* module/language/lua/compile-tree-il.scm (compile): Add clause for
ast-variable-arguments.
* module/language/lua/parser.scm (define-ast, make-parser): Add
vararg-gensym field to functions, gensym field to variable-arguments.
Propagate *vararg-gensym* from functions to variable-arguments.
* test-suite/tests/lua-eval-2.test ("lua-eval"): Check for #nil
values resulting from a function call as the last argument.
doc/ref/api-languages.texi: Add a small blurb about Lua.
module/language/lua/compile-tree-il.scm: Function calls now properly
handle multiple values resulting from a function call as the last
argument.
* module/language/lua/lexer.scm: Some tweaks and reindentations. Remove
the define/init lexer interface; I don't like separating declaration
and initialization.
* module/language/lua/parser.scm:
* test-suite/tests/lua-lexer.test: Adapt to lexer interface change.
What is missing:
+ Functions: module, getfenv, setfenv, math.modf, table.sort
+ Parser: needs to be more flexible
+ Compiler: needs more extensive work to properly handle all possible
cases of variable arguments, multiple returns, and loops
+ Language: Variable arguments and unpacking of multiple returns. (For
example we need to be able to handle something as complex as
print(unpack({...})), which is easy with Lua's explicit stack but will
require lots of tree-il gymnastics, or perhaps modifications to better
allow different calling conventions. (For instance -- how would we
support Python or Ruby, where keyword arguments are gathered into a
hashtable and passed as a single argument?)
What is there:
A fair shot at supporting Lua 5.1, not quite a drop-in replacement, but
not far from that goal either.
* libguile/vm-engine.c (return-values): Remove NVALUES operand.
Don't reset the frame.
* test-suite/tests/rtl.test ("cached-toplevel-set!"): Adapt to the fact
that 'return-values' has no operand now, and that 'reset-frame' must
be done first.
Fixes <http://bugs.gnu.org/15100>.
Reported by Göran Weinholt <goran@weinholt.se>.
* libguile/bytevectors.c (INTEGERS_TO_LIST): Enforce the R6RS
requirement that SIZE divides the bytevector length.
* test-suite/tests/bytevectors.test: Add test. Remove tests that
assumes that this "divides" check is not enforced.
Partially fixes <http://bugs.gnu.org/15100>.
Reported by Göran Weinholt <goran@weinholt.se>.
* libguile/bytevectors.c (INTEGERS_TO_LIST): Make sure SIZE isn't 0.
Allow SIZE to be greater than the bytevector length, for consistency
with allowing extra bytes at the end when the bytevector length is
non-zero. Use scm_from_size_t instead of scm_from_uint.
* test-suite/tests/bytevectors.test: Add tests. Remove a test that
checks for an exception when SIZE is greater than the bytevector
length.
* test-suite/tests/peval.test ("partial evaluation"): Adapt tests to the
fact that arithmetic operators are now expanded into binary ops with
left-to-right associativity.
* libguile/instructions.c (FOR_EACH_INSTRUCTION_WORD_TYPE): Allow for
five-word instructions, and for new instruction word types.
* libguile/vm-engine.c (RETURN_ONE_VALUE): Instead of returning the
value in the fixed part of the call frame, return it in the same place
multiple-value returns go: from slot 1.
(BR_ARITHMETIC): Allow arithmetic tests to be negated.
(rtl_vm_engine): Change calling convention to use the same location
for single and multiple-value returns. Renumber all instructions.
(halt, halt/values): Fold into a single instruction (halt).
(call): Take the location of the procedure instead of the location of
the call frame. Also take the number of args, and reset the sp before
jumping to the procedure, so as to indicate the number of arguments.
(call/values): Remove, as the new calling convention has RA == MVRA.
(tail-call): Require the procedure to be shuffled down already, and
take "nlocals" as an arg instead of "nargs".
(receive, receive-values): New instructions, for receiving returned
values from calls.
(return-values): Rename from return/values. Remove "values".
(alloc-frame): Rename from reserve-locals.
(reset-frame): New instruction.
(drop-locals): Remove.
(br-if-=, br-if-<, br-if-<=): Allow these instructions to be
negatable.
(br-if->, br-if->=): Remove. Probably a bad idea, given NaN.
(box-ref): Don't bother trying to do a reverse lookup -- the
toplevel-box, module-box, and resolve instructions should handle
that.
(resolve): Add arg to check that the variable is bound.
(toplevel-box, module-box): New instructions, replacing toplevel-ref,
toplevel-set, module-ref, and module-set.
* libguile/vm.c (rtl_boot_continuation_code, rtl_values_code): Adapt to
instruction set changes.
* module/Makefile.am: Make the assembler and disassembler dependent on
vm-operations.h.
* module/system/vm/assembler.scm:
* module/system/vm/disassembler.scm: Adapt to instruction changes and
new instruction word kinds.
* test-suite/tests/rtl.test: Adapt to instruction set changes.
* module/language/tree-il/peval.scm (peval): Remove abort optimization;
the CPS compiler will do much better here, and it is complicating
things in the meantime.
* libguile/numbers.c (scm_numerator, scm_denominator): Handle signed
zeroes and infinities in accordance with the corresponding R6RS flonum
procedures.
* module/rnrs/arithmetic/flonums.scm (flnumerator, fldenominator):
Remove special handling of infinities.
* test-suite/tests/numbers.test (numerator, denominator): Add tests.
Convert existing tests to use 'pass-if-equal'.
* test-suite/tests/r6rs-arithmetic-flonums.test (flnumerator): Fix
broken test of (flnumerator -0.0).
* libguile/strports.c (scm_mkstrport): Use UTF-8; ignore
%default-port-encoding. Rename 'str_len' and 'c_pos' to
'num_bytes' and 'c_byte_pos'. Interpret 'pos' argument
as a character index instead of a byte index.
* module/ice-9/boot-9.scm (%cond-expand-features): Add srfi-6 to the
list of core features.
* module/srfi/srfi-6.scm (open-input-string, open-output-string): Simply
re-export these, since the core versions are now compliant.
* doc/ref/api-io.texi (String Ports): Remove text that describes
non-compliant behavior of string ports with regard to encoding.
* doc/ref/srfi-modules.texi (SRFI-0): Add srfi-6 to the list of
core features.
(SRFI-6): Remove text that mentions non-compliant behavior of
core string ports.
* module/ice-9/format.scm (format):
* module/ice-9/pretty-print.scm (truncated-print):
* module/rnrs/io/ports.scm (open-string-input-port,
open-string-output-port):
* test-suite/test-suite/lib.scm (format-test-name):
* test-suite/tests/chars.test ("combining accent is pretty-printed",
"combining X is pretty-printed"):
* test-suite/tests/ecmascript.test (eread, eread/1):
* test-suite/tests/rdelim.test:
* test-suite/tests/reader.test (read-string):
* test-suite/tests/regexp.test:
* test-suite/tests/srfi-105.test (read-string): Don't set
%default-port-encoding before creating string ports.
* benchmark-suite/benchmarks/ports.bm (%latin1-port): Use
'set-port-encoding!' to set the string port encoding.
(%utf8/ascii-port, %utf8/wide-port, "rdelim"): Don't set
%default-port-encoding before creating string ports.
* test-suite/tests/r6rs-ports.test ("lookahead-u8 non-ASCII"): Don't set
%default-port-encoding before creating string ports.
("put-bytevector with UTF-16 string port", "put-bytevector with
wrong-encoding string port"): Use 'set-port-encoding!' to set the
string port encoding.
* test-suite/tests/print.test (tprint): Use 'set-port-encoding!' to set
the string port encoding.
("truncated-print"): Use 'pass-if-equal'.
* test-suite/tests/ports.test ("encoding failure leads to exception",
"%default-port-encoding is honored", "peek-char [latin-1]", "peek-char
[utf-8]", "peek-char [utf-16]"): Remove tests.
("%default-port-encoding is ignored", "peek-char"): Add tests.
("suitable encoding [latin-1]", "suitable encoding [latin-3]",
"wrong encoding, error", "wrong encoding, substitute",
"wrong encoding, escape"): Use 'set-port-encoding!' to set the
string port encoding.
("%default-port-encoding, wrong encoding"): Rewrite to use
a file port instead of a string port.
* libguile/vm-engine.c (push_frame, call): Separate out push_frame from
call, and expect the caller to arrange the proc and arguments.
* test-suite/tests/rtl.test ("call"): Update tests.
* libguile/vm-engine.c: Change the RTL VM to number the procedure as
local 0, and other locals from 1. In the future we will want the FP
to point to local 0 instead of local 1. In the future also we can
elide the procedure for well-known closures (closures in which all
references are known call sites).
(make_closure, free_set): Instead of taking rest arguments, we add a
new free-set! op that initializes closures.
(free_ref): Take the closure as an argument.
* libguile/vm.c (rtl_boot_continuation_code): Remove comments, which
were out of date.
(rtl_apply_code, rtl_values_code): Update comments.
* module/system/vm/assembler.scm (intern-constant, emit-init-constants):
Adapt to locals numbering change.
(begin-kw-arity): For assert-nargs-ee purposes, nreq includes the
procedure.
* module/system/vm/disassembler.scm (code-annotation): Adapt annotation
for assert-nargs-ee/locals.
* test-suite/tests/rtl.test: Adapt tests.
* module/rnrs/arithmetic/bitwise.scm (bitwise-if, bitwise-length,
bitwise-first-bit-set, bitwise-bit-field, bitwise-reverse-bit-field):
Replace these with aliases to the identical SRFI-60 operators
'bitwise-if', 'integer-length', 'first-set-bit', 'bit-field', and
'reverse-bit-field'.
(bitwise-copy-bit, bitwise-copy-bit-field, bitwise-rotate-bit-field):
Reimplement these based upon the similar SRFI-60 operators 'copy-bit',
'copy-bit-field', and 'rotate-bit-field'.
* test-suite/tests/r6rs-arithmetic-bitwise.test (bitwise-copy-bit): Fix
test to conform to the specification, which requires the third
argument to be either 0 or 1.
* test-suite/tests/r6rs-arithmetic-fixnums.test (fxcopy-bit): Fix test
to conform to the specification, which requires the third argument to
be either 0 or 1.
Fixes <http://bugs.gnu.org/14905>.
Reported by Göran Weinholt <goran@weinholt.se>.
* libguile/numbers.c (scm_rationalize): Rewrite. Previously an
incorrect algorithm was used which failed in many cases.
* test-suite/tests/numbers.test (rationalize): Add tests.
Reported by Göran Weinholt <goran@weinholt.se>.
* module/rnrs/arithmetic/fixnums.scm (fxbit-count): If the argument is
negative, return the 'bitwise-not' of the result of 'logcount', as per
R6RS. Previously, 'fxbit-count' was identical to 'logcount'.
* test-suite/tests/r6rs-arithmetic-fixnums.test (fxbit-count): Add test.
Fixes <http://bugs.gnu.org/14868>.
Reported by Göran Weinholt <goran@weinholt.se>.
* module/rnrs/arithmetic/flonums.scm (flfinite?): If the argument is a
NaN, return false.
* test-suite/tests/r6rs-arithmetic-flonums.test (flfinite?): Add test.
* libguile/numbers.c (scm_num_eq_p): Fix bug comparing fractions to
infinities (reported by Göran Weinholt <goran@weinholt.se>). Fix
erroneous comment describing the logic behind inum/flonum comparison.
Use similar logic for inum/complex comparison to avoid rounding
errors. Make minor indentation fixes and simplifications.
* test-suite/tests/numbers.test (=): Add tests.
* libguile/numbers.c (INUM_LOSSLESSLY_CONVERTIBLE_TO_DOUBLE):
New macro.
(scm_i_divide2double): Use INUM_LOSSLESSLY_CONVERTIBLE_TO_DOUBLE to
determine if our fast path is safe. Previously, negative arguments
were not checked properly.
* test-suite/tests/numbers.test (exact->inexact): Add tests.
Fixes <http://bugs.gnu.org/14864>.
Reported by Göran Weinholt <goran@weinholt.se>.
* module/rnrs/arithmetic/bitwise.scm (bitwise-bit-count): If the
argument is negative, return the 'bitwise-not' of the result of
'logcount', as per R6RS. Previously, 'bitwise-bit-count' was
identical to 'logcount'.
* module/language/tree-il.scm (<prompt>): Change to have the body and
handler be lambdas, and add an "escape-only?" field. This will make
generic prompts work better in CPS or ANF with the RTL VM, as it
doesn't make sense in that context to capture only part of a frame.
Escape-only prompts can still be fully inlined.
(parse-tree-il, unparse-tree-il): Add escape-only? to the
serialization.
(make-tree-il-folder, pre-post-order): Deal with escape-only?.
* module/language/tree-il/analyze.scm (analyze-lexicals): Handle
escape-only?, and the new expectations for the body and handler.
* module/language/tree-il/canonicalize.scm (canonicalize): Ensure that
the body of an escape-only continuation is a thunk, and that the
handler is always a lambda.
* module/language/tree-il/debug.scm (verify-tree-il): Assert that
escape-only? is a boolean.
* module/language/tree-il/cse.scm (cse):
* module/language/tree-il/effects.scm (make-effects-analyzer):
* module/language/tree-il/peval.scm (peval):
* module/language/tree-il/primitives.scm (*primitive-expand-table*):
* test-suite/tests/peval.test ("partial evaluation"):
* module/language/tree-il/compile-glil.scm (flatten-lambda-case): Adapt
to <prompt> change.
* doc/ref/compiler.texi: Remove mention of <dynwind>.
* libguile/eval.c (eval): Remove SCM_M_DYNWIND case.
* libguile/expand.c: Remove scm_sym_at_dynamic_wind.
* libguile/memoize.c (do_wind, do_unwind): A couple of hacky subrs. If
we see a wind or unwind primcall, we expand to a call of a quoted subr
value. It works and removes a kind of memoized value from the
interpreter. For the compiler,primcalls to wind and unwind are
handled specially.
(MAKMEMO_DYNWIND): Remove.
(scm_tc16_memoizer): Remove. Yay!
(memoize): Remove speculative lookup for toplevels to see if they are
memoizers: there are no more memoizers. Memoize calls to the wind and
unwind primitives.
(m_dynamic_wind): Remove.
(unmemoize): Remove dynwind case.
(scm_init_memoize): Add wind and unwind local definitions.
* module/ice-9/boot-9.scm (dynamic-wind): Reimplement in terms of "wind"
and "unwind" primitives. These primitives are not exposed to other
modules.
* module/ice-9/eval.scm (primitive-eval): Remove dynwind case.
* module/language/scheme/decompile-tree-il.scm (do-decompile):
(choose-output-names): Remove dynwind cases.
* module/language/tree-il.scm: Remove <dynwind>. Yaaay!
* module/language/tree-il/analyze.scm (analyze-lexicals): Remove dynwind
cases.
* module/language/tree-il/compile-glil.scm (*primcall-ops*): Add wind
and unwind.
(flatten-lambda-case): Remove dynwind case. Yay!
* module/language/tree-il/cse.scm (cse):
* module/language/tree-il/debug.scm (verify-tree-il):
* module/language/tree-il/effects.scm (make-effects-analyzer):
* module/language/tree-il/peval.scm (singly-valued-expression?, peval):
Remove <dywind> cases. Inline primcalls to dynamic-wind. Add
constant folding for thunk?.
* module/language/tree-il/primitives.scm (*interesting-primitive-names*):
Remove @dynamic-wind, and add procedure? and thunk?.
(*effect+exception-free-primitives*): Add procedure? and thunk?.
(*multiply-valued-primitives*): Remove @dynamic-wind.
Remove @dynamic-wind expander.
* test-suite/tests/peval.test ("partial evaluation"): Update tests for
dynwind desugaring.
