[Webkit-unassigned] [Bug 45594] Add AtomicString::fromUTF8

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=45594





--- Comment #7 from Patrick R. Gansterer <paroga at paroga.com>  2010-09-12 12:30:37 PST ---
(In reply to comment #6)
> (From update of attachment 67334 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=67334&action=prettypatch
> 
> I don't get it.  Is this more efficient than making a string and then converting that to an atomic string?  Maybe it saves you a memcpy in the case where you have multibyte characters?
XMLParser will uses the version witout a explicit length most of the time. Ther we can calcualte the hash, utf8length and utf16 length in one step:
+AtomicString AtomicString::fromUTF8(const char* characters)
+{
+    if (!characters)
+        return AtomicString();
+
+    HashAndUTF8Characters buffer;
+    buffer.characters = characters;
+    buffer.string = 0;
+
+    if (!calculateHashAndLength(characters, buffer.hash, buffer.lengthUTF8, buffer.lengthUTF16))
+        return AtomicString(); // characters contains invalid UTF-8 characters.
+
+    pair<HashSet<StringImpl*>::iterator, bool> addResult = stringTable().add<HashAndUTF8Characters, HashAndUTF8CharactersTranslator>(buffer);
+
+    // If the string is newly-translated, then we need to adopt it.
+    // The boolean in the pair tells us if that is so.
+    AtomicString atomicString;
+    atomicString.m_string = addResult.second ? adoptRef(*addResult.first) : *addResult.first;
+    return atomicString;
+}

This will avoid unnecessary operations, but requires additonal changes to the stringHash and UTF8 functions.
I'd like to commit this in a first step to start using the "new API" in XMLParser.

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.