trying to make qse_scm_print() non-recursive

This commit is contained in:
hyung-hwan 2011-03-03 08:41:24 +00:00
parent 1d478104b3
commit 5703d4c58a
7 changed files with 984 additions and 917 deletions

View File

@ -164,6 +164,7 @@ int scm_main (int argc, qse_char_t* argv[])
qse_scm_attachio (scm, &io); qse_scm_attachio (scm, &io);
} }
//qse_scm_read (scm);
qse_scm_print (scm, qse_scm_read (scm)); qse_scm_print (scm, qse_scm_read (scm));
#if 0 #if 0

View File

@ -8,6 +8,6 @@ AM_CPPFLAGS = \
lib_LTLIBRARIES = libqsescm.la lib_LTLIBRARIES = libqsescm.la
libqsescm_la_SOURCES = scm.h scm.c mem.c print.c err.c libqsescm_la_SOURCES = scm.h scm.c mem.c read.c print.c err.c
libqsescm_la_LDFLAGS = -L../cmn -L$(libdir) -version-info 1:0:0 -no-undefined libqsescm_la_LDFLAGS = -L../cmn -L$(libdir) -version-info 1:0:0 -no-undefined
libqsescm_la_LIBADD = -lqsecmn libqsescm_la_LIBADD = -lqsecmn

View File

@ -71,7 +71,7 @@ am__base_list = \
am__installdirs = "$(DESTDIR)$(libdir)" am__installdirs = "$(DESTDIR)$(libdir)"
LTLIBRARIES = $(lib_LTLIBRARIES) LTLIBRARIES = $(lib_LTLIBRARIES)
libqsescm_la_DEPENDENCIES = libqsescm_la_DEPENDENCIES =
am_libqsescm_la_OBJECTS = scm.lo mem.lo print.lo err.lo am_libqsescm_la_OBJECTS = scm.lo mem.lo read.lo print.lo err.lo
libqsescm_la_OBJECTS = $(am_libqsescm_la_OBJECTS) libqsescm_la_OBJECTS = $(am_libqsescm_la_OBJECTS)
libqsescm_la_LINK = $(LIBTOOL) --tag=CC $(AM_LIBTOOLFLAGS) \ libqsescm_la_LINK = $(LIBTOOL) --tag=CC $(AM_LIBTOOLFLAGS) \
$(LIBTOOLFLAGS) --mode=link $(CCLD) $(AM_CFLAGS) $(CFLAGS) \ $(LIBTOOLFLAGS) --mode=link $(CCLD) $(AM_CFLAGS) $(CFLAGS) \
@ -238,7 +238,7 @@ AM_CPPFLAGS = \
-I$(includedir) -I$(includedir)
lib_LTLIBRARIES = libqsescm.la lib_LTLIBRARIES = libqsescm.la
libqsescm_la_SOURCES = scm.h scm.c mem.c print.c err.c libqsescm_la_SOURCES = scm.h scm.c mem.c read.c print.c err.c
libqsescm_la_LDFLAGS = -L../cmn -L$(libdir) -version-info 1:0:0 -no-undefined libqsescm_la_LDFLAGS = -L../cmn -L$(libdir) -version-info 1:0:0 -no-undefined
libqsescm_la_LIBADD = -lqsecmn libqsescm_la_LIBADD = -lqsecmn
all: all-am all: all-am
@ -318,6 +318,7 @@ distclean-compile:
@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/err.Plo@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/err.Plo@am__quote@
@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/mem.Plo@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/mem.Plo@am__quote@
@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/print.Plo@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/print.Plo@am__quote@
@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/read.Plo@am__quote@
@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/scm.Plo@am__quote@ @AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/scm.Plo@am__quote@
.c.o: .c.o:

View File

@ -114,12 +114,30 @@ static qse_size_t long_to_str (
return ret; return ret;
} }
static QSE_INLINE push (qse_scm_t* scm, qse_scm_ent_t* obj)
{
qse_scm_ent_t* top;
top = qse_scm_makepairent (scm, obj, scm->p.s);
if (top == QSE_NULL) return -1;
scm->p.s = top;
return 0;
}
static QSE_INLINE qse_scm_ent_t* pop (qse_scm_t* scm)
{
qse_scm_ent_t* top = scm->p.s;
scm->p.s = PAIR_CDR(scm->p.s);
return PAIR_CAR(top);
}
static int print_entity ( static int print_entity (
qse_scm_t* scm, const qse_scm_ent_t* obj, int prt_cons_par) qse_scm_t* scm, const qse_scm_ent_t* obj, int prt_cons_par)
{ {
qse_char_t buf[256];
qse_long_t nval; qse_long_t nval;
retry:
if (IS_SMALLINT(scm,obj)) if (IS_SMALLINT(scm,obj))
{ {
nval = FROM_SMALLINT(scm,obj); nval = FROM_SMALLINT(scm,obj);
@ -155,6 +173,8 @@ static int print_entity (
#if 0 #if 0
case QSE_SCM_ENT_REAL: case QSE_SCM_ENT_REAL:
{
qse_char_t buf[256];
scm->prm.sprintf ( scm->prm.sprintf (
scm->prm.udd, scm->prm.udd,
buf, QSE_COUNTOF(buf), buf, QSE_COUNTOF(buf),
@ -168,6 +188,7 @@ static int print_entity (
OUTPUT_STR (scm, buf); OUTPUT_STR (scm, buf);
break; break;
}
#endif #endif
case QSE_SCM_ENT_SYM: case QSE_SCM_ENT_SYM:
@ -185,6 +206,35 @@ static int print_entity (
{ {
const qse_scm_ent_t* p = obj; const qse_scm_ent_t* p = obj;
if (prt_cons_par) OUTPUT_STR (scm, QSE_T("(")); if (prt_cons_par) OUTPUT_STR (scm, QSE_T("("));
do
{
if (push (scm, PAIR_CDR(p)) <= -1) return -1;
obj = PAIR_CAR(p);
goto retry;
resume:
p = pop (scm);
if (!IS_NIL(scm,p))
{
OUTPUT_STR (scm, QSE_T(" "));
if (IS_SMALLINT(scm,p) || TYPE(p) != QSE_SCM_ENT_PAIR)
{
OUTPUT_STR (scm, QSE_T(". "));
// push resume location
// push ....
if (push (scm, 1) <= -1) return -1;
obj = p;
goto retry;
//qse_scm_print (scm, p);
}
}
}
while (!IS_NIL(scm,p) && !IS_SMALLINT(scm,p) && TYPE(p) == QSE_SCM_ENT_PAIR);
if (prt_cons_par) OUTPUT_STR (scm, QSE_T(")"));
#if 0
do do
{ {
qse_scm_print (scm, PAIR_CAR(p)); qse_scm_print (scm, PAIR_CAR(p));
@ -201,29 +251,13 @@ static int print_entity (
} }
while (p != scm->nil && TYPE(p) == QSE_SCM_ENT_PAIR); while (p != scm->nil && TYPE(p) == QSE_SCM_ENT_PAIR);
if (prt_cons_par) OUTPUT_STR (scm, QSE_T(")")); if (prt_cons_par) OUTPUT_STR (scm, QSE_T(")"));
#endif
break; break;
} }
#if 0 #if 0
case QSE_SCM_ENT_FUNC: case QSE_SCM_ENT_PROC:
/*OUTPUT_STR (scm, QSE_T("func"));*/
OUTPUT_STR (scm, QSE_T("(lambda "));
if (print_entity (scm, QSE_SCM_FFORMAL(obj), 1) == -1) return -1;
OUTPUT_STR (scm, QSE_T(" "));
if (print_entity (scm, QSE_SCM_FBODY(obj), 0) == -1) return -1;
OUTPUT_STR (scm, QSE_T(")"));
break;
case QSE_SCM_ENT_MACRO:
OUTPUT_STR (scm, QSE_T("(macro "));
if (print_entity (scm, QSE_SCM_FFORMAL(obj), 1) == -1) return -1;
OUTPUT_STR (scm, QSE_T(" "));
if (print_entity (scm, QSE_SCM_FBODY(obj), 0) == -1) return -1;
OUTPUT_STR (scm, QSE_T(")"));
break;
case QSE_SCM_ENT_PRIM:
OUTPUT_STR (scm, QSE_T("prim"));
break; break;
#endif #endif
@ -233,6 +267,10 @@ static int print_entity (
return -1; return -1;
} }
/* if the print stack is not empty, we still got more to print */
if (!IS_NIL(scm,scm->p.s)) goto resume;
return 0; return 0;
} }

912
qse/lib/scm/read.c Normal file
View File

@ -0,0 +1,912 @@
/*
* $Id$
*
Copyright 2006-2009 Chung, Hyung-Hwan.
This file is part of QSE.
QSE is free software: you can redistribute it and/or modify
it under the terms of the GNU Lesser General Public License as
published by the Free Software Foundation, either version 3 of
the License, or (at your option) any later version.
QSE is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public
License along with QSE. If not, see <http://www.gnu.org/licenses/>.
*/
#include "scm.h"
enum list_flag_t
{
QUOTED = (1 << 0),
DOTTED = (1 << 1),
CLOSED = (1 << 2)
};
enum tok_type_t
{
TOK_END = 0,
TOK_T = 1,
TOK_F = 2,
TOK_INT = 3,
TOK_REAL = 4,
TOK_SYMBOL = 5,
TOK_STRING = 6,
TOK_LPAREN = 7,
TOK_RPAREN = 8,
TOK_DOT = 9,
TOK_QUOTE = 10,
TOK_QQUOTE = 11, /* quasiquote */
TOK_COMMA = 12,
TOK_COMMAAT = 13,
#if 0
TOK_INVALID = 50
#endif
};
#define TOK_CLR(scm) qse_str_clear(&(scm)->r.t.name)
#define TOK_TYPE(scm) (scm)->r.t.type
#define TOK_IVAL(scm) (scm)->r.t.ival
#define TOK_RVAL(scm) (scm)->r.t.rval
#define TOK_NAME(scm) (&(scm)->r.t.name)
#define TOK_NAME_PTR(scm) TOK_NAME(scm)->ptr
#define TOK_NAME_LEN(scm) TOK_NAME(scm)->len
#define TOK_LOC(scm) (scm)->r.t.loc
#define TOK_ADD_CHAR(scm,ch) QSE_BLOCK (\
if (qse_str_ccat(TOK_NAME(scm), ch) == -1) \
{ \
qse_scm_seterror (scm, QSE_SCM_ENOMEM, QSE_NULL, &scm->r.curloc); \
return -1; \
} \
)
#define IS_DIGIT(ch) ((ch) >= QSE_T('0') && (ch) <= QSE_T('9'))
#define IS_SPACE(ch) ((ch) == QSE_T(' ') || (ch) == QSE_T('\t'))
#define IS_NEWLINE(ch) ((ch) == QSE_T('\n') || (ch) == QSE_T('\r'))
#define IS_WHSPACE(ch) IS_SPACE(ch) || IS_NEWLINE(ch)
#define IS_DELIM(ch) \
(IS_WHSPACE(ch) || (ch) == QSE_T('(') || (ch) == QSE_T(')') || \
(ch) == QSE_T('\"') || (ch) == QSE_T(';') || (ch) == QSE_CHAR_EOF)
#define READ_CHAR(scm) QSE_BLOCK(if (read_char(scm) <= -1) return -1;)
#define READ_TOKEN(scm) QSE_BLOCK(if (read_token(scm) <= -1) return -1;)
static int read_char (qse_scm_t* scm)
{
qse_ssize_t n;
qse_char_t c;
/* TODO: do bufferring */
scm->err.num = QSE_SCM_ENOERR;
n = scm->io.fns.in (scm, QSE_SCM_IO_READ, &scm->io.arg.in, &c, 1);
if (n <= -1)
{
if (scm->err.num == QSE_SCM_ENOERR)
qse_scm_seterror (scm, QSE_SCM_EIO, QSE_NULL, QSE_NULL);
return -1;
}
/* TODO: handle the case when a new file is included or loaded ...
* stacking of curloc is needed??? see qseawk for reference
*/
if (n == 0) scm->r.curc = QSE_CHAR_EOF;
else
{
scm->r.curc = c;
if (c == QSE_T('\n'))
{
scm->r.curloc.colm = 0;
scm->r.curloc.line++;
}
else scm->r.curloc.colm++;
}
qse_printf (QSE_T("[%c]\n"), scm->r.curc);
return 0;
}
static int read_string_token (qse_scm_t* scm)
{
qse_cint_t c;
int escaped = 0;
int digit_count = 0;
qse_cint_t c_acc = 0;
while (1)
{
READ_CHAR (scm);
c = scm->r.curc;
if (c == QSE_CHAR_EOF)
{
qse_scm_seterror (
scm, QSE_SCM_EENDSTR,
QSE_NULL, &scm->r.curloc);
return -1;
}
if (escaped == 3)
{
if (c >= QSE_T('0') && c <= QSE_T('7'))
{
c_acc = c_acc * 8 + c - QSE_T('0');
digit_count++;
if (digit_count >= escaped)
{
TOK_ADD_CHAR (scm, c_acc);
escaped = 0;
}
continue;
}
else
{
TOK_ADD_CHAR (scm, c_acc);
escaped = 0;
}
}
else if (escaped == 2 || escaped == 4 || escaped == 8)
{
if (c >= QSE_T('0') && c <= QSE_T('9'))
{
c_acc = c_acc * 16 + c - QSE_T('0');
digit_count++;
if (digit_count >= escaped)
{
TOK_ADD_CHAR (scm, c_acc);
escaped = 0;
}
continue;
}
else if (c >= QSE_T('A') && c <= QSE_T('F'))
{
c_acc = c_acc * 16 + c - QSE_T('A') + 10;
digit_count++;
if (digit_count >= escaped)
{
TOK_ADD_CHAR (scm, c_acc);
escaped = 0;
}
continue;
}
else if (c >= QSE_T('a') && c <= QSE_T('f'))
{
c_acc = c_acc * 16 + c - QSE_T('a') + 10;
digit_count++;
if (digit_count >= escaped)
{
TOK_ADD_CHAR (scm, c_acc);
escaped = 0;
}
continue;
}
else
{
qse_char_t rc;
rc = (escaped == 2)? QSE_T('x'):
(escaped == 4)? QSE_T('u'): QSE_T('U');
if (digit_count == 0) TOK_ADD_CHAR (scm, rc);
else TOK_ADD_CHAR (scm, c_acc);
escaped = 0;
}
}
if (escaped == 0 && c == QSE_T('\"'))
{
/* terminating quote */
/*NEXT_CHAR_TO (scm, c);*/
READ_CHAR (scm);
break;
}
if (escaped == 0 && c == QSE_T('\\'))
{
escaped = 1;
continue;
}
if (escaped == 1)
{
if (c == QSE_T('n')) c = QSE_T('\n');
else if (c == QSE_T('r')) c = QSE_T('\r');
else if (c == QSE_T('t')) c = QSE_T('\t');
else if (c == QSE_T('f')) c = QSE_T('\f');
else if (c == QSE_T('b')) c = QSE_T('\b');
else if (c == QSE_T('v')) c = QSE_T('\v');
else if (c == QSE_T('a')) c = QSE_T('\a');
else if (c >= QSE_T('0') && c <= QSE_T('7'))
{
escaped = 3;
digit_count = 1;
c_acc = c - QSE_T('0');
continue;
}
else if (c == QSE_T('x'))
{
escaped = 2;
digit_count = 0;
c_acc = 0;
continue;
}
#ifdef QSE_CHAR_IS_WCHAR
else if (c == QSE_T('u') && QSE_SIZEOF(qse_char_t) >= 2)
{
escaped = 4;
digit_count = 0;
c_acc = 0;
continue;
}
else if (c == QSE_T('U') && QSE_SIZEOF(qse_char_t) >= 4)
{
escaped = 8;
digit_count = 0;
c_acc = 0;
continue;
}
#endif
escaped = 0;
}
TOK_ADD_CHAR (scm, c);
}
TOK_TYPE(scm) = TOK_STRING;
return 0;
}
enum read_number_token_flag_t
{
RNT_NEGATIVE = (1 << 0),
RNT_SKIP_TO_FRACTION = (1 << 1)
};
static int read_number_token (qse_scm_t* scm, int flags)
{
qse_long_t ival = 0;
qse_real_t rval = .0;
qse_real_t fraction;
if (flags & RNT_SKIP_TO_FRACTION) goto fraction_part;
do
{
ival = ival * 10 + (scm->r.curc - QSE_T('0'));
TOK_ADD_CHAR (scm, scm->r.curc);
READ_CHAR (scm);
}
while (IS_DIGIT(scm->r.curc));
/* TODO: extend parsing floating point number */
if (scm->r.curc == QSE_T('.'))
{
fraction_part:
fraction = 0.1;
TOK_ADD_CHAR (scm, scm->r.curc);
READ_CHAR (scm);
rval = (qse_real_t)ival;
while (IS_DIGIT(scm->r.curc))
{
rval += (qse_real_t)(scm->r.curc - QSE_T('0')) * fraction;
fraction *= 0.1;
TOK_ADD_CHAR (scm, scm->r.curc);
READ_CHAR (scm);
}
TOK_RVAL(scm) = rval;
TOK_TYPE(scm) = TOK_REAL;
if (flags & RNT_NEGATIVE) rval *= -1;
}
else
{
TOK_IVAL(scm) = ival;
TOK_TYPE(scm) = TOK_INT;
if (flags & RNT_NEGATIVE) ival *= -1;
}
return 0;
}
static int read_sharp_token (qse_scm_t* scm)
{
/* TODO: read a token beginning with #.*/
TOK_ADD_CHAR (scm, scm->r.curc); /* add # to the token name */
READ_CHAR (scm);
switch (scm->r.curc)
{
case QSE_T('t'):
TOK_ADD_CHAR (scm, scm->r.curc);
READ_CHAR (scm);
if (!IS_DELIM(scm->r.curc)) goto charname;
TOK_TYPE(scm) = TOK_T;
break;
case QSE_T('f'):
TOK_ADD_CHAR (scm, scm->r.curc);
READ_CHAR (scm);
if (!IS_DELIM(scm->r.curc)) goto charname;
TOK_TYPE(scm) = TOK_F;
break;
case QSE_T('\\'):
break;
case QSE_T('b'):
break;
case QSE_T('o'):
break;
case QSE_T('d'):
break;
case QSE_T('x'):
break;
}
return 0;
charname:
do
{
TOK_ADD_CHAR (scm, scm->r.curc);
READ_CHAR (scm);
}
while (!IS_DELIM(scm->r.curc));
/* TODO: character name comparison... */
qse_scm_seterror (scm, QSE_SCM_ESHARP, QSE_NULL, &scm->r.curloc);
return -1;
}
static int read_token (qse_scm_t* scm)
{
int flags = 0;
TOK_CLR (scm);
/* skip a series of white spaces and comment lines */
do
{
/* skip white spaces */
while (IS_WHSPACE(scm->r.curc)) READ_CHAR (scm);
if (scm->r.curc != QSE_T(';')) break;
/* skip a comment line */
do { READ_CHAR (scm); }
while (scm->r.curc != QSE_T('\n') &&
scm->r.curc != QSE_CHAR_EOF);
}
while (1);
TOK_LOC(scm) = scm->r.curloc;
if (scm->r.curc == QSE_CHAR_EOF)
{
TOK_TYPE(scm) = TOK_END;
return 0;
}
switch (scm->r.curc)
{
case QSE_T('('):
TOK_ADD_CHAR (scm, scm->r.curc);
TOK_TYPE(scm) = TOK_LPAREN;
READ_CHAR (scm);
return 0;
case QSE_T(')'):
TOK_ADD_CHAR (scm, scm->r.curc);
TOK_TYPE(scm) = TOK_RPAREN;
READ_CHAR (scm);
return 0;
case QSE_T('.'):
TOK_ADD_CHAR (scm, scm->r.curc);
READ_CHAR (scm);
if (!IS_DELIM(scm->r.curc))
{
flags |= RNT_SKIP_TO_FRACTION;
goto try_number;
}
TOK_TYPE(scm) = TOK_DOT;
return 0;
case QSE_T('\''):
TOK_ADD_CHAR (scm, scm->r.curc);
TOK_TYPE(scm) = TOK_QUOTE;
READ_CHAR (scm);
return 0;
case QSE_T('`'):
TOK_ADD_CHAR (scm, scm->r.curc);
TOK_TYPE(scm) = TOK_QQUOTE;
READ_CHAR (scm);
return 0;
case QSE_T(','):
TOK_ADD_CHAR (scm, scm->r.curc);
READ_CHAR (scm);
if (scm->r.curc == QSE_T('@'))
{
TOK_TYPE(scm) = TOK_COMMAAT;
READ_CHAR (scm);
}
else TOK_TYPE(scm) = TOK_COMMA;
return 0;
case QSE_T('#'):
return read_sharp_token (scm);
case QSE_T('\"'):
return read_string_token (scm);
}
if (scm->r.curc == QSE_T('+') || scm->r.curc == QSE_T('-'))
{
/* a number can begin with + or -. we don't know
* if it is the part of a number or not yet.
* let's set the NEGATIVE bit in 'flags' if the sign is
* negative for later use in case it is followed by a digit.
* we also add the sign character to the token name
* so that we can form a complete symbol if the word turns
* out to be a symbol eventually.
*/
if (scm->r.curc == QSE_T('-')) flags |= RNT_NEGATIVE;
TOK_ADD_CHAR (scm, scm->r.curc);
READ_CHAR (scm);
}
if (IS_DIGIT(scm->r.curc))
{
try_number:
/* we got a digit, maybe or maybe not following a sign.
* call read_number_token() to read the current token
* as a number. */
if (read_number_token (scm, flags) <= -1) return -1;
/* the read_number() function exits once it sees a character
* that can not compose a number. if it is a delimiter,
* the token is numeric. */
if (IS_DELIM(scm->r.curc)) return 0;
/* otherwise, we carry on reading trailing characters to
* compose a symbol token */
}
/* we got here as the current token does not begin with special
* token characters. treat it as a symbol token. */
do
{
TOK_ADD_CHAR (scm, scm->r.curc);
READ_CHAR (scm);
}
while (!IS_DELIM(scm->r.curc));
TOK_TYPE(scm) = TOK_SYMBOL;
return 0;
#if 0
TOK_TYPE(scm) = TOK_INVALID;
READ_CHAR (scm); /* consume */
return 0;
#endif
}
static QSE_INLINE qse_scm_ent_t* push (qse_scm_t* scm, qse_scm_ent_t* obj)
{
qse_scm_ent_t* pair;
pair = qse_scm_makepairent (scm, obj, scm->r.s);
if (pair == QSE_NULL) return QSE_NULL;
scm->r.s = pair;
/* return the top of the stack which is the containing pair */
return pair;
}
static QSE_INLINE_ALWAYS void pop (qse_scm_t* scm)
{
QSE_ASSERTX (
!IS_NIL(scm,scm->r.s),
"You've called pop() more times than push()"
);
scm->r.s = PAIR_CDR(scm->r.s);
}
static QSE_INLINE qse_scm_ent_t* enter_list (qse_scm_t* scm, int flagv)
{
/* upon entering a list, it pushes three cells into a stack.
*
* rstack -------+
* V
* +---cons--+
* +------ | -------+
* car| +---------+ |cdr
* V |
* nil#1 V
* +---cons--+
* +------ | --------+
* car| +---------+ |cdr
* v |
* nil#2 V
* +---cons--+
* +------ | --------+
* car| +---------+ |cdr
* V |
* flag number V
* previous stack top
*
* nil#1 to store the first element in the list.
* nil#2 to store the last element in the list.
* both to be updated in chain_to_list() as items are added.
*/
return (push (scm, TO_SMALLINT(scm,flagv)) == QSE_NULL ||
push (scm, scm->nil) == QSE_NULL ||
push (scm, scm->nil) == QSE_NULL)? QSE_NULL: scm->r.s;
}
static QSE_INLINE_ALWAYS qse_scm_ent_t* leave_list (qse_scm_t* scm, int* flagv)
{
qse_scm_ent_t* head;
/* the stack must not be empty */
QSE_ASSERTX (
!IS_NIL(scm,scm->r.s),
"You cannot leave a list without entering it"
);
/* remember the current list head */
head = PAIR_CAR(PAIR_CDR(scm->r.s));
/* upon leaving a list, it pops the three cells off the stack */
pop (scm);
pop (scm);
pop (scm);
if (IS_NIL(scm,scm->r.s))
{
/* the stack is empty after popping.
* it is back to the top level.
* the top level can never be quoted. */
*flagv = 0;
}
else
{
/* restore the flag for the outer returning level */
qse_scm_ent_t* flag = PAIR_CDR(PAIR_CDR(scm->r.s));
QSE_ASSERT (IS_SMALLINT(scm,PAIR_CAR(flag)));
*flagv = FROM_SMALLINT(scm,PAIR_CAR(flag));
}
/* return the head of the list being left */
return head;
}
static QSE_INLINE_ALWAYS void dot_list (qse_scm_t* scm)
{
qse_scm_ent_t* pair;
int flagv;
QSE_ASSERT (!IS_NIL(scm,scm->r.s));
/* mark the state that a dot has appeared in the list */
pair = PAIR_CDR(PAIR_CDR(scm->r.s));
flagv = FROM_SMALLINT(scm,PAIR_CAR(pair));
PAIR_CAR(pair) = TO_SMALLINT(scm,flagv|DOTTED);
}
static qse_scm_ent_t* chain_to_list (qse_scm_t* scm, qse_scm_ent_t* obj)
{
qse_scm_ent_t* cell, * head, * tail, *flag;
int flagv;
/* the stack top is the pair pointing to the list tail */
tail = scm->r.s;
QSE_ASSERT (!IS_NIL(scm,tail));
/* the pair pointing to the list head is below the tail cell
* connected via cdr. */
head = PAIR_CDR(tail);
QSE_ASSERT (!IS_NIL(scm,head));
/* the pair pointing to the flag is below the head cell
* connected via cdr */
flag = PAIR_CDR(head);
/* retrieve the numeric flag value */
QSE_ASSERT(IS_SMALLINT(scm,PAIR_CAR(flag)));
flagv = (int)FROM_SMALLINT(scm,PAIR_CAR(flag));
if (flagv & CLOSED)
{
/* the list has already been closed. cannot add more items. */
qse_scm_seterror (scm, QSE_SCM_ERPAREN, QSE_NULL, &TOK_LOC(scm));
return QSE_NULL;
}
else if (flagv & DOTTED)
{
/* the list must not be empty to have reached the dotted state */
QSE_ASSERT (!IS_NIL(scm,PAIR_CAR(tail)));
/* chain the object via 'cdr' of the tail cell */
PAIR_CDR(PAIR_CAR(tail)) = obj;
/* update the flag to CLOSED so that you can have more than
* one item after the dot. */
PAIR_CAR(flag) = TO_SMALLINT(scm,flagv|CLOSED);
}
else
{
cell = qse_scm_makepairent (scm, obj, scm->nil);
if (cell == QSE_NULL) return QSE_NULL;
if (PAIR_CAR(head) == scm->nil)
{
/* the list head is not set yet. it is the first
* element added to the list. let both head and tail
* point to the new cons cell */
QSE_ASSERT (PAIR_CAR(tail) == scm->nil);
PAIR_CAR(head) = cell;
PAIR_CAR(tail) = cell;
}
else
{
/* the new cons cell is not the first element.
* append it to the list */
PAIR_CDR(PAIR_CAR(tail)) = cell;
PAIR_CAR(tail) = cell;
}
}
return obj;
}
static QSE_INLINE_ALWAYS int is_list_empty (qse_scm_t* scm)
{
/* the stack must not be empty */
QSE_ASSERTX (
!IS_NIL(scm,scm->r.s),
"You can not call this function while the stack is empty"
);
/* if the tail pointer is pointing to nil, the list is empty */
return IS_NIL(scm,PAIR_CAR(scm->r.s));
}
static int read_entity (qse_scm_t* scm)
{
/* this function read an s-expression non-recursively
* by manipulating its own stack. */
int level = 0, flagv = 0;
qse_scm_ent_t* obj;
while (1)
{
redo:
switch (TOK_TYPE(scm))
{
default:
QSE_ASSERT (!"should never happen - invalid token type");
qse_scm_seterror (scm, QSE_SCM_EINTERN, QSE_NULL, QSE_NULL);
return -1;
#if 0
case TOK_INVALID:
qse_scm_seterror (
scm, QSE_SCM_ESYNTAX,
QSE_NULL, &TOK_LOC(scm));
return -1;
#endif
case TOK_END:
qse_scm_seterror (
scm, QSE_SCM_EEND,
QSE_NULL, &TOK_LOC(scm));
return -1;
case TOK_QUOTE:
if (level >= QSE_TYPE_MAX(int))
{
/* the nesting level has become too deep */
qse_scm_seterror (
scm, QSE_SCM_ELSTDEEP,
QSE_NULL, &TOK_LOC(scm));
return -1;
}
/* enter a quoted string */
flagv |= QUOTED;
if (enter_list (scm, flagv) == QSE_NULL) return -1;
level++;
/* force-chain the quote symbol to the new list entered */
if (chain_to_list (scm, scm->quote) == QSE_NULL) return -1;
/* read the next token */
READ_TOKEN (scm);
goto redo;
case TOK_LPAREN:
if (level >= QSE_TYPE_MAX(int))
{
/* the nesting level has become too deep */
qse_scm_seterror (
scm, QSE_SCM_ELSTDEEP,
QSE_NULL, &TOK_LOC(scm));
return -1;
}
/* enter a normal string */
flagv = 0;
if (enter_list (scm, flagv) == QSE_NULL) return -1;
level++;
/* read the next token */
READ_TOKEN (scm);
goto redo;
case TOK_DOT:
if (level <= 0 || is_list_empty (scm))
{
qse_scm_seterror (
scm, QSE_SCM_EDOT,
QSE_NULL, &TOK_LOC(scm));
return -1;
}
dot_list (scm);
READ_TOKEN (scm);
goto redo;
case TOK_RPAREN:
if ((flagv & QUOTED) || level <= 0)
{
/* the right parenthesis can never appear while
* 'quoted' is true. 'quoted' is set to false when
* entering a normal list. 'quoted' is set to true
* when entering a quoted list. a quoted list does
* not have an explicit right parenthesis.
* so the right parenthesis can only pair up with
* the left parenthesis for the normal list.
*
* For example, '(1 2 3 ') 5 6)
*
* this condition is triggerred when the first ) is
* met after the second quote.
*
* also it is illegal to have the right parenthesis
* with no opening(left) parenthesis, which is
* indicated by level<=0.
*/
qse_scm_seterror (
scm, QSE_SCM_ELPAREN,
QSE_NULL, &TOK_LOC(scm));
return -1;
}
obj = leave_list (scm, &flagv);
level--;
break;
case TOK_T:
obj = scm->t;
break;
case TOK_F:
obj = scm->f;
break;
case TOK_INT:
obj = qse_scm_makenument (scm, TOK_IVAL(scm));
break;
case TOK_REAL:
obj = qse_scm_makerealent (scm, TOK_RVAL(scm));
break;
case TOK_STRING:
obj = qse_scm_makestrent (
scm, TOK_NAME_PTR(scm), TOK_NAME_LEN(scm));
break;
case TOK_SYMBOL:
obj = qse_scm_makesyment (scm, TOK_NAME_PTR(scm));
break;
}
/* check if the element is read for a quoted list */
while (flagv & QUOTED)
{
QSE_ASSERT (level > 0);
/* if so, append the element read into the quote list */
if (chain_to_list (scm, obj) == QSE_NULL) return -1;
/* exit out of the quoted list. the quoted list can have
* one element only. */
obj = leave_list (scm, &flagv);
/* one level up toward the top */
level--;
}
/* check if we are at the top level */
if (level <= 0) break; /* yes */
/* if not, append the element read into the current list.
* if we are not at the top level, we must be in a list */
if (chain_to_list (scm, obj) == QSE_NULL) return -1;
/* read the next token */
READ_TOKEN (scm);
}
/* upon exit, we must be at the top level */
QSE_ASSERT (level == 0);
scm->r.e = obj;
return 0;
}
qse_scm_ent_t* qse_scm_read (qse_scm_t* scm)
{
QSE_ASSERTX (
scm->io.fns.in != QSE_NULL,
"Specify input function before calling qse_scm_read()"
);
if (read_char(scm) <= -1) return QSE_NULL;
if (read_token(scm) <= -1) return QSE_NULL;
#if 0
scm.r.state = READ_NORMAL;
do
{
if (func[scm.r.state] (scm) <= -1) return QSE_NULL;
}
while (scm.r.state != READ_DONE)
#endif
#if 0
do
{
qse_printf (QSE_T("TOKEN: [%s]\n"), TOK_NAME_PTR(scm));
if (read_token(scm) <= -1) return QSE_NULL;
}
while (TOK_TYPE(scm) != TOK_END);
#endif
if (read_entity (scm) <= -1) return QSE_NULL;
#if 0
{
int i;
for (i = 0; i < 100; i++)
{
qse_printf (QSE_T("%p\n"), alloc_entity(scm, QSE_NULL, QSE_NULL));
}
}
#endif
return scm->r.e;
}

View File

@ -210,6 +210,8 @@ static qse_scm_t* qse_scm_init (
scm->gloenv = scm->nil; scm->gloenv = scm->nil;
scm->r.s = scm->nil; scm->r.s = scm->nil;
scm->r.e = scm->nil; scm->r.e = scm->nil;
scm->p.s = scm->nil;
scm->p.e = scm->nil;
/* build the global environment entity as a pair */ /* build the global environment entity as a pair */
scm->gloenv = qse_scm_makepairent (scm, scm->nil, scm->nil); scm->gloenv = qse_scm_makepairent (scm, scm->nil, scm->nil);
@ -234,896 +236,3 @@ static void qse_scm_fini (qse_scm_t* scm)
} }
/*---------------------------------------------------------------------------
* READER
*---------------------------------------------------------------------------*/
enum list_flag_t
{
QUOTED = (1 << 0),
DOTTED = (1 << 1),
CLOSED = (1 << 2)
};
enum tok_type_t
{
TOK_END = 0,
TOK_T = 1,
TOK_F = 2,
TOK_INT = 3,
TOK_REAL = 4,
TOK_SYMBOL = 5,
TOK_STRING = 6,
TOK_LPAREN = 7,
TOK_RPAREN = 8,
TOK_DOT = 9,
TOK_QUOTE = 10,
TOK_QQUOTE = 11, /* quasiquote */
TOK_COMMA = 12,
TOK_COMMAAT = 13,
#if 0
TOK_INVALID = 50
#endif
};
#define TOK_CLR(scm) qse_str_clear(&(scm)->r.t.name)
#define TOK_TYPE(scm) (scm)->r.t.type
#define TOK_IVAL(scm) (scm)->r.t.ival
#define TOK_RVAL(scm) (scm)->r.t.rval
#define TOK_NAME(scm) (&(scm)->r.t.name)
#define TOK_NAME_PTR(scm) TOK_NAME(scm)->ptr
#define TOK_NAME_LEN(scm) TOK_NAME(scm)->len
#define TOK_LOC(scm) (scm)->r.t.loc
#define TOK_ADD_CHAR(scm,ch) QSE_BLOCK (\
if (qse_str_ccat(TOK_NAME(scm), ch) == -1) \
{ \
qse_scm_seterror (scm, QSE_SCM_ENOMEM, QSE_NULL, &scm->r.curloc); \
return -1; \
} \
)
#define IS_DIGIT(ch) ((ch) >= QSE_T('0') && (ch) <= QSE_T('9'))
#define IS_SPACE(ch) ((ch) == QSE_T(' ') || (ch) == QSE_T('\t'))
#define IS_NEWLINE(ch) ((ch) == QSE_T('\n') || (ch) == QSE_T('\r'))
#define IS_WHSPACE(ch) IS_SPACE(ch) || IS_NEWLINE(ch)
#define IS_DELIM(ch) \
(IS_WHSPACE(ch) || (ch) == QSE_T('(') || (ch) == QSE_T(')') || \
(ch) == QSE_T('\"') || (ch) == QSE_T(';') || (ch) == QSE_CHAR_EOF)
#define READ_CHAR(scm) QSE_BLOCK(if (read_char(scm) <= -1) return -1;)
#define READ_TOKEN(scm) QSE_BLOCK(if (read_token(scm) <= -1) return -1;)
static int read_char (qse_scm_t* scm)
{
qse_ssize_t n;
qse_char_t c;
/* TODO: do bufferring */
scm->err.num = QSE_SCM_ENOERR;
n = scm->io.fns.in (scm, QSE_SCM_IO_READ, &scm->io.arg.in, &c, 1);
if (n <= -1)
{
if (scm->err.num == QSE_SCM_ENOERR)
qse_scm_seterror (scm, QSE_SCM_EIO, QSE_NULL, QSE_NULL);
return -1;
}
/* TODO: handle the case when a new file is included or loaded ...
* stacking of curloc is needed??? see qseawk for reference
*/
if (n == 0) scm->r.curc = QSE_CHAR_EOF;
else
{
scm->r.curc = c;
if (c == QSE_T('\n'))
{
scm->r.curloc.colm = 0;
scm->r.curloc.line++;
}
else scm->r.curloc.colm++;
}
return 0;
}
static int read_string_token (qse_scm_t* scm)
{
qse_cint_t c;
int escaped = 0;
int digit_count = 0;
qse_cint_t c_acc = 0;
while (1)
{
READ_CHAR (scm);
c = scm->r.curc;
if (c == QSE_CHAR_EOF)
{
qse_scm_seterror (
scm, QSE_SCM_EENDSTR,
QSE_NULL, &scm->r.curloc);
return -1;
}
if (escaped == 3)
{
if (c >= QSE_T('0') && c <= QSE_T('7'))
{
c_acc = c_acc * 8 + c - QSE_T('0');
digit_count++;
if (digit_count >= escaped)
{
TOK_ADD_CHAR (scm, c_acc);
escaped = 0;
}
continue;
}
else
{
TOK_ADD_CHAR (scm, c_acc);
escaped = 0;
}
}
else if (escaped == 2 || escaped == 4 || escaped == 8)
{
if (c >= QSE_T('0') && c <= QSE_T('9'))
{
c_acc = c_acc * 16 + c - QSE_T('0');
digit_count++;
if (digit_count >= escaped)
{
TOK_ADD_CHAR (scm, c_acc);
escaped = 0;
}
continue;
}
else if (c >= QSE_T('A') && c <= QSE_T('F'))
{
c_acc = c_acc * 16 + c - QSE_T('A') + 10;
digit_count++;
if (digit_count >= escaped)
{
TOK_ADD_CHAR (scm, c_acc);
escaped = 0;
}
continue;
}
else if (c >= QSE_T('a') && c <= QSE_T('f'))
{
c_acc = c_acc * 16 + c - QSE_T('a') + 10;
digit_count++;
if (digit_count >= escaped)
{
TOK_ADD_CHAR (scm, c_acc);
escaped = 0;
}
continue;
}
else
{
qse_char_t rc;
rc = (escaped == 2)? QSE_T('x'):
(escaped == 4)? QSE_T('u'): QSE_T('U');
if (digit_count == 0) TOK_ADD_CHAR (scm, rc);
else TOK_ADD_CHAR (scm, c_acc);
escaped = 0;
}
}
if (escaped == 0 && c == QSE_T('\"'))
{
/* terminating quote */
/*NEXT_CHAR_TO (scm, c);*/
READ_CHAR (scm);
break;
}
if (escaped == 0 && c == QSE_T('\\'))
{
escaped = 1;
continue;
}
if (escaped == 1)
{
if (c == QSE_T('n')) c = QSE_T('\n');
else if (c == QSE_T('r')) c = QSE_T('\r');
else if (c == QSE_T('t')) c = QSE_T('\t');
else if (c == QSE_T('f')) c = QSE_T('\f');
else if (c == QSE_T('b')) c = QSE_T('\b');
else if (c == QSE_T('v')) c = QSE_T('\v');
else if (c == QSE_T('a')) c = QSE_T('\a');
else if (c >= QSE_T('0') && c <= QSE_T('7'))
{
escaped = 3;
digit_count = 1;
c_acc = c - QSE_T('0');
continue;
}
else if (c == QSE_T('x'))
{
escaped = 2;
digit_count = 0;
c_acc = 0;
continue;
}
#ifdef QSE_CHAR_IS_WCHAR
else if (c == QSE_T('u') && QSE_SIZEOF(qse_char_t) >= 2)
{
escaped = 4;
digit_count = 0;
c_acc = 0;
continue;
}
else if (c == QSE_T('U') && QSE_SIZEOF(qse_char_t) >= 4)
{
escaped = 8;
digit_count = 0;
c_acc = 0;
continue;
}
#endif
escaped = 0;
}
TOK_ADD_CHAR (scm, c);
}
TOK_TYPE(scm) = TOK_STRING;
return 0;
}
enum read_number_token_flag_t
{
RNT_NEGATIVE = (1 << 0),
RNT_SKIP_TO_FRACTION = (1 << 1)
};
static int read_number_token (qse_scm_t* scm, int flags)
{
qse_long_t ival = 0;
qse_real_t rval = .0;
qse_real_t fraction;
if (flags & RNT_SKIP_TO_FRACTION) goto fraction_part;
do
{
ival = ival * 10 + (scm->r.curc - QSE_T('0'));
TOK_ADD_CHAR (scm, scm->r.curc);
READ_CHAR (scm);
}
while (IS_DIGIT(scm->r.curc));
/* TODO: extend parsing floating point number */
if (scm->r.curc == QSE_T('.'))
{
fraction_part:
fraction = 0.1;
TOK_ADD_CHAR (scm, scm->r.curc);
READ_CHAR (scm);
rval = (qse_real_t)ival;
while (IS_DIGIT(scm->r.curc))
{
rval += (qse_real_t)(scm->r.curc - QSE_T('0')) * fraction;
fraction *= 0.1;
TOK_ADD_CHAR (scm, scm->r.curc);
READ_CHAR (scm);
}
TOK_RVAL(scm) = rval;
TOK_TYPE(scm) = TOK_REAL;
if (flags & RNT_NEGATIVE) rval *= -1;
}
else
{
TOK_IVAL(scm) = ival;
TOK_TYPE(scm) = TOK_INT;
if (flags & RNT_NEGATIVE) ival *= -1;
}
return 0;
}
static int read_sharp_token (qse_scm_t* scm)
{
/* TODO: read a token beginning with #.*/
TOK_ADD_CHAR (scm, scm->r.curc); /* add # to the token name */
READ_CHAR (scm);
switch (scm->r.curc)
{
case QSE_T('t'):
TOK_ADD_CHAR (scm, scm->r.curc);
READ_CHAR (scm);
if (!IS_DELIM(scm->r.curc)) goto charname;
TOK_TYPE(scm) = TOK_T;
break;
case QSE_T('f'):
TOK_ADD_CHAR (scm, scm->r.curc);
READ_CHAR (scm);
if (!IS_DELIM(scm->r.curc)) goto charname;
TOK_TYPE(scm) = TOK_F;
break;
case QSE_T('\\'):
break;
case QSE_T('b'):
break;
case QSE_T('o'):
break;
case QSE_T('d'):
break;
case QSE_T('x'):
break;
}
return 0;
charname:
do
{
TOK_ADD_CHAR (scm, scm->r.curc);
READ_CHAR (scm);
}
while (!IS_DELIM(scm->r.curc));
/* TODO: character name comparison... */
qse_scm_seterror (scm, QSE_SCM_ESHARP, QSE_NULL, &scm->r.curloc);
return -1;
}
static int read_token (qse_scm_t* scm)
{
int flags = 0;
TOK_CLR (scm);
/* skip a series of white spaces and comment lines */
do
{
/* skip white spaces */
while (IS_WHSPACE(scm->r.curc)) READ_CHAR (scm);
if (scm->r.curc != QSE_T(';')) break;
/* skip a comment line */
do { READ_CHAR (scm); }
while (scm->r.curc != QSE_T('\n') &&
scm->r.curc != QSE_CHAR_EOF);
}
while (1);
TOK_LOC(scm) = scm->r.curloc;
if (scm->r.curc == QSE_CHAR_EOF)
{
TOK_TYPE(scm) = TOK_END;
return 0;
}
switch (scm->r.curc)
{
case QSE_T('('):
TOK_ADD_CHAR (scm, scm->r.curc);
TOK_TYPE(scm) = TOK_LPAREN;
READ_CHAR (scm);
return 0;
case QSE_T(')'):
TOK_ADD_CHAR (scm, scm->r.curc);
TOK_TYPE(scm) = TOK_RPAREN;
READ_CHAR (scm);
return 0;
case QSE_T('.'):
TOK_ADD_CHAR (scm, scm->r.curc);
READ_CHAR (scm);
if (!IS_DELIM(scm->r.curc))
{
flags |= RNT_SKIP_TO_FRACTION;
goto try_number;
}
TOK_TYPE(scm) = TOK_DOT;
return 0;
case QSE_T('\''):
TOK_ADD_CHAR (scm, scm->r.curc);
TOK_TYPE(scm) = TOK_QUOTE;
READ_CHAR (scm);
return 0;
case QSE_T('`'):
TOK_ADD_CHAR (scm, scm->r.curc);
TOK_TYPE(scm) = TOK_QQUOTE;
READ_CHAR (scm);
return 0;
case QSE_T(','):
TOK_ADD_CHAR (scm, scm->r.curc);
READ_CHAR (scm);
if (scm->r.curc == QSE_T('@'))
{
TOK_TYPE(scm) = TOK_COMMAAT;
READ_CHAR (scm);
}
else TOK_TYPE(scm) = TOK_COMMA;
return 0;
case QSE_T('#'):
return read_sharp_token (scm);
case QSE_T('\"'):
return read_string_token (scm);
}
if (scm->r.curc == QSE_T('+') || scm->r.curc == QSE_T('-'))
{
/* a number can begin with + or -. we don't know
* if it is the part of a number or not yet.
* let's set the NEGATIVE bit in 'flags' if the sign is
* negative for later use in case it is followed by a digit.
* we also add the sign character to the token name
* so that we can form a complete symbol if the word turns
* out to be a symbol eventually.
*/
if (scm->r.curc == QSE_T('-')) flags |= RNT_NEGATIVE;
TOK_ADD_CHAR (scm, scm->r.curc);
READ_CHAR (scm);
}
if (IS_DIGIT(scm->r.curc))
{
try_number:
/* we got a digit, maybe or maybe not following a sign.
* call read_number_token() to read the current token
* as a number. */
if (read_number_token (scm, flags) <= -1) return -1;
/* the read_number() function exits once it sees a character
* that can not compose a number. if it is a delimiter,
* the token is numeric. */
if (IS_DELIM(scm->r.curc)) return 0;
/* otherwise, we carry on reading trailing characters to
* compose a symbol token */
}
/* we got here as the current token does not begin with special
* token characters. treat it as a symbol token. */
do
{
TOK_ADD_CHAR (scm, scm->r.curc);
READ_CHAR (scm);
}
while (!IS_DELIM(scm->r.curc));
TOK_TYPE(scm) = TOK_SYMBOL;
return 0;
#if 0
TOK_TYPE(scm) = TOK_INVALID;
READ_CHAR (scm); /* consume */
return 0;
#endif
}
static QSE_INLINE qse_scm_ent_t* push (qse_scm_t* scm, qse_scm_ent_t* obj)
{
qse_scm_ent_t* pair;
pair = qse_scm_makepairent (scm, obj, scm->r.s);
if (pair == QSE_NULL) return QSE_NULL;
scm->r.s = pair;
/* return the top of the stack which is the containing pair */
return pair;
}
static QSE_INLINE_ALWAYS void pop (qse_scm_t* scm)
{
QSE_ASSERTX (
!IS_NIL(scm,scm->r.s),
"You've called pop() more times than push()"
);
scm->r.s = PAIR_CDR(scm->r.s);
}
static QSE_INLINE qse_scm_ent_t* enter_list (qse_scm_t* scm, int flagv)
{
/* upon entering a list, it pushes three cells into a stack.
*
* rstack -------+
* V
* +---cons--+
* +------ | -------+
* car| +---------+ |cdr
* V |
* nil#1 V
* +---cons--+
* +------ | --------+
* car| +---------+ |cdr
* v |
* nil#2 V
* +---cons--+
* +------ | --------+
* car| +---------+ |cdr
* V |
* flag number V
* previous stack top
*
* nil#1 to store the first element in the list.
* nil#2 to store the last element in the list.
* both to be updated in chain_to_list() as items are added.
*/
return (push (scm, TO_SMALLINT(scm,flagv)) == QSE_NULL ||
push (scm, scm->nil) == QSE_NULL ||
push (scm, scm->nil) == QSE_NULL)? QSE_NULL: scm->r.s;
}
static QSE_INLINE_ALWAYS qse_scm_ent_t* leave_list (qse_scm_t* scm, int* flagv)
{
qse_scm_ent_t* head;
/* the stack must not be empty */
QSE_ASSERTX (
!IS_NIL(scm,scm->r.s),
"You cannot leave a list without entering it"
);
/* remember the current list head */
head = PAIR_CAR(PAIR_CDR(scm->r.s));
/* upon leaving a list, it pops the three cells off the stack */
pop (scm);
pop (scm);
pop (scm);
if (IS_NIL(scm,scm->r.s))
{
/* the stack is empty after popping.
* it is back to the top level.
* the top level can never be quoted. */
*flagv = 0;
}
else
{
/* restore the flag for the outer returning level */
qse_scm_ent_t* flag = PAIR_CDR(PAIR_CDR(scm->r.s));
QSE_ASSERT (TYPE(PAIR_CAR(flag)) == QSE_SCM_ENT_NUM);
*flagv = NUM_VALUE(PAIR_CAR(flag));
}
/* return the head of the list being left */
return head;
}
static QSE_INLINE_ALWAYS void dot_list (qse_scm_t* scm)
{
qse_scm_ent_t* pair;
int flagv;
QSE_ASSERT (!IS_NIL(scm,scm->r.s));
/* mark the state that a dot has appeared in the list */
pair = PAIR_CDR(PAIR_CDR(scm->r.s));
flagv = FROM_SMALLINT(scm,PAIR_CAR(pair));
PAIR_CAR(pair) = TO_SMALLINT(scm,flagv|DOTTED);
}
static qse_scm_ent_t* chain_to_list (qse_scm_t* scm, qse_scm_ent_t* obj)
{
qse_scm_ent_t* cell, * head, * tail, *flag;
int flagv;
/* the stack top is the pair pointing to the list tail */
tail = scm->r.s;
QSE_ASSERT (!IS_NIL(scm,tail));
/* the pair pointing to the list head is below the tail cell
* connected via cdr. */
head = PAIR_CDR(tail);
QSE_ASSERT (!IS_NIL(scm,head));
/* the pair pointing to the flag is below the head cell
* connected via cdr */
flag = PAIR_CDR(head);
/* retrieve the numeric flag value */
QSE_ASSERT(IS_SMALLINT(scm,PAIR_CAR(flag)));
flagv = (int)FROM_SMALLINT(scm,PAIR_CAR(flag));
if (flagv & CLOSED)
{
/* the list has already been closed. cannot add more items. */
qse_scm_seterror (scm, QSE_SCM_ERPAREN, QSE_NULL, &TOK_LOC(scm));
return QSE_NULL;
}
else if (flagv & DOTTED)
{
/* the list must not be empty to have reached the dotted state */
QSE_ASSERT (!IS_NIL(scm,PAIR_CAR(tail)));
/* chain the object via 'cdr' of the tail cell */
PAIR_CDR(PAIR_CAR(tail)) = obj;
/* update the flag to CLOSED so that you can have more than
* one item after the dot. */
PAIR_CAR(flag) = TO_SMALLINT(scm,flagv|CLOSED);
}
else
{
cell = qse_scm_makepairent (scm, obj, scm->nil);
if (cell == QSE_NULL) return QSE_NULL;
if (PAIR_CAR(head) == scm->nil)
{
/* the list head is not set yet. it is the first
* element added to the list. let both head and tail
* point to the new cons cell */
QSE_ASSERT (PAIR_CAR(tail) == scm->nil);
PAIR_CAR(head) = cell;
PAIR_CAR(tail) = cell;
}
else
{
/* the new cons cell is not the first element.
* append it to the list */
PAIR_CDR(PAIR_CAR(tail)) = cell;
PAIR_CAR(tail) = cell;
}
}
return obj;
}
static QSE_INLINE_ALWAYS int is_list_empty (qse_scm_t* scm)
{
/* the stack must not be empty */
QSE_ASSERTX (
!IS_NIL(scm,scm->r.s),
"You can not call this function while the stack is empty"
);
/* if the tail pointer is pointing to nil, the list is empty */
return IS_NIL(scm,PAIR_CAR(scm->r.s));
}
static int read_entity (qse_scm_t* scm)
{
/* this function read an s-expression non-recursively
* by manipulating its own stack. */
int level = 0, flagv = 0;
qse_scm_ent_t* obj;
while (1)
{
redo:
switch (TOK_TYPE(scm))
{
default:
QSE_ASSERT (!"should never happen - invalid token type");
qse_scm_seterror (scm, QSE_SCM_EINTERN, QSE_NULL, QSE_NULL);
return -1;
#if 0
case TOK_INVALID:
qse_scm_seterror (
scm, QSE_SCM_ESYNTAX,
QSE_NULL, &TOK_LOC(scm));
return -1;
#endif
case TOK_END:
qse_scm_seterror (
scm, QSE_SCM_EEND,
QSE_NULL, &TOK_LOC(scm));
return -1;
case TOK_QUOTE:
if (level >= QSE_TYPE_MAX(int))
{
/* the nesting level has become too deep */
qse_scm_seterror (
scm, QSE_SCM_ELSTDEEP,
QSE_NULL, &TOK_LOC(scm));
return -1;
}
/* enter a quoted string */
flagv |= QUOTED;
if (enter_list (scm, flagv) == QSE_NULL) return -1;
level++;
/* force-chain the quote symbol to the new list entered */
if (chain_to_list (scm, scm->quote) == QSE_NULL) return -1;
/* read the next token */
READ_TOKEN (scm);
goto redo;
case TOK_LPAREN:
if (level >= QSE_TYPE_MAX(int))
{
/* the nesting level has become too deep */
qse_scm_seterror (
scm, QSE_SCM_ELSTDEEP,
QSE_NULL, &TOK_LOC(scm));
return -1;
}
/* enter a normal string */
flagv = 0;
if (enter_list (scm, flagv) == QSE_NULL) return -1;
level++;
/* read the next token */
READ_TOKEN (scm);
goto redo;
case TOK_DOT:
if (level <= 0 || is_list_empty (scm))
{
qse_scm_seterror (
scm, QSE_SCM_EDOT,
QSE_NULL, &TOK_LOC(scm));
return -1;
}
dot_list (scm);
READ_TOKEN (scm);
goto redo;
case TOK_RPAREN:
if ((flagv & QUOTED) || level <= 0)
{
/* the right parenthesis can never appear while
* 'quoted' is true. 'quoted' is set to false when
* entering a normal list. 'quoted' is set to true
* when entering a quoted list. a quoted list does
* not have an explicit right parenthesis.
* so the right parenthesis can only pair up with
* the left parenthesis for the normal list.
*
* For example, '(1 2 3 ') 5 6)
*
* this condition is triggerred when the first ) is
* met after the second quote.
*
* also it is illegal to have the right parenthesis
* with no opening(left) parenthesis, which is
* indicated by level<=0.
*/
qse_scm_seterror (
scm, QSE_SCM_ELPAREN,
QSE_NULL, &TOK_LOC(scm));
return -1;
}
obj = leave_list (scm, &flagv);
level--;
break;
case TOK_T:
obj = scm->t;
break;
case TOK_F:
obj = scm->f;
break;
case TOK_INT:
obj = qse_scm_makenument (scm, TOK_IVAL(scm));
break;
case TOK_REAL:
obj = qse_scm_makerealent (scm, TOK_RVAL(scm));
break;
case TOK_STRING:
obj = qse_scm_makestrent (
scm, TOK_NAME_PTR(scm), TOK_NAME_LEN(scm));
break;
case TOK_SYMBOL:
obj = qse_scm_makesyment (scm, TOK_NAME_PTR(scm));
break;
}
/* check if the element is read for a quoted list */
while (flagv & QUOTED)
{
QSE_ASSERT (level > 0);
/* if so, append the element read into the quote list */
if (chain_to_list (scm, obj) == QSE_NULL) return -1;
/* exit out of the quoted list. the quoted list can have
* one element only. */
obj = leave_list (scm, &flagv);
/* one level up toward the top */
level--;
}
/* check if we are at the top level */
if (level <= 0) break; /* yes */
/* if not, append the element read into the current list.
* if we are not at the top level, we must be in a list */
if (chain_to_list (scm, obj) == QSE_NULL) return -1;
/* read the next token */
READ_TOKEN (scm);
}
/* upon exit, we must be at the top level */
QSE_ASSERT (level == 0);
scm->r.e = obj;
return 0;
}
qse_scm_ent_t* qse_scm_read (qse_scm_t* scm)
{
QSE_ASSERTX (
scm->io.fns.in != QSE_NULL,
"Specify input function before calling qse_scm_read()"
);
if (read_char(scm) <= -1) return QSE_NULL;
if (read_token(scm) <= -1) return QSE_NULL;
#if 0
scm.r.state = READ_NORMAL;
do
{
if (func[scm.r.state] (scm) <= -1) return QSE_NULL;
}
while (scm.r.state != READ_DONE)
#endif
#if 0
do
{
qse_printf (QSE_T("TOKEN: [%s]\n"), TOK_NAME_PTR(scm));
if (read_token(scm) <= -1) return QSE_NULL;
}
while (TOK_TYPE(scm) != TOK_END);
#endif
if (read_entity (scm) <= -1) return QSE_NULL;
#if 0
{
int i;
for (i = 0; i < 100; i++)
{
qse_printf (QSE_T("%p\n"), alloc_entity(scm, QSE_NULL, QSE_NULL));
}
}
#endif
return scm->r.e;
}

View File

@ -180,6 +180,12 @@ struct qse_scm_t
qse_scm_ent_t* e; /* last entity read */ qse_scm_ent_t* e; /* last entity read */
} r; } r;
struct
{
qse_scm_ent_t* s; /* stack for printing */
qse_scm_ent_t* e; /* top entity being printed */
} p;
/* common values */ /* common values */
qse_scm_ent_t* nil; qse_scm_ent_t* nil;
qse_scm_ent_t* t; qse_scm_ent_t* t;