Squeak-dev May 1999

squeak-dev@lists.squeakfoundation.org

126 participants
289 discussions

[FIX] Cancelling 'save as...' fix, plus enhancement
by Doug Way 04 May '99

04 May '99

This change set fixes a bug in which cancelling an image "save as..." operation from the main menu causes a "zero length filename" error. This bug appears to have been around since at least 2.0, if not longer. Hitting the Cancel button on the Morphic FillInTheBlank dialog for "save as..." will now simply cancel the save. Deleting the entire string and accepting will also have the effect of cancelling the save, for both Morphic and MVC. (Which makes sense, because hitting Cancel on a FillInTheBlank returns an empty string.) An additional enhancement is also included, which changes the default image name in the 'save as...' FillInTheBlank from 'NewImageName' to the actual current name of the image. This is a standard UI convention for Save As dialogs, with good reason... I often want to make a slight modification to the name of the image I'm saving, so I'd rather not have to type the whole thing in again. Sometimes I don't even remember the current name for the image (especially in full-screen mode), so including it in this dialog helps. Also, 'NewImageName' is a redundant hint in the Morphic FillInTheBlank, since the title already says 'New File Name?' Both of these changes are in a single method, SystemDictionary>>saveAs. The changeset is at: http://www.mindspring.com/~dway/smalltalk/SaveAsFixes-dew.cs - Doug Way EAI/Transom Technogies, Ann Arbor, MI dway(a)eai.com, dway(a)mat.net http://www.transom.com Smalltalk: Guaranteed Y2T Compliant

1 0

[FIX] 2.4 diffs for Acorn plat stuff
by Tim Rowledge 04 May '99

04 May '99

Couple of minor diffs in two platform spec files that need folding in:- Diff files 'raFS::sq24.$.sqConfig/h' and 'raFS::sq24.$.h.sqConfig' after raFS::sq24.$.h.sqConfig line 27: # define USE_CLOCK_MSECS remove raFS::sq24.$.sqConfig/h line 28: # define LITTLE_ENDIAN_BITBLT after raFS::sq24.$.h.sqConfig line 210: # endif remove raFS::sq24.$.sqConfig/h line 212: # if defined(ITSY) remove raFS::sq24.$.sqConfig/h line 213: # define LITTLE_ENDIAN_BITBLT remove raFS::sq24.$.sqConfig/h line 214: # endif Diff files 'raFS::sq24.$.sqPlatformSpecific/h' and 'raFS::sq24.$.h.sqPlatformSpecific' change raFS::sq24.$.sqPlatformSpecific/h, line 46 to 46 line 46: #define sqFilenameFromString(dst, src, num) sqFilenameFromString(dst, src, num) to raFS::sq24.$.h.sqPlatformSpecific, line 46 to 46 line 46: #define sqFilenameFromString(dst, src, num) sqFilenameFromString(dst, src, num) after raFS::sq24.$.h.sqPlatformSpecific line 48: remove raFS::sq24.$.sqPlatformSpecific/h line 49: #ifdef LITTLE_ENDIAN_BITBLT remove raFS::sq24.$.sqPlatformSpecific/h line 50: #define ifLittleEndianDoelseDo(a, b) a remove raFS::sq24.$.sqPlatformSpecific/h line 51: #else remove raFS::sq24.$.sqPlatformSpecific/h line 52: #define ifLittleEndianDoelseDo(a, b) b remove raFS::sq24.$.sqPlatformSpecific/h line 53: #endif - basically leftovers from the withdrawal of the little-endian bitblt stuff. tim -- Useful random insult:- A black-and-white mind working on a color-coded problem. Tim Rowledge: rowledge(a)interval.com (w) +1 (650) 842-6110 (w) tim(a)sumeru.stanford.edu (h) <http://sumeru.stanford.edu/tim>

1 0

RE: [PWS] PWS only meant for Swiki?
by Lex Spoon 04 May '99

04 May '99

> >I am curious how, in practice, a Swiki performs. In particular, how > >practical are these things for communities of several hundred people? > The current Swikis haven't been optimized much at all, but as Mark Guzdial posted earlier, they are handling his ~100 student sophomore class without trouble. I can quantify this a little more. We did some quick tests on a Mac (Bolot, Je77, do you know what kind of Mac it is?), and found two things that are bottlenecks right now: 1. Reading a Swiki's source text, ie SwikiPage>>text, takes over 100 milliseconds by itself. This could obviously be improved: contentsOfEntireFile on the samy file took like 15-20 milliseconds. 2. "swikifying" a page is currently dominated by the check for whether a particular line break is in between < and >. This check is needed, but it could be made more efficient. With the check, the time taken is about 200 milliseconds. Without the check, the time is down in the 10-20 millisecond range, if I remember right. It was definately a LOT less. So in summary, on this modest Macintosh: 1. Currently, it can sustain about 2 Swiki page requests per second, which is reasonable for a group of a few hundred people. 2. It can be easily improved to at least 20 regular requests per second (50 ms per hit). Lex

4 3

Re: [? Windows] CR instead of CR/LF (was: RE: update 1164)
by Reinier van Loon 04 May '99

04 May '99

Chris, > I've tried Notepad, WordPad and MS Word 98 so far. NotePad and >WordPad fail miserably, but MS Word seems to work. Do any of you Windoze >Squeakers have a better tool for reading these files from the File Manager >or from the Explorer (besides Squeak itself)? I use PFE (programmers file editor). It distinguishes between CR, CRLF and Unix conventions. Things look good in PFE. It's small and fast and I have it as a link in my Sent To folder. That way, I can send any file to PFE without registering a file type. Reinier.

2 1

update 1164
by Richard L. Peskin 03 May '99

03 May '99

I get a "selector changed" error when tring to load update "1164projectFlapPref-sw " into 2.4b. Has anyone else encountered this? --dick peskin ================================= R. L. Peskin, Rutgers Univ. ;<peskin(a)caip.rutgers.edu>;<http://www.caip.rutgers.edu/~peskin> RLP Consulting <rlpcon(a)sover.net>; <http://www.sover.net/~rlpcon>

5 7

Re: update 1164
by Lawrence Press 03 May '99

03 May '99

> >Why does my image not shrink after Smalltalk majorShrink? > >Reinier. Have you tried doing the majorShrink in the 2.3 VM, rather than the 2.4 VM? The problem seems to be with Symbol rehash (though I have no idea why VM version should affect it). Lawrence Press

1 0

Re: update 1164
by Reinier van Loon 03 May '99

03 May '99

> This is almost certainly a problem with your .changes file, most likely due to erroneous text conversion during > download or unpacking. Why was CR chosen for Squeak instead of CR LF? BTW How can I generate the sources anew? Why does my image not shrink after Smalltalk majorShrink? Reinier.

2 1

(BUG?) morphic click down
by Don Charnley 03 May '99

03 May '99

Hi, This issue seems a little specific for this group, and maybe someone can shed some light on it. I have a morphic application with several layers of morphs present. When I start it, all works well, and clicks (mouseDown) are received by the target morph, which is inside another morph. If I then change the focus to another window, say a browser, and then return to the application, subsequent clicks are not passed through to the target morph. I can tell this by inserting halts in the mouseDown routine in the target. Of course, the handlesMouseDown always returns true. Anyone have clues what I might be doing wrong, or a way to fix this. It would appear to be a side effect of something, and I can't tell what. As far as I can tell, I am not manipulating any of the morphic structures or instance variables with my own code. don

1 0

[PWS] PWS only meant for Swiki?
by Reinier van Loon 03 May '99

03 May '99

I tried to use a PWS for something else then a Swiki. I created a Server directory, but did not copy the files as mentioned. I copied the method initializeAll and stripped some actions. I tried to start the server but I got the message 'The path to the Server folder is wrong. Please modify the following method ...' But the path to the Server folder wasn't wrong, there was a folder Server. It turned out that the path to ...\Server\Swiki\page.html did not exist. This would have been no problem if I had copied the server folder and it's easily fixed by stripping the checkVersion message send. But the message is misleading. And I think the checkVersion method is bit Swiki minded. We should put something like that in a method called checkSwiki. Or is PWS only meant for Swiki's? Macarena.

6 7

Re: JPEGReadWriter, Differences between Squeak and VisualWorks
by Leandro Caniglia 02 May '99

02 May '99

Tim -- >Unfortunately, the Interval>>to:by:do: use that you point out in > > JPEGReadWriter>>idctBlockInt:component: > >is there because that method is on the edge of not compiling -- the >required frame size is close to exceeding the maximum of 32 for a large >frame. Changing it to use the compiler in-lined to:by:do: increases the >requirements to over that limit. Tim. I think that Andres is talking about JPEGReadWriter>>makeDerivedTables where the (to:by:)do: is inside another do: /Leandro

1 0

Jump to page: