A thread-safe Media Query Parser

Source/core/css/parser/MediaQueryTokenizer.h

+// Copyright 2014 The Chromium Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style license that can be
+// found in the LICENSE file.
+
+#ifndef MediaQueryTokenizer_h
+#define MediaQueryTokenizer_h
+
+#include "core/css/parser/MediaQueryToken.h"
+#include "core/html/parser/InputStreamPreprocessor.h"
+#include "wtf/text/WTFString.h"
+
+#define CODE_POINTS_NUM 128

[eseidel, 2014/03/11 20:14:19]

What is this for?  Can it be a const instead of a macro?

This is just the size of a signed char, no?  sizeof(char)?  sizeof(LChar)?

+
+namespace WebCore {
+
+class MediaQueryInputStream;
+
+class MediaQueryTokenizer {
+    WTF_MAKE_NONCOPYABLE(MediaQueryTokenizer);
+    WTF_MAKE_FAST_ALLOCATED;
+public:
+    MediaQueryTokenizer();
+
+    MediaQueryToken nextToken(MediaQueryInputStream&);
+    static void tokenize(String, Vector<MediaQueryToken>&);
+
+private:
+    UChar consume();
+    void consume(unsigned);
+    void reconsume(UChar);
+
+    MediaQueryToken consumeNumericToken();
+    MediaQueryToken consumeIdentLikeToken();
+    MediaQueryToken consumeNumber();
+
+    void consumeUntilNonWhitespace();
+
+    bool consumeIfNext(UChar);
+    String consumeName();
+    UChar consumeEscape();
+
+    bool nextTwoCharsAreValidEscape();
+    bool nextCharsAreNumber();
+    bool nextCharsAreIdentifier();
+
+    typedef MediaQueryToken (MediaQueryTokenizer::*CodePoint)(UChar);
+
+    class CodePoints {
+    public:
+        CodePoint codePoints[CODE_POINTS_NUM];
+
+        CodePoints()
+        {
+            memset(codePoints, 0, CODE_POINTS_NUM);

[eseidel, 2014/03/11 20:14:19]

This could be static and stored in the binary instead of allocated on the heap,
no?

[Yoav Weiss, 2014/03/12 20:39:00]

I'm not sure how to do that. I'd love some guidance here.

[eseidel, 2014/03/13 00:49:36]

You'd probably have to do it with static initializers:

CodePoint codePoints[128] {
  whiteSpace, // null
  whiteSpace, // start heading
  leftParen, // '('
  .. for all the ascii chars.
};

It would be tedious, and again may be best deferred until a later CL.  But the
current implementation will allocate a fresh copy of this array for each render
process, which is convenient, but needless.

+            codePoints['\n'] = &MediaQueryTokenizer::whiteSpace;

[eseidel, 2014/03/11 20:14:19]

Why are we using a jump-table instead of an if-cascade?

I guess it doens't really matter one way or the other.  I don't think this
should be in the header.

[Yoav Weiss, 2014/03/12 20:39:00]

I implemented it as a table following a comment by abarth on the original CL
(that a lookup table will be faster).

[eseidel, 2014/03/13 00:49:36]

I'm told that modern compilers are able to convert switches as well as
if-cascades into lookup tables when needed, but I could also be putting waaay
too much faith in the compiler. :)

+            codePoints['\r'] = &MediaQueryTokenizer::whiteSpace;
+            codePoints['\t'] = &MediaQueryTokenizer::whiteSpace;
+            codePoints[' '] = &MediaQueryTokenizer::whiteSpace;
+            codePoints['\f'] = &MediaQueryTokenizer::whiteSpace;
+            codePoints['('] = &MediaQueryTokenizer::leftParenthesis;
+            codePoints[')'] = &MediaQueryTokenizer::rightParenthesis;
+            codePoints['+'] = &MediaQueryTokenizer::plusOrFullStop;
+            codePoints['.'] = &MediaQueryTokenizer::plusOrFullStop;
+            codePoints[','] = &MediaQueryTokenizer::comma;
+            codePoints['-'] = &MediaQueryTokenizer::hyphenMinus;
+            codePoints['/'] = &MediaQueryTokenizer::solidus;
+            codePoints[':'] = &MediaQueryTokenizer::colon;
+            codePoints[';'] = &MediaQueryTokenizer::semiColon;
+            codePoints['\\'] = &MediaQueryTokenizer::reverseSolidus;
+            for (unsigned char digit = '0'; digit <= '9'; ++digit)
+                codePoints[digit] = &MediaQueryTokenizer::asciiDigit;
+            for (unsigned char alpha = 'a'; alpha <= 'z'; ++alpha)
+                codePoints[alpha] = &MediaQueryTokenizer::nameStart;
+            for (unsigned char alpha = 'A'; alpha <= 'Z'; ++alpha)
+                codePoints[alpha] = &MediaQueryTokenizer::nameStart;
+            codePoints['_'] = &MediaQueryTokenizer::nameStart;
+            codePoints[kEndOfFileMarker] = &MediaQueryTokenizer::endOfFile;
+        }
+    };
+
+    CodePoints* codePoints()
+    {
+        static CodePoints codePoints;
+        return &codePoints;
+    }
+
+    MediaQueryToken whiteSpace(UChar);
+    MediaQueryToken leftParenthesis(UChar);
+    MediaQueryToken rightParenthesis(UChar);
+    MediaQueryToken plusOrFullStop(UChar);
+    MediaQueryToken comma(UChar);
+    MediaQueryToken hyphenMinus(UChar);
+    MediaQueryToken solidus(UChar);
+    MediaQueryToken colon(UChar);
+    MediaQueryToken semiColon(UChar);
+    MediaQueryToken reverseSolidus(UChar);
+    MediaQueryToken asciiDigit(UChar);
+    MediaQueryToken nameStart(UChar);
+    MediaQueryToken endOfFile(UChar);
+
+    MediaQueryInputStream* m_input;
+};
+
+
+
+} // namespace WebCore
+
+#endif // MediaQueryTokenizer_h