From a2e8e2022bcab4dd87ee5cf6f1ef1d9df1cf4c98 Mon Sep 17 00:00:00 2001
From: akr <akr@b2dd03c8-39d4-4d8f-98ff-823fe69b080e>
Date: Sun, 27 Jan 2008 14:27:07 +0000
Subject: * include/ruby/oniguruma.h: precise mbclen API redesigned to avoid  
 inline functions.   (onigenc_mbclen_charfound): removed.  
 (onigenc_mbclen_needmore): removed.   (onigenc_mbclen_recover): removed.  
 (ONIGENC_MBCLEN_CHARFOUND): removed.   (ONIGENC_MBCLEN_CHARFOUND_P): defined.
   (ONIGENC_MBCLEN_CHARFOUND_LEN): defined.   (ONIGENC_MBCLEN_INVALID):
 removed.   (ONIGENC_MBCLEN_INVALID_P): defined.   (ONIGENC_MBCLEN_NEEDMORE):
 removed.   (ONIGENC_MBCLEN_NEEDMORE_P): defined.  
 (ONIGENC_MBCLEN_NEEDMORE_LEN): defined.   (ONIGENC_MBC_ENC_LEN): use
 onigenc_mbclen_approximate.

* regenc.c (onigenc_mbclen_approximate): defined.

* include/ruby/encoding.h (MBCLEN_CHARFOUND): removed.
  (MBCLEN_INVALID): removed.
  (MBCLEN_NEEDMORE): removed.
  (MBCLEN_CHARFOUND_P): defined.
  (MBCLEN_INVALID_P): defined.
  (MBCLEN_NEEDMORE_P): defined.
  (MBCLEN_CHARFOUND_LEN): defined.
  (MBCLEN_NEEDMORE_LEN): defined.

* encoding.c: use new API.

* re.c: ditto.

* string.c: ditto.

* parse.y: ditto.



git-svn-id: http://svn.ruby-lang.org/repos/ruby/trunk@15280 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
---
 string.c | 29 +++++++++++++++--------------
 1 file changed, 15 insertions(+), 14 deletions(-)

(limited to 'string.c')

diff --git a/string.c b/string.c
index 452e4affc..3c4e2fc98 100644
--- a/string.c
+++ b/string.c
@@ -170,11 +170,10 @@ coderange_scan(const char *p, long len, rb_encoding *enc)
         }
         while (p < e) {
             int ret = rb_enc_precise_mbclen(p, e, enc);
-            int len = MBCLEN_CHARFOUND(ret);
-            if (!len) {
+            if (!MBCLEN_CHARFOUND_P(ret)) {
                 return ENC_CODERANGE_BROKEN;
             }
-            p += len;
+            p += MBCLEN_CHARFOUND_LEN(ret);
             if (p < e) {
                 p = search_nonascii(p, e);
                 if (!p) {
@@ -190,12 +189,11 @@ coderange_scan(const char *p, long len, rb_encoding *enc)
 
     while (p < e) {
         int ret = rb_enc_precise_mbclen(p, e, enc);
-        int len = MBCLEN_CHARFOUND(ret);
 
-        if (!len) {
+        if (!MBCLEN_CHARFOUND_P(ret)) {
             return ENC_CODERANGE_BROKEN;
         }
-        p += len;
+        p += MBCLEN_CHARFOUND_LEN(ret);
     }
     if (e < p) {
         return ENC_CODERANGE_BROKEN;
@@ -2017,7 +2015,8 @@ enc_succ_char(char *p, int len, rb_encoding *enc)
             return NEIGHBOR_WRAPPED;
         ++((unsigned char*)p)[i];
         l = rb_enc_precise_mbclen(p, p+len, enc);
-        if (MBCLEN_CHARFOUND(l)) {
+        if (MBCLEN_CHARFOUND_P(l)) {
+            l = MBCLEN_CHARFOUND_LEN(l);
             if (l == len) {
                 return NEIGHBOR_FOUND;
             }
@@ -2025,11 +2024,11 @@ enc_succ_char(char *p, int len, rb_encoding *enc)
                 memset(p+l, 0xff, len-l);
             }
         }
-        if (MBCLEN_INVALID(l) && i < len-1) {
+        if (MBCLEN_INVALID_P(l) && i < len-1) {
             int len2, l2;
             for (len2 = len-1; 0 < len2; len2--) {
                 l2 = rb_enc_precise_mbclen(p, p+len2, enc);
-                if (!MBCLEN_INVALID(l2))
+                if (!MBCLEN_INVALID_P(l2))
                     break;
             }
             memset(p+len2+1, 0xff, len-(len2+1));
@@ -2048,7 +2047,8 @@ enc_pred_char(char *p, int len, rb_encoding *enc)
             return NEIGHBOR_WRAPPED;
         --((unsigned char*)p)[i];
         l = rb_enc_precise_mbclen(p, p+len, enc);
-        if (MBCLEN_CHARFOUND(l)) {
+        if (MBCLEN_CHARFOUND_P(l)) {
+            l = MBCLEN_CHARFOUND_LEN(l);
             if (l == len) {
                 return NEIGHBOR_FOUND;
             }
@@ -2056,11 +2056,11 @@ enc_pred_char(char *p, int len, rb_encoding *enc)
                 memset(p+l, 0, len-l);
             }
         }
-        if (MBCLEN_INVALID(l) && i < len-1) {
+        if (MBCLEN_INVALID_P(l) && i < len-1) {
             int len2, l2;
             for (len2 = len-1; 0 < len2; len2--) {
                 l2 = rb_enc_precise_mbclen(p, p+len2, enc);
-                if (!MBCLEN_INVALID(l2))
+                if (!MBCLEN_INVALID_P(l2))
                     break;
             }
             memset(p+len2+1, 0, len-(len2+1));
@@ -3300,11 +3300,12 @@ rb_str_inspect(VALUE str)
 	int cc;
 
         n = rb_enc_precise_mbclen(p, pend, enc);
-        if (!MBCLEN_CHARFOUND(n)) {
+        if (!MBCLEN_CHARFOUND_P(n)) {
             p++;
             n = 1;
             goto escape_codepoint;
         }
+        n = MBCLEN_CHARFOUND_LEN(n);
 
 	c = rb_enc_codepoint(p, pend, enc);
 	n = rb_enc_codelen(c, enc);
@@ -3313,7 +3314,7 @@ rb_str_inspect(VALUE str)
 	if (c == '"'|| c == '\\' ||
 	    (c == '#' &&
              p < pend &&
-             MBCLEN_CHARFOUND(rb_enc_precise_mbclen(p,pend,enc)) &&
+             MBCLEN_CHARFOUND_P(rb_enc_precise_mbclen(p,pend,enc)) &&
              (cc = rb_enc_codepoint(p,pend,enc),
               (cc == '$' || cc == '@' || cc == '{')))) {
 	    prefix_escape(result, c, enc);
-- 
cgit