mirror/nasm
2
0
Fork 0
mirror of https://github.com/netwide-assembler/nasm.git synced 2024-12-27 09:29:06 +08:00
Code Issues Packages Projects Releases Wiki Activity
19f15a50f6
nasm/stdscan.c

229 lines
7.1 KiB
C
Raw Normal View History

Portability fixes Concentrate compiler dependencies to compiler.h; make sure compiler.h is included first in every .c file (since some prototypes may depend on the presence of feature request macros.) Actually use the conditional inclusion of various functions (totally broken in previous releases.)
2007-10-03 12:53:51 +08:00
#include "compiler.h"
Finishing touches on perfect hash tokenizer; actually turn the thing on Finish the perfect hash tokenizer, and actually enable it. Move stdscan() et al to a separate file, since it's not needed in any of the clients of nasmlib other than nasm itself. Run make alldeps.
2007-08-31 06:35:34 +08:00
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <ctype.h>
#include <inttypes.h>
#include "nasm.h"
#include "nasmlib.h"
qstring: backquoted strings seem to work now... Hopefully backquoted strings should work correctly now.
2008-06-02 12:34:49 +08:00
#include "quote.h"
Finishing touches on perfect hash tokenizer; actually turn the thing on Finish the perfect hash tokenizer, and actually enable it. Move stdscan() et al to a separate file, since it's not needed in any of the clients of nasmlib other than nasm itself. Run make alldeps.
2007-08-31 06:35:34 +08:00
#include "stdscan.h"
#include "insns.h"
/*
* Standard scanner routine used by parser.c and some output
* formats. It keeps a succession of temporary-storage strings in
* stdscan_tempstorage, which can be cleared using stdscan_reset.
*/
static char **stdscan_tempstorage = NULL;
static int stdscan_tempsize = 0, stdscan_templen = 0;
#define STDSCAN_TEMP_DELTA 256
static void stdscan_pop(void)
{
nasm_free(stdscan_tempstorage[--stdscan_templen]);
}
void stdscan_reset(void)
{
while (stdscan_templen > 0)
stdscan_pop();
}
/*
* Unimportant cleanup is done to avoid confusing people who are trying
* to debug real memory leaks
*/
void stdscan_cleanup(void)
{
stdscan_reset();
nasm_free(stdscan_tempstorage);
}
static char *stdscan_copy(char *p, int len)
{
char *text;
text = nasm_malloc(len + 1);
Fix memory management issues with expanded %include Ownership of the filename string was a bit fuzzy, with the result that we were freeing it even though it was retained for use by __FILE__. Clean up a number of other memory management issues with the new quoting code, and change the stdscan implementation to one pass over the string.
2008-06-05 02:26:59 +08:00
memcpy(text, p, len);
Finishing touches on perfect hash tokenizer; actually turn the thing on Finish the perfect hash tokenizer, and actually enable it. Move stdscan() et al to a separate file, since it's not needed in any of the clients of nasmlib other than nasm itself. Run make alldeps.
2007-08-31 06:35:34 +08:00
text[len] = '\0';
if (stdscan_templen >= stdscan_tempsize) {
stdscan_tempsize += STDSCAN_TEMP_DELTA;
stdscan_tempstorage = nasm_realloc(stdscan_tempstorage,
stdscan_tempsize *
sizeof(char *));
}
stdscan_tempstorage[stdscan_templen++] = text;
return text;
}
char *stdscan_bufptr = NULL;
int stdscan(void *private_data, struct tokenval *tv)
{
char ourcopy[MAX_KEYWORD + 1], *r, *s;
(void)private_data; /* Don't warn that this parameter is unused */
ctype.h: wrapper ctype functions with a cast to (unsigned char) ctype functions take an *int*, which the user is expected to have taken the input character from getc() and friends, or taken a character and cast it to (unsigned char). We don't care about EOF (-1), so use macros that cast to (unsigned char) for us.
2008-06-22 01:23:17 +08:00
while (nasm_isspace(*stdscan_bufptr))
Finishing touches on perfect hash tokenizer; actually turn the thing on Finish the perfect hash tokenizer, and actually enable it. Move stdscan() et al to a separate file, since it's not needed in any of the clients of nasmlib other than nasm itself. Run make alldeps.
2007-08-31 06:35:34 +08:00
stdscan_bufptr++;
if (!*stdscan_bufptr)
return tv->t_type = 0;
/* we have a token; either an id, a number or a char */
if (isidstart(*stdscan_bufptr) ||
(*stdscan_bufptr == '$' && isidstart(stdscan_bufptr[1]))) {
/* now we've got an identifier */
Use the compiler-provided booleans if available, otherwise emulate Both C and C++ have "bool", "true" and "false" in lower case; C requires <stdbool.h> for this, in C++ it is an inherent type built into the compiler. Use those instead of the old macros; emulate with a simple typedef enum if unavailable.
2007-10-11 05:58:45 +08:00
bool is_sym = false;
Finishing touches on perfect hash tokenizer; actually turn the thing on Finish the perfect hash tokenizer, and actually enable it. Move stdscan() et al to a separate file, since it's not needed in any of the clients of nasmlib other than nasm itself. Run make alldeps.
2007-08-31 06:35:34 +08:00
if (*stdscan_bufptr == '$') {
Use the compiler-provided booleans if available, otherwise emulate Both C and C++ have "bool", "true" and "false" in lower case; C requires <stdbool.h> for this, in C++ it is an inherent type built into the compiler. Use those instead of the old macros; emulate with a simple typedef enum if unavailable.
2007-10-11 05:58:45 +08:00
is_sym = true;
Finishing touches on perfect hash tokenizer; actually turn the thing on Finish the perfect hash tokenizer, and actually enable it. Move stdscan() et al to a separate file, since it's not needed in any of the clients of nasmlib other than nasm itself. Run make alldeps.
2007-08-31 06:35:34 +08:00
stdscan_bufptr++;
}
r = stdscan_bufptr++;
/* read the entire buffer to advance the buffer pointer but... */
while (isidchar(*stdscan_bufptr))
stdscan_bufptr++;
/* ... copy only up to IDLEN_MAX-1 characters */
tv->t_charptr = stdscan_copy(r, stdscan_bufptr - r < IDLEN_MAX ?
stdscan_bufptr - r : IDLEN_MAX - 1);
if (is_sym || stdscan_bufptr - r > MAX_KEYWORD)
return tv->t_type = TOKEN_ID; /* bypass all other checks */
for (s = tv->t_charptr, r = ourcopy; *s; s++)
Use an explicit table for tolower() to avoid a function call On some platforms, tolower() is implemented as a function call, in order to handle locale support. We never change locales, so can the result of tolower() into a table, so we don't have to sit through the function call every time. ~1.3% overall performance improvement on a macro-heavy benchmark under Linux x86-64.
2008-06-12 06:49:41 +08:00
*r++ = nasm_tolower(*s);
Finishing touches on perfect hash tokenizer; actually turn the thing on Finish the perfect hash tokenizer, and actually enable it. Move stdscan() et al to a separate file, since it's not needed in any of the clients of nasmlib other than nasm itself. Run make alldeps.
2007-08-31 06:35:34 +08:00
*r = '\0';
/* right, so we have an identifier sitting in temp storage. now,
* is it actually a register or instruction name, or what? */
Slightly optimize the interface to nasm_token_hash() Instead of returning -1 from nasm_token_hash, set tv->t_type to TOKEN_ID and return TOKEN_ID, since that's what stdscan.c wants to do with it anyway. This allows us to simply tailcall nasm_token_hash().
2007-09-19 13:54:40 +08:00
return nasm_token_hash(ourcopy, tv);
Finishing touches on perfect hash tokenizer; actually turn the thing on Finish the perfect hash tokenizer, and actually enable it. Move stdscan() et al to a separate file, since it's not needed in any of the clients of nasmlib other than nasm itself. Run make alldeps.
2007-08-31 06:35:34 +08:00
} else if (*stdscan_bufptr == '$' && !isnumchar(stdscan_bufptr[1])) {
/*
* It's a $ sign with no following hex number; this must
* mean it's a Here token ($), evaluating to the current
* assembly location, or a Base token ($$), evaluating to
* the base of the current segment.
*/
stdscan_bufptr++;
if (*stdscan_bufptr == '$') {
stdscan_bufptr++;
return tv->t_type = TOKEN_BASE;
}
return tv->t_type = TOKEN_HERE;
} else if (isnumstart(*stdscan_bufptr)) { /* now we've got a number */
Additional uses of bool and enum Proper use of bool and enum makes code easier to debug. Do more of it. In particular, we really should stomp out any residual uses of magic constants that aren't enums or, in some cases, even #defines.
2007-10-11 15:05:31 +08:00
bool rn_error;
Allow underscores in numbers; better detection of FP - Allow underscores as group separators in numbers, for example: 0x1234_5678 is now a legal number. The underscore is just ignored, it adds no meaning. - Recognize dotless floating-point numbers, such as "1e30". This entails distinguishing hexadecimal numbers in the scanner, since e.g. 0x1e30 is a perfectly legitimate hex constant.
2007-10-20 04:10:46 +08:00
bool is_hex = false;
bool is_float = false;
Don't confuse suffixed hexadecimal with floating-point 1e30 is a floating-point constant, but 1e30h is not. The scanner won't know that until it sees the "h", so make sure we keep enough state to be able to distinguish "1e30" (a possible hex constant) from "1.e30", "1e+30" or "1.0" (unabiguously floating-point.)
2007-10-20 05:10:35 +08:00
bool has_e = false;
Allow underscores in numbers; better detection of FP - Allow underscores as group separators in numbers, for example: 0x1234_5678 is now a legal number. The underscore is just ignored, it adds no meaning. - Recognize dotless floating-point numbers, such as "1e30". This entails distinguishing hexadecimal numbers in the scanner, since e.g. 0x1e30 is a perfectly legitimate hex constant.
2007-10-20 04:10:46 +08:00
char c;
Finishing touches on perfect hash tokenizer; actually turn the thing on Finish the perfect hash tokenizer, and actually enable it. Move stdscan() et al to a separate file, since it's not needed in any of the clients of nasmlib other than nasm itself. Run make alldeps.
2007-08-31 06:35:34 +08:00
More consistent handling of radix letters Allow any radix letter from the set [bydtoqhx] to be used either "Intel-style" (0...x) or "C-style" (0x...). In Intel style, the leading 0 remains optional as long as the first digit is in the range 0-9. As a consequence, allow the prefix "0h" for hexadecimal floating point.
2007-10-23 07:53:48 +08:00
r = stdscan_bufptr;
Finishing touches on perfect hash tokenizer; actually turn the thing on Finish the perfect hash tokenizer, and actually enable it. Move stdscan() et al to a separate file, since it's not needed in any of the clients of nasmlib other than nasm itself. Run make alldeps.
2007-08-31 06:35:34 +08:00
More consistent handling of radix letters Allow any radix letter from the set [bydtoqhx] to be used either "Intel-style" (0...x) or "C-style" (0x...). In Intel style, the leading 0 remains optional as long as the first digit is in the range 0-9. As a consequence, allow the prefix "0h" for hexadecimal floating point.
2007-10-23 07:53:48 +08:00
if (*stdscan_bufptr == '$') {
stdscan_bufptr++;
Allow underscores in numbers; better detection of FP - Allow underscores as group separators in numbers, for example: 0x1234_5678 is now a legal number. The underscore is just ignored, it adds no meaning. - Recognize dotless floating-point numbers, such as "1e30". This entails distinguishing hexadecimal numbers in the scanner, since e.g. 0x1e30 is a perfectly legitimate hex constant.
2007-10-20 04:10:46 +08:00
is_hex = true;
More consistent handling of radix letters Allow any radix letter from the set [bydtoqhx] to be used either "Intel-style" (0...x) or "C-style" (0x...). In Intel style, the leading 0 remains optional as long as the first digit is in the range 0-9. As a consequence, allow the prefix "0h" for hexadecimal floating point.
2007-10-23 07:53:48 +08:00
}
Allow underscores in numbers; better detection of FP - Allow underscores as group separators in numbers, for example: 0x1234_5678 is now a legal number. The underscore is just ignored, it adds no meaning. - Recognize dotless floating-point numbers, such as "1e30". This entails distinguishing hexadecimal numbers in the scanner, since e.g. 0x1e30 is a perfectly legitimate hex constant.
2007-10-20 04:10:46 +08:00
for (;;) {
c = *stdscan_bufptr++;
Don't confuse suffixed hexadecimal with floating-point 1e30 is a floating-point constant, but 1e30h is not. The scanner won't know that until it sees the "h", so make sure we keep enough state to be able to distinguish "1e30" (a possible hex constant) from "1.e30", "1e+30" or "1.0" (unabiguously floating-point.)
2007-10-20 05:10:35 +08:00
if (!is_hex && (c == 'e' || c == 'E')) {
has_e = true;
if (*stdscan_bufptr == '+' || *stdscan_bufptr == '-') {
/* e can only be followed by +/- if it is either a
prefixed hex number or a floating-point number */
is_float = true;
stdscan_bufptr++;
}
More consistent handling of radix letters Allow any radix letter from the set [bydtoqhx] to be used either "Intel-style" (0...x) or "C-style" (0x...). In Intel style, the leading 0 remains optional as long as the first digit is in the range 0-9. As a consequence, allow the prefix "0h" for hexadecimal floating point.
2007-10-23 07:53:48 +08:00
} else if (c == 'H' || c == 'h' || c == 'X' || c == 'x') {
is_hex = true;
Support binary and octal floating-point For consistency, support binary and octal floating-point, and accept a "0d" or "0t" prefix for decimal floating-point. However, we do not accept a binary exponent (p) for a decimal mantissa, or vice versa.
2007-10-23 08:34:10 +08:00
} else if (c == 'P' || c == 'p') {
Allow underscores in numbers; better detection of FP - Allow underscores as group separators in numbers, for example: 0x1234_5678 is now a legal number. The underscore is just ignored, it adds no meaning. - Recognize dotless floating-point numbers, such as "1e30". This entails distinguishing hexadecimal numbers in the scanner, since e.g. 0x1e30 is a perfectly legitimate hex constant.
2007-10-20 04:10:46 +08:00
is_float = true;
if (*stdscan_bufptr == '+' || *stdscan_bufptr == '-')
stdscan_bufptr++;
} else if (isnumchar(c) || c == '_')
; /* just advance */
else if (c == '.')
is_float = true;
else
break;
}
stdscan_bufptr--; /* Point to first character beyond number */
More consistent handling of radix letters Allow any radix letter from the set [bydtoqhx] to be used either "Intel-style" (0...x) or "C-style" (0x...). In Intel style, the leading 0 remains optional as long as the first digit is in the range 0-9. As a consequence, allow the prefix "0h" for hexadecimal floating point.
2007-10-23 07:53:48 +08:00
if (has_e && !is_hex) {
Don't confuse suffixed hexadecimal with floating-point 1e30 is a floating-point constant, but 1e30h is not. The scanner won't know that until it sees the "h", so make sure we keep enough state to be able to distinguish "1e30" (a possible hex constant) from "1.e30", "1e+30" or "1.0" (unabiguously floating-point.)
2007-10-20 05:10:35 +08:00
/* 1e13 is floating-point, but 1e13h is not */
is_float = true;
}
Allow underscores in numbers; better detection of FP - Allow underscores as group separators in numbers, for example: 0x1234_5678 is now a legal number. The underscore is just ignored, it adds no meaning. - Recognize dotless floating-point numbers, such as "1e30". This entails distinguishing hexadecimal numbers in the scanner, since e.g. 0x1e30 is a perfectly legitimate hex constant.
2007-10-20 04:10:46 +08:00
if (is_float) {
tv->t_charptr = stdscan_copy(r, stdscan_bufptr - r);
return tv->t_type = TOKEN_FLOAT;
} else {
r = stdscan_copy(r, stdscan_bufptr - r);
tv->t_integer = readnum(r, &rn_error);
stdscan_pop();
qstring: backquoted strings seem to work now... Hopefully backquoted strings should work correctly now.
2008-06-02 12:34:49 +08:00
if (rn_error) {
/* some malformation occurred */
return tv->t_type = TOKEN_ERRNUM;
}
Allow underscores in numbers; better detection of FP - Allow underscores as group separators in numbers, for example: 0x1234_5678 is now a legal number. The underscore is just ignored, it adds no meaning. - Recognize dotless floating-point numbers, such as "1e30". This entails distinguishing hexadecimal numbers in the scanner, since e.g. 0x1e30 is a perfectly legitimate hex constant.
2007-10-20 04:10:46 +08:00
tv->t_charptr = NULL;
return tv->t_type = TOKEN_NUM;
}
qstring: backquoted strings seem to work now... Hopefully backquoted strings should work correctly now.
2008-06-02 12:34:49 +08:00
} else if (*stdscan_bufptr == '\'' || *stdscan_bufptr == '"' ||
*stdscan_bufptr == '`') {
Fix memory management issues with expanded %include Ownership of the filename string was a bit fuzzy, with the result that we were freeing it even though it was retained for use by __FILE__. Clean up a number of other memory management issues with the new quoting code, and change the stdscan implementation to one pass over the string.
2008-06-05 02:26:59 +08:00
/* a quoted string */
char start_quote = *stdscan_bufptr;
tv->t_charptr = stdscan_bufptr;
tv->t_inttwo = nasm_unquote(tv->t_charptr, &stdscan_bufptr);
if (*stdscan_bufptr != start_quote)
Make strings a first-class token type; defer evaluation Make strings a proper, first-class token type, instead of relying on the "TOKEN_NUM with tv_charptr" hack. Only convert a string to a number if requested in an expression context; this also makes it possible to actually issue a warning when it overflows.
2008-06-10 11:45:19 +08:00
return tv->t_type = TOKEN_ERRSTR;
Fix memory management issues with expanded %include Ownership of the filename string was a bit fuzzy, with the result that we were freeing it even though it was retained for use by __FILE__. Clean up a number of other memory management issues with the new quoting code, and change the stdscan implementation to one pass over the string.
2008-06-05 02:26:59 +08:00
stdscan_bufptr++; /* Skip final quote */
Make strings a first-class token type; defer evaluation Make strings a proper, first-class token type, instead of relying on the "TOKEN_NUM with tv_charptr" hack. Only convert a string to a number if requested in an expression context; this also makes it possible to actually issue a warning when it overflows.
2008-06-10 11:45:19 +08:00
return tv->t_type = TOKEN_STR;
qstring: backquoted strings seem to work now... Hopefully backquoted strings should work correctly now.
2008-06-02 12:34:49 +08:00
} else if (*stdscan_bufptr == ';') {
/* a comment has happened - stay */
Finishing touches on perfect hash tokenizer; actually turn the thing on Finish the perfect hash tokenizer, and actually enable it. Move stdscan() et al to a separate file, since it's not needed in any of the clients of nasmlib other than nasm itself. Run make alldeps.
2007-08-31 06:35:34 +08:00
return tv->t_type = 0;
} else if (stdscan_bufptr[0] == '>' && stdscan_bufptr[1] == '>') {
stdscan_bufptr += 2;
return tv->t_type = TOKEN_SHR;
} else if (stdscan_bufptr[0] == '<' && stdscan_bufptr[1] == '<') {
stdscan_bufptr += 2;
return tv->t_type = TOKEN_SHL;
} else if (stdscan_bufptr[0] == '/' && stdscan_bufptr[1] == '/') {
stdscan_bufptr += 2;
return tv->t_type = TOKEN_SDIV;
} else if (stdscan_bufptr[0] == '%' && stdscan_bufptr[1] == '%') {
stdscan_bufptr += 2;
return tv->t_type = TOKEN_SMOD;
} else if (stdscan_bufptr[0] == '=' && stdscan_bufptr[1] == '=') {
stdscan_bufptr += 2;
return tv->t_type = TOKEN_EQ;
} else if (stdscan_bufptr[0] == '<' && stdscan_bufptr[1] == '>') {
stdscan_bufptr += 2;
return tv->t_type = TOKEN_NE;
} else if (stdscan_bufptr[0] == '!' && stdscan_bufptr[1] == '=') {
stdscan_bufptr += 2;
return tv->t_type = TOKEN_NE;
} else if (stdscan_bufptr[0] == '<' && stdscan_bufptr[1] == '=') {
stdscan_bufptr += 2;
return tv->t_type = TOKEN_LE;
} else if (stdscan_bufptr[0] == '>' && stdscan_bufptr[1] == '=') {
stdscan_bufptr += 2;
return tv->t_type = TOKEN_GE;
} else if (stdscan_bufptr[0] == '&' && stdscan_bufptr[1] == '&') {
stdscan_bufptr += 2;
return tv->t_type = TOKEN_DBL_AND;
} else if (stdscan_bufptr[0] == '^' && stdscan_bufptr[1] == '^') {
stdscan_bufptr += 2;
return tv->t_type = TOKEN_DBL_XOR;
} else if (stdscan_bufptr[0] == '|' && stdscan_bufptr[1] == '|') {
stdscan_bufptr += 2;
return tv->t_type = TOKEN_DBL_OR;
} else /* just an ordinary char */
return tv->t_type = (uint8_t)(*stdscan_bufptr++);
}
Reference in New Issue Copy Permalink