Remove url::ReadUTFChar's dependency on icu. Use base instead.

BUG=362608

Review URL: https://codereview.chromium.org/250233002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@267800 0039d316-1c4b-4281-b951-d872f2087c98

3 files changed, 30 insertions, 49 deletions

diff --git a/url/url_canon_icu.cc b/url/url_canon_icu.cc
index d53d7cc..741bed2 100644
--- a/url/url_canon_icu.cc
+++ b/url/url_canon_icu.cc
@@ -184,47 +184,4 @@ bool IDNToASCII(const base::char16* src, int src_len, CanonOutputW* output) {
   }
 }
 
-bool ReadUTFChar(const char* str, int* begin, int length,
-                 unsigned* code_point_out) {
-  int code_point;  // Avoids warning when U8_NEXT writes -1 to it.
-  U8_NEXT(str, *begin, length, code_point);
-  *code_point_out = static_cast<unsigned>(code_point);
-
-  // The ICU macro above moves to the next char, we want to point to the last
-  // char consumed.
-  (*begin)--;
-
-  // Validate the decoded value.
-  if (U_IS_UNICODE_CHAR(code_point))
-    return true;
-  *code_point_out = kUnicodeReplacementCharacter;
-  return false;
-}
-
-bool ReadUTFChar(const base::char16* str, int* begin, int length,
-                 unsigned* code_point) {
-  if (U16_IS_SURROGATE(str[*begin])) {
-    if (!U16_IS_SURROGATE_LEAD(str[*begin]) || *begin + 1 >= length ||
-        !U16_IS_TRAIL(str[*begin + 1])) {
-      // Invalid surrogate pair.
-      *code_point = kUnicodeReplacementCharacter;
-      return false;
-    } else {
-      // Valid surrogate pair.
-      *code_point = U16_GET_SUPPLEMENTARY(str[*begin], str[*begin + 1]);
-      (*begin)++;
-    }
-  } else {
-    // Not a surrogate, just one 16-bit word.
-    *code_point = str[*begin];
-  }
-
-  if (U_IS_UNICODE_CHAR(*code_point))
-    return true;
-
-  // Invalid code point.
-  *code_point = kUnicodeReplacementCharacter;
-  return false;
-}
-
 }  // namespace url
diff --git a/url/url_canon_internal.cc b/url/url_canon_internal.cc
index bbd1484..1554814c 100644
--- a/url/url_canon_internal.cc
+++ b/url/url_canon_internal.cc
@@ -2,13 +2,15 @@
 // Use of this source code is governed by a BSD-style license that can be
 // found in the LICENSE file.
 
+#include "url/url_canon_internal.h"
+
 #include <errno.h>
 #include <stdlib.h>
 
 #include <cstdio>
 #include <string>
 
-#include "url/url_canon_internal.h"
+#include "base/strings/utf_string_conversion_utils.h"
 
 namespace url {
 
@@ -245,6 +247,32 @@ void AppendStringOfType(const base::char16* source, int length,
       source, length, type, output);
 }
 
+bool ReadUTFChar(const char* str, int* begin, int length,
+                 unsigned* code_point_out) {
+  // This depends on ints and int32s being the same thing.  If they're not, it
+  // will fail to compile.
+  // TODO(mmenke):  This should probably be fixed.
+  if (!base::ReadUnicodeCharacter(str, length, begin, code_point_out) ||
+      !base::IsValidCharacter(*code_point_out)) {
+    *code_point_out = kUnicodeReplacementCharacter;
+    return false;
+  }
+  return true;
+}
+
+bool ReadUTFChar(const base::char16* str, int* begin, int length,
+                 unsigned* code_point_out) {
+  // This depends on ints and int32s being the same thing.  If they're not, it
+  // will fail to compile.
+  // TODO(mmenke):  This should probably be fixed.
+  if (!base::ReadUnicodeCharacter(str, length, begin, code_point_out) ||
+      !base::IsValidCharacter(*code_point_out)) {
+    *code_point_out = kUnicodeReplacementCharacter;
+    return false;
+  }
+  return true;
+}
+
 void AppendInvalidNarrowString(const char* spec, int begin, int end,
                                CanonOutput* output) {
   DoAppendInvalidNarrowString<char, unsigned char>(spec, begin, end, output);
diff --git a/url/url_canon_internal.h b/url/url_canon_internal.h
index e8efd82..71bfc40 100644
--- a/url/url_canon_internal.h
+++ b/url/url_canon_internal.h
@@ -148,8 +148,6 @@ extern const base::char16 kUnicodeReplacementCharacter;
 // |*begin| will be updated to point to the last character consumed so it
 // can be incremented in a loop and will be ready for the next character.
 // (for a single-byte ASCII character, it will not be changed).
-//
-// Implementation is in url_canon_icu.cc.
 URL_EXPORT bool ReadUTFChar(const char* str, int* begin, int length,
                             unsigned* code_point_out);
 
@@ -225,10 +223,8 @@ inline void AppendUTF8EscapedValue(unsigned char_value, CanonOutput* output) {
 // |*begin| will be updated to point to the last character consumed so it
 // can be incremented in a loop and will be ready for the next character.
 // (for a single-16-bit-word character, it will not be changed).
-//
-// Implementation is in url_canon_icu.cc.
 URL_EXPORT bool ReadUTFChar(const base::char16* str, int* begin, int length,
-                            unsigned* code_point);
+                            unsigned* code_point_out);
 
 // Equivalent to U16_APPEND_UNSAFE in ICU but uses our output method.
 inline void AppendUTF16Value(unsigned code_point,