8423610f45
call-with-port does not use dynamic-wind to close the port.
2008-07-07 00:56:52 -07:00
b15c7063f9
repl exception handler now returns for non-serious conditions
...
(as per bug 242900)
2008-07-07 00:50:19 -07:00
badf83557f
fixed rounding for rationals.
2008-07-07 00:22:14 -07:00
579b823f44
WIP on tag analysis, annotations, and utilization.
2008-07-06 23:48:16 -07:00
d73dfd1287
- More work on the new optimizer
...
- Deleted old optimizer and changed the meaning of optimize level to
be:
0: bypass the optimizer
1: copy propagation, constant folding, no inlining
1: copy propagation, constant folding, with inlining
The default is set to 1.
2008-06-29 12:35:34 -07:00
66464cc348
Fixed a bug when doing (/ 1 1 1 ...).
2008-06-28 13:02:05 -07:00
2c8cb7dda2
fixed a bug in matching library subversion in psyntax.
2008-06-28 03:16:21 -07:00
1cd581de70
fixed char encoding error message and condition as per bug 243662.
2008-06-28 02:49:50 -07:00
45346ef865
- Added cp0! including:
...
- (optimize-level [0,1,2]) and ikarus -O[0,1,2]
where -O0 = no optimizations
-O1 = using old optimizer
-O2 = using the new cp0 optimizer
defaults to -O1 for now.
- (cp0-size-limit n) which is the limit of the residual size for
each inlining attempt
- (cp0-effort-limit n) which is the limit on the effort expended
for each inlining attempt
- Rewrote the syntax-match macro to make use of the same technology
used in syntax-case itself resulting in reduced code size.
- Added (system-value <symbol>) which returns the system value.
E.g., (system-value 'car) => #<procedure car>
This is pretty much the same as
(eval <symbol> (environment '(ikarus)))
except that it does not involve compiling the expression or
consulting the library/expander systems.
- Fixed the fasl loader to make it understand complex numbers.
2008-06-28 02:25:44 -07:00
7d9ed176ac
More work on cp0 which can now swallow the compiler but cannot yet
...
fold any primitives.
2008-06-22 22:10:05 -07:00
5a2501d4bb
Added port-position and port-has-port-position?
2008-06-19 21:49:24 -07:00
e65b39d95d
All parameters that are bound to variables now have proper names.
...
E.g., pretty-width now prints as #<procedure pretty-width>.
2008-06-19 01:58:59 -07:00
4bb7e170b5
Some procedures (like console-output-port) did not get names when
...
printed (instead, they were just #<procedure>). Names for
procedures that are defined like
(define foo
(let ([something ---])
(lamdba () ---)))
now works.
2008-06-18 22:47:56 -07:00
1bd699349a
Fixed a problem with port position accounting which was using fixnum
...
arithmetic instead of generic arithmetic.
2008-06-17 21:05:01 -07:00
333dc03f8f
(waitpid -1 #f) now returns #f if no child has died yet.
2008-06-13 22:11:51 -07:00
8f0f0abbca
Applied a patch supplied by Derick Eddington that provides enhanced
...
functionality for waitpid (nonblocking waitpid and more status info)
as well as the ability to send a signal to a process, e.g.,
(kill pid 'HUP).
2008-06-13 05:43:17 -07:00
041f9fdafa
Fixed a problem in code generator that was producing invalid memory
...
addressing for predicate operations.
2008-06-13 04:15:25 -07:00
3bddca30c4
allocation overflow check sequence is a little tighter now.
2008-06-10 23:01:22 -07:00
3ee75bece0
fixed incorrect handling of ".ddd" when reading from a port.
2008-06-10 18:38:06 -07:00
0939370a07
suppressing printing of file names during bootstrap.
2008-06-10 13:21:00 -07:00
fe1f7077ff
fixed minor bug in syntax-case where the wraps and marks of
...
top-marked wrapped syntax objects were incorrectly combined.
E.g., it used to be that:
(syntax-case (datum->syntax #'foo #'(x y)) ()
[(x y) 'shouldntmatch] [_ 'ok])
yields shouldntmatch; it's now ok.
2008-06-10 12:35:56 -07:00
ef50e9f515
dropping input-port-byte-position from (ikarus)'s exports.
2008-06-09 01:36:27 -07:00
c53ffd5314
fixed lookahead-char bug that I just made.
2008-06-08 15:53:42 -07:00
665ca784d1
Fixed a couple of typos in ikarus.io.ss.
2008-06-08 15:48:52 -07:00
ee950fcaf2
Added test to exercise last bug fixed (refilling io read buffer
...
drops bytes already in the buffer).
2008-06-08 06:42:58 -07:00
432e1d9b87
replaced &i/o-write with &i/o-read for IO read errors. Duuh.
2008-06-08 06:25:22 -07:00
a0c8346840
fixed bug in IO where the system was silently eating up any
...
remaining bytes while refilling a buffer.
2008-06-08 06:19:29 -07:00
1a5ab01e26
The bottom-most exception handler now only exits.
2008-06-07 08:32:00 -07:00
9438328f55
bytevector size for doing select has been fixed. It used to be 8
...
times larger than needed.
2008-06-06 08:22:45 -07:00
df4b31af1f
io-error and raise/strerr now include an &error condition.
2008-06-06 08:10:17 -07:00
af5472bfb2
exporting fork
2008-06-06 07:52:40 -07:00
6741ac2817
better error message when a library file "foo.ss" does not contain
...
the expected library name (foo).
2008-06-05 00:58:04 -07:00
9f7196d11a
better contextual error messages for invalid numeric sequences.
2008-06-03 22:27:33 -07:00
be37f629c5
the reader now understands some complex numbers.
2008-06-02 00:01:59 -07:00
a492d318e1
- fixed minor bug in current-directory
...
- changed implementation of string->number.
2008-05-31 20:10:17 -07:00
53905b9eea
Changes from Derick Eddington supporting file system info:
...
- Added file-regular?, file-directory?, file-symbolic-link?,
directory-list, make-directory, delete-directory, change-mode,
and make-symbolic-link.
- Bug fix in ikrt_open_input_fd and ikrt_open_output_fd which
considered a 0 FD as an error.
- Bug fix in ikrt_getcwd about len+1.
- Bug fix of &i/o-file-protection which didn't have &i/o-filename as
its parent.
- There is a new src/ikarus-errno.c file.
- Made tcp-connect and friends tell when they "failed to resolve
host name or connect".
- To support the uses of &i/o-filename subtypes, I modified
print-condition so it will cleanly print the parents' fields of a
condition;
2008-05-31 10:43:55 -07:00
6437aa98e0
- exp now understands complex numbers.
...
- expt of complex exponent is implemented in terms of exp.
2008-05-28 23:50:36 -07:00
fec5dcd419
sqrt, log, and expt now understand complex numbers.
2008-05-28 22:21:08 -07:00
b74e3976df
allocation check was still incorrect when the ap+size overflows and
...
shows up to be less than the end of allocation pointer.
2008-05-25 15:32:22 -07:00
7f0f127081
Allocation overflow check was incorrectly using signed comparison
...
instead of an unsigned one.
2008-05-25 14:39:03 -07:00
467095677e
+, -, *, and / now handle inexact complex numbers.
2008-05-25 10:37:41 -07:00
ab2e05e8b0
Added inexact complex numbers.
2008-05-24 10:13:01 -07:00
8827b98972
fasl reader/writer now understands complex numbers.
2008-05-21 00:40:42 -07:00
0a6df79e3b
ikarus.reader is now loadable separately.
2008-05-20 23:21:37 -07:00
ebcc042ea1
= now handles complex numbers.
2008-05-18 22:33:49 -07:00
4cb8165181
+, -, *, / now handle complex numbers.
2008-05-18 21:41:53 -07:00
82140f87ba
- Added complex numbers representation (exact complex only).
...
- Added make-rectangular.
- Added the ability to display complex numbers.
2008-05-18 20:39:41 -07:00
da7cedfe64
Added bitwise-xor.
2008-05-18 03:21:05 -07:00
de4276124b
added bitwise-ior.
2008-05-18 02:27:55 -07:00
0cd61369b2
added some preliminary work for known-types.
2008-05-16 23:27:08 -07:00
71640d11bf
Fixed a problem with forward references in the repl which used to
...
get a location different from the one assigned when the variable is
defined.
2008-05-15 06:40:07 -07:00
6511dd1002
Fixed a paren mismatch in "macro?" helper in printer.
2008-05-13 06:09:58 -07:00
5d3ab96c2c
fixed "inaccurate error message" bug in map when applied a non-list
...
argument.
2008-05-12 02:39:28 -07:00
4133bd73d3
Added copyright notice to pretty-formats.ss
2008-05-12 00:37:55 -07:00
4acf71d6d2
moved pretty-printing formats to their own library.
2008-05-12 00:35:19 -07:00
f05f171f28
exported pretty-format, a parameter maker for setting the pretty
...
format for various forms. The pretty format structure is not
documented yet and there is no error-checking mechanism for
verifying proper input.
2008-05-12 01:00:01 -04:00
f8b0e949d9
- syntax-case now raises a syntax violation when _ appears in the
...
literals
- syntax-rules points to the invalid literal when raising an
exception.
2008-05-10 19:05:40 -04:00
5c21f9995e
source annotations for r6rs-script and r6rs-libraries are now in
...
chars instead of bytes.
2008-05-06 15:38:05 -04:00
b5fc5624ec
Fixed a bug in enum-set-universe not returning an enum type.
2008-05-05 23:53:48 -04:00
3bcc3249e5
Fixed another source of "annotations being revealed". This time,
...
it's a reader bug of not inserting enough annotations for reader
macro forms.
2008-05-05 03:52:08 -04:00
c3b8e50b23
Fixed a minor exposure of annotation records.
2008-05-03 06:49:36 -04:00
3f9b567a5b
with-syntax now gives more informative error messages:
...
> (with-syntax ([(x y) #'(a 1)] [(q) #'(a b c)]) 12)
Unhandled exception
Condition components:
1. &assertion
2. &who: with-syntax
3. &message: "pattern does not match value"
4. &irritants: ((q) #<syntax (a b c)>)
instead of the previous behavior:
> (with-syntax ([(x y) #'(a 1)] [(q) #'(a b c)]) 12)
Unhandled exception
Condition components:
1. &message: "invalid syntax"
2. &syntax:
form: ((a 1) (a b c))
subform: #f
2008-05-03 06:23:35 -04:00
373edfc969
Fixes typos in (rnrs enumeration) error messages.
2008-05-03 04:25:58 -04:00
5f92e4b96d
Added current-directory parameter that gets/sets cwd.
2008-05-03 02:39:49 -04:00
7e5d053fb2
Fixed a bug causing syntax-error traces to print as
...
#[annotation ---]
instead of
#<syntax --- [position ---]>
2008-05-02 14:34:03 -04:00
d87ced74d1
cleaned up some stale primitive names.
2008-05-01 18:13:11 -04:00
db70f9751f
Made all unimplemented procedures raise errors when called.
2008-05-01 17:52:52 -04:00
b56b0fbd85
fixed a problem causing free-id=? to intern unbound identifiers
...
into the interaction environment.
2008-05-01 15:04:43 -04:00
d49aed209a
Added (interaction-environment).
2008-05-01 06:02:36 -04:00
e7386cd7d7
Major changes to the interactive repl:
...
- We can now redefine imported identifiers.
- We can now use let-syntax, letrec-syntax, and modules at
top-level.
Hand-wavey Repl Semantics:
- Imported identifiers mean what they meant previously. In
particular, you cannot set! imported identifiers.
- An imported identifier may be redefined using define. Once
something is defined at the top-level, a specific location for it
is created, and all definitions, references, and set!s to that
variable go through the top-level location.
- Re-importing an identifier shadows the top-level location.
- Redefining re-exposes the top-level location.
- and do on.
- A reference to an unbound variable also fabricates a top-level
location for that variable.
Let's see how this goes.
2008-05-01 04:21:07 -04:00
1389f239fe
Added process-nonblocking which is like process but returns
...
nonblocking ports.
2008-04-30 22:55:59 -04:00
f69e82e6c5
Added with-output-to-port:
...
> (let-values ([(p e) (open-string-output-port)])
(with-output-to-port p (lambda () (printf "Hello\n")))
(e))
"Hello\n"
2008-04-29 00:20:29 -04:00
d8cd4f0acf
Added engine-handler parameter and a ($swap-engine-counter! <neg-fx>)
...
to (ikarus system $interrupts).
2008-04-29 00:10:49 -04:00
478719cf32
Fixed names of procedures <, <=, =, >, >=. They used to be
...
printed as #<procedure f>.
2008-04-28 16:52:44 -04:00
b97b568e36
Reader now explicitly checks for valid unicode range when reading
...
#\xHHHH "\xHHHH;" and foo\xHHHH;bar sequences.
2008-04-28 15:01:45 -04:00
89def78c3c
Fixed the definition of identifier? and id? in expander so that only
...
top-make-wrapped (possibly annotated) symbols are identifiers. It
used to be that symbols were considered identifiers by the previous
definition.
2008-04-28 14:01:49 -04:00
3ce6d037e4
Fixed a bug in module imports where some marks were missing in action.
...
(Note from Kent, extracted from chez scheme release notes, follows)
This always worked:
(let ()
(define-syntax from
(syntax-rules ()
[(_ m v) (let () (import m) v)]))
(module a (x) (define x 'x-of-a))
(from a x))
Didn't work before this change:
(let ()
(define-syntax x-from-a
(syntax-rules ()
[(_) (let () (import a) x)]))
(module a (x) (define x 'x-of-a))
(x-from-a))
2008-04-19 17:55:30 -04:00
66d085713f
Made interrupted conditions &serious so that you can interrupt R6RS
...
scripts.
2008-04-19 16:55:40 -04:00
177f6b84a7
Fixed a reader bug that causes '#!eof not to read properly.
...
Used to be:
> '#!eof
Unhandled exception
Condition components:
1. &lexical
2. &message: "invalid eof after quote read macro"
3. &lexical-position:
file-name: *stdin*
character: 7
Now:
> '#!eof
#!eof
2008-04-14 22:18:56 -04:00
139ff2f33c
First complete (but broken) build of ikarus in 64-bit.
2008-04-14 22:02:03 -04:00
d02e9fe035
Passing 709 tests in 64-bit mode
2008-04-12 15:06:55 -04:00
df0aa2de1f
Added input-socket-buffer-size and output-socket-buffer-size
...
parameters.
2008-04-11 14:02:43 -04:00
bdd81e2b05
accept and accept-nonblocking now set the port-id to a string
...
representing the incoming address like "nnn.nnn.nnn.nnn:pppp"
2008-04-11 07:01:27 -04:00
25344fa1d0
Added nanosleep as per bug 210678.
2008-04-11 05:36:54 -04:00
191a82e007
Passing 685 tests in 64bit including handling of some foreign calls.
2008-04-09 07:04:02 -04:00
866b2b1c17
Now passing 682 tests in 64-bit mode.
2008-04-09 05:34:36 -04:00
f63f85e1cc
570 tests in 64-bit mode
2008-04-09 03:05:19 -04:00
aa44ce2733
Passing 548 tests in 64-bit mode.
2008-04-08 03:57:13 -04:00
5eaa2ff469
fixed a couple of bugs in 64bit assembler.
2008-04-08 02:22:26 -04:00
6bdb38ca16
fixed bug in ($make-vector 0)
2008-04-07 12:32:55 -04:00
42e3d53d00
I CAN HAZ CONS in 64BIT!
2008-04-07 10:20:05 -04:00
bf6138f86f
- Added reset-output-port! which sets the write index to 0 and
...
unregisters any callbacks associated with the buffer.
- commented out the call for unblocking the stdin for now
(for reason yet unknown to me, it causes stdout to be unblocked
as well, making writes to the console to come out in bizarre
mixed order)
2008-04-06 10:57:56 -04:00
afd7592ae1
Added &i/o-read, &i/o-write, and &i/o conditions to read/write
...
errors.
2008-04-05 05:15:30 -04:00
8d8f6d39f7
Fixing bug 210744: Interrupts and IO callbacks not interacting
...
properly
2008-04-02 20:28:45 -04:00
30bc1b7be9
apply error now shows the procedure before the number of args.
2008-03-25 22:15:10 -04:00
398ff23992
Fixes bug 206847: default exception handler expected to return for
...
non-&serious conditions
2008-03-25 21:38:11 -04:00
c01f3049e4
Fixes bug 206842: syntax tracing not working by default when
...
syntax-case fenders fail
2008-03-25 21:23:23 -04:00
c2e20a3a68
Fixes bug 206839: map not failing with invalid arguments
2008-03-25 21:03:26 -04:00
48f7c88e2d
made time macro not depend on format.
2008-03-24 21:50:37 -04:00
81c17da20b
Fixes bug 190373: `time' macro's "running stats for" can be misleading
2008-03-24 21:45:36 -04:00
3c3fec3908
Fixed problem shutting down servers on cygwin.
2008-03-24 13:25:59 -04:00
12a3f7e19b
Fixes bug 205988: sockaddr_in not recognized on cygwin.
2008-03-24 13:18:39 -04:00
fc92ec8e9f
Fixes bug 205437: socket errors not being reported
2008-03-24 00:01:22 -04:00
884f3fe921
Added tcp-server-socket-nonblocking, tcp-accept-connection-nonblocking
...
and register-callback for handling nonblocking servers and connections.
2008-03-23 05:02:12 -04:00
0f55361b19
Fixes bug 205427 in letrec*.
2008-03-23 04:14:53 -04:00
2119f44125
Added a simple tcp server facility. See lab/greeting-server.ss.
2008-03-23 03:44:20 -04:00
9aaf306f16
Nonblocking sockets no longer raise a continuable exception when a
...
read or write operation would block. Instead, they are schedules
with an event handler that `selects' on the pending file descriptors
and dispatches the appropriate callback.
2008-03-23 02:14:00 -04:00
6e599c4c72
refactored nonblocking code in ikarus-io.c and ikarus.io.ss
2008-03-23 00:41:49 -04:00
ec69dc4a7c
Added more tests for div0 and mod0.
2008-03-22 21:25:42 -04:00
48a86cbaa3
Fixed bug in (mod 3 5/6).
2008-03-22 21:23:51 -04:00
ce496aebaf
* Better error message when a tcp connection is refused.
...
* implemented udp connections (not working yet).
2008-03-22 19:29:41 -04:00
85d09cbc1c
fxarithmetic-shift-left now detects overflows properly.
2008-03-18 00:49:24 -04:00
97507bce08
Fixes bug in reader now recognizing the following comment:
...
> #|| hello ||# '12
2008-03-15 21:12:43 -04:00
50888bef94
Fixes issues/warnings in bug 193327.
2008-03-15 21:06:47 -04:00
d35566081b
import now recognizes and ignore ``for'' spec.
2008-03-15 21:00:16 -04:00
a3f6e3e039
Cut down bootstrap time by 10% by caching the values of scheme-stx.
...
Macroexpansion time is reduced by 25%.
2008-03-12 18:12:57 -04:00
1943212436
- Added tests for fasl objects
...
- fasl-read can now read bignums, flonums, and ratnums.
2008-03-09 00:25:03 -05:00
8f0b606609
sin, cos, tan, asin, acos, and atan now support rational numbers.
2008-03-03 13:49:47 -05:00
5210f1448b
Fixed part of bug 162334. Reason: (inexact -1/2) lost the sign and
...
returned 0.5. It now correctly return -0.5.
2008-03-01 21:54:27 -05:00
f6ee3618ba
(+ 1/2), (+ 0.5), (* 1/2) and (* 0.5) used to raise "not a number" errors.
2008-03-01 21:48:42 -05:00
db54cd0074
Fixes bug 191116: rationalize broken
2008-03-01 21:45:48 -05:00
755843e251
196894: --compile-dependencies misnamed in error message
2008-03-01 21:14:35 -05:00
ed9f85ea7c
Fixes bug 194232: fasl-write error message wrong about textual
...
output ports
2008-02-26 04:07:33 -05:00
e36b0029f6
Fixes bug 194259. R6RS records can now be written and read back
...
through fasl-write and fasl-read.
2008-02-26 03:53:00 -05:00
c8f9b1dc3d
Added fluid-let-syntax.
...
See http://www.scheme.com/csug7/syntax.html#./syntax:s15
for details.
2008-02-25 02:15:51 -05:00
7a9526ec91
removed a macos-specific dependencies in ikrt_file_ctime procedure.
2008-02-19 03:28:40 -05:00
523fa2e588
minor fixes to warning messages.
2008-02-19 00:15:18 -05:00
81f6c50341
fixed struct initialization problem in fasl-read.
2008-02-18 23:37:01 -05:00
259d43ca37
Changed fasl extension from ".ikfasl" to ".ikarus-fasl"
2008-02-18 23:29:49 -05:00
405c7dc9e2
fixed a missing put-mark call in fasl-read.
2008-02-18 23:26:54 -05:00
f45a74c115
removed the separate-compilation.ss file because it was not needed.
2008-02-18 22:07:58 -05:00
8564000d0d
ikarus --compile-dependencies seems to be working now.
2008-02-18 22:05:29 -05:00
8a809e2f58
Added (file-ctime filename) which returns the time of last change
...
(in nanoseconds)
2008-02-18 21:58:11 -05:00
e751c15bc4
library record now contains original file name for libraries loaded
...
from files.
2008-02-18 20:39:42 -05:00
c430a91bb8
more preparations for separate compilation
2008-02-18 20:28:54 -05:00
fa08c543bb
library records now cache expanded code.
2008-02-18 19:15:47 -05:00
163809f667
Ikarus can now compile, save, and reload a simple hello-world library.
2008-02-18 02:02:00 -05:00
e02b646d6e
Added separate compilation stubs.
2008-02-17 04:08:38 -05:00
5c56450c3e
removed some stale junk from compiler.ss
2008-02-17 03:11:12 -05:00
314f80eee9
include now attached annotations to read files (using read-annotated
...
instead of read).
2008-02-17 02:29:36 -05:00
1d6d5bec61
Some work on optimizer (still not working).
2008-02-17 02:11:04 -05:00
19e5d2bacd
Fixes bug 192222: Quasiquote broken
2008-02-15 15:25:20 -05:00
30aae80c5d
removed all symbolicly linked files.
2008-02-14 17:45:15 -05:00
f4fb08adba
Added an experimental macro expansion stack trace that looks like:
...
Unhandled exception:
Condition components:
1. &who: let
2. &message: "not an identifier"
3. &syntax:
form: (let ((12 a)) (printf "a=~s\n" a))
subform: 12
4. &trace: #<syntax (let ((12 a)) (printf "a=~s\n" a))>
5. &trace: #<syntax (right 12 a (printf "a=~s\n" a))>
6. &trace: #<syntax (wrong a 12 (printf "a=~s\n" a)) [byte 216 of test.ss]>
For a file containing:
(define-syntax right
(syntax-rules ()
[(_ a b c) (let ([a b]) c)]))
(define-syntax wrong
(syntax-rules ()
[(_ a b c) (right b a c)]))
(right x 17 (wrong a 12 (printf "a=~s\n" a)))
2008-02-14 04:01:09 -05:00
f6b35c4506
Fixes bug 191659: add1 and sub1 are slow
2008-02-13 18:12:00 -05:00
66b9f6968e
Work in progress on reimplementing the optimizer based on
...
Oscar Waddell's dissertation (chapter 4). The existing
optimizer in Ikarus is just a joke.
2008-02-13 03:29:34 -05:00
08fda34ebe
Added a single argument (fx- n) case that was missing.
2008-02-12 16:43:15 -05:00
bde9000c06
fx- now checks for overflow.
2008-02-12 15:03:43 -05:00
1683997a12
fx* now detects overflow.
2008-02-12 14:46:18 -05:00
3811d0a4c2
Loops with a single free variable no longer allocate a closure.
2008-02-11 22:18:32 -05:00
Abdulaziz Ghuloum