Squeak-dev February 2000

squeak-dev@lists.squeakfoundation.org

163 participants
655 discussions

Re: How Do You Do Business Apps? (Squeak/Pro Proposal)
by Lex Spoon 01 Mar '00

01 Mar '00

Most of your proposal sounds great (except that if I could personally afford to pay someone to develop squeak stuff, I'd probably pay myself :)). I have issues with a few of your issues, however: > * either a couple of PD or licensed fonts or maybe OS loadable fonts (if > legal), The current fonts are for all practical purpsose free, right? Now maybe we need *better* fonts, but that's a different issue. > * command line scripting, Squeak already has this. "Squeak myimage.image myfile.st". Now it could be more *useful*, but that seems like a different issue. Also, I don't think Squeak's biggest selling point is scripting, anyway. Even for an app that is ultimately headless, I'd generally prefer to use the GUI to *develop* the app. > * VM GUI events (maybe), and What do you mean here, exactly? Morphic uses events for most everything. If you are talking about having the VM pick up mouse click events instead of polling the mouse state, well, that would be nice, but it hardly seems like a critical issue for most people. Lex

3 2

Re: [CCC] False class comment
by Lex Spoon 01 Mar '00

01 Mar '00

Bijan Parsia <bparsia(a)email.unc.edu> wrote: > > [snip] > >> Oh yeah, and this is true for both True and False. So maybe we should > >> just put this stuff in class Boolean, instead of duplicating it in two > >> places. > > > > Even more important IMHO is to always thing about that when reading a > > comment it should be self contained. That means we don't want to move > > parts of the comment to the superclass only because that part is true > for > > multiple subclasses. If the part is really large, we might thing about > an > > explicit "see superclass for more details" but othervise, I'd vote to a > > small redundancy in favor for comments you can understand without > knowing > > the context. Keep in mind, the comment is for references not just a > part > > of a larger explaination. > > Even in the spirit of anti-redundancy, I do agree that comments should be > reasonably non-fragmented. But actual duplication leads to a degree of > maintenance pain. So perhaps an inclusion mechanism would be > feasible? After all, we already have links, it can't be *that* hard to > pull the actual text in. > Morph and its 245 subclasses would form a rather striking example.... Granted, True/False is just two clasess, and so it isn't such a big deal. IMHO, though, people reading Smalltalk code should just expect to have to look at the superclasses for the whole picture. But of course, redundent comments are better than none. :) Lex

2 1

Virus???!!!: "Check this" message posted earlier
by Michael Rueger 01 Mar '00

01 Mar '00

Hi all, I'm not quite sure what to make of this, but it seems our mail system has detected a virus in this message posted to the mailing list? Or what do others see when looking at this message? Michael "Jerome E. Garcia" wrote: > > **** WARNING **** WARNING **** WARNING **** WARNING **** > > A virus was detected in this message and the part(s) of the > message which were infected have been automatically deleted. > > DO NOT resend the message until the virus infection has been > removed up from your PC or the infected content will be > deleted again. > > The virus name was VBS/FreeLink > > ----------------------------------------------------------------------------- > Have fun with these links. > Bye. > > ----------------------------------------------------------------------------- > Name: LINKS.VBS > LINKS.VBS Type: application/x-unknown-content-type-vbsfile > Encoding: quoted-printable -- "To improve is to change, to be perfect is to change often." Winston Churchill +------------------------------------------------------------+ | Michael Rueger | | Phone: ++1 (818) 623 3283 Fax: ++1 (818) 623 3559 | +---------- Michael.Rueger.-ND(a)corp.go.com ------------------+

3 2

[VIRUS WARNING] Re: Check this
by Duane Maxwell 29 Feb '00

29 Feb '00

This appears to be a Visual Basic virus. DON'T EXECUTE IT! >Have fun with these links. >Bye.

1 0

[CCC] MarqueeMorph (was Re: [CCC] False class comment)
by Stefan Matthias Aust 29 Feb '00

29 Feb '00

>And now we'll both owe some comments :) You want a comment? Here is one... ;-> MarqueeMorph ---------------- The MarqueeMorph is a subclass of the BorderedMorph which quickly cycles its border color. It's only used in The FatBitsPaint morph and should be moved out of the Morphic-Kernel category. Its implementation could be simplified and it is generally not very useful. The color values and cycle speed are hard-coded for example. Finally, it's anoying to look at. ------------------ Okay, okay, I know I said that one shouldn't critic a class but this one - well - I stumbled across it when I investigated that really strange cut/copy selection mechanism of the FatBitsPaint morph. I'm so sorry. It's getting late here. bye -- Stefan Matthias Aust // Bevor wir fallen, fallen wir lieber auf.

1 0

Re: [ENH] Warning when overwriting a clean image
by Doug Way 29 Feb '00

29 Feb '00

On Mon, 28 Feb 2000, Tim Rowledge wrote: > Nice idea! Add in my other-than-startup-directory saving and things > ought to be much safer. Thanks. Yeah, being able to save the image in a subdirectory would be handy. Oddly enough, I was behind on my Squeak mail and didn't even see the discussion about this until after I posted my changeset. I liked Bert's idea of bringing up a Save As... when first saving, but given Jesse's comments, I think bringing up the warning (plus the preference) is probably the best compromise. - Doug Way EAI/Transom Technologies, Ann Arbor, MI http://www.transom.com dway(a)mat.net, @eai.com

1 0

RE: AIDA/Web license (was RE: Zope)
by agree＠carltonfields.com 29 Feb '00

29 Feb '00

> Hello Squeakers, > > As an author of AIDA/Web (http://www.eranova.si/aida/) I > like to clarify a bit the license issues mentioned in this > thread. > > To be honest, I just didn't care about those license > problems too much and AIDA was released under for me in that > moment most appropriate license. Also, here we don't > complicate so much about licensing issues so I really need > help from you o find a more appropriate license. > > My intention with AIDA is to help spread Smalltalk in web > applications and on the other side that community help me > developing AIDA further. We all can then make money > providing services and custom solutions based on AIDA. > > Therefore, I'm willing to change license to be more lawyer > proof :)) > > Please help me find such a license. Can we agree for a > common license for all Smalltalk OS projects? Should I start > with LGPL? Three ways to go: BSD-based: Your product is so similar in nature to the Apache server, and Apache's license is so reasonable, that it rates to be a good fit. http://www.apache.org/LICENSE.txt I note that this license (the so-called "old-BSD" license retains the "advertising clause" which RMS considers to be a "bad thing." He does approve of the far simpler "new BSD" license http://www.xfree86.org/3.3.3/COPYRIGHT5.html, which would serve just as well. Squeak-L: An alternative license, which is also highly liberal and decidedly Smalltalk friendly, is the Squeak license. http://www.squeak.org/license.html, removing the Apple-specific terminology, since your product is not bound by Apple code at its base. Squeak-L, like Mozilla and Apple's Open Source license, requires publication of source code to modifications, regardless of whether the changes have been distributed to others. However, Squeak-L is probably not compatible with GPL code, while BSD or the apache licenses certainly are. IMHO, until the community settles around a Smalltalk- (image-) based license (essentially GPL, solving the linking issues), BSD is probably more than adequate. A fair number of us (myself included) affirmatively like the requirement to publish source to modifications, although RMS considers such a limitation to make software "unfree." Other licenses: LGPL may well be adequate for these purposes. I will need to look at it with this question in mind, but it may well serve the purpose. I'll take another look and see if anything there is hurtful to a Smalltalk "in-image" product. Disclaimer: Please note that I offer only general information and am not in a position, one way or the other, to offer legal advises for your particular purposes. Sound legal advice requires a detailed understanding of the particular facts of your project, and the application of the law to those facts, which I have not undertaken here. If you require legal advice, you should rely on the advice of counsel you have retained for that purpose.

1 0

AIDA/Web license (was RE: Zope)
by Janko Mivsek 29 Feb '00

29 Feb '00

Hello Squeakers, As an author of AIDA/Web (http://www.eranova.si/aida/) I like to clarify a bit the license issues mentioned in this thread. To be honest, I just didn't care about those license problems too much and AIDA was released under for me in that moment most appropriate license. Also, here we don't complicate so much about licensing issues so I really need help from you o find a more appropriate license. My intention with AIDA is to help spread Smalltalk in web applications and on the other side that community help me developing AIDA further. We all can then make money providing services and custom solutions based on AIDA. Therefore, I'm willing to change license to be more lawyer proof :)) Please help me find such a license. Can we agree for a common license for all Smalltalk OS projects? Should I start with LGPL? Best regards Janko -- Janko Mivsek EraNova d.o.o. Pod hribom 55 1000 Ljubljana Slovenia phone: +386 61 18 16 232 fax: +386 61 18 16 231 janko.mivsek(a)eranova.si www.eranova.si

1 0

Re: [CCC] False class comment
by Lex Spoon 29 Feb '00

29 Feb '00

Dwight Hughes <dwighth(a)ipa.net> wrote: > ----------------------------------------------------------------- > False defines the behavior of its single instance, false -- logical > negation. Notice how the truth-value checks become direct message sends, > without the need for explicit testing. > > Be aware however that most of these methods are not sent as real > messages in normal use. Most are inline coded by the compiler as test > and jump bytecodes - avoiding the overhead of the full message sends. So > simply redefining these methods here will have no effect. > > If you want to actually send them as messages, you will need to send > them using #perform:with: and #perform:with:with:. > > ----------------------------------------------------------------- > Looks cool. Arguably, though, a better way to call the relevant methods for real is to tell the compiler to stop inlining them, and then to recompile everything. There's no point in calling them with #perform: etc unless you have actually changed them, and if you have actually changed them, then you probably want the whole system to enjoy the modifications. So maybe we should just leave in the warning and leave out the circumvention. Oh yeah, and this is true for both True and False. So maybe we should just put this stuff in class Boolean, instead of duplicating it in two places. Lex

2 1

Re: [UPDATES] 26 more for 2.8a
by Bruce ONeel 29 Feb '00

29 Feb '00

Hi, Now ftp://st.cs.uiuc.edu//Smalltalk/Squeak/2.7/files/Squeak2.8a-beo.zip is updated with these changes. cheers bruce Dan Ingalls <Dan.Ingalls(a)disney.com> wrote: > Folks - > > These incorporate many of the submissions from SqueakVille in the last month. Once again, much credit goes to Stefan Matthias Aust for taking the trouble to cull through them and check for various conflicts. > > Enjoy > - Dan > --------------------- > 1874cleanup-tkNE -- Ted Kaehler -- 23 February 2000 > Defines (SmartRefStream cleanUpCategories) to remove obsolete conversion methods. > Includes 29 removals or currently obsolete conversion methods." > > 1875projectEnv-tkNJ > > 1876Explain-th -- Torge Husfeldt -- 2 February 2000 > Explain in code pain was broken because doitFlag in Parser was sometimes not initialized correctly. I think I found the right place to fix it." > > 1877FontDisplayEnh-acg -- Andrew C. Greenberg -- 2 February 2000 > A small enhancement to display all characters of a strikefont." > > 1878SysWindowReframe-dns -- David N Smith -- 2 February 2000 > A small update to SystemWindow which provides an option to suppress reframe handles of the window. This can be desirable when an application wants control over its own layout and wants no one (like an 8-year old) to resize the window. > The default allows reframing; the option can be set and reset during execution however one wishes. > Please note the postscript." > > 1879FillInTheBlankFix-sma -- Stefan Matthias Aust -- 10 February 2000 > Stewart MacLean noticed a bug in the MVC version FillInTheBlankController." > > 1880GraphicalMenuBug-dns -- David N Smith -- 4 February 2000 > Small bug in the comment of GraphicalMenu>>initializeFor:... No big deal." > > 1881PrefFix-ssa -- Sam S. Adams -- 9 February 2000 > Spelling error in preferences fixed." > > 1882LastUpdString-sma -- Stefan Matthias Aust -- 6 February 2000 > I tried a ''Smalltalk majorShrink; abandonSources; lastRemoval'' on a 2.8a/1789 image and noticed that here's an error in SystemDictionary>>lastUpdateString which is called from condenseChanges during abandonSources. When this method tries to determine the patch level by calling ''ChangeSorter highestNumberedChangeSet'' the ChangeSorter class has already been stripped. > This changeset works around this problem." > > 1883ColorPrintOn-dns -- David N Smith -- 6 February 2000 > Print named colors with names instead of (Color r: 0.833 g: 0.588 b: 0.524)" > > 1884BlobEnh-dns -- David N Smith -- 9 February 2000 > Blobs stay upfront now and you can click through them (use Cmd-click to get them). They also don't show handles when opened from the ''new morph'' menu (sma)." > > 1885DecompileUnknown-th -- Torge Husfeldt & SMA -- 7 February 2000 > A smaller (and enhanced) version of Torge's submission to decompile unbound methods. Torge wrote: 'Here's a little snippet I wrote to decompile CMs that are no longer in any method dictionary (such as DoIts that are about to be deleted, but there are other examples). Why shouldn't we want do do this sometimes.' " > > 1886ProtoObjComment-di -- Dan Ingalls -- 8 February 2000 > Please note the postscript! > Dan's comment template the ProtoObject's comment which started up a long discussion on the mailing list. There were better solutions provided by Dan and Vassili Bykov but this is a first step --sma." > > 1887UncatchableHalt-sma -- Stefan Matthias Aust -- 12 February 2000 > This is my proposal for an uncatchable halt, that is > [self halt] on: Error do: [] > will not silently do nothing but still open a debug notifier. > On the other hand, > [self halt] on: Halt do: [] > will catch the halt." > > 1888BrowseIt-slr -- Steve Rees -- 11 February 2000 > Minor fix to browseIt on ParagraphEditor. > Previously ParagraphEditor assumed that a selection beginning with an uppercase letter represented a Class name and that a selection beginning with a lowercase letter was a method name. > Whilst the second assumption is valid, the first is not, since although conventionally method names begin with a lowercase letter they can also begin with an uppercase letter. > This changeset modifies the assumptions in ParagraphEditor>>browseIt so that if the selected name begins with an uppercase letter, but no entry is found in Smalltalk a search is done for implementing methods with that name." > > 1889HtmlFmtFix-jcc -- John Clonts -- 11 February 2000 > A small correction in the example comment. But while I was testing the code, I noticed a problem in DisplayScanner using MVC and fixed it --sma" > > 1890CmdKeyDesc-sma -- Stefan Matthias Aust -- 12 February 2000 > Updated command key description as well as the disablement of the enablement of Ctrl+( enclosement keys for Windows because this disabled this function on Macs. Okay? > Unfortunately, the different platforms are quite different in what key codes they generate. I tried to work around this problem but failed. I'd suggest to double check the VM code for all platforms and make sure that they generate identical codes. This simplify the paragraph editor code and also enables a few more keyboard shortcuts for Non-Macs. Frankly, I'm a bit annoyed that often platform-independence for Squeak actually means Mac-dependence. > Added back Cmd-Shift+I for exploreIt." > > 1891ClassComments-dwh -- Dwight Hughs and Göran Hultgreen -- 26 February 2000 > Added method categories to the PenRecorder --sma" > > 1892TTStringMorph-sma -- Stefan Matthias Aust -- 26 February 2000 > Brian Mc Enery noticed: If multiple TTSampleString's are being copied (where each instance uses the same font) then each copy will automatically have a new version of the TTFontDescription including the glyphs and the mapping table. This occurs if a TTSampleString is created out of a parts bin, or if it is duplicated using the halo menu." > > 1893EnhTranscript-sma -- Stefan Matthias Aust -- 26 February 2000 > Now, show: can show any object, not just Strings. > Also added method categories to the Transcripter class." > > 1894FillInFixes-ssa -- Sam S. Adams -- 14 February 2000 > Here's a fix for FillInTheBlank so it complies with Preference settings for > scroll bars and rounded corners." > > 1895EmptySymbol-sma -- Stefan Matthias Aust -- 26 February 2000 > Richard A. Harmon noticed that the empty symbol ('' asSymbol) didn't work." > > 1896RmEmtyCatFix-sma -- Stefan Matthias Aust -- 27 February 2000 > The browser was missing the function #removeEmptyCategories. Added." > > 1897NameOfClass-sma -- Stefan Matthias Aust -- 27 February 2000 > It seems that somebody accidentally created a 'NameOfClass' class. Removed." > > 1898PPBlockTemps-sma -- Stefan Matthias Aust -- 27 February 2000 > Pretty printing of block temporaries didn't work. Fixed." > > 1899segBug-tkNL -- Ted Kaehler -- 28 February 2000 > Update 1860 added debugging class variables to ImageSegmentRootStub. They allow the user to see the stack of what caused a segment fault. Very useful. However, the new class variables were never defined in the fileOut. Nor were they initialized. Any fault caused an error. We fix it here."

2 1

Jump to page: