Changeset 131836 in webkit

Timestamp:

Oct 18, 2012, 6:22:21 PM (13 years ago)

Author:

msaboff@apple.com

Message:

convertUTF8ToUTF16() Should Check for ASCII Input
ihttps://bugs.webkit.org/show_bug.cgi?id=99739

Reviewed by Geoffrey Garen.

Source/JavaScriptCore:

Using the updated convertUTF8ToUTF16() , we can determine if is makes more sense to
create a string using the 8 bit source. Added a new OpaqueJSString::create(LChar*, unsigned).
Had to add a cast n JSStringCreateWithCFString to differentiate which create() to call.

• API/JSStringRef.cpp:

(JSStringCreateWithUTF8CString):

• API/JSStringRefCF.cpp:

(JSStringCreateWithCFString):

• API/OpaqueJSString.h:

(OpaqueJSString::create):
(OpaqueJSString):
(OpaqueJSString::OpaqueJSString):

Source/WTF:

Added code to accumulate the "or" of all characters seen during the UTF8 to UTF16 conversion. This is
used to check to see if all characters are ASCII and is returned via a bool*.

• wtf/unicode/UTF8.cpp:

(WTF::Unicode::convertUTF8ToUTF16):

• wtf/unicode/UTF8.h:

Location:

trunk/Source

Files:

• JavaScriptCore/API/JSStringRef.cpp (modified) (1 diff)
• JavaScriptCore/API/JSStringRefCF.cpp (modified) (1 diff)
• JavaScriptCore/API/OpaqueJSString.h (modified) (2 diffs)
• JavaScriptCore/ChangeLog (modified) (1 diff)
• WTF/ChangeLog (modified) (1 diff)
• WTF/wtf/unicode/UTF8.cpp (modified) (4 diffs)
• WTF/wtf/unicode/UTF8.h (modified) (1 diff)

trunk/Source/JavaScriptCore/API/JSStringRef.cpp

@@ -47 +47 @@
         Vector<UChar, 1024> buffer(length);
         UChar* p = buffer.data();
-        if (conversionOK == convertUTF8ToUTF16(&string, string + length, &p, p + length))
+        bool sourceIsAllASCII;
+        if (conversionOK == convertUTF8ToUTF16(&string, string + length, &p, p + length, &sourceIsAllASCII)) {
+            if (sourceIsAllASCII)
+                return OpaqueJSString::create(reinterpret_cast<const LChar*>(string), length).leakRef();
             return OpaqueJSString::create(buffer.data(), p - buffer.data()).leakRef();
+        }
     }
 

trunk/Source/JavaScriptCore/API/JSStringRefCF.cpp

@@ -47 +47 @@
         return OpaqueJSString::create(reinterpret_cast<UChar*>(buffer.get()), length).leakRef();
     } else {
-        return OpaqueJSString::create(0, 0).leakRef();
+        return OpaqueJSString::create(static_cast<const LChar*>(0), 0).leakRef();
     }
 }

trunk/Source/JavaScriptCore/API/OpaqueJSString.h

@@ -42 +42 @@
     }
 
+    static PassRefPtr<OpaqueJSString> create(const LChar* characters, unsigned length)
+    {
+        return adoptRef(new OpaqueJSString(characters, length));
+    }
+
     static PassRefPtr<OpaqueJSString> create(const UChar* characters, unsigned length)
     {
@@ -71 +76 @@
     }
 
+    OpaqueJSString(const LChar* characters, unsigned length)
+    {
+        m_string = String(characters, length);
+    }
+
     OpaqueJSString(const UChar* characters, unsigned length)
     {

trunk/Source/JavaScriptCore/ChangeLog

@@ +1 @@
+2012-10-18  Michael Saboff  <msaboff@apple.com>
+
+        convertUTF8ToUTF16() Should Check for ASCII Input
+        ihttps://bugs.webkit.org/show_bug.cgi?id=99739
+
+        Reviewed by Geoffrey Garen.
+
+        Using the updated convertUTF8ToUTF16() , we can determine if is makes more sense to 
+        create a string using the 8 bit source.  Added a new OpaqueJSString::create(LChar*, unsigned).
+        Had to add a cast n JSStringCreateWithCFString to differentiate which create() to call.
+
+        * API/JSStringRef.cpp:
+        (JSStringCreateWithUTF8CString):
+        * API/JSStringRefCF.cpp:
+        (JSStringCreateWithCFString):
+        * API/OpaqueJSString.h:
+        (OpaqueJSString::create):
+        (OpaqueJSString):
+        (OpaqueJSString::OpaqueJSString):
+
 2012-10-18  Oliver Hunt  <oliver@apple.com>
 

trunk/Source/WTF/ChangeLog

@@ +1 @@
+2012-10-18  Michael Saboff  <msaboff@apple.com>
+
+        convertUTF8ToUTF16() Should Check for ASCII Input
+        ihttps://bugs.webkit.org/show_bug.cgi?id=99739
+
+        Reviewed by Geoffrey Garen.
+
+        Added code to accumulate the "or" of all characters seen during the UTF8 to UTF16 conversion.  This is
+        used to check to see if all characters are ASCII and is returned via a bool*.
+
+        * wtf/unicode/UTF8.cpp:
+        (WTF::Unicode::convertUTF8ToUTF16):
+        * wtf/unicode/UTF8.h:
+
 2012-10-18  Michael Saboff  <msaboff@apple.com>
 

trunk/Source/WTF/wtf/unicode/UTF8.cpp

@@ -298 +298 @@
 ConversionResult convertUTF8ToUTF16(
     const char** sourceStart, const char* sourceEnd, 
-    UChar** targetStart, UChar* targetEnd, bool strict)
+    UChar** targetStart, UChar* targetEnd, bool* sourceAllASCII, bool strict)
 {
     ConversionResult result = conversionOK;
     const char* source = *sourceStart;
     UChar* target = *targetStart;
+    UChar orAllData = 0;
     while (source < sourceEnd) {
         int utf8SequenceLength = inlineUTF8SequenceLength(*source);
@@ -330 +331 @@
                     result = sourceIllegal;
                     break;
-                } else
+                } else {
                     *target++ = replacementCharacter;
-            } else
+                    orAllData |= replacementCharacter;
+                }
+            } else {
                 *target++ = character; // normal case
+                orAllData |= character;
+            }
         } else if (U_IS_SUPPLEMENTARY(character)) {
             // target is a character in range 0xFFFF - 0x10FFFF
@@ -343 +348 @@
             *target++ = U16_LEAD(character);
             *target++ = U16_TRAIL(character);
+            orAllData = 0xffff;
         } else {
             if (strict) {
@@ -348 +354 @@
                 result = sourceIllegal;
                 break; // Bail out; shouldn't continue
-            } else
+            } else {
                 *target++ = replacementCharacter;
+                orAllData |= replacementCharacter;
+            }
         }
     }
     *sourceStart = source;
     *targetStart = target;
+
+    if (sourceAllASCII)
+        *sourceAllASCII = !(orAllData & 0x7f);
+
     return result;
 }

trunk/Source/WTF/wtf/unicode/UTF8.h

@@ -65 +65 @@
     WTF_EXPORT_PRIVATE ConversionResult convertUTF8ToUTF16(
                     const char** sourceStart, const char* sourceEnd, 
-                    UChar** targetStart, UChar* targetEnd, bool strict = true);
+                    UChar** targetStart, UChar* targetEnd, bool* isSourceAllASCII = 0, bool strict = true);
 
     WTF_EXPORT_PRIVATE ConversionResult convertLatin1ToUTF8(