1berry/ruby
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Use lex_eol_p family

Browse Source
This commit is contained in:
Nobuyoshi Nakada 2023-06-30 16:12:56 +09:00
parent 6500f85927
commit 26aef1c736
Notes: git 2023-07-26 02:39:49 +00:00 
Merged: https://github.com/ruby/ruby/pull/8119
1 changed files with 20 additions and 18 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

38
parse.y
View File

@ -841,8 +841,10 @@ static int parser_yyerror0(struct parser_params*, const char*);
#define yyerror(yylloc, p, msg) parser_yyerror(p, yylloc, msg) #define yyerror(yylloc, p, msg) parser_yyerror(p, yylloc, msg)
#define token_flush(ptr) ((ptr)->lex.ptok = (ptr)->lex.pcur) #define token_flush(ptr) ((ptr)->lex.ptok = (ptr)->lex.pcur)
#define lex_goto_eol(p) ((p)->lex.pcur = (p)->lex.pend) #define lex_goto_eol(p) ((p)->lex.pcur = (p)->lex.pend)
#define lex_eol_p(p) ((p)->lex.pcur >= (p)->lex.pend) #define lex_eol_p(p) lex_eol_n_p(p, 0)
#define lex_eol_n_p(p,n) ((p)->lex.pcur+(n) >= (p)->lex.pend) #define lex_eol_n_p(p,n) lex_eol_ptr_n_p(p, (p)->lex.pcur, n)
#define lex_eol_ptr_p(p,ptr) lex_eol_ptr_n_p(p,ptr,0)
#define lex_eol_ptr_n_p(p,ptr,n) ((ptr)+(n) >= (p)->lex.pend)
static void token_info_setup(token_info *ptinfo, const char *ptr, const rb_code_location_t *loc); static void token_info_setup(token_info *ptinfo, const char *ptr, const rb_code_location_t *loc);
static void token_info_push(struct parser_params*, const char *token, const rb_code_location_t *loc); static void token_info_push(struct parser_params*, const char *token, const rb_code_location_t *loc);
@ -7142,7 +7144,7 @@ nextline(struct parser_params *p, int set_encoding)
if (p->eofp) if (p->eofp)
return -1; return -1;
if (p->lex.pend > p->lex.pbeg && *(p->lex.pend-1) != '\n') { if (!lex_eol_ptr_p(p, p->lex.pbeg) && *(p->lex.pend-1) != '\n') {
goto end_of_input; goto end_of_input;
} }
@ -7190,7 +7192,7 @@ nextc0(struct parser_params *p, int set_encoding)
{ {
int c; int c;
if (UNLIKELY((p->lex.pcur == p->lex.pend) || p->eofp || RTEST(p->lex.nextline))) { if (UNLIKELY(lex_eol_p(p) || p->eofp || RTEST(p->lex.nextline))) {
if (nextline(p, set_encoding)) return -1; if (nextline(p, set_encoding)) return -1;
} }
c = (unsigned char)*p->lex.pcur++; c = (unsigned char)*p->lex.pcur++;
@ -7223,7 +7225,7 @@ static int
looking_at_eol_p(struct parser_params *p) looking_at_eol_p(struct parser_params *p)
{ {
const char *ptr = p->lex.pcur; const char *ptr = p->lex.pcur;
while (ptr < p->lex.pend) { while (!lex_eol_ptr_p(p, ptr)) {
int c = (unsigned char)*ptr++; int c = (unsigned char)*ptr++;
int eol = (c == '\n' || c == '#'); int eol = (c == '\n' || c == '#');
if (eol || !ISSPACE(c)) { if (eol || !ISSPACE(c)) {
@ -7411,8 +7413,8 @@ tokadd_utf8(struct parser_params *p, rb_encoding **encp,
else { else {
const char *second = NULL; const char *second = NULL;
int c, last = nextc(p); int c, last = nextc(p);
if (p->lex.pcur >= p->lex.pend) goto unterminated; if (lex_eol_p(p)) goto unterminated;
while (ISSPACE(c = *p->lex.pcur) && ++p->lex.pcur < p->lex.pend); while (ISSPACE(c = peekc(p)) && !lex_eol_ptr_p(p, ++p->lex.pcur));
while (c != close_brace) { while (c != close_brace) {
if (c == term) goto unterminated; if (c == term) goto unterminated;
if (second == multiple_codepoints) if (second == multiple_codepoints)
@ -7421,8 +7423,8 @@ tokadd_utf8(struct parser_params *p, rb_encoding **encp,
if (!tokadd_codepoint(p, encp, regexp_literal, TRUE)) { if (!tokadd_codepoint(p, encp, regexp_literal, TRUE)) {
break; break;
} }
while (ISSPACE(c = *p->lex.pcur)) { while (ISSPACE(c = peekc(p))) {
if (++p->lex.pcur >= p->lex.pend) goto unterminated; if (lex_eol_ptr_p(p, ++p->lex.pcur)) goto unterminated;
last = c; last = c;
} }
if (term == -1 && !second) if (term == -1 && !second)
@ -7781,7 +7783,7 @@ tokadd_string(struct parser_params *p,
} }
--*nest; --*nest;
} }
else if ((func & STR_FUNC_EXPAND) && c == '#' && p->lex.pcur < p->lex.pend) { else if ((func & STR_FUNC_EXPAND) && c == '#' && !lex_eol_p(p)) {
int c2 = *p->lex.pcur; int c2 = *p->lex.pcur;
if (c2 == '$' || c2 == '@' || c2 == '{') { if (c2 == '$' || c2 == '@' || c2 == '{') {
pushback(p, c); pushback(p, c);
@ -7974,12 +7976,12 @@ parser_peek_variable_name(struct parser_params *p)
int c; int c;
const char *ptr = p->lex.pcur; const char *ptr = p->lex.pcur;
if (ptr + 1 >= p->lex.pend) return 0; if (lex_eol_ptr_n_p(p, ptr, 1)) return 0;
c = *ptr++; c = *ptr++;
switch (c) { switch (c) {
case '$': case '$':
if ((c = *ptr) == '-') { if ((c = *ptr) == '-') {
if (++ptr >= p->lex.pend) return 0; if (lex_eol_ptr_p(p, ++ptr)) return 0;
c = *ptr; c = *ptr;
} }
else if (is_global_name_punct(c) || ISDIGIT(c)) { else if (is_global_name_punct(c) || ISDIGIT(c)) {
@ -7988,7 +7990,7 @@ parser_peek_variable_name(struct parser_params *p)
break; break;
case '@': case '@':
if ((c = *ptr) == '@') { if ((c = *ptr) == '@') {
if (++ptr >= p->lex.pend) return 0; if (lex_eol_ptr_p(p, ++ptr)) return 0;
c = *ptr; c = *ptr;
} }
break; break;
@ -8333,7 +8335,7 @@ static int
word_match_p(struct parser_params *p, const char *word, long len) word_match_p(struct parser_params *p, const char *word, long len)
{ {
if (strncmp(p->lex.pcur, word, len)) return 0; if (strncmp(p->lex.pcur, word, len)) return 0;
if (p->lex.pcur + len == p->lex.pend) return 1; if (lex_eol_n_p(p, len)) return 1;
int c = (unsigned char)p->lex.pcur[len]; int c = (unsigned char)p->lex.pcur[len];
if (ISSPACE(c)) return 1; if (ISSPACE(c)) return 1;
switch (c) { switch (c) {
@ -8528,7 +8530,7 @@ here_document(struct parser_params *p, rb_strterm_heredoc_t *here)
rb_str_cat(str, ptr, ptr_end - ptr); rb_str_cat(str, ptr, ptr_end - ptr);
else else
str = STR_NEW(ptr, ptr_end - ptr); str = STR_NEW(ptr, ptr_end - ptr);
if (ptr_end < p->lex.pend) rb_str_cat(str, "\n", 1); if (!lex_eol_ptr_p(p, ptr_end)) rb_str_cat(str, "\n", 1);
lex_goto_eol(p); lex_goto_eol(p);
if (p->heredoc_indent > 0) { if (p->heredoc_indent > 0) {
goto flush_str; goto flush_str;
@ -9043,7 +9045,7 @@ parser_prepare(struct parser_params *p)
if (peek(p, '!')) p->has_shebang = 1; if (peek(p, '!')) p->has_shebang = 1;
break; break;
case 0xef: /* UTF-8 BOM marker */ case 0xef: /* UTF-8 BOM marker */
if (p->lex.pend - p->lex.pcur >= 2 && if (!lex_eol_n_p(p, 2) &&
(unsigned char)p->lex.pcur[0] == 0xbb && (unsigned char)p->lex.pcur[0] == 0xbb &&
(unsigned char)p->lex.pcur[1] == 0xbf) { (unsigned char)p->lex.pcur[1] == 0xbf) {
p->enc = rb_utf8_encoding(); p->enc = rb_utf8_encoding();
@ -9368,14 +9370,14 @@ parse_qmark(struct parser_params *p, int space_seen)
if (tokadd_mbchar(p, c) == -1) return 0; if (tokadd_mbchar(p, c) == -1) return 0;
} }
else if ((rb_enc_isalnum(c, p->enc) || c == '_') && else if ((rb_enc_isalnum(c, p->enc) || c == '_') &&
p->lex.pcur < p->lex.pend && is_identchar(p, p->lex.pcur, p->lex.pend, p->enc)) { !lex_eol_p(p) && is_identchar(p, p->lex.pcur, p->lex.pend, p->enc)) {
if (space_seen) { if (space_seen) {
const char *start = p->lex.pcur - 1, *ptr = start; const char *start = p->lex.pcur - 1, *ptr = start;
do { do {
int n = parser_precise_mbclen(p, ptr); int n = parser_precise_mbclen(p, ptr);
if (n < 0) return -1; if (n < 0) return -1;
ptr += n; ptr += n;
} while (ptr < p->lex.pend && is_identchar(p, ptr, p->lex.pend, p->enc)); } while (!lex_eol_ptr_p(p, ptr) && is_identchar(p, ptr, p->lex.pend, p->enc));
rb_warn2("`?' just followed by `%.*s' is interpreted as" \ rb_warn2("`?' just followed by `%.*s' is interpreted as" \
" a conditional operator, put a space after `?'", " a conditional operator, put a space after `?'",
WARN_I((int)(ptr - start)), WARN_S_L(start, (ptr - start))); WARN_I((int)(ptr - start)), WARN_S_L(start, (ptr - start)));