Consolidate most of net/'s icu_string+conversions dependencies.

TBR=brettw@chromium.org
BUG=362608

Review URL: https://codereview.chromium.org/252733002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266957 0039d316-1c4b-4281-b951-d872f2087c98

net/DEPS

@@ -23,7 +23,7 @@ specific_include_rules = {
 
   # Within net, only used by file: requests.
   "filename_util\.cc": [
-    "+base/i18n",
+    "+base/i18n/file_util_icu.h",
   ],
   
   # Functions largely not used by the rest of net.
@@ -37,26 +37,17 @@ specific_include_rules = {
     "+base/i18n",
   ],
 
-  # Only use icu for string conversions.
-  "escape_unittest\.cc": [
-    "+base/i18n/icu_string_conversions.h",
-  ],
-  "http_auth_handler_basic\.cc": [
-    "+base/i18n/icu_string_conversions.h",
-  ],
-  "http_auth_handler_digest\.cc": [
+  # Consolidated string functions that depend on icu.
+  "net_string_util_icu\.cc": [
+    "+base/i18n/i18n_constants.h",
     "+base/i18n/icu_string_conversions.h",
+    "+third_party/icu/source/common/unicode/ucnv.h"
   ],
+
+  # Only use icu for string conversions.
   "proxy_script_fetcher_impl\.cc": [
     "+base/i18n/icu_string_conversions.h",
   ],
-  "x509_cert_types_mac\.cc": [
-    "+base/i18n/icu_string_conversions.h",
-  ],
-  "http_content_disposition\.cc": [
-    "+base/i18n/icu_string_conversions.h",
-    "+third_party/icu",
-  ],
   "websocket_channel\.h": [
     "+base/i18n",
   ],

net/base/escape_unittest.cc

@@ -8,7 +8,6 @@
 #include "net/base/escape.h"
 
 #include "base/basictypes.h"
-#include "base/i18n/icu_string_conversions.h"
 #include "base/strings/string_util.h"
 #include "base/strings/stringprintf.h"
 #include "base/strings/utf_string_conversions.h"

net/base/filename_util.cc

@@ -7,7 +7,6 @@
 #include "base/file_util.h"
 #include "base/files/file_path.h"
 #include "base/i18n/file_util_icu.h"
-#include "base/i18n/icu_string_conversions.h"
 #include "base/path_service.h"
 #include "base/strings/string_util.h"
 #include "base/strings/sys_string_conversions.h"
@@ -15,6 +14,7 @@
 #include "base/threading/thread_restrictions.h"
 #include "net/base/escape.h"
 #include "net/base/mime_util.h"
+#include "net/base/net_string_util.h"
 #include "net/http/http_content_disposition.h"
 #include "url/gurl.h"
 
@@ -77,9 +77,8 @@ std::string GetFileNameFromURL(const GURL& url,
     // encoding detection.
     base::string16 utf16_output;
     if (!referrer_charset.empty() &&
-        base::CodepageToUTF16(unescaped_url_filename,
+        net::ConvertToUTF16(unescaped_url_filename,
                             referrer_charset.c_str(),
-                              base::OnStringConversionError::FAIL,
                             &utf16_output)) {
       decoded_filename = base::UTF16ToUTF8(utf16_output);
     } else {

net/base/net_string_util.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 NET_BASE_NET_STRING_UTIL_H__
+#define NET_BASE_NET_STRING_UTIL_H__
+
+#include <string>
+
+#include "base/strings/string16.h"
+
+// String conversion functions.  By default, they're implemented with ICU, but
+// when building with USE_ICU_ALTERNATIVES, they use platform functions instead.
+
+namespace net {
+
+// Converts |text| using |charset| to UTF-8, and writes it to |output|.
+// On failure, returns false and |output| is cleared.
+bool ConvertToUtf8(const std::string& text, const char* charset,
+                   std::string* output);
+
+// Converts |text| using |charset| to UTF-8, normalizes the result, and writes
+// it to |output|.  On failure, returns false and |output| is cleared.
+bool ConvertToUtf8AndNormalize(const std::string& text, const char* charset,
+                               std::string* output);
+
+// Converts |text| encoded in Latin-1 to UTF-8, normalizes the result, and
+// writes it to |output|.  On failure, returns false and |output| is cleared.
+bool ConvertLatin1ToUtf8AndNormalize(const std::string& text,
+                                     std::string* output);
+
+// Converts |text| using |charset| to UTF-16, and writes it to |output|.
+// On failure, returns false and |output| is cleared.
+bool ConvertToUTF16(const std::string& text, const char* charset,
+                    base::string16* output);
+
+// Converts |text| encoded in Latin-1 to UTF-16, and writes it to |output|.
+// On failure, returns false and |output| is cleared.
+bool ConvertLatin1ToUTF16(const std::string& text, base::string16* output);
+
+}  // namespace net
+
+#endif  // NET_BASE_NET_STRING_UTIL_H__

net/base/net_string_util_icu.cc

+// 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.
+
+#include "net/base/net_string_util.h"
+
+#include "base/i18n/i18n_constants.h"
+#include "base/i18n/icu_string_conversions.h"
+#include "base/strings/string_util.h"
+#include "third_party/icu/source/common/unicode/ucnv.h"
+
+namespace net {
+
+bool ConvertToUtf8(const std::string& text, const char* charset,
+                   std::string* output) {
+  output->clear();
+
+  UErrorCode err = U_ZERO_ERROR;
+  UConverter* converter(ucnv_open(charset, &err));
+  if (U_FAILURE(err))
+    return false;
+
+  // A single byte in a legacy encoding can be expanded to 3 bytes in UTF-8.
+  // A 'two-byte character' in a legacy encoding can be expanded to 4 bytes
+  // in UTF-8. Therefore, the expansion ratio is 3 at most. Add one for a
+  // trailing '\0'.
+  size_t output_length = text.length() * 3 + 1;
+  char* buf = WriteInto(output, output_length);
+  output_length = ucnv_toAlgorithmic(UCNV_UTF8, converter, buf, output_length,
+                                     text.data(), text.length(), &err);
+  ucnv_close(converter);
+  if (U_FAILURE(err)) {
+    output->clear();
+    return false;
+  }
+
+  output->resize(output_length);
+  return true;
+}
+
+bool ConvertToUtf8AndNormalize(const std::string& text, const char* charset,
+                               std::string* output) {
+  return base::ConvertToUtf8AndNormalize(text,  charset, output);
+}
+
+bool ConvertLatin1ToUtf8AndNormalize(const std::string& text,
+                                    std::string* output) {
+  return net::ConvertToUtf8AndNormalize(text,  base::kCodepageLatin1, output);
+}
+
+bool ConvertToUTF16(const std::string& text, const char* charset,
+                    base::string16* output) {
+  return base::CodepageToUTF16(text, charset,
+                               base::OnStringConversionError::FAIL, output);
+}
+
+bool ConvertLatin1ToUTF16(const std::string& text, base::string16* output) {
+  return base::CodepageToUTF16(text, base::kCodepageLatin1,
+                               base::OnStringConversionError::FAIL, output);
+}
+
+}  // namespace net

net/cert/x509_cert_types_mac.cc

@@ -10,10 +10,10 @@
 #include <Security/SecAsn1Coder.h>
 #include <Security/Security.h>
 
-#include "base/i18n/icu_string_conversions.h"
 #include "base/logging.h"
 #include "base/mac/mac_logging.h"
 #include "base/strings/utf_string_conversions.h"
+#include "net/base/net_string_util.h"
 
 namespace net {
 
@@ -115,8 +115,7 @@ std::string DataToString(CSSM_DATA data) {
 // Converts raw CSSM_DATA in ISO-8859-1 to a std::string in UTF-8.
 std::string Latin1DataToUTF8String(CSSM_DATA data) {
   base::string16 utf16;
-  if (!CodepageToUTF16(DataToString(data), base::kCodepageLatin1,
-                       base::OnStringConversionError::FAIL, &utf16))
+  if (!ConvertLatin1ToUTF16(DataToString(data), &utf16))
     return "";
   return base::UTF16ToUTF8(utf16);
 }

net/http/http_auth_handler_basic.cc

@@ -7,10 +7,10 @@
 #include <string>
 
 #include "base/base64.h"
-#include "base/i18n/icu_string_conversions.h"
 #include "base/strings/string_util.h"
 #include "base/strings/utf_string_conversions.h"
 #include "net/base/net_errors.h"
+#include "net/base/net_string_util.h"
 #include "net/http/http_auth.h"
 #include "net/http/http_auth_challenge_tokenizer.h"
 
@@ -43,8 +43,7 @@ bool ParseRealm(const HttpAuthChallengeTokenizer& tokenizer,
     if (!LowerCaseEqualsASCII(parameters.name(), "realm"))
       continue;
 
-    if (!base::ConvertToUtf8AndNormalize(
-            parameters.value(), base::kCodepageLatin1, realm))
+    if (!net::ConvertLatin1ToUtf8AndNormalize(parameters.value(), realm))
       return false;
   }
   return parameters.valid();

net/http/http_auth_handler_digest.cc

@@ -6,7 +6,6 @@
 
 #include <string>
 
-#include "base/i18n/icu_string_conversions.h"
 #include "base/logging.h"
 #include "base/md5.h"
 #include "base/rand_util.h"
@@ -14,6 +13,7 @@
 #include "base/strings/stringprintf.h"
 #include "base/strings/utf_string_conversions.h"
 #include "net/base/net_errors.h"
+#include "net/base/net_string_util.h"
 #include "net/base/net_util.h"
 #include "net/http/http_auth.h"
 #include "net/http/http_auth_challenge_tokenizer.h"
@@ -227,7 +227,7 @@ bool HttpAuthHandlerDigest::ParseChallengeProperty(const std::string& name,
                                                    const std::string& value) {
   if (LowerCaseEqualsASCII(name, "realm")) {
     std::string realm;
-    if (!base::ConvertToUtf8AndNormalize(value, base::kCodepageLatin1, &realm))
+    if (!net::ConvertLatin1ToUtf8AndNormalize(value, &realm))
       return false;
     realm_ = realm;
     original_realm_ = value;

net/http/http_content_disposition.cc

@@ -5,15 +5,14 @@
 #include "net/http/http_content_disposition.h"
 
 #include "base/base64.h"
-#include "base/i18n/icu_string_conversions.h"
 #include "base/logging.h"
 #include "base/strings/string_tokenizer.h"
 #include "base/strings/string_util.h"
 #include "base/strings/sys_string_conversions.h"
 #include "base/strings/utf_string_conversions.h"
+#include "net/base/net_string_util.h"
 #include "net/base/net_util.h"
 #include "net/http/http_util.h"
-#include "third_party/icu/source/common/unicode/ucnv.h"
 
 namespace {
 
@@ -65,32 +64,16 @@ bool DecodeBQEncoding(const std::string& part,
                       std::string* output) {
   std::string decoded;
   if (!((enc_type == B_ENCODING) ?
-        base::Base64Decode(part, &decoded) : DecodeQEncoding(part, &decoded)))
+        base::Base64Decode(part, &decoded) : DecodeQEncoding(part, &decoded))) {
     return false;
+  }
 
   if (decoded.empty()) {
     output->clear();
     return true;
   }
 
-  UErrorCode err = U_ZERO_ERROR;
-  UConverter* converter(ucnv_open(charset.c_str(), &err));
-  if (U_FAILURE(err))
-    return false;
-
-  // A single byte in a legacy encoding can be expanded to 3 bytes in UTF-8.
-  // A 'two-byte character' in a legacy encoding can be expanded to 4 bytes
-  // in UTF-8. Therefore, the expansion ratio is 3 at most. Add one for a
-  // trailing '\0'.
-  size_t output_length = decoded.length() * 3 + 1;
-  char* buf = WriteInto(output, output_length);
-  output_length = ucnv_toAlgorithmic(UCNV_UTF8, converter, buf, output_length,
-                                     decoded.data(), decoded.length(), &err);
-  ucnv_close(converter);
-  if (U_FAILURE(err))
-    return false;
-  output->resize(output_length);
-  return true;
+  return net::ConvertToUtf8(decoded, charset.c_str(), output);
 }
 
 bool DecodeWord(const std::string& encoded_word,
@@ -110,8 +93,7 @@ bool DecodeWord(const std::string& encoded_word,
     } else {
       base::string16 utf16_output;
       if (!referrer_charset.empty() &&
-          base::CodepageToUTF16(encoded_word, referrer_charset.c_str(),
-                                base::OnStringConversionError::FAIL,
+          net::ConvertToUTF16(encoded_word, referrer_charset.c_str(),
                               &utf16_output)) {
         *output = base::UTF16ToUTF8(utf16_output);
       } else {
@@ -343,7 +325,7 @@ bool DecodeExtValue(const std::string& param_value, std::string* decoded) {
   std::string unescaped = net::UnescapeURLComponent(
       value, net::UnescapeRule::SPACES | net::UnescapeRule::URL_SPECIAL_CHARS);
 
-  return base::ConvertToUtf8AndNormalize(unescaped, charset, decoded);
+  return net::ConvertToUtf8AndNormalize(unescaped, charset.c_str(), decoded);
 }
 
 } // namespace

net/net.gypi

@@ -44,6 +44,8 @@
       'base/net_log_source_type_list.h',
       'base/net_module.cc',
       'base/net_module.h',
+      'base/net_string_util.h',
+      'base/net_string_util_icu.cc',
       'base/net_util.cc',
       'base/net_util.h',
       'base/net_util_icu.cc',