Bug report

[henry@rxf.ibp.fr ( Henry Lieberman )]

Bugs in Coral 2.0 a1 patch p1:


The manual says it is compatible with 6.0.3 and later.  The
application failed to launch on a Mac 2 cx with 6.0.3, but succeeded
when the system was upgrated to 6.0.4.  Incidentally, while trying to
install I got a cryptic "Installation failed -- error #54" message
from the Installer, and the problem turned out to be a locked System
file, which Installer should detect.

I'm really sad to see that even the limited stepper that came with
earlier versions is now gone.  Hope it comes back and works with CLOS
methods. Improvement in debugging tools is the most pressing need for MACL.

If you resize the Apropos window, the first time it displays it does
not show the scroll bar properly.  After you try to use the invisible
scroll bar, it redisplays.

Meta-Close-Paren is still not fixed, a bug I reported a long time ago.
It should delete any whitespace before the cursor before performing
the rest of its job.  This is necessary for several successive
Meta-Close-Parens to work properly.  The bug in Meta-U is also still
present.

Control-Meta-Rubout in Lispm and Unix Emacses is delete s-expression
backward.

The Quickdraw file requires RECORDS, but there is no records.lisp file
in the distribution that I got.  Seems to work despite absence of
RECORDS [is this now built in?]. The proceed options for REQUIRE
doesn't let you continue ignoring the require.


If Init.Lisp exists and Init.fasl doesn't (load "init") will fail to
find the .Lisp file.  LOAD is supposed to load either the fasl file or
the source, whichever is newer.

Fred stuff:

BUFFER-MARKP doesn't exist.  Marks are of type BUFFER-MARK, not MARK [p.318]
BUFFER-INSERT-STRING is in CCL::.

Does (FRED-BUFFER <a Fred Window>) officially replace
(WINDOW-CURSOR-MARK)?  Manual is unclear on this point, and on
terminology "A buffer is not a data type but a vector.  There are no
buffer and mark data types, only BUFFER-MARK, consisting of the mark
and the vector" Huh?

Naming of functions seems to be inconsistent.  Why are some functions
prefixed FRED- others ED-, others WINDOW- [when they act on Fred
windows], mark functions MOVE-MARK, MARK-BACKWARD-P, others not
prefixed, etc.

The file in Examples;font-menus.lisp doesn't seem to work reliably.
Example: Select All in the Listener, and set to Courier 12.  Sometimes
when the font does change, the spacing gets messed up.

On a bright note, CLOS seems to work well.  I tried a substantial
program that I wrote using PCL and after removing PCL-specifics, it
worked straight away, and of course much faster than PCL.  My informal
timing puts the raw speed of creating objects at 4x slower than Object
Lisp [it would be nice if this were improved], and accessing slots at
50% faster than Object Lisp.