mit-scheme.git
16 years agoHandle NaN objects specially when comparing them against rational
Chris Hanson [Fri, 25 Apr 2008 01:20:24 +0000 (01:20 +0000)]
Handle NaN objects specially when comparing them against rational
numbers.  We should probably do something similar for infinities.

16 years agoEliminate special cases for flonums in generic predicates. These were
Chris Hanson [Fri, 25 Apr 2008 01:19:04 +0000 (01:19 +0000)]
Eliminate special cases for flonums in generic predicates.  These were
doing the wrong thing for NaNs.

16 years agoReword missing-support error message to clarify its meaning.
Chris Hanson [Tue, 11 Mar 2008 04:00:52 +0000 (04:00 +0000)]
Reword missing-support error message to clarify its meaning.

16 years agoEliminate type-size errors and warnings; some NO_CHANNEL and
Chris Hanson [Sun, 9 Mar 2008 20:24:33 +0000 (20:24 +0000)]
Eliminate type-size errors and warnings; some NO_CHANNEL and
NO_PROCESS were long types, while the corresponding Tchannel and
Tprocess were int types.  Now all are int types.

16 years agoCan't use UCODE-PRIMITIVE in this file; it breaks the initial build.
Chris Hanson [Sat, 8 Mar 2008 18:46:41 +0000 (18:46 +0000)]
Can't use UCODE-PRIMITIVE in this file; it breaks the initial build.

16 years agoPrioritize .so files over .bin files -- we were loading .bin files
Chris Hanson [Sat, 8 Mar 2008 17:43:12 +0000 (17:43 +0000)]
Prioritize .so files over .bin files -- we were loading .bin files
when .so files were available.  (Thanks to Don Allen for catching
this.)

16 years agoUnsyntax conditionals with the primitive NOT procedure, not the global
Taylor R. Campbell [Fri, 29 Feb 2008 16:06:34 +0000 (16:06 +0000)]
Unsyntax conditionals with the primitive NOT procedure, not the global
binding of NOT.

16 years agoRegenerate tables with proper 0x prefix.
Chris Hanson [Thu, 21 Feb 2008 03:54:11 +0000 (03:54 +0000)]
Regenerate tables with proper 0x prefix.

16 years agoDoh... Forgot to add missing 0x prefix to hex constants.
Chris Hanson [Thu, 21 Feb 2008 03:52:10 +0000 (03:52 +0000)]
Doh... Forgot to add missing 0x prefix to hex constants.

16 years agoEliminate now-unused exports.
Chris Hanson [Fri, 15 Feb 2008 05:06:01 +0000 (05:06 +0000)]
Eliminate now-unused exports.

16 years agoWhen calling %MAYBE-TOGGLE-THREAD-TIMER from WAIT-FOR-IO, disregard
Taylor R. Campbell [Fri, 15 Feb 2008 04:19:00 +0000 (04:19 +0000)]
When calling %MAYBE-TOGGLE-THREAD-TIMER from WAIT-FOR-IO, disregard
the value of TIMER-INTERVAL, since we are waiting only until I/O or
the next timer event.

16 years agoExport MHASH-HMAC-UPDATE from (RUNTIME CRYPTO).
Taylor R. Campbell [Thu, 14 Feb 2008 09:51:36 +0000 (09:51 +0000)]
Export MHASH-HMAC-UPDATE from (RUNTIME CRYPTO).

16 years agoIron out two remaining little bugs in the recent flurry of changes
Taylor R. Campbell [Thu, 14 Feb 2008 08:20:22 +0000 (08:20 +0000)]
Iron out two remaining little bugs in the recent flurry of changes
surrounding errors during compiled invocation setup from primitives.

16 years agoFix errors in output for rewriting example.
Chris Hanson [Thu, 14 Feb 2008 06:48:45 +0000 (06:48 +0000)]
Fix errors in output for rewriting example.

16 years agoUndo previous change that removed primitive apply optimizations. (It
Chris Hanson [Thu, 14 Feb 2008 06:47:37 +0000 (06:47 +0000)]
Undo previous change that removed primitive apply optimizations.  (It
turns out that my analysis was incomplete and further changes would
have been required.)  Fix the problem with the apply hacks by
introducing a new exception code PRIM_APPLY_ERROR, which provides an
alternate path for signaling an error to the interpreter, _without_
backing out of the primitive (since we've already made all the
necessary changes to the stack).

16 years agoFix various places where variables defined with DEFINE-PRIMITIVES were
Chris Hanson [Thu, 14 Feb 2008 03:34:11 +0000 (03:34 +0000)]
Fix various places where variables defined with DEFINE-PRIMITIVES were
assumed have primitive-procedure values.

16 years agoOn Taylor's suggestion, change DEFINE-PRIMITIVES so that it generates
Chris Hanson [Thu, 14 Feb 2008 02:35:05 +0000 (02:35 +0000)]
On Taylor's suggestion, change DEFINE-PRIMITIVES so that it generates
compound procedures rather than just binding the variables to the
primitive-procedure objects.  This allows us to drop the new macros
needed by fixart, eliminating the need to patch older compilers to
compile this code.

16 years agoDon't insert top-level bindings of procedures into code, assuming that
Chris Hanson [Thu, 14 Feb 2008 02:12:52 +0000 (02:12 +0000)]
Don't insert top-level bindings of procedures into code, assuming that
they are primitives.  Use explicit references to primitives.

16 years agoFix a few bad instances of (ACCESS ...).
Chris Hanson [Thu, 14 Feb 2008 00:45:17 +0000 (00:45 +0000)]
Fix a few bad instances of (ACCESS ...).

16 years agoDon't write (ACCESS <foo> ()).
Chris Hanson [Wed, 13 Feb 2008 23:36:15 +0000 (23:36 +0000)]
Don't write (ACCESS <foo> ()).

16 years agoAdd SYSTEM-GLOBAL-ENVIRONMENT to global constants; this will suppress
Chris Hanson [Wed, 13 Feb 2008 23:30:18 +0000 (23:30 +0000)]
Add SYSTEM-GLOBAL-ENVIRONMENT to global constants; this will suppress
warnings about (ACCESS <foo> SYSTEM-GLOBAL-ENVIRONMENT).

16 years agoFix fatal error: FASDUMP wasn't cleaning up correctly when signalling
Chris Hanson [Wed, 13 Feb 2008 23:26:23 +0000 (23:26 +0000)]
Fix fatal error: FASDUMP wasn't cleaning up correctly when signalling
an error.

16 years agoAdd an error check to MAKE-COMBINATION for an operator that's a
Chris Hanson [Wed, 13 Feb 2008 14:26:47 +0000 (14:26 +0000)]
Add an error check to MAKE-COMBINATION for an operator that's a
non-primitive procedure.  This is the most likely screw case from the
recent change in the handling of primitive integrations.

16 years agoRewrite some more variable references as explicit primitive constants.
Chris Hanson [Wed, 13 Feb 2008 14:25:33 +0000 (14:25 +0000)]
Rewrite some more variable references as explicit primitive constants.
This includes one instance, in MAKE-MACRO-REFERENCE-TRAP-EXPRESSION,
that was causing a fatal error.

16 years agoRedo the previous change: it was causing VERY BAD things to happen.
Chris Hanson [Wed, 13 Feb 2008 06:21:07 +0000 (06:21 +0000)]
Redo the previous change: it was causing VERY BAD things to happen.
This one should be considerably safer.

16 years agoEliminate three special cases (compiled_with_interrupt_mask,
Chris Hanson [Wed, 13 Feb 2008 04:28:27 +0000 (04:28 +0000)]
Eliminate three special cases (compiled_with_interrupt_mask,
compiled_with_stack_marker, and apply_compiled_from_primitive) that
were adding hairy edge cases to gain a small amount of efficiency.

16 years agoFix bug: setup_compiled_invocation_from_primitive wasn't calling
Chris Hanson [Tue, 12 Feb 2008 19:10:13 +0000 (19:10 +0000)]
Fix bug: setup_compiled_invocation_from_primitive wasn't calling
PRIMITIVE_ABORT when it should have been.

16 years agoFix bug: setup_compiled_invocation_from_primitive wasn't calling
Chris Hanson [Tue, 12 Feb 2008 19:09:44 +0000 (19:09 +0000)]
Fix bug: setup_compiled_invocation_from_primitive wasn't calling
PRIMITIVE_ABORT when it should have been.

16 years agoFix bug in last commit. %MESSAGE-ORDER-KEY yields a pair of the key
Taylor R. Campbell [Tue, 12 Feb 2008 00:37:54 +0000 (00:37 +0000)]
Fix bug in last commit.  %MESSAGE-ORDER-KEY yields a pair of the key
and index already, so MESSAGE-ORDER-KEY doesn't need to add the index.

16 years agoOne more stab at it: turns out that what I thought APPLY wanted to do
Taylor R. Campbell [Tue, 12 Feb 2008 00:16:57 +0000 (00:16 +0000)]
One more stab at it: turns out that what I thought APPLY wanted to do
is actually what WITH-INTERRUPT-MASK and WITH-STACK-MARKER also want
to do, instead of aborting the primitive.  So we can, after all, use
`setup_compiled_invocation_from_primitive' in
`apply_compiled_from_primitive'.

16 years agoFix think-o in previous commit -- wetware pattern matcher failed in
Taylor R. Campbell [Mon, 11 Feb 2008 23:59:24 +0000 (23:59 +0000)]
Fix think-o in previous commit -- wetware pattern matcher failed in
`apply_compiled_from_primitive', which obviously can't use the new
`setup_compiled_invocation_from_primitive' because the latter does a
primitive abort if an error happens, whereas APPLY just wants to
return.

16 years agoMeter the progress of summary buffer generation.
Taylor R. Campbell [Mon, 11 Feb 2008 22:49:10 +0000 (22:49 +0000)]
Meter the progress of summary buffer generation.

16 years agoCache the keys by which messages are ordered in a hash table, so that
Taylor R. Campbell [Mon, 11 Feb 2008 22:48:02 +0000 (22:48 +0000)]
Cache the keys by which messages are ordered in a hash table, so that
we need not repeatedly fetch and parse header fields and such for the
same messages.  (The cache may be optionally disabled by passing a
switch to MAKE-FOLDER-ORDER if it is concluded superfluous.)

16 years agoImplement incremental processing of IMAP responses, so that long lists
Taylor R. Campbell [Mon, 11 Feb 2008 22:45:43 +0000 (22:45 +0000)]
Implement incremental processing of IMAP responses, so that long lists
of large responses, such as all the header fields in a large folder,
will not be queued up to exhaust the heap space.  Exploit this when
preloading folder outlines, and don't fetch the header fields when
only the size is needed.  (Preloading folder outlines no longer even
fetches the size -- but to do this right requires a lot more work and
would probably make preloading the outlines slower anyway, and I can
live with slightly more network-intensive summary buffer generation
for folders never before summarized.)  Store messages' header fields
and envelopes only weakly in memory to conserve space.

16 years agoSimplify code to invoke compiled procedures from primitives. Fix bug
Taylor R. Campbell [Mon, 11 Feb 2008 21:07:21 +0000 (21:07 +0000)]
Simplify code to invoke compiled procedures from primitives.  Fix bug
in APPLY that would cause the stack to be munged when errors or
interrupts occur during the invocation setup.

16 years agoActually do avoid the race condition.
Taylor R. Campbell [Mon, 11 Feb 2008 07:52:06 +0000 (07:52 +0000)]
Actually do avoid the race condition.

16 years agoFix bug: it's necessary to close the I/O channels of a subprocess when
Chris Hanson [Mon, 11 Feb 2008 07:23:21 +0000 (07:23 +0000)]
Fix bug: it's necessary to close the I/O channels of a subprocess when
it has finished running.  Otherwise we can end up blocked on one of
those channels, with no way to unblock.

16 years agoAvoid a race condition with the GC daemon in LOOKUP-UNCOMPRESSED-FILE.
Taylor R. Campbell [Sun, 10 Feb 2008 21:35:46 +0000 (21:35 +0000)]
Avoid a race condition with the GC daemon in LOOKUP-UNCOMPRESSED-FILE.

16 years agoImplement SMTP AUTH LOGIN and AUTH PLAIN extensions for authentication
Taylor R. Campbell [Sun, 10 Feb 2008 10:44:13 +0000 (10:44 +0000)]
Implement SMTP AUTH LOGIN and AUTH PLAIN extensions for authentication
when sending mail by SMTP.  New Edwin variable MAIL-AUTHENTICATION
controls whether to authenticate; set it to #T to enable authentication
by default.  New Edwin variable SMTP-USER-NAME can be set to specify a
user name.

Still to do:  CRAM-MD5 (shouldn't be hard, but I don't need it now),
and automatic detection of necessity of authentication.  Is there an
SMTP reply code to tell us that we must authenticate?

16 years agoMove support for temporarily stored pass-phrases into Edwin from
Taylor R. Campbell [Sun, 10 Feb 2008 10:06:51 +0000 (10:06 +0000)]
Move support for temporarily stored pass-phrases into Edwin from
IMAIL's front end.

16 years agoChange lots of global primitives to be defined as compound procedures.
Chris Hanson [Sun, 10 Feb 2008 06:14:19 +0000 (06:14 +0000)]
Change lots of global primitives to be defined as compound procedures.
The compiler will still open-code them appropriately when the appear
in the operator position.  However, this change means that
non-operator references to these primitives avoid passing a primitive
object around, which improves performance when the object is
eventually called.

16 years agoAdd LAMBDA-TAG:UNNAMED, needed for cold load.
Chris Hanson [Sun, 10 Feb 2008 06:12:08 +0000 (06:12 +0000)]
Add LAMBDA-TAG:UNNAMED, needed for cold load.

16 years agoAdd FIXNUM? to list of primitives.
Chris Hanson [Sun, 10 Feb 2008 05:02:21 +0000 (05:02 +0000)]
Add FIXNUM? to list of primitives.

16 years agoChange handling of global primitives; they are now integrated only
Chris Hanson [Sun, 10 Feb 2008 04:42:43 +0000 (04:42 +0000)]
Change handling of global primitives; they are now integrated only
when they appear in operator position.

16 years agoMake SAVE-RESOURCE on file folders gracefully handle failure to write
Taylor R. Campbell [Sat, 9 Feb 2008 10:29:03 +0000 (10:29 +0000)]
Make SAVE-RESOURCE on file folders gracefully handle failure to write
the file.  Previously killing an IMAIL file folder buffer for a
read-only file would create an *error* buffer and fail to kill the
IMAIL buffer!

(Perhaps it would be better to add a new value to FOLDER-SYNC-STATUS,
to indicate that the persistant storage is immutable.  But putting an
error handler here is useful anyway, because the state of the world
may change between the call to FOLDER-SYNC-STATUS and the actual write
to the disk.)

16 years agoConditionally defined OS/2 packages.
Chris Hanson [Sat, 2 Feb 2008 18:21:30 +0000 (18:21 +0000)]
Conditionally defined OS/2 packages.

16 years agoWhen initializing packages, ignore missing packages.
Chris Hanson [Sat, 2 Feb 2008 18:20:59 +0000 (18:20 +0000)]
When initializing packages, ignore missing packages.

16 years agoGive FIND-PACKAGE an optional argument for controlling error behavior.
Chris Hanson [Sat, 2 Feb 2008 18:20:19 +0000 (18:20 +0000)]
Give FIND-PACKAGE an optional argument for controlling error behavior.

16 years agoNow that linking works, change WRITE-NOTIFICATION-LINE to be an alias
Chris Hanson [Sat, 2 Feb 2008 17:59:59 +0000 (17:59 +0000)]
Now that linking works, change WRITE-NOTIFICATION-LINE to be an alias
for WITH-NOTIFICATION, as originally planned.

16 years agoSimpler fix for variable-linking problem: clobber the source trap with
Chris Hanson [Sat, 2 Feb 2008 17:57:37 +0000 (17:57 +0000)]
Simpler fix for variable-linking problem: clobber the source trap with
the target cache, so that both traps share the same cache.

16 years agoFix bug: a relatively rare circumstance was causing linked variables
Chris Hanson [Sat, 2 Feb 2008 17:26:28 +0000 (17:26 +0000)]
Fix bug: a relatively rare circumstance was causing linked variables
to become unlinked.  Specifically:

    1. Load compiled code that creates a cache to variable A.
    2. Link variable B to variable C.
    3. Link variable B to variable A.

After step (3), variable C was no longer linked to variable B,
although B and A were still linked.  The problem is that step (3)
overwrites the cache in B with the cache in A, but C is unchanged.

This has been fixed by leaving a forwarding link in the "old" cache
and snapping the link on reference.  Any outstanding copies of the
"old" cache, such as that in C, are updated to point to the "new"
cache the next time they're referenced.

16 years agoMake sure there's only one instance of a name in each package.
Chris Hanson [Sat, 2 Feb 2008 06:54:01 +0000 (06:54 +0000)]
Make sure there's only one instance of a name in each package.

16 years agoWork around bug in variable linking mechanism.
Chris Hanson [Sat, 2 Feb 2008 06:53:37 +0000 (06:53 +0000)]
Work around bug in variable linking mechanism.

16 years agoEliminate unused variable.
Chris Hanson [Sat, 2 Feb 2008 06:51:48 +0000 (06:51 +0000)]
Eliminate unused variable.

16 years agoDoh.
Chris Hanson [Sat, 2 Feb 2008 06:01:59 +0000 (06:01 +0000)]
Doh.

16 years agoFix typo.
Chris Hanson [Sat, 2 Feb 2008 06:00:14 +0000 (06:00 +0000)]
Fix typo.

16 years agoFix some bugs in the previous change.
Chris Hanson [Sat, 2 Feb 2008 05:48:57 +0000 (05:48 +0000)]
Fix some bugs in the previous change.

16 years agoChange WITH-NOTIFICATION so that the port it passes to the MESSAGE is
Chris Hanson [Sat, 2 Feb 2008 05:35:33 +0000 (05:35 +0000)]
Change WITH-NOTIFICATION so that the port it passes to the MESSAGE is
smart about newlines and writes the prefix correctly.  Also, make the
THUNK optional, and consequently WRITE-NOTIFICATION-LINE is an alias
for WITH-NOTIFICATION.

16 years agoEliminate cross-file dependency on gstate structure. Generic I/O port
Chris Hanson [Sat, 2 Feb 2008 04:28:49 +0000 (04:28 +0000)]
Eliminate cross-file dependency on gstate structure.  Generic I/O port
now provides abstraction for managing additional state elements.

16 years agoExport GUARANTEE-THUNK.
Chris Hanson [Sat, 2 Feb 2008 03:44:52 +0000 (03:44 +0000)]
Export GUARANTEE-THUNK.

16 years agoFix comment.
Chris Hanson [Sat, 2 Feb 2008 02:08:48 +0000 (02:08 +0000)]
Fix comment.

16 years agoEliminate direct use of make-gstate in file I/O ports.
Chris Hanson [Sat, 2 Feb 2008 02:07:58 +0000 (02:07 +0000)]
Eliminate direct use of make-gstate in file I/O ports.

16 years agoEliminate direct use of make-gstate in string I/O ports, by extending
Chris Hanson [Sat, 2 Feb 2008 02:02:53 +0000 (02:02 +0000)]
Eliminate direct use of make-gstate in string I/O ports, by extending
MAKE-GENERIC-I/O-PORT to take extra arguments.

16 years agoRename and export MAKE-NON-CHANNEL-PORT-{SOURCE,SINK}.
Chris Hanson [Sat, 2 Feb 2008 01:48:56 +0000 (01:48 +0000)]
Rename and export MAKE-NON-CHANNEL-PORT-{SOURCE,SINK}.

16 years agoImplement PORT-TYPE/PARENT.
Chris Hanson [Sat, 2 Feb 2008 01:19:13 +0000 (01:19 +0000)]
Implement PORT-TYPE/PARENT.

16 years agoUpdate current copyright year.
Chris Hanson [Thu, 31 Jan 2008 03:49:49 +0000 (03:49 +0000)]
Update current copyright year.

16 years agoUpdate to match Debian directory.
Chris Hanson [Thu, 31 Jan 2008 02:53:58 +0000 (02:53 +0000)]
Update to match Debian directory.

16 years agoFix xscheme-prompt-for-expression-exit with patch from debian
Chris Hanson [Thu, 31 Jan 2008 02:18:57 +0000 (02:18 +0000)]
Fix xscheme-prompt-for-expression-exit with patch from debian
Bug#375858.

16 years agoUpdate for some of the srfi-1 procedures.
Chris Hanson [Thu, 31 Jan 2008 00:54:36 +0000 (00:54 +0000)]
Update for some of the srfi-1 procedures.

16 years agoUpdate copyright for 2008.
Chris Hanson [Wed, 30 Jan 2008 20:06:12 +0000 (20:06 +0000)]
Update copyright for 2008.

16 years agoOptimize handling of #F events, so that they are added to the event
Chris Hanson [Wed, 30 Jan 2008 14:33:12 +0000 (14:33 +0000)]
Optimize handling of #F events, so that they are added to the event
queue only when necessary.

16 years agoFix bug in previous change. Repaginate.
Chris Hanson [Wed, 30 Jan 2008 08:02:20 +0000 (08:02 +0000)]
Fix bug in previous change.  Repaginate.

16 years agoIn INFERIOR-THREAD-OUTPUT!, signal one thread event, then no more
Chris Hanson [Wed, 30 Jan 2008 07:50:32 +0000 (07:50 +0000)]
In INFERIOR-THREAD-OUTPUT!, signal one thread event, then no more
until after this batch of output is accepted.  The previous behavior
was causing significant delays when the inferior thread generated a
lot of output.

16 years agoMake sure %MAYBE-TOGGLE-THREAD-TIMER is called at every point where it
Chris Hanson [Wed, 30 Jan 2008 07:45:17 +0000 (07:45 +0000)]
Make sure %MAYBE-TOGGLE-THREAD-TIMER is called at every point where it
might be needed.

16 years agoUpdate RC-code tables for version 4.6. Implement Scheme program to
Chris Hanson [Tue, 29 Jan 2008 06:09:55 +0000 (06:09 +0000)]
Update RC-code tables for version 4.6.  Implement Scheme program to
generate the tables automatically.

16 years agoEliminate mixed fix/flo cases in binary operations. These do the
Chris Hanson [Tue, 29 Jan 2008 03:12:25 +0000 (03:12 +0000)]
Eliminate mixed fix/flo cases in binary operations.  These do the
wrong thing for multiply, so we might as well let the Scheme code
handle it.

16 years agoChange GCC restriction to be versioned.
Chris Hanson [Tue, 29 Jan 2008 02:27:47 +0000 (02:27 +0000)]
Change GCC restriction to be versioned.

16 years agoOops, older GCC used different "guard" macro name.
Chris Hanson [Tue, 29 Jan 2008 02:26:33 +0000 (02:26 +0000)]
Oops, older GCC used different "guard" macro name.

16 years agoAdd "guard" to prevent use of <stdbool.h> in old versions of GCC.
Chris Hanson [Tue, 29 Jan 2008 02:21:46 +0000 (02:21 +0000)]
Add "guard" to prevent use of <stdbool.h> in old versions of GCC.
Although we don't load that file, it is indirectly loaded by some
other include files.

16 years agoEliminate incorrect reference to HAVE__BOOL.
Chris Hanson [Tue, 29 Jan 2008 02:01:56 +0000 (02:01 +0000)]
Eliminate incorrect reference to HAVE__BOOL.

16 years agoEliminate compilation error.
Chris Hanson [Tue, 29 Jan 2008 02:01:37 +0000 (02:01 +0000)]
Eliminate compilation error.

16 years agoChange WEAK-LIST->LIST and LIST->WEAK-LIST to use iterative
Chris Hanson [Mon, 28 Jan 2008 04:26:49 +0000 (04:26 +0000)]
Change WEAK-LIST->LIST and LIST->WEAK-LIST to use iterative
algorithms.

16 years agoWhen parsing local variables, convert editor errors into failures.
Chris Hanson [Sun, 27 Jan 2008 02:25:37 +0000 (02:25 +0000)]
When parsing local variables, convert editor errors into failures.

16 years agoAdd definition of MAKE-TOP-LEVEL-ENVIRONMENT.
Chris Hanson [Sat, 26 Jan 2008 00:20:07 +0000 (00:20 +0000)]
Add definition of MAKE-TOP-LEVEL-ENVIRONMENT.

16 years agoDon't hand non-positive values to REAL-TIMER-SET!. (Thanks Taylor!)
Chris Hanson [Tue, 22 Jan 2008 22:46:08 +0000 (22:46 +0000)]
Don't hand non-positive values to REAL-TIMER-SET!.  (Thanks Taylor!)

16 years agoBase URI of READ-RDF/TURTLE-FILE should never have a file type.
Chris Hanson [Tue, 15 Jan 2008 04:39:27 +0000 (04:39 +0000)]
Base URI of READ-RDF/TURTLE-FILE should never have a file type.

16 years agoTweak handling of the thread timer so that it is set only when needed.
Chris Hanson [Mon, 14 Jan 2008 03:14:10 +0000 (03:14 +0000)]
Tweak handling of the thread timer so that it is set only when needed.
This saves power by avoiding unnecessary interrupts, and additionally
saves computation.

16 years agoFix bug: VC-CONTROL-DIRECTORIES must handle shadowing of inherited
Chris Hanson [Wed, 9 Jan 2008 19:23:43 +0000 (19:23 +0000)]
Fix bug: VC-CONTROL-DIRECTORIES must handle shadowing of inherited
control directories.  In other words, if directory A has a .bzr
subdirectory, and A/B has a .svn directory, then files in A/B/C
should ignore the .bzr directory in A.

16 years agoModify handling of EVALUATE-IN-INFERIOR-REPL and RUN-LIGHT so that
Chris Hanson [Mon, 7 Jan 2008 14:31:04 +0000 (14:31 +0000)]
Modify handling of EVALUATE-IN-INFERIOR-REPL and RUN-LIGHT so that
they adapt when SCHEME-ENVIRONMENT is set to a procedure.

16 years agoDon't allocate tables based on the values of POSIX configuration
Chris Hanson [Thu, 3 Jan 2008 00:30:47 +0000 (00:30 +0000)]
Don't allocate tables based on the values of POSIX configuration
variables.  These values may be undefined, or unreasonably large.
Instead, allocate small tables and grow them as needed.

16 years agoRework last change to allow full type-specific mode-line status
Chris Hanson [Thu, 20 Dec 2007 02:49:18 +0000 (02:49 +0000)]
Rework last change to allow full type-specific mode-line status
rendering.  Tweak bzr mode-line to be a little clearer.

16 years agoOptimize performance of bzr-controlled buffers. It turns out that
Chris Hanson [Thu, 20 Dec 2007 01:24:29 +0000 (01:24 +0000)]
Optimize performance of bzr-controlled buffers.  It turns out that
"bzr log" is very slow (several seconds on a Core Duo), even with
"--limit=1".  So disable display of the revision in the mode line, and
reduce the number of places that need to know the revision.  Also,
cache the output of "bzr ls", which is also slow-ish.

16 years agoFix brokenness in previous change: can't pass "." for PATH argument to
Chris Hanson [Tue, 18 Dec 2007 03:59:33 +0000 (03:59 +0000)]
Fix brokenness in previous change: can't pass "." for PATH argument to
"bzr ls --non-recursive".  Instead pass no PATH arg and it does the
right thing.  Also: don't forget to quote the filename before passing
it to RE-STRING-SEARCH-FORWARD.

16 years agoPass "--non-recursive" when calling "bzr ls", since we don't care
Chris Hanson [Sun, 16 Dec 2007 01:10:38 +0000 (01:10 +0000)]
Pass "--non-recursive" when calling "bzr ls", since we don't care
about files in sub-directories.

16 years agoUse notification mechanism for SHOW-TIME.
Chris Hanson [Mon, 10 Dec 2007 04:24:57 +0000 (04:24 +0000)]
Use notification mechanism for SHOW-TIME.

16 years agoBe more specific that column tracking works only for ASCII graphic
Chris Hanson [Sun, 9 Dec 2007 06:12:14 +0000 (06:12 +0000)]
Be more specific that column tracking works only for ASCII graphic
characters.

16 years agoImprove line breaking to have a more accurate idea where the break
Chris Hanson [Sun, 9 Dec 2007 06:09:32 +0000 (06:09 +0000)]
Improve line breaking to have a more accurate idea where the break
should go.

16 years agoChange indentation style to put open paren inline, contents indented,
Chris Hanson [Sun, 9 Dec 2007 05:53:04 +0000 (05:53 +0000)]
Change indentation style to put open paren inline, contents indented,
and close paren unindented.  Change indentation increment from 2 to 4.
Generate optional trailing ";" in bnodes and "." in subgraphs.  Don't
automatically indent multiple objects; instead try to break lines
where needed.  Add space after an unquoted literal, to prevent
confusion should it be followed by ".".

16 years agoSimplify column tracking: track columns as long as only simple
Chris Hanson [Sun, 9 Dec 2007 05:45:39 +0000 (05:45 +0000)]
Simplify column tracking: track columns as long as only simple
characters are used, independent of the port coding.

16 years agoImplement WRITE-RDF/TURTLE-SUBGRAPH.
Chris Hanson [Sun, 9 Dec 2007 05:09:28 +0000 (05:09 +0000)]
Implement WRITE-RDF/TURTLE-SUBGRAPH.