Initial import of the Chrome Frame codebase. Integration in chrome.gyp coming in a separate CL.

BUG=None
TEST=None

Review URL: http://codereview.chromium.org/218019

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@27042 0039d316-1c4b-4281-b951-d872f2087c98

1 files changed, 120 insertions, 0 deletions

diff --git a/chrome_frame/html_utils.h b/chrome_frame/html_utils.h
new file mode 100644
index 0000000..c670b00
--- /dev/null
+++ b/chrome_frame/html_utils.h
@@ -0,0 +1,120 @@
+// Copyright (c) 2009 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.

+

+// This file defines utility functions for working with html.

+

+#ifndef CHROME_FRAME_HTML_UTILS_H_

+#define CHROME_FRAME_HTML_UTILS_H_

+

+#include <string>

+#include <vector>

+

+#include "base/basictypes.h"

+#include "testing/gtest/include/gtest/gtest_prod.h"

+

+// Forward declarations

+class HtmlUtilUnittest;

+

+//

+// Class designed to take a string of HTML and extract from it named

+// attribute values from named tags.

+//

+// Caveat: this class currently doesn't handle multi-word UTF-16 encoded

+// characters. Doesn't handle implies that any data following such a

+// character could possibly be misinterpreted.

+//

+class HTMLScanner {

+ public:

+  typedef std::wstring::const_iterator StrPos;

+

+  // Structure maintaining const_iterators into html_string_.

+  class StringRange {

+    friend class HTMLScanner;

+   public:

+    StringRange();

+    StringRange(StrPos start, StrPos end);

+

+    bool LowerCaseEqualsASCII(const char* other) const;

+    bool Equals(const wchar_t* other) const;

+

+    // Copies the data described by StringRange into destination.

+    std::wstring Copy() const;

+

+    // If this StringRange represents a tag, this method extracts the name of

+    // the tag and sticks it in tag_name.

+    // Returns true if the tag name was successfully extracted.

+    // Returns false if this string doesn't look like a valid tag.

+    bool GetTagName(std::wstring* tag_name) const;

+

+    // From a given string range, uses a string tokenizer to extract the value

+    // of the named attribute if a simple scan finds that the attribute name is

+    // present.

+    //

+    // Returns true if the named attribute can be located and it has a value

+    // which has been placed in attribute_value.

+    //

+    // Note that the attribute value is unquoted here as well, so that

+    // GetTagAttribute(*<foo bar="baz">*, L"bar", *out_value*) will stick

+    // 'bar' in out_value and not '"bar"'.

+    //

+    // Returns false if the named attribute is not present in the tag or if it

+    // did not have a value.

+    //

+    bool GetTagAttribute(const wchar_t* attribute_name,

+                         StringRange* attribute_value) const;

+

+    // Unquotes a StringRange by removing a matching pair of either ' or "

+    // characters from the beginning and end of the string if present.

+    // Returns true if string was modified, false otherwise.

+    bool UnQuote();

+   private:

+     StrPos start_;

+     StrPos end_;

+  };

+

+  typedef std::vector<StringRange> StringRangeList;

+

+  // html_string must be a null-terminated string containing the HTML

+  // to be scanned.

+  explicit HTMLScanner(const wchar_t* html_string);

+

+  // Returns the set of ranges denoting HTML tags that match the given name.

+  // If stop_tag_name is given, then as soon as a tag with this name is

+  // encountered this method will return.

+  void GetTagsByName(const wchar_t* name, StringRangeList* tag_list,

+                     const wchar_t* stop_tag_name);

+

+ private:

+  friend class HtmlUtilUnittest;

+  FRIEND_TEST(HtmlUtilUnittest, BasicTest);

+

+  // Given html_string which represents the remaining html range, this method

+  // returns the next tag in tag and advances html_string to one character after

+  // the end of tag. This method is intended to be called repeatedly to extract

+  // all of the tags in sequence.

+  //

+  // Returns true if another tag was found and 'tag' was populated with a valid

+  // range.

+  // Returns false if we have reached the end of the html data.

+  bool NextTag(StringRange* html_string, StringRange* tag);

+

+  // Returns true if c can be found in quotes_, false otherwise

+  bool IsQuote(wchar_t c);

+

+  // Returns true if pos refers to the last character in an HTML comment in a

+  // string described by html_string, false otherwise.

+  // For example with html_string describing <!-- foo> -->, pos must refer to

+  // the last > for this method to return true.

+  bool IsHTMLCommentClose(StringRange* html_string, StrPos pos);

+

+  // We store a (CollapsedWhitespace'd) copy of the html data.

+  const std::wstring html_string_;

+

+  // Store the string of quote characters to avoid repeated construction.

+  const std::wstring quotes_;

+

+  DISALLOW_COPY_AND_ASSIGN(HTMLScanner);

+};

+

+#endif  // CHROME_FRAME_HTML_UTILS_H_