Fwd: [Openmcl-devel] openmcl 1.0 vs. GUI
gb at clozure.com
Mon Aug 29 00:54:54 UTC 2005
On Sun, 28 Aug 2005, alex crain wrote:
> On Aug 28, 2005, at 1:32 PM, Hamilton Link wrote:
>> To that end, Alex, do you have a version of the GUI in which the editor,
>> backtrace, listener, and inspector all approximately load and can be
>> invoked? Or is the version that comes in the distro preferred?
> Without putting too much of a point on it, my #1 problem with the GUI project
> is the level feedback that I've received.
> None. Nada. Nothing. Zip.
One change that I'd made to the bleeding-edge version around the time
that you announced your version is to try to defer redisplay until
after a simple editing command completes. (As the code is written,
redisplay is attempted after every buffer modification; that's clearly
wrong, and the change I introduced at least muffles that so that you
don't see a lot of hyterical redisplay when an editing command (like
indentation) makes a lot of individual buffer changes.)
That (fairly simple) change does seem to improve usability quite a
bit. To some degree, it's a bandaid on a deeper problem, and I have
some concerns about whether it'd be simpler and better to run
editor commands in the event thread. We've discussed this a little,
and should probably discuss it more.
> For anyone how is interested - http://ren.widgetworks.com/~alex/ccl-
> This is a disk image that contains a relatively recent version of the
> bleeding edge CCL code and a launchable Hemlock application.
> It is a bit buggy and there have been a few reported problems with the
> launcher but I think that the only catch is that you can't have
> another version of CCL in your load path or any of the CCL* environment
> variables set.
> I've tested in on Tiger and 10.3.9 and it boots fine in both cases as long as
> there isn't any other ccl distributions around.
I -think- that the problem that I've seen may have to do with
LaunchServices' caching (or something very similar). I've tried a few
times to debug it, but haven't had any luck in simulating the effect
of a launch of the bundle under GDB.
My best guess is that whatever the problem is, it's pretty mundane: we
may have to worry about whether bundle identifiers in the Info.plist file
are unique, or something like that.
> This version has a editor, listener, backtrace, inspector, trace monitor and
> a version of the ANSI standard documentation that appears to be copyright
> (Unlike the HyperSpec). It supports syntax coloring, drag and drop and sexpr
> friendly selections.
I'm sure that it's possible to distribute the spec (without proprietary
HyperSpec formatting); I'm pretty sure that GCL's "gcl.info" file is
basically the ANSI CL spec in GNU Info format. (I don't know if they
had to jump through any legal hoops to do that or not; that'd be worth
You seem to have translated the ANSI spec into a lisp markup language,
and I assume that the IDE's documentation browser contains an
interpreter for that language. NSTextViews contain support for RTF
and HTML (and WebKit contains richer support for HTML); it might be
worth leveraging that (I'm sure that there are tradeoffs involved, and
that there must have been some advantages to the lisp markup language;
there are obviously advantages to dealing with RTF and HTML, as well.)
> There is precious little actual documentation - I'm working on extracting
> something useful from the scribe sources for the hemlock docs.
I'd find a few paragraphs that describe how things are organized very
helpful. I haven't poked around in this code as much as I should have,
and part of the reason is that poking's slow going (I don't have a big-
picture overview that'd tell me where to look.)
I think that Gilbert Baumann (the person who originally made Portable
Hemlock more portable) had a tool that converted from SCRIBE to PostScript;
I have PDF versions of the Users Guide and Command Implementor's Guide
that were derived via ps2pdf; I'll put them on the FTP server in a minute.
[They're in <ftp://clozure.com/pub/hemlock*.pdf>]
I don't know offhand of tools that'd translate from SCRIBE to something
more maintainable (texinfo, XML, ...), though it seems reasonable to
assume that such tools exist.
> Please, feel free to download it, play with it and even comment on it.
> Openmcl-devel mailing list
> Openmcl-devel at clozure.com
More information about the Openmcl-devel