Add option to build googleurl as dll...

src/url_util.cc

 // Copyright 2007, Google Inc.
 // All rights reserved.
 //
 // Redistribution and use in source and binary forms, with or without
 // modification, are permitted provided that the following conditions are
 // met:
 //
 //     * Redistributions of source code must retain the above copyright
 // notice, this list of conditions and the following disclaimer.
 //     * Redistributions in binary form must reproduce the above
@@ skipping 15 matching lines @@
 // THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
 // (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
 // OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 
 #include <string.h>
 #include <vector>
 
 #include "googleurl/src/url_util.h"
 
 #include "base/logging.h"
+#include "googleurl/src/url_canon_internal.h"
 #include "googleurl/src/url_file.h"
 
 namespace url_util {
 
 namespace {
 
 // ASCII-specific tolower.  The standard library's tolower is locale sensitive,
 // so we don't want to use it here.
 template <class Char> inline Char ToLowerASCII(Char c) {
   return (c >= 'A' && c <= 'Z') ? (c + ('a' - 'A')) : c;
@@ skipping 447 matching lines @@
   }
   return a_begin == a_end && b_begin == b_end;
 }
 
 bool LowerCaseEqualsASCII(const char16* a_begin,
                           const char16* a_end,
                           const char* b) {
   return DoLowerCaseEqualsASCII(a_begin, a_end, b);
 }
 
+void DecodeURLEscapeSequences(const char* input, int length,
+                              url_canon::CanonOutputW* output) {
+  url_canon::RawCanonOutputT<char> unescaped_chars;
+  for (int i = 0; i < length; i++) {
+    if (input[i] == '%') {
+      unsigned char ch;
+      if (url_canon::DecodeEscaped(input, &i, length, &ch)) {
+        unescaped_chars.push_back(ch);
+      } else {
+        // Invalid escape sequence, copy the percent literal.
+        unescaped_chars.push_back('%');
+      }
+    } else {
+      // Regular non-escaped 8-bit character.
+      unescaped_chars.push_back(input[i]);
+    }
+  }
+
+  // Convert that 8-bit to UTF-16. It's not clear IE does this at all to
+  // JavaScript URLs, but Firefox and Safari do.
+  for (int i = 0; i < unescaped_chars.length(); i++) {
+    unsigned char uch = static_cast<unsigned char>(unescaped_chars.at(i));
+    if (uch < 0x80) {
+      // Non-UTF-8, just append directly
+      output->push_back(uch);
+    } else {
+      // next_ch will point to the last character of the decoded
+      // character.
+      int next_character = i;
+      unsigned code_point;
+      if (url_canon::ReadUTFChar(unescaped_chars.data(), &next_character,
+                                 unescaped_chars.length(), &code_point)) {
+        // Valid UTF-8 character, convert to UTF-16.
+        url_canon::AppendUTF16Value(code_point, output);
+        i = next_character;
+      } else {
+        // If there are any sequences that are not valid UTF-8, we keep
+        // invalid code points and promote to UTF-16. We copy all characters
+        // from the current position to the end of the identified sequence.
+        while (i < next_character) {
+          output->push_back(static_cast<unsigned char>(unescaped_chars.at(i)));
+          i++;
+        }
+        output->push_back(static_cast<unsigned char>(unescaped_chars.at(i)));
+      }
+    }
+  }
+}
+
 }  // namespace url_util