* array.c (rb_ary_cycle): typo in rdoc. a patch from Yugui

  <[email protected]>.  [ruby-dev:31748]

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@13348 b2dd03c8-39d4-4d8f-98ff-823fe69b080e

ChangeLog

@@ -1,3 +1,8 @@
+Thu Sep  6 21:31:49 2007  Yukihiro Matsumoto  <[email protected]>
+
+	* array.c (rb_ary_cycle): typo in rdoc.  a patch from Yugui
+	  <[email protected]>.  [ruby-dev:31748]
+
 Thu Sep  6 12:42:10 2007  Nobuyoshi Nakada  <[email protected]>
 
 	* string.c (rb_str_succ, rb_str_chop_bang, rb_str_chop): m17n support.

array.c

@@ -2929,7 +2929,7 @@ rb_ary_choice(VALUE ary)
  *  Calls <i>block</i> repeatedly forever.
  *     
  *     a = ["a", "b", "c"]
- *     a.each {|x| puts x }  # print, a, b, c, a, b, c,.. forever.
+ *     a.cycle {|x| puts x }  # print, a, b, c, a, b, c,.. forever.
  *     
  */
 

encoding.c

@@ -189,7 +189,7 @@ rb_enc_nth(const char *p, const char *e, int nth, rb_encoding *enc)
     }
     else {
 	for (c=0; p<e && nth--; c++) {
-	    int n = rb_enc_mbclen(p, enc);
+	    int n = rb_enc_mbclen(p, e, enc);
 
 	    if (n == 0) return 0;
 	    p += n;
@@ -208,7 +208,7 @@ rb_enc_strlen(const char *p, const char *e, rb_encoding *enc)
     }
 
     for (c=0; p<e; c++) {
-	int n = rb_enc_mbclen(p, enc);
+	int n = rb_enc_mbclen(p, e, enc);
 
 	if (n == 0) return -1;
 	p += n;
@@ -217,9 +217,9 @@ rb_enc_strlen(const char *p, const char *e, rb_encoding *enc)
 }
 
 int
-rb_enc_mbclen(const char *p, rb_encoding *enc)
+rb_enc_mbclen(const char *p, const char *e, rb_encoding *enc)
 {
-    int n = ONIGENC_MBC_ENC_LEN(enc, (UChar*)p);
+    int n = ONIGENC_MBC_ENC_LEN(enc, (UChar*)p, (UChar*)e);
     if (n == 0) {
 	rb_raise(rb_eArgError, "invalid mbstring sequence");
     }

euc_jp.c

@@ -51,7 +51,7 @@ static const int EncLen_EUCJP[] = {
 };
 
 static int
-mbc_enc_len(const UChar* p)
+mbc_enc_len(const UChar* p, const UChar* e)
 {
   return EncLen_EUCJP[*p];
 }
@@ -62,7 +62,7 @@ mbc_to_code(const UChar* p, const UChar* end)
   int c, i, len;
   OnigCodePoint n;
 
-  len = enc_len(ONIG_ENCODING_EUC_JP, p);
+  len = enc_len(ONIG_ENCODING_EUC_JP, p, end);
   n = (OnigCodePoint )*p++;
   if (len == 1) return n;
 
@@ -113,7 +113,7 @@ code_to_mbc(OnigCodePoint code, UChar *buf)
   *p++ = (UChar )(code & 0xff);
 
 #if 1
-  if (enc_len(ONIG_ENCODING_EUC_JP, buf) != (p - buf))
+  if (enc_len(ONIG_ENCODING_EUC_JP, buf, p) != (p - buf))
     return ONIGENC_ERR_INVALID_WIDE_CHAR_VALUE;
 #endif  
   return p - buf;
@@ -134,7 +134,7 @@ mbc_case_fold(OnigCaseFoldType flag,
   else {
     int i;
 
-    len = enc_len(ONIG_ENCODING_EUC_JP, p);
+    len = enc_len(ONIG_ENCODING_EUC_JP, p, end);
     for (i = 0; i < len; i++) {
       *lower++ = *p++;
     }
@@ -156,7 +156,7 @@ left_adjust_char_head(const UChar* start, const UChar* s)
   p = s;
 
   while (!eucjp_islead(*p) && p > start) p--;
-  len = enc_len(ONIG_ENCODING_EUC_JP, p);
+  len = enc_len(ONIG_ENCODING_EUC_JP, p, s);
   if (p + len > s) return (UChar* )p;
   p += len;
   return (UChar* )(p + ((s - p) & ~1));

ext/strscan/strscan.c

@@ -661,7 +661,7 @@ strscan_getch(VALUE self)
     if (EOS_P(p))
         return Qnil;
 
-    len = rb_enc_mbclen(CURPTR(p), enc);
+    len = rb_enc_mbclen(CURPTR(p), S_PEND(p), enc);
     if (p->curr + len > S_LEN(p)) {
         len = S_LEN(p) - p->curr;
     }

include/ruby/encoding.h

@@ -50,7 +50,7 @@ rb_encoding * rb_enc_find(const char *name);
 #define rb_enc_mbmaxlen(enc) (enc)->max_enc_len
 
 /* ptr,encoding -> mbclen */
-int rb_enc_mbclen(const char*, rb_encoding*);
+int rb_enc_mbclen(const char*, const char *, rb_encoding*);
 
 /* code,encoding -> codelen */
 int rb_enc_codelen(int, rb_encoding*);

include/ruby/oniguruma.h

@@ -144,7 +144,7 @@ typedef struct {
 typedef int (*OnigApplyAllCaseFoldFunc)(OnigCodePoint from, OnigCodePoint* to, int to_len, void* arg);
 
 typedef struct OnigEncodingTypeST {
-  int    (*mbc_enc_len)(const OnigUChar* p);
+  int    (*mbc_enc_len)(const OnigUChar* p,const OnigUChar* e);
   const char*   name;
   int           max_enc_len;
   int           min_enc_len;
@@ -255,11 +255,11 @@ ONIG_EXTERN OnigEncodingType OnigEncodingGB18030;
 #define ONIGENC_MAX_STD_CTYPE  ONIGENC_CTYPE_ASCII
 
 
-#define enc_len(enc,p)                ONIGENC_MBC_ENC_LEN(enc, p)
+#define enc_len(enc,p,e)                ONIGENC_MBC_ENC_LEN(enc, p, e)
 
 #define ONIGENC_IS_UNDEF(enc)          ((enc) == ONIG_ENCODING_UNDEF)
 #define ONIGENC_IS_SINGLEBYTE(enc)     (ONIGENC_MBC_MAXLEN(enc) == 1)
-#define ONIGENC_IS_MBC_HEAD(enc,p)     (ONIGENC_MBC_ENC_LEN(enc,p) != 1)
+#define ONIGENC_IS_MBC_HEAD(enc,p,e)   (ONIGENC_MBC_ENC_LEN(enc,p,e) != 1)
 #define ONIGENC_IS_MBC_ASCII(p)           (*(p)   < 128)
 #define ONIGENC_IS_CODE_ASCII(code)       ((code) < 128)
 #define ONIGENC_IS_MBC_WORD(enc,s,end) \
@@ -281,7 +281,7 @@ ONIG_EXTERN OnigEncodingType OnigEncodingGB18030;
 #define ONIGENC_STEP_BACK(enc,start,s,n) \
         onigenc_step_back((enc),(start),(s),(n))
 
-#define ONIGENC_MBC_ENC_LEN(enc,p)             (enc)->mbc_enc_len(p)
+#define ONIGENC_MBC_ENC_LEN(enc,p,e)           (enc)->mbc_enc_len(p,e)
 #define ONIGENC_MBC_MAXLEN(enc)               ((enc)->max_enc_len)
 #define ONIGENC_MBC_MAXLEN_DIST(enc)           ONIGENC_MBC_MAXLEN(enc)
 #define ONIGENC_MBC_MINLEN(enc)               ((enc)->min_enc_len)

include/ruby/regex.h

@@ -29,8 +29,8 @@ extern "C" {
 
 ONIG_EXTERN OnigEncoding    OnigEncDefaultCharEncoding;
 
-#define ismbchar(p, enc) (mbclen((p),(enc)) != 1)
-#define mbclen(p,enc)  rb_enc_mbclen((p), (enc))
+#define ismbchar(p, e, enc) (mbclen((p),(e),(enc)) != 1)
+#define mbclen(p,e,enc)  rb_enc_mbclen((p),(e),(enc))
 
 #endif /* ifndef ONIG_RUBY_M17N */
 

parse.y

@@ -4558,10 +4558,10 @@ ripper_dispatch_delayed_token(struct parser_params *parser, int t)
 # define SIGN_EXTEND_CHAR(c) ((((unsigned char)(c)) ^ 128) - 128)
 #endif
 
-#define parser_mbclen()  mbclen((lex_p-1),parser->enc)
-#define is_identchar(p, enc) (rb_enc_isalnum(*p, enc) || (*p) == '_' || ismbchar(p, enc))
-#define parser_ismbchar() ismbchar((lex_p-1), parser->enc)
-#define parser_is_identchar() (!parser->eofp && is_identchar((lex_p-1),parser->enc))
+#define parser_mbclen()  mbclen((lex_p-1),lex_pend,parser->enc)
+#define is_identchar(p,e,enc) (rb_enc_isalnum(*p,enc) || (*p) == '_' || ismbchar(p,e,enc))
+#define parser_ismbchar() ismbchar((lex_p-1), lex_pend, parser->enc)
+#define parser_is_identchar() (!parser->eofp && is_identchar((lex_p-1),lex_pend,parser->enc))
 
 static int
 parser_yyerror(struct parser_params *parser, const char *msg)
@@ -5995,7 +5995,7 @@ parser_yylex(struct parser_params *parser)
 	    }
 	}
 	else if ((rb_enc_isalnum(c, parser->enc) || c == '_') &&
-		 lex_p < lex_pend && is_identchar(lex_p, parser->enc)) {
+		 lex_p < lex_pend && is_identchar(lex_p, lex_pend, parser->enc)) {
 	    goto ternary;
 	}
 	else if (c == '\\') {
@@ -8328,7 +8328,7 @@ internal_id_gen(struct parser_params *parser)
 }
 
 static int
-is_special_global_name(const char *m, rb_encoding *enc)
+is_special_global_name(const char *m, const char *e, rb_encoding *enc)
 {
     switch (*m) {
       case '~': case '*': case '$': case '?': case '!': case '@':
@@ -8340,7 +8340,7 @@ is_special_global_name(const char *m, rb_encoding *enc)
 	break;
       case '-':
 	++m;
-	if (is_identchar(m, enc)) m += rb_enc_mbclen(m, enc);
+	if (is_identchar(m, e, enc)) m += rb_enc_mbclen(m, e, enc);
 	break;
       default:
 	if (!rb_enc_isdigit(*m, enc)) return 0;
@@ -8353,6 +8353,7 @@ int
 rb_symname_p(const char *name)
 {
     const char *m = name;
+    const char *e = m + strlen(m);
     int localid = Qfalse;
     rb_encoding *enc = rb_enc_from_index(0);
 
@@ -8362,7 +8363,7 @@ rb_symname_p(const char *name)
 	return Qfalse;
 
       case '$':
-	if (is_special_global_name(++m, enc)) return Qtrue;
+	if (is_special_global_name(++m, e, enc)) return Qtrue;
 	goto id;
 
       case '@':
@@ -8411,8 +8412,9 @@ rb_symname_p(const char *name)
       default:
 	localid = !rb_enc_isupper(*m, enc);
       id:
-	if (*m != '_' && !rb_enc_isalpha(*m, enc) && !ismbchar(m, enc)) return Qfalse;
-	while (is_identchar(m, enc)) m += rb_enc_mbclen(m, enc);
+	if (*m != '_' && !rb_enc_isalpha(*m, enc) && !ismbchar(m, e, enc))
+		  return Qfalse;
+	while (is_identchar(m, e, enc)) m += rb_enc_mbclen(m, e, enc);
 	if (localid) {
 	    switch (*m) {
 	      case '!': case '?': case '=': ++m;
@@ -8427,6 +8429,7 @@ ID
 rb_intern3(const char *name, long len, rb_encoding *enc)
 {
     const char *m = name;
+    const char *e = m + len;
     VALUE str;
     ID id;
     int last;
@@ -8445,7 +8448,7 @@ rb_intern3(const char *name, long len, rb_encoding *enc)
     switch (*m) {
       case '$':
 	id |= ID_GLOBAL;
-	if (is_special_global_name(++m, enc)) goto new_id;
+	if (is_special_global_name(++m, e, enc)) goto new_id;
 	break;
       case '@':
 	if (m[1] == '@') {
@@ -8490,8 +8493,8 @@ rb_intern3(const char *name, long len, rb_encoding *enc)
 	break;
     }
     if (!rb_enc_isdigit(*m, enc)) {
-	while (m <= name + last && is_identchar(m, enc)) {
-	    m += rb_enc_mbclen(m, enc);
+	while (m <= name + last && is_identchar(m, e, enc)) {
+	    m += rb_enc_mbclen(m, e, enc);
 	}
     }
     if (m - name < len) id = ID_JUNK;

re.c

@@ -393,11 +393,11 @@ rb_reg_expr_str(VALUE str, const char *s, long len)
 
     p = s; pend = p + len;
     while (p<pend) {
-	if (*p == '/' || (!rb_enc_isprint(*p, enc) && !ismbchar(p, enc))) {
+	if (*p == '/' || (!rb_enc_isprint(*p, enc) && !ismbchar(p, pend, enc))) {
 	    need_escape = 1;
 	    break;
 	}
-	p += mbclen(p, enc);
+	p += mbclen(p, pend, enc);
     }
     if (!need_escape) {
 	rb_str_buf_cat(str, s, len);
@@ -406,7 +406,7 @@ rb_reg_expr_str(VALUE str, const char *s, long len)
 	p = s;
 	while (p<pend) {
 	    if (*p == '\\') {
-		int n = mbclen(p+1, enc) + 1;
+		int n = mbclen(p+1, pend, enc) + 1;
 		rb_str_buf_cat(str, p, n);
 		p += n;
 		continue;
@@ -416,9 +416,9 @@ rb_reg_expr_str(VALUE str, const char *s, long len)
 		rb_str_buf_cat(str, &c, 1);
 		rb_str_buf_cat(str, p, 1);
 	    }
-	    else if (ismbchar(p, enc)) {
-	    	rb_str_buf_cat(str, p, mbclen(p, enc));
-		p += mbclen(p, enc);
+	    else if (ismbchar(p, pend, enc)) {
+	    	rb_str_buf_cat(str, p, mbclen(p, pend, enc));
+		p += mbclen(p, pend, enc);
 		continue;
 	    }
 	    else if (rb_enc_isprint(*p, enc)) {
@@ -1906,8 +1906,8 @@ rb_reg_quote(VALUE str)
     send = s + RSTRING_LEN(str);
     for (; s < send; s++) {
 	c = *s;
-	if (ismbchar(s, enc)) {
-	    int n = mbclen(s, enc);
+	if (ismbchar(s, send, enc)) {
+	    int n = mbclen(s, send, enc);
 
 	    while (n-- && s < send)
 		s++;
@@ -1935,8 +1935,8 @@ rb_reg_quote(VALUE str)
 
     for (; s < send; s++) {
 	c = *s;
-	if (ismbchar(s, enc)) {
-	    int n = mbclen(s, enc);
+	if (ismbchar(s, send, enc)) {
+	    int n = mbclen(s, send, enc);
 
 	    while (n-- && s < send)
 		*t++ = *s++;
@@ -2180,8 +2180,8 @@ rb_reg_regsub(VALUE str, VALUE src, struct re_registers *regs, VALUE regexp)
     while (s < e) {
 	char *ss = s++;
 
-	if (ismbchar(ss, enc)) {
-	    s += mbclen(ss, enc) - 1;
+	if (ismbchar(ss, e, enc)) {
+	    s += mbclen(ss, e, enc) - 1;
 	    continue;
 	}
 	if (*ss != '\\' || s == e) continue;
@@ -2214,7 +2214,7 @@ rb_reg_regsub(VALUE str, VALUE src, struct re_registers *regs, VALUE regexp)
               name_end = name = s + 1;
               while (name_end < e) {
                 if (*name_end == '>') break;
-                name_end += mbclen(name_end, enc);
+                name_end += mbclen(name_end, e, enc);
               }
               if (name_end < e) {
                 no = name_to_backref_number(regs, regexp, name, name_end);