diff options
Diffstat (limited to 'reader.c')
| -rw-r--r-- | reader.c | 263 |
1 files changed, 212 insertions, 51 deletions
diff --git a/reader.c b/reader.c index efcbaa53bdfbe..3bbf9a79c1873 100644 --- a/reader.c +++ b/reader.c @@ -1,11 +1,11 @@ -/* $Id: reader.c,v 1.68 2017/02/02 01:05:36 tom Exp $ */ +/* $Id: reader.c,v 1.79 2020/03/30 23:54:13 tom Exp $ */ #include "defs.h" /* The line size must be a positive integer. One hundred was chosen */ /* because few lines in Yacc input grammars exceed 100 characters. */ /* Note that if a line exceeds LINESIZE characters, the line buffer */ -/* will be expanded to accomodate it. */ +/* will be expanded to accommodate it. */ #define LINESIZE 100 @@ -62,6 +62,13 @@ char line_format[] = "#line %d \"%s\"\n"; param *lex_param; param *parse_param; +static const char *code_keys[] = +{ + "", "requires", "provides", "top", "imports", +}; + +struct code_lines code_lines[CODE_MAX]; + #if defined(YYBTYACC) int destructor = 0; /* =1 if at least one %destructor */ @@ -144,7 +151,7 @@ line_directive(void) switch (ld) { case ldSPC1: - if (isspace(ch)) + if (isspace(UCH(ch))) { break; } @@ -153,7 +160,7 @@ line_directive(void) ld = ldSPC2; break; case ldSPC2: - if (isspace(ch)) + if (isspace(UCH(ch))) { break; } @@ -171,26 +178,26 @@ line_directive(void) ld = ldSPC3; break; case ldSPC3: - if (isspace(ch)) + if (isspace(UCH(ch))) { break; } else - UNLESS(!isdigit(ch)); + UNLESS(!isdigit(UCH(ch))); line_1st = n; ld = ldNUM; /* FALLTHRU */ case ldNUM: - if (isdigit(ch)) + if (isdigit(UCH(ch))) { break; } else - UNLESS(!isspace(ch)); + UNLESS(!isspace(UCH(ch))); ld = ldSPC4; break; case ldSPC4: - if (isspace(ch)) + if (isspace(UCH(ch))) { break; } @@ -217,19 +224,33 @@ line_directive(void) if (ld == ldOK) { size_t need = (size_t) (name_end - name_1st); - if (need > input_file_name_len) + if ((long)need > (long)input_file_name_len) { - input_file_name_len = need; - input_file_name = TREALLOC(char, input_file_name, need + 1); + input_file_name_len = ((need + 1) * 3) / 2; + input_file_name = TREALLOC(char, input_file_name, input_file_name_len); NO_SPACE(input_file_name); } - memcpy(input_file_name, line + name_1st + 1, need - 1); - input_file_name[need - 1] = '\0'; + if ((long)need > 0) + { + memcpy(input_file_name, line + name_1st + 1, need - 1); + input_file_name[need - 1] = '\0'; + } + else + { + input_file_name[0] = '\0'; + } } if (ld >= ldNUM && ld < ldERR) { - lineno = (int)strtol(line + line_1st, NULL, 10) - 1; + if (line_1st >= 0) + { + lineno = (int)strtol(line + line_1st, NULL, 10) - 1; + } + else + { + lineno = 0; + } } return (ld == ldOK); @@ -423,11 +444,12 @@ nextc(void) /* *INDENT-OFF* */ static struct keyword { - char name[14]; + char name[16]; int token; } keywords[] = { { "binary", NONASSOC }, + { "code", XCODE }, { "debug", XXXDEBUG }, #if defined(YYBTYACC) { "destructor", DESTRUCTOR }, @@ -435,7 +457,7 @@ keywords[] = { { "error-verbose",ERROR_VERBOSE }, { "expect", EXPECT }, { "expect-rr", EXPECT_RR }, - { "ident", IDENT }, + { "ident", IDENT }, #if defined(YYBTYACC) { "initial-action", INITIAL_ACTION }, #endif @@ -447,11 +469,11 @@ keywords[] = { { "nonassoc", NONASSOC }, { "parse-param", PARSE_PARAM }, { "pure-parser", PURE_PARSER }, - { "right", RIGHT }, + { "right", RIGHT }, { "start", START }, { "term", TOKEN }, { "token", TOKEN }, - { "token-table", TOKEN_TABLE }, + { "token-table", TOKEN_TABLE }, { "type", TYPE }, { "union", UNION }, { "yacc", POSIX_YACC }, @@ -474,18 +496,18 @@ keyword(void) struct keyword *key; c = *++cptr; - if (isalpha(c)) + if (isalpha(UCH(c))) { cinc = 0; for (;;) { - if (isalpha(c)) + if (isalpha(UCH(c))) { - if (isupper(c)) + if (isupper(UCH(c))) c = tolower(c); cachec(c); } - else if (isdigit(c) + else if (isdigit(UCH(c)) || c == '-' || c == '.' || c == '$') @@ -646,6 +668,127 @@ copy_comment(void) return msdone(temp); } +static int +check_key(int pos) +{ + const char *key = code_keys[pos]; + while (*cptr && *key) + if (*key++ != *cptr++) + return 0; + if (*key || (!isspace(UCH(*cptr)) && *cptr != L_CURL)) + return 0; + cptr--; + return 1; +} + +static void +copy_code(void) +{ + int c; + int curl; + int cline; + int on_line = 0; + int pos = CODE_HEADER; + struct mstring *code_mstr; + + /* read %code <keyword> { */ + for (;;) + { + c = *++cptr; + if (c == EOF) + unexpected_EOF(); + if (isspace(UCH(c))) + continue; + + if (c == L_CURL) + break; + + if (pos == CODE_HEADER) + { + switch (UCH(c)) + { + case 'r': + pos = CODE_REQUIRES; + break; + case 'p': + pos = CODE_PROVIDES; + break; + case 't': + pos = CODE_TOP; + break; + case 'i': + pos = CODE_IMPORTS; + break; + default: + break; + } + + if (pos == -1 || !check_key(pos)) + { + syntax_error(lineno, line, cptr); + return; + } + } + } + + cptr++; /* skip initial curl */ + while (*cptr && isspace(UCH(*cptr))) /* skip space */ + cptr++; + curl = 1; /* nesting count */ + + /* gather text */ + code_lines[pos].name = code_keys[pos]; + if ((cline = (int)code_lines[pos].num) != 0) + { + code_mstr = msrenew(code_lines[pos].lines); + } + else + { + code_mstr = msnew(); + } + cline++; + msprintf(code_mstr, line_format, lineno, input_file_name); + for (;;) + { + c = *cptr++; + switch (c) + { + case '\0': + get_line(); + if (line == NULL) + { + unexpected_EOF(); + return; + } + continue; + case '\n': + cline++; + on_line = 0; + break; + case L_CURL: + curl++; + break; + case R_CURL: + if (--curl == 0) + { + if (on_line > 1) + { + mputc(code_mstr, '\n'); + cline++; + } + code_lines[pos].lines = msdone(code_mstr); + code_lines[pos].num = (size_t) cline; + return; + } + break; + default: + break; + } + mputc(code_mstr, c); + on_line++; + } +} + static void copy_text(void) { @@ -1235,7 +1378,7 @@ get_literal(void) cachec('\\'); cachec(c); } - else if (isprint(c)) + else if (isprint(UCH(c))) cachec(c); else { @@ -1329,13 +1472,21 @@ static Value_t get_number(void) { int c; - Value_t n; + long n; + char *base = cptr; n = 0; - for (c = *cptr; isdigit(c); c = *++cptr) - n = (Value_t)(10 * n + (c - '0')); + for (c = *cptr; isdigit(UCH(c)); c = *++cptr) + { + n = (10 * n + (c - '0')); + if (n > MAXYYINT) + { + syntax_error(lineno, line, base); + /*NOTREACHED */ + } + } - return (n); + return (Value_t)(n); } static char * @@ -1382,7 +1533,7 @@ get_tag(void) c = nextc(); if (c == EOF) unexpected_EOF(); - if (!isalpha(c) && c != '_' && c != '$') + if (!IS_NAME1(c)) illegal_tag(t_lineno, t_line, t_cptr); cinc = 0; @@ -1412,7 +1563,7 @@ scan_id(void) { char *b = cptr; - while (isalnum(UCH(*cptr)) || *cptr == '_' || *cptr == '$') + while (IS_NAME2(UCH(*cptr))) cptr++; return cache_tag(b, (size_t) (cptr - b)); } @@ -1442,7 +1593,7 @@ declare_tokens(int assoc) for (;;) { - if (isalpha(c) || c == '_' || c == '.' || c == '$') + if (isalpha(UCH(c)) || c == '_' || c == '.' || c == '$') bp = get_name(); else if (c == '\'' || c == '"') bp = get_literal(); @@ -1472,8 +1623,7 @@ declare_tokens(int assoc) if (c == EOF) unexpected_EOF(); - value = UNDEFINED; - if (isdigit(c)) + if (isdigit(UCH(c))) { value = get_number(); if (bp->value != UNDEFINED && value != bp->value) @@ -1509,7 +1659,7 @@ declare_expect(int assoc) for (;;) { - if (isdigit(c)) + if (isdigit(UCH(c))) { if (assoc == EXPECT) SRexpect = get_number(); @@ -1522,7 +1672,7 @@ declare_expect(int assoc) * Spaces, tabs, and numbers are ok, * words, punc., etc. are syntax errors. */ - else if (c == '\n' || isalpha(c) || !isspace(c)) + else if (c == '\n' || isalpha(UCH(c)) || !isspace(UCH(c))) { syntax_error(lineno, line, cptr); } @@ -1591,7 +1741,7 @@ declare_types(void) c = nextc(); if (c == EOF) unexpected_EOF(); - if (isalpha(c) || c == '_' || c == '.' || c == '$') + if (isalpha(UCH(c)) || c == '_' || c == '.' || c == '$') { bp = get_name(); #if defined(YYBTYACC) @@ -1629,7 +1779,7 @@ declare_start(void) c = nextc(); if (c == EOF) unexpected_EOF(); - if (!isalpha(c) && c != '_' && c != '.' && c != '$') + if (!isalpha(UCH(c)) && c != '_' && c != '.' && c != '$') syntax_error(lineno, line, cptr); bp = get_name(); if (bp->class == TERM) @@ -1664,6 +1814,10 @@ read_declarations(void) copy_ident(); break; + case XCODE: + copy_code(); + break; + case TEXT: copy_text(); break; @@ -1871,7 +2025,7 @@ parse_id(char *p, char **save) if (!isalpha(UCH(*p)) && *p != '_') return NULL; b = p; - while (isalnum(UCH(*p)) || *p == '_' || *p == '$') + while (IS_NAME2(UCH(*p))) p++; if (save) { @@ -2129,7 +2283,10 @@ can_elide_arg(char **theptr, char *yyvaltag) { char *arg; if (!(p = parse_id(p, &arg))) + { + FREE(offsets); return 0; + } for (i = plhs[nrules]->args - 1; i >= 0; i--) if (arg == plhs[nrules]->argnames[i]) break; @@ -2149,7 +2306,7 @@ can_elide_arg(char **theptr, char *yyvaltag) rv = 0; if (maxoffset > 0) FREE(offsets); - if (*p || rv <= 0) + if (p == 0 || *p || rv <= 0) return 0; *theptr = p + 1; return rv; @@ -2228,6 +2385,10 @@ advance_to_start(void) s_cptr = cptr; switch (keyword()) { + case XCODE: + copy_code(); + break; + case MARK: no_grammar(); @@ -2245,7 +2406,7 @@ advance_to_start(void) } c = nextc(); - if (!isalpha(c) && c != '_' && c != '.' && c != '_') + if (!isalpha(UCH(c)) && c != '_' && c != '.' && c != '_') syntax_error(lineno, line, cptr); bp = get_name(); if (goal == 0) @@ -2572,7 +2733,7 @@ copy_action(void) FREE(d_line); goto loop; } - else if (isdigit(c)) + else if (isdigit(UCH(c))) { i = get_number(); if (i == 0) @@ -2596,7 +2757,7 @@ copy_action(void) goto loop; } #if defined(YYBTYACC) - else if (isalpha(c) || c == '_') + else if (isalpha(UCH(c)) || c == '_') { char *arg = scan_id(); for (i = plhs[nrules]->args - 1; i >= 0; i--) @@ -2727,14 +2888,14 @@ copy_action(void) } } #endif - if (isalpha(c) || c == '_' || c == '$') + if (IS_NAME1(c)) { do { putc(c, f); c = *++cptr; } - while (isalnum(c) || c == '_' || c == '$'); + while (IS_NAME2(c)); goto loop; } ++cptr; @@ -2923,14 +3084,14 @@ get_code(struct ainfo *a, const char *loc) cptr += 2; goto loop; } - if (isalpha(c) || c == '_' || c == '$') + if (IS_NAME1(c)) { do { mputc(code_mstr, c); c = *++cptr; } - while (isalnum(c) || c == '_' || c == '$'); + while (IS_NAME2(c)); goto loop; } ++cptr; @@ -3049,7 +3210,7 @@ copy_destructor(void) bp->destructor = process_destructor_XX(code_text, tag); } } - else if (isalpha(c) || c == '_' || c == '.' || c == '$') + else if (isalpha(UCH(c)) || c == '_' || c == '.' || c == '$') { /* "symbol" destructor */ bp = get_name(); if (bp->destructor != NULL) @@ -3090,14 +3251,14 @@ process_destructor_XX(char *code, char *tag) msprintf(new_code, "(*val).%s", tag); goto loop; } - if (isalpha(c) || c == '_' || c == '$') + if (IS_NAME1(c)) { do { mputc(new_code, c); c = *++codeptr; } - while (isalnum(c) || c == '_' || c == '$'); + while (IS_NAME2(c)); goto loop; } ++codeptr; @@ -3180,7 +3341,7 @@ mark_symbol(void) syntax_error(lineno, line, cptr); c = nextc(); - if (isalpha(c) || c == '_' || c == '.' || c == '$') + if (isalpha(UCH(c)) || c == '_' || c == '.' || c == '$') bp = get_name(); else if (c == '\'' || c == '"') bp = get_literal(); @@ -3211,7 +3372,7 @@ read_grammar(void) c = nextc(); if (c == EOF) break; - if (isalpha(c) + if (isalpha(UCH(c)) || c == '_' || c == '.' || c == '$' |