Reviewed by Richard.

	<rdar://problem/3493140> REGRESSION (85-100): cedille displays %-escaped in JavaScript message at hotmail.com

	* kjs/function.cpp:
        (KJS::GlobalFuncImp::call): Replace our escape() and unescape() implementations with
	ones from KDE KJS, which have the proper latin-1 behavior to match Win IE.
        * kjs/lexer.cpp:
        (Lexer::isHexDigit): Made static and non-const.
        * kjs/lexer.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@7507 268f45cc-cd09-0410-ab3c-d52691b4dbfc

diff --git a/JavaScriptCore/ChangeLog b/JavaScriptCore/ChangeLog
index 48e35c2f6246b42515758dee83c1ac0e7a694055..368a68f4a61826365261c276153d2cf50bf6476b 100644 (file)
--- a/JavaScriptCore/ChangeLog
+++ b/JavaScriptCore/ChangeLog
@@ -1,3 +1,16 @@
+2004-09-09  Maciej Stachowiak  <mjs@apple.com>
+
+        Reviewed by Richard.
+
+       <rdar://problem/3493140> REGRESSION (85-100): cedille displays %-escaped in JavaScript message at hotmail.com
+
+       * kjs/function.cpp:
+        (KJS::GlobalFuncImp::call): Replace our escape() and unescape() implementations with
+       ones from KDE KJS, which have the proper latin-1 behavior to match Win IE.
+        * kjs/lexer.cpp:
+        (Lexer::isHexDigit): Made static and non-const.
+        * kjs/lexer.h:
+
 === Safari-163 ===
 
 2004-09-06  Darin Adler  <darin@apple.com>

diff --git a/JavaScriptCore/kjs/function.cpp b/JavaScriptCore/kjs/function.cpp
index fa8cd1efbe1b99234aa8d04344eb17d27913762c..b4b596c2751f9cba22e0bb0b54244f6dbfbfa8d3 100644 (file)
--- a/JavaScriptCore/kjs/function.cpp
+++ b/JavaScriptCore/kjs/function.cpp
@@ -613,6 +613,7 @@ Value GlobalFuncImp::call(ExecState *exec, Object &/*thisObj*/, const List &args
     "abcdefghijklmnopqrstuvwxyz"
     "0123456789"
     "*+-./@_";
+
   static const char do_not_escape_when_encoding_URI_component[] =
     "ABCDEFGHIJKLMNOPQRSTUVWXYZ"
     "abcdefghijklmnopqrstuvwxyz"
@@ -710,11 +711,54 @@ Value GlobalFuncImp::call(ExecState *exec, Object &/*thisObj*/, const List &args
     res = encode(exec, args, do_not_escape_when_encoding_URI_component);
     break;
   case Escape:
-    res = encode(exec, args, do_not_escape);
-    break;
+    {
+      UString r = "", s, str = args[0].toString(exec);
+      const UChar *c = str.data();
+      for (int k = 0; k < str.size(); k++, c++) {
+        int u = c->uc;
+        if (u > 255) {
+          char tmp[7];
+          sprintf(tmp, "%%u%04X", u);
+          s = UString(tmp);
+        } else if (strchr(do_not_escape, (char)u)) {
+          s = UString(c, 1);
+        } else {
+          char tmp[4];
+          sprintf(tmp, "%%%02X", u);
+          s = UString(tmp);
+        }
+        r += s;
+      }
+      res = String(r);
+      break;
+    }
   case UnEscape:
-    res = decode(exec, args, "", false);
-    break;
+    {
+      UString s, str = args[0].toString(exec);
+      int k = 0, len = str.size();
+      while (k < len) {
+        const UChar *c = str.data() + k;
+        UChar u;
+        if (*c == UChar('%') && k <= len - 6 && *(c+1) == UChar('u')) {
+          if (Lexer::isHexDigit((c+2)->uc) && Lexer::isHexDigit((c+3)->uc) &&
+              Lexer::isHexDigit((c+4)->uc) && Lexer::isHexDigit((c+5)->uc)) {
+         u = Lexer::convertUnicode((c+2)->uc, (c+3)->uc,
+                                   (c+4)->uc, (c+5)->uc);
+         c = &u;
+         k += 5;
+          }
+        } else if (*c == UChar('%') && k <= len - 3 &&
+                   Lexer::isHexDigit((c+1)->uc) && Lexer::isHexDigit((c+2)->uc)) {
+          u = UChar(Lexer::convertHex((c+1)->uc, (c+2)->uc));
+          c = &u;
+          k += 2;
+        }
+        k++;
+        s += UString(c, 1);
+      }
+      res = String(s);
+      break;
+    }
 #ifndef NDEBUG
   case KJSPrint:
     puts(args[0].toString(exec).ascii());

diff --git a/JavaScriptCore/kjs/lexer.cpp b/JavaScriptCore/kjs/lexer.cpp
index 91594c7b004d9ca34c9011ab1689dc28b9557cf1..612f4c46c5a13fd6d0fc25b3807be642216b65f0 100644 (file)
--- a/JavaScriptCore/kjs/lexer.cpp
+++ b/JavaScriptCore/kjs/lexer.cpp
@@ -577,7 +577,7 @@ bool Lexer::isDecimalDigit(unsigned short c)
   return (c >= '0' && c <= '9');
 }
 
-bool Lexer::isHexDigit(unsigned short c) const
+bool Lexer::isHexDigit(unsigned short c)
 {
   return (c >= '0' && c <= '9' ||
           c >= 'a' && c <= 'f' ||

diff --git a/JavaScriptCore/kjs/lexer.h b/JavaScriptCore/kjs/lexer.h
index 898b96881d90211c7524ae49514e63dbfd5216cf..2ce85a1269c73d7c0bc5b61dc067cb25dc7fb9be 100644 (file)
--- a/JavaScriptCore/kjs/lexer.h
+++ b/JavaScriptCore/kjs/lexer.h
@@ -100,7 +100,6 @@ namespace KJS {
 
     bool isWhiteSpace() const;
     bool isLineTerminator();
-    bool isHexDigit(unsigned short c) const;
     bool isOctalDigit(unsigned short c) const;
 
     int matchPunctuator(unsigned short c1, unsigned short c2,
@@ -115,6 +114,7 @@ namespace KJS {
                                 unsigned short c3, unsigned short c4);
     static bool isIdentLetter(unsigned short c);
     static bool isDecimalDigit(unsigned short c);
+    static bool isHexDigit(unsigned short c);
 
 #ifdef KJS_DEBUG_MEM
     /**