Updated *.pak file format to support both UTF8 and UTF16

ui/base/resource/data_pack.cc

Index: ui/base/resource/data_pack.cc
diff --git a/ui/base/resource/data_pack.cc b/ui/base/resource/data_pack.cc
index 1fcdfe1dff58a8629bf7d0858b2f512680562872..34c8892a11d1c7913e836b7f8abcfc41ee8bf88d 100644
--- a/ui/base/resource/data_pack.cc
+++ b/ui/base/resource/data_pack.cc
@@ -17,9 +17,9 @@
 
 namespace {
 
-static const uint32 kFileFormatVersion = 3;
-// Length of file header: version and entry count.
-static const size_t kHeaderLength = 2 * sizeof(uint32);
+static const uint32 kFileFormatVersion = 4;
+// Length of file header: version, entry count and text encoding type.
+static const size_t kHeaderLength = 2 * sizeof(uint32) + sizeof(uint8);
 
 #pragma pack(push,2)
 struct DataPackEntry {
@@ -60,7 +60,7 @@ enum LoadErrors {
 namespace ui {
 
 // In .cc for MemoryMappedFile dtor.
-DataPack::DataPack() : resource_count_(0) {
+DataPack::DataPack() : resource_count_(0), text_encoding_type_(BINARY) {
 }
 DataPack::~DataPack() {
 }
@@ -76,7 +76,7 @@ bool DataPack::Load(const FilePath& path) {
   }
 
   // Parse the header of the file.

[tony, 2011/08/30 21:19:58]

We should probably make sure that kHeaderLength <= mmap_->length() here (a bug
in the existing code).

[marcvs, 2011/08/30 22:17:22]

Done.

-  // First uint32: version; second: resource count.
+  // First uint32: version; second: resource count;
   const uint32* ptr = reinterpret_cast<const uint32*>(mmap_->data());
   uint32 version = ptr[0];
   if (version != kFileFormatVersion) {
@@ -89,6 +89,17 @@ bool DataPack::Load(const FilePath& path) {
   }
   resource_count_ = ptr[1];
 
+  // third: text encoding.
+  const uint8* ptr_encoding = reinterpret_cast<const uint8*>(ptr + 2);
+  text_encoding_type_ = (TextEncodingType)(*ptr_encoding);

[tony, 2011/08/30 21:19:58]

Nit: static_cast<TextEncodingType>

[marcvs, 2011/08/30 22:17:22]

Done.

+  if (text_encoding_type_ != UTF8 && text_encoding_type_ != UTF16 &&
+      text_encoding_type_ != BINARY) {
+    LOG(ERROR) << "Bad data pack text encoding: got " << text_encoding_type_
+               << ", expected between " << BINARY << " and " << UTF16;
+    mmap_.reset();
+    return false;
+  }
+
   // Sanity check the file.
   // 1) Check we have enough entries.
   if (kHeaderLength + resource_count_ * sizeof(DataPackEntry) >
@@ -156,7 +167,8 @@ RefCountedStaticMemory* DataPack::GetStaticMemory(uint16 resource_id) const {
 
 // static
 bool DataPack::WritePack(const FilePath& path,
-                         const std::map<uint16, base::StringPiece>& resources) {
+                         const std::map<uint16, base::StringPiece>& resources,
+                         TextEncodingType textEncodingType) {
   FILE* file = file_util::OpenFile(path, "wb");
   if (!file)
     return false;
@@ -176,6 +188,21 @@ bool DataPack::WritePack(const FilePath& path,
     return false;
   }
 
+  if (textEncodingType != UTF8 && textEncodingType != UTF16 &&
+      textEncodingType != BINARY) {
+    LOG(ERROR) << "Invalid text encoding type, got " << textEncodingType
+               << ", expected between " << BINARY << " and " << UTF16;
+    file_util::CloseFile(file);
+    return false;
+  }
+
+  uint8 write_buffer = textEncodingType;
+  if (fwrite(&write_buffer, sizeof(uint8), 1, file) != 1) {
+    LOG(ERROR) << "Failed to write file text resources encoding";
+    file_util::CloseFile(file);
+    return false;
+  }
+
   // Each entry is a uint16 + a uint32. We have an extra entry after the last
   // item so we can compute the size of the list item.
   uint32 index_length = (entry_count + 1) * sizeof(DataPackEntry);