality works (the classifications model and an inspector and class editor, and lots of other features.) Drag and drop work (with thanks to Ned for the help). Ned re-factored the menu handling and button handling into more service classes, and added DVS, Morphic, and SqueakMap support as examples of extension. The goal of the Star Browser is to allow you to browse your Smalltalk environment and classify anything you encounter while doing so. Therefore it has full drag&drop support to allow you to classify software elements such as classes and methods while you are browsing. It is easily extended to handle browsing most anything in the Squeak environment. I'm interested in hearing any bug reports, and will gladly consider extensions. Enjoy! -- Ned Konz http://bike-nomad.com GPG key ID: BEEA7EFE

4 12

SUnit swiki broken links
by Ross Boylan 01 Dec '02

01 Dec '02

http://minnow.cc.gatech.edu/squeak/1547 on SUnit has its first 2 links broken. I'm looking for a basic description of SUnit and how to use it, if anyone has any suggestions. I'm wandering around the other links now....

2 1

[FIX][VM][UNIX] Directory deletion fix
by Ned Konz 01 Dec '02

01 Dec '02

There is a problem exhibited by the Mac and Unix VMs: If you access the contents of a directory, then delete it, then try to access its contents again, the directory is not properly diagnosed as being nonexistent. This is caught by the FileDirectoryTests. In the Unix VM, this is due to the opened directory handle not being invalidated when the directory is deleted. The attached patch to the SF 3.2-6 sources fixes this problem by marking the opened directory as invalid when it's deleted. This does require the name to be the same (i.e. watch out for trailing slashes!). -- Ned Konz http://bike-nomad.com GPG key ID: BEEA7EFE

1 1

[ENH][GOODIE] IOHandle V1.2
by David T. Lewis 01 Dec '02

01 Dec '02

IOHandle V1.2 for Squeak3.4alpha. This is beta level code, bugs are likely. IOHandle adds classes which represent input/output on external channels such as files or sockets. The FileStream and Socket hierarchies are modified to use IOHandle rather than to directly invoke primitives. Instances of FileStream or Socket collaborate with instances of IOHandle to separate streaming and protocol behaviors from behaviors related to external device and network access. The purpose of IOHandle is to: - Separate the representation of external IO channels from the represention of streams and communication protocols. - Provide a uniform representation of IO channels similar to the unix notion of treating everything as a 'file'. - Simplify future refactoring of Socket and FileStream. - Provide a place for handling asynchronous IO events. Refer to the aio handling in the unix VM. Files, Sockets, and AsyncFiles could (should) use a common IO event handling mechanism (aio event signaling a Smalltalk Semaphore). If used with OSProcess, please update to OSProcessV3.0.4. OSProcess should be loaded prior to installing IOHandle (if loading OSProcess after IOHandle, just proceed through several error notifiers and all will be well). I have not yet integrated this with Flow (it's a lot of work). If there is interest, I'll have a bash at it as soon as Flow can be loaded into an image along side the existing classes. These change sets put my initials (dtl) on a lot of methods which were really written by other people. This needs to be fixed; I can take credit for lots of new errors but not for the code. I moved the registries for StandardFileStream and Socket into class IOHandle. I'm not sure I got this all right, and I probably screwed up Andreas' StandardFileStream>>retryWithGC:until:forFileNamed: when I replaced it with IOHandle>>retryWithGC:until:forRegisterObject. Also, I moved the registry threshold methods from Socket to IOHandle. They were not being used in Socket, but these appear to be in place to prevent the registry from becoming too large. Removing things from a large registry is extremely slow, so it would probably be a good idea to put these methods to use. There are some minimal hooks for aio event handling in class IOHandle. I have done an implementation of aio event forwarding for Unix pipes, and it works quite well (I have not released that code yet). Class IOHandle would be a good place to put this to allow the event handling to work identically for sockets, pipes, and files. IOHandle cannot be easily removed from the system, so please begin with a fresh copy of your image before loading IOHandle. Dave

1 0

[ENH][GOODIE]FIX] OSProcess V3.0.4 (updates to support IOHandle)
by David T. Lewis 01 Dec '02

01 Dec '02

This is an incremental patch for OSProcess 3.0. Provides fixes to permit OSProcess to function correctly with IOHandle V1.2. Dave

1 0

Jump to page: