Squeak-dev March 2002

squeak-dev@lists.squeakfoundation.org

169 participants
610 discussions

Re: Behaviors vs Modules
by Anthony Hannan 24 May '02

24 May '02

Jecel Assumpcao Jr <jecel(a)merlintec.com> wrote: > On Friday 22 February 2002 23:38, Anthony Hannan wrote: > > - A Smalltalk expression now knows, via the selectors it calls, which > > protocols and corresponding default behaviors it depends on. So > > starting from a Smalltalk expression we can trace selectors and > > methods recursively and load all default behaviors that will be > > needed. > > Two slight problems: > > 1) #perform: perform: will have to take a selector object that already knows its protocol instead of just a symbol. The selector's protocol will be loaded at runtime when the perform is triggered. > 2) if #foo is used in only one protocol in your system, you don't > have to say which one that is, right? Now imagine that you send your > module to me and I load it into my system which also has a #foo in only > one protocol, but it happens to be a different protocol from yours! The compiled method will have the selector object "foo" in its literals which knows its protocol, not the symbol #foo. So if you send the compiled method to another image (via image segments) then the protocol will go with it. If you just send the source code, your right, "foo" will be looked up in the new environment. But this is the same problem we have with globals now. How do I know "String" in my image means the same as "String" in your image? Cheers, Anthony

3 2

Internet Time...
by Mayuresh Kathe 01 May '02

01 May '02

Hi, I just realised there is nothing to get 'Internet Time' in Squeak. If I write the code, what will I have to do to submit it for updation in the final image? ~Mayuresh

7 13

[ENH] 3.2 NuBlu events - already in 3.3a update stream (was RE: [BUG]Morph>>updateableActionMap)
by rwithers12＠attbi.com 24 Apr '02

24 Apr '02

here 'tis. Rob "Withers, Robert" <rwithers(a)quallaby.com> wrote: > I can post those this evening,

2 1

CPU-Usage again
by Markus Fritsche 18 Apr '02

18 Apr '02

Hi, it's me again... :-) So, I've made the following findings: starting squeak with the ComSwiki-Image on a K6-2/400/linux in Windowing mode: speed not killing, but cpu-usage okay. PIII/900/W2k: cpu-usage not noticeable (when running in background) headless on a k6-2/233/linux: squeak takes every little timeslice it could get? Is this really the effect of "gettimeofday()" as I thougt before? The Linuxsystem (debian-testing) is the same on the k6-2/233 and k6-2/400. Any hints what I can do? I've already tried shrinking, but it's hard to "majorShrink" the Image to somewhat useful for a newbie :-)) Regards, Markus -- SIGSTOP GMX - Die Kommunikationsplattform im Internet. http://www.gmx.net

6 6

Squeakers in Seattle?
by Martin McClure 16 Apr '02

16 Apr '02

Hi all, I'm looking for a 'show of hands' of Squeakers based in the immediate area of Seattle. I'm working on a Seattle-based project that involves Squeak, and we'd like to know what kind of pool of local Squeakers there is to draw on in the unlikely event that I become unavailable to work on the Squeak code. If you are in the Seattle area and would conceivably be interested in such a thing, please mail me off-list. Thanks, -Martin

2 1

[updates] 14 for Squeak 3.3alpha
by Dan Ingalls 10 Apr '02

10 Apr '02

Here are the summaries for the 14. Methinks it may be time for some harvesting fairly soon. - Dan ---------------------- 4800gZipContents-sw -- Scott Wallace -- 12 March 2002 * Makes the window into which a gZipped file's decompressed contents are projected be a Workspace rather than a plain StringHolder, so that its window-menu will offer 'save contents to a file...', which is frequently useful here. * Do not automatically append '.text' to the specified filename when a workspace's contents are saved to a file -- instead, *suggest* a .text suffix (and not even that, in the case of a gZipped file being viewed decompressed,) but don't insist on it." 4801jumpHierarch-jla -- Jerry L. Archibald -- May 28, 2001 Modifies the 'jump to project' menu such that the primary project-list is ordered as a tree that reflects the hierarchical structure of the set of projects. Jerry Archibald (1940-2001) died in June 2001, and this code was his final contribution to Squeak. Published, with minimal modification, by sw, 3/12/2002" 4802Ev-Essay-fix 4803MakeActorFix-ar -- Andreas Raab -- 14 March 2002 Fixes a problem reading MDL files." 4804classRefs-sw -- Scott Wallace -- 17 March 2002 Cleans up the problem that references-to-symbol queries were putting up an inappropriate informer on any symbol that was both a submodule-name and a class name -- e.g. Browser, Debugger, FileList" 4805alignSelection-sw -- Scott Wallace -- 20 March 2002 Adds six alignment commands to SelectionMorph's halo manu, allowing groups of morphs to be statically aligned (relative to each other) in various ways." 4806fastFrame-bmk -- Brian Keefer -- 22 March 2002 When fastDragWindowForMorphic is in effect, makes window dragging start relative to where you first clicked, not wherever the mouse jerked to while the SystemWindow was waiting to see if this drag was merely a sloppy click." 4807HTMLTable-tk -- Ted Kaehler -- 22 March 2002 After you parse an incoming web pabe, class HTMLTable is able to deliver its data to you in an Array2D. Very handy. When grubbing around in web pages, the match: method is very handy. (strm match: 'foo'), advances the stream to just after then next 'foo', and returns true if is found one. Now added backUpTo:, for going backwards." 4808colorScrollBars-dew -- Doug Way -- 23 March 2002 - Changes the alternativeScrollbarLook so that the inboard scrollbar has the flat, non-rounded look of the original scrollbar, but it matches the window color. (The flop-out scrollbar is still rounded, since the rounded look looks better with the rounded arrows, and the sharper highlighting makes more sense when only one scrollbar appears on the screen at one time.) - Removes border around this new inboard scrollbar, which looks cleaner, and also saves space. - Improves the Color>>lighter/darker family to work consistently for lighter and darker shades of color. Basically, #lighter and #darker will now adjust by the same amount, no matter what shade you start from (similar to what #dansDarker attempted). This makes the #raised/inset bevelling of buttons/etc look a lot better. Also tweaks a few users of these methods in the system to adjust for the change. Mostly everything looks fine as is. (The new methods are actually about 10% slower than the old ones on micro benchmarks, but this doesn't seem to be enough to be noticeable in macro benchmarks or in normal use.) - Fixes keyboard focus highlighting so that the rectangle isn't hidden behind inboard scrollbars. (cleaned up ScrollPane>>innerBounds) - Permanently hides scrollbars in a few short panes, since they're not needed, and look ugly when inboard. These panes include the top pane of a hierarchy/filecontents browser, the pattern pane of the filelist, and the top pane of the methodfinder. - Makes the scrollbars fit correctly within the panes, without overlapping borders (when inboard). - Increases lightening and darkening of window titlebars when active/inactive, so the difference is more noticeable. - Moves alternativeWindowLook preference to windows category (instead of browsing), since it is window-related. - Fixes alternativeScrollbarLook scrollbar so that the thumb doesn't darken after you drag it the first time. - Includes a small fix since the March 14 version, so that translucent/transparent colors are handled correctly. 4809recentChangesColor-dew -- Doug Way -- 24 March 2002 A tweak to correct the multi-selection color in the Recent Changes window, which was trying to darker a transparent color, and was affected by the Color changes." 4810slideToTrash-sw -- Scott Wallace -- 25 March 2002 Applies only when the #preserveTrash preference is true: When dismissing by halo, slide the object being dismissed to the trash-can. If no trash can is visible on the screen, put one there first. This update also fixes the problem that when haloTransitions and preserveTrash were both true and you dismissed a morph via halo, a spurious momentary fading halo would appear *in the wrong place* at the topleft of the screen." 4811classListScroll-sw -- Scott Wallace -- 25 March 2002 Addresses two long-standing bugs, both relating to whether the selected class is properly scrolled into view in a browser or Hierarchy Browser freshly launched by some browser-requesting or hierarchy-requesting command. * One was an off-by-one bug affecting the browse-it and browse-full commands, when the first class just *below* the last one visible in the unscrolled class-list-pane would not properly get scrolled into view. * The other is worked around rather than fixed; it relates to hierarchy-browsers launched by browse-hierarchy or by browse-it when #alternativeBrowseIt preference is true. The workaround is to force the list selections to be made a second time, after launch; this takes extra time but results in selections being visible. * Also (unrelated) fixed here is a bug wherein the hitting of the Enter key into a morphic PluggableTextMorph that was visible in an mvc project could generate an error." 4812projSemiFix-hg -- Henrik Gedenryd -- 24 March 2002 This makes Project windows closable but simply comments out a chunk of Environments-related code that I translated into something equivalent for modules, but really couldn't make sense of." 4813ZipCheckNot-di -- Dan Ingalls -- 27 March 2002 Defeated the content-based Zip Archive test in ArchiveViewer because it caused an error when using the FileList to view a remote server. The problem was because of a bug that can probably be repaired (looking for the remote fileName on the local disk), but I couldn't take the time to fix it so I simply disabled it for now."

2 1

[SqueakEnd] and modules
by Stephen Pair 09 Apr '02

09 Apr '02

Since Dan is tentatively scheduled to give a presentation on modules at the SqueakEnd/MMWorkshop, I thought it would be a good idea to have a project oriented around that. I've only been following modules off and on for the last 5 or 6 months and don't really know what issues are critical at the moment. Can anyone comment on the current state of modularity and what they think might make for a good module oriented project at SqueakEnd? - Stephen

4 5

RE: Comments welcome: designer look for squeak
by Norton, Chris 08 Apr '02

08 Apr '02

Hi Martin. Having been on this list for several years now, it seems to me that a newish person proposes a "new look" for Squeak every six months or so. That's not a bad thing, mind you, it's just a fact that change happens very slowly in this area. If you look at the Swiki, you'll see lots of cool screenshots of Squeak-tweaks that people have done over the years. Unfortunately, most of these tweaks never got past the "ain't it cool" stage. The really hard part of any project is to find someone who will stay the course and finish it. If you are interested in changing Squeak's look, I encourage you! More power to you. But you may want to see why those other efforts failed before you spend a lot of time on this. Cheers, ---==> Chris PS> Many of the people who were interested in changing Squeak's "look" are still on the list. Perhaps you can round up a few of them and really turn some heads.

11 12

[BUG] Hugely inefficient FTP code: coding opportunity
by Henrik Gedenryd 08 Apr '02

08 Apr '02

I've now made remote up-and downloading of module work properly. However, Squeak's underlying FTP code is so painfully inefficient that I recommend that you avoid using it, and instead keep the accessRemoteRepostories preference off, and upload from the local repository cache using some other ftp program. An upload that takes say 5-15 seconds with another FTP program takes some *5 minutes* with Squeak according to the file time stamps. I've made downloads avoid this by using HTTP access. The problem is that Squeak's ftp code logs in and logs out between *every* CWD, load, store, etc. -- every single server action. It is this same deficiency that makes project loading so slow: 0.25 seconds to load an image segment from disk *really quickly*, but eons spent on redundantly logging in and out of the ftp server. It wouldn't be too hard to change the code to allow multiple commands within a single session: Put a socket instvar in ServerDirectory, and make the code either login and logout if that socket instvar is nil, or if not, just use it to send the command without logging in and out. Then wrap sessions in a login and ensured logout. The other option is supporting HTTP PUT, Cees would be willing to make the server support this, but this may not be trivial. I am not planning to spend my time on this. Henrik

3 6

[SqF]Report of VM Construction Project for Feb '02
by Tim Rowledge 06 Apr '02

06 Apr '02

This is the initial report for the VM Construction Project group. Since this is the first such report, I'll start with my view of the goals of the project. Goal: The purpose of this project is to create a vm building process and associated code that makes vm building as simple, reliable and maintainable as we can manage. Participants: tim, John McIntosh, Rob Withers, David Lewis, and of course Ian Piuumarta and Andreas Raab. Other participants to represent the other vm ports are sought. Stefan Rudlof, Goran Hultgren and others have provided advise and help. Background: From the initial release of Squeak, the image has included all the code needed to create a vm for a Mac. For other platforms there has been a variety of places and approaches used over time, mostly the provision of an archive of some sort with a file tree suited to the platform. Sometimes this has included 'common' files (files such as sq.h etc) that are derived from but not identical to those in the release image. This situation could continue into the future without being a major embarassment to the Squeak community, but some time ago I felt that it could be improved with some fairly simple changes. One of the original driving forces for the work that has been released as the VMMaker and VMMakerTool was the need at exobox to be able to produce frequent vm builds automatically without extensive user intervention. Another original aim was to find a way to produce a file tree that reasonably suited all known platforms and that was compatible with automated vm production. The time-honoured vm production routine was not very suited to thise aims. For example, generating the core of the vm required two unrelated incantations and produced a dump of files in the current wrking directory; these files needed moving, sometimes editing, before being compiled. Producing a vm for a Mac required a different incantation and produced a different group of files. Changing which set of plugins were produced and whether they were internal or external, required editing a method. None of this prevented a rudimentary script from being used succesfully, but it apearred untidy and suboptimal to some users. A secondary concern was the scattered provenance of the code for non-Mac machines and the varied update levels of that code. Using some common repository was suggested and seemed a sensible experiment. The initial attempts to use SourceForge mainly served to demonstrate that a better file organisation would be very helpful. Current status: The VMMaker has been the major focus of the work for this project thus far. It is currently able to produce vms for any of the four most active platform from a codebase stored on SourceForge. A UI exists that makes it easy to configure the plugins and to save and use said configurations. The basic tool is scriptable and has been documented on the swiki and in SqueakNews, as well as having been demonstrated at OOPSLA. Work is still in progress to bring all platforms to the same level of compliance (for want of a better word). Future Plans: The VMMaker codebase is reasonably complete but there are enhancements that might be worth pursuing - makefile (or at least makefile fragment) production, integration with sqCVS, vm internal configuration (cache size etc) amongst them. The codebase currently on SF is in good order, barring the mostly makefile related updating needed by non-unix platforms, but tools to make it easier to fetch the code in release-matched sets, to tag release sets, deal with branches and bug fixes are all needed. The most important goal is to reach agreement on adopting VMMaker or on an acceptable alternative. No other interesting alternative is currently proposed. Once a reasonable level of agreement is reached the artifact can be integrated into the main release, presumably by creating a module for it. tim. chair, VM construction project. "We live for the VM, we debug for the VM. But we don't debug stupidly"

10 33

Jump to page: