Browse Source

lexer: Convert type (u)int to mp_(u)int_t.

pull/744/head
Damien George 10 years ago
parent
commit
54eb4e723e
  1. 55
      py/lexer.c
  2. 10
      py/lexer.h
  3. 4
      py/lexerstr.c
  4. 2
      py/parsehelper.c
  5. 2
      teensy/lexermemzip.c

55
py/lexer.c

@ -50,25 +50,25 @@ struct _mp_lexer_t {
unichar chr0, chr1, chr2; // current cached characters from source
uint line; // source line
uint column; // source column
mp_uint_t line; // source line
mp_uint_t column; // source column
int emit_dent; // non-zero when there are INDENT/DEDENT tokens to emit
int nested_bracket_level; // >0 when there are nested brackets over multiple lines
mp_int_t emit_dent; // non-zero when there are INDENT/DEDENT tokens to emit
mp_int_t nested_bracket_level; // >0 when there are nested brackets over multiple lines
uint alloc_indent_level;
uint num_indent_level;
mp_uint_t alloc_indent_level;
mp_uint_t num_indent_level;
uint16_t *indent_level;
vstr_t vstr;
mp_token_t tok_cur;
};
uint mp_optimise_value;
mp_uint_t mp_optimise_value;
// TODO replace with a call to a standard function
bool str_strn_equal(const char *str, const char *strn, int len) {
uint i = 0;
bool str_strn_equal(const char *str, const char *strn, mp_uint_t len) {
mp_uint_t i = 0;
while (i < len && *str == *strn) {
++i;
@ -81,7 +81,7 @@ bool str_strn_equal(const char *str, const char *strn, int len) {
#ifdef MICROPY_DEBUG_PRINTERS
void mp_token_show(const mp_token_t *tok) {
printf("(%d:%d) kind:%d str:%p len:%d", tok->src_line, tok->src_column, tok->kind, tok->str, tok->len);
printf("(" UINT_FMT ":" UINT_FMT ") kind:%u str:%p len:" UINT_FMT, tok->src_line, tok->src_column, tok->kind, tok->str, tok->len);
if (tok->str != NULL && tok->len > 0) {
const byte *i = (const byte *)tok->str;
const byte *j = (const byte *)i + tok->len;
@ -175,7 +175,7 @@ STATIC void next_char(mp_lexer_t *lex) {
return;
}
int advance = 1;
mp_uint_t advance = 1;
if (lex->chr0 == '\n') {
// LF is a new line
@ -210,7 +210,7 @@ STATIC void next_char(mp_lexer_t *lex) {
}
}
void indent_push(mp_lexer_t *lex, uint indent) {
void indent_push(mp_lexer_t *lex, mp_uint_t indent) {
if (lex->num_indent_level >= lex->alloc_indent_level) {
// TODO use m_renew_maybe and somehow indicate an error if it fails... probably by using MP_TOKEN_MEMORY_ERROR
lex->indent_level = m_renew(uint16_t, lex->indent_level, lex->alloc_indent_level, lex->alloc_indent_level + MICROPY_ALLOC_LEXEL_INDENT_INC);
@ -219,7 +219,7 @@ void indent_push(mp_lexer_t *lex, uint indent) {
lex->indent_level[lex->num_indent_level++] = indent;
}
uint indent_top(mp_lexer_t *lex) {
mp_uint_t indent_top(mp_lexer_t *lex) {
return lex->indent_level[lex->num_indent_level - 1];
}
@ -308,9 +308,9 @@ STATIC const char *tok_kw[] = {
"__debug__",
};
STATIC int hex_digit(unichar c) {
STATIC mp_uint_t hex_digit(unichar c) {
// c is assumed to be hex digit
int n = c - '0';
mp_uint_t n = c - '0';
if (n > 9) {
n &= ~('a' - 'A');
n -= ('A' - ('9' + 1));
@ -320,8 +320,9 @@ STATIC int hex_digit(unichar c) {
// This is called with CUR_CHAR() before first hex digit, and should return with
// it pointing to last hex digit
STATIC bool get_hex(mp_lexer_t *lex, int num_digits, uint *result) {
uint num = 0;
// num_digits must be greater than zero
STATIC bool get_hex(mp_lexer_t *lex, mp_uint_t num_digits, mp_uint_t *result) {
mp_uint_t num = 0;
while (num_digits-- != 0) {
next_char(lex);
unichar c = CUR_CHAR(lex);
@ -394,7 +395,7 @@ STATIC void mp_lexer_next_token_into(mp_lexer_t *lex, mp_token_t *tok, bool firs
} else if (had_physical_newline && lex->nested_bracket_level == 0) {
tok->kind = MP_TOKEN_NEWLINE;
uint num_spaces = lex->column - 1;
mp_uint_t num_spaces = lex->column - 1;
lex->emit_dent = 0;
if (num_spaces == indent_top(lex)) {
} else if (num_spaces > indent_top(lex)) {
@ -463,7 +464,7 @@ STATIC void mp_lexer_next_token_into(mp_lexer_t *lex, mp_token_t *tok, bool firs
next_char(lex);
// work out if it's a single or triple quoted literal
int num_quotes;
mp_uint_t num_quotes;
if (is_char_and(lex, quote_char, quote_char)) {
// triple quotes
next_char(lex);
@ -475,7 +476,7 @@ STATIC void mp_lexer_next_token_into(mp_lexer_t *lex, mp_token_t *tok, bool firs
}
// parse the literal
int n_closing = 0;
mp_uint_t n_closing = 0;
while (!is_end(lex) && (num_quotes > 1 || !is_char(lex, '\n')) && n_closing < num_quotes) {
if (is_char(lex, quote_char)) {
n_closing += 1;
@ -512,7 +513,7 @@ STATIC void mp_lexer_next_token_into(mp_lexer_t *lex, mp_token_t *tok, bool firs
// Otherwise fall through.
case 'x':
{
uint num = 0;
mp_uint_t num = 0;
if (!get_hex(lex, (c == 'x' ? 2 : c == 'u' ? 4 : 8), &num)) {
// TODO error message
assert(0);
@ -531,8 +532,8 @@ STATIC void mp_lexer_next_token_into(mp_lexer_t *lex, mp_token_t *tok, bool firs
default:
if (c >= '0' && c <= '7') {
// Octal sequence, 1-3 chars
int digits = 3;
int num = c - '0';
mp_uint_t digits = 3;
mp_uint_t num = c - '0';
while (is_following_odigit(lex) && --digits != 0) {
next_char(lex);
num = num * 8 + (CUR_CHAR(lex) - '0');
@ -627,7 +628,7 @@ STATIC void mp_lexer_next_token_into(mp_lexer_t *lex, mp_token_t *tok, bool firs
// search for encoded delimiter or operator
const char *t = tok_enc;
uint tok_enc_index = 0;
mp_uint_t tok_enc_index = 0;
for (; *t != 0 && !is_char(lex, *t); t += 1) {
if (*t == 'e' || *t == 'c') {
t += 1;
@ -649,7 +650,7 @@ STATIC void mp_lexer_next_token_into(mp_lexer_t *lex, mp_token_t *tok, bool firs
// get the maximum characters for a valid token
t += 1;
uint t_index = tok_enc_index;
mp_uint_t t_index = tok_enc_index;
for (;;) {
for (; *t == 'e'; t += 1) {
t += 1;
@ -712,8 +713,8 @@ STATIC void mp_lexer_next_token_into(mp_lexer_t *lex, mp_token_t *tok, bool firs
// the parser gives a syntax error on, eg, x.__debug__. Otherwise, we
// need to check for this special token in many places in the compiler.
// TODO improve speed of these string comparisons
//for (int i = 0; tok_kw[i] != NULL; i++) {
for (int i = 0; i < MP_ARRAY_SIZE(tok_kw); i++) {
//for (mp_int_t i = 0; tok_kw[i] != NULL; i++) {
for (mp_int_t i = 0; i < MP_ARRAY_SIZE(tok_kw); i++) {
if (str_strn_equal(tok_kw[i], tok->str, tok->len)) {
if (i == MP_ARRAY_SIZE(tok_kw) - 1) {
// tok_kw[MP_ARRAY_SIZE(tok_kw) - 1] == "__debug__"

10
py/lexer.h

@ -131,12 +131,12 @@ typedef enum _mp_token_kind_t {
} mp_token_kind_t;
typedef struct _mp_token_t {
uint src_line; // source line
uint src_column; // source column
mp_uint_t src_line; // source line
mp_uint_t src_column; // source column
mp_token_kind_t kind; // kind of token
const char *str; // string of token (valid only while this token is current token)
uint len; // (byte) length of string of token
mp_uint_t len; // (byte) length of string of token
} mp_token_t;
// the next-char function must return the next character in the stream
@ -151,7 +151,7 @@ typedef struct _mp_lexer_t mp_lexer_t;
void mp_token_show(const mp_token_t *tok);
mp_lexer_t *mp_lexer_new(qstr src_name, void *stream_data, mp_lexer_stream_next_char_t stream_next_char, mp_lexer_stream_close_t stream_close);
mp_lexer_t *mp_lexer_new_from_str_len(qstr src_name, const char *str, uint len, uint free_len);
mp_lexer_t *mp_lexer_new_from_str_len(qstr src_name, const char *str, mp_uint_t len, mp_uint_t free_len);
void mp_lexer_free(mp_lexer_t *lex);
qstr mp_lexer_source_name(mp_lexer_t *lex);
@ -177,4 +177,4 @@ typedef enum {
mp_import_stat_t mp_import_stat(const char *path);
mp_lexer_t *mp_lexer_new_from_file(const char *filename);
extern uint mp_optimise_value;
extern mp_uint_t mp_optimise_value;

4
py/lexerstr.c

@ -30,7 +30,7 @@
#include "lexer.h"
typedef struct _mp_lexer_str_buf_t {
uint free_len; // if > 0, src_beg will be freed when done by: m_free(src_beg, free_len)
mp_uint_t free_len; // if > 0, src_beg will be freed when done by: m_free(src_beg, free_len)
const char *src_beg; // beginning of source
const char *src_cur; // current location in source
const char *src_end; // end (exclusive) of source
@ -51,7 +51,7 @@ STATIC void str_buf_free(mp_lexer_str_buf_t *sb) {
m_del_obj(mp_lexer_str_buf_t, sb);
}
mp_lexer_t *mp_lexer_new_from_str_len(qstr src_name, const char *str, uint len, uint free_len) {
mp_lexer_t *mp_lexer_new_from_str_len(qstr src_name, const char *str, mp_uint_t len, mp_uint_t free_len) {
mp_lexer_str_buf_t *sb = m_new_obj(mp_lexer_str_buf_t);
sb->free_len = free_len;
sb->src_beg = str;

2
py/parsehelper.c

@ -43,7 +43,7 @@
#define STR_INVALID_SYNTAX "invalid syntax"
void mp_parse_show_exception(mp_lexer_t *lex, mp_parse_error_kind_t parse_error_kind) {
printf(" File \"%s\", line %d, column %d\n", qstr_str(mp_lexer_source_name(lex)), mp_lexer_cur(lex)->src_line, mp_lexer_cur(lex)->src_column);
printf(" File \"%s\", line " UINT_FMT ", column " UINT_FMT "\n", qstr_str(mp_lexer_source_name(lex)), mp_lexer_cur(lex)->src_line, mp_lexer_cur(lex)->src_column);
switch (parse_error_kind) {
case MP_PARSE_ERROR_MEMORY:
printf("MemoryError: %s\n", STR_MEMORY);

2
teensy/lexermemzip.c

@ -16,6 +16,6 @@ mp_lexer_t *mp_lexer_new_from_file(const char *filename)
return NULL;
}
return mp_lexer_new_from_str_len(qstr_from_str(filename), (const char *)data, (uint)len, 0);
return mp_lexer_new_from_str_len(qstr_from_str(filename), (const char *)data, (mp_uint_t)len, 0);
}

Loading…
Cancel
Save