Move ByteString::FromUnicode() to WideString::ToDefANSI()

Turns out that "FromUnicode" is misleading in that, on linux, it simply
removes any characters beyond 0xFF and passes the rest unchanged, so
no unicode decoding actually takes place. On Windows, it passes it into
the system function specifying FX_CODEPAGE_DefANSI, converting it into
the so-called "default ANSI code plane", passing some characters,
converting others to '?' and still others to 'A'. Either way, nothing
resembling UTF8 comes out of this, so pick a better name.

These now immediately look suspicious, so a follow-up CL will see
which ones should really be WideString::UTF8Encode() instead.

Making this a normal method on a widestring rather than a static
method on a bytestring feels more natural; this is parallel to
the UTF8Encode and UTF16LE_Encode functions.

Add a test that shows these conversions.

Change-Id: Ia7551b47199eba61b5c328a97bfe9176ac8e583c
Reviewed-on: https://pdfium-review.googlesource.com/39690
Reviewed-by: Lei Zhang <thestig@chromium.org>
Commit-Queue: Tom Sepez <tsepez@chromium.org>

5 files changed, 16 insertions, 19 deletions

diff --git a/fxjs/cfxjse_class.cpp b/fxjs/cfxjse_class.cpp
index e7e54c8867..9207fd5094 100644
--- a/fxjs/cfxjse_class.cpp
+++ b/fxjs/cfxjse_class.cpp
@@ -112,8 +112,7 @@ void DynPropGetterAdapter_MethodCallback(
     WideString err = JSFormatErrorString(pClassDescriptor->name, *szPropName,
                                          result.Error());
     v8::MaybeLocal<v8::String> str = v8::String::NewFromUtf8(
-        info.GetIsolate(), ByteString::FromUnicode(err).c_str(),
-        v8::NewStringType::kNormal);
+        info.GetIsolate(), err.ToDefANSI().c_str(), v8::NewStringType::kNormal);
     info.GetIsolate()->ThrowException(str.ToLocalChecked());
     return;
   }
diff --git a/fxjs/cjs_field.cpp b/fxjs/cjs_field.cpp
index 19e6712f2a..cd97372ad0 100644
--- a/fxjs/cjs_field.cpp
+++ b/fxjs/cjs_field.cpp
@@ -731,7 +731,7 @@ CJS_Return CJS_Field::set_border_style(CJS_Runtime* pRuntime,
   if (!m_bCanSet)
     return CJS_Return(JSMessage::kReadOnlyError);
 
-  ByteString byte_str = ByteString::FromUnicode(pRuntime->ToWideString(vp));
+  ByteString byte_str = pRuntime->ToWideString(vp).ToDefANSI();
   if (m_bDelay) {
     AddDelay_String(FP_BORDERSTYLE, byte_str);
   } else {
@@ -2002,7 +2002,7 @@ CJS_Return CJS_Field::set_text_font(CJS_Runtime* pRuntime,
 
   if (!m_bCanSet)
     return CJS_Return(JSMessage::kReadOnlyError);
-  if (ByteString::FromUnicode(pRuntime->ToWideString(vp)).IsEmpty())
+  if (pRuntime->ToWideString(vp).ToDefANSI().IsEmpty())
     return CJS_Return(JSMessage::kValueError);
   return CJS_Return();
 }
diff --git a/fxjs/cjs_global.cpp b/fxjs/cjs_global.cpp
index 567d54853c..1865f57ddb 100644
--- a/fxjs/cjs_global.cpp
+++ b/fxjs/cjs_global.cpp
@@ -232,7 +232,7 @@ CJS_Return CJS_Global::QueryProperty(const wchar_t* propname) {
 
 CJS_Return CJS_Global::DelProperty(CJS_Runtime* pRuntime,
                                    const wchar_t* propname) {
-  auto it = m_MapGlobal.find(ByteString::FromUnicode(propname));
+  auto it = m_MapGlobal.find(WideString(propname).ToDefANSI());
   if (it == m_MapGlobal.end())
     return CJS_Return(JSMessage::kUnknownProperty);
 
@@ -242,7 +242,7 @@ CJS_Return CJS_Global::DelProperty(CJS_Runtime* pRuntime,
 
 CJS_Return CJS_Global::GetProperty(CJS_Runtime* pRuntime,
                                    const wchar_t* propname) {
-  auto it = m_MapGlobal.find(ByteString::FromUnicode(propname));
+  auto it = m_MapGlobal.find(WideString(propname).ToDefANSI());
   if (it == m_MapGlobal.end())
     return CJS_Return();
 
@@ -272,7 +272,7 @@ CJS_Return CJS_Global::GetProperty(CJS_Runtime* pRuntime,
 CJS_Return CJS_Global::SetProperty(CJS_Runtime* pRuntime,
                                    const wchar_t* propname,
                                    v8::Local<v8::Value> vp) {
-  ByteString sPropName = ByteString::FromUnicode(propname);
+  ByteString sPropName = WideString(propname).ToDefANSI();
   if (vp->IsNumber()) {
     return SetGlobalVariables(sPropName, JS_GlobalDataType::NUMBER,
                               pRuntime->ToDouble(vp), false, "",
@@ -284,10 +284,9 @@ CJS_Return CJS_Global::SetProperty(CJS_Runtime* pRuntime,
                               v8::Local<v8::Object>(), false);
   }
   if (vp->IsString()) {
-    return SetGlobalVariables(
-        sPropName, JS_GlobalDataType::STRING, 0, false,
-        ByteString::FromUnicode(pRuntime->ToWideString(vp)),
-        v8::Local<v8::Object>(), false);
+    return SetGlobalVariables(sPropName, JS_GlobalDataType::STRING, 0, false,
+                              pRuntime->ToWideString(vp).ToDefANSI(),
+                              v8::Local<v8::Object>(), false);
   }
   if (vp->IsObject()) {
     return SetGlobalVariables(sPropName, JS_GlobalDataType::OBJECT, 0, false,
@@ -310,8 +309,7 @@ CJS_Return CJS_Global::setPersistent(
   if (params.size() != 2)
     return CJS_Return(JSMessage::kParamError);
 
-  auto it = m_MapGlobal.find(
-      ByteString::FromUnicode(pRuntime->ToWideString(params[0])));
+  auto it = m_MapGlobal.find(pRuntime->ToWideString(params[0]).ToDefANSI());
   if (it == m_MapGlobal.end() || it->second->bDeleted)
     return CJS_Return(JSMessage::kGlobalNotFoundError);
 
@@ -432,7 +430,7 @@ void CJS_Global::ObjectToArray(CJS_Runtime* pRuntime,
       continue;
     }
     if (v->IsString()) {
-      ByteString sValue = ByteString::FromUnicode(pRuntime->ToWideString(v));
+      ByteString sValue = pRuntime->ToWideString(v).ToDefANSI();
       CJS_KeyValue* pObjElement = new CJS_KeyValue;
       pObjElement->nType = JS_GlobalDataType::STRING;
       pObjElement->sKey = sKey;
diff --git a/fxjs/cjs_publicmethods.cpp b/fxjs/cjs_publicmethods.cpp
index a86ece3461..f2cd3ff7c1 100644
--- a/fxjs/cjs_publicmethods.cpp
+++ b/fxjs/cjs_publicmethods.cpp
@@ -327,7 +327,7 @@ v8::Local<v8::Array> CJS_PublicMethods::AF_MakeArrayFromList(
     return pRuntime->ToArray(val);
 
   WideString wsStr = pRuntime->ToWideString(val);
-  ByteString t = ByteString::FromUnicode(wsStr);
+  ByteString t = wsStr.ToDefANSI();
   const char* p = t.c_str();
 
   int nIndex = 0;
@@ -882,7 +882,7 @@ CJS_Return CJS_PublicMethods::AFNumber_Format(
     return CJS_Return(L"No event handler");
 
   WideString& Value = pEvent->Value();
-  ByteString strValue = StrTrim(ByteString::FromUnicode(Value));
+  ByteString strValue = StrTrim(Value.ToDefANSI());
   if (strValue.IsEmpty())
     return CJS_Return();
 
@@ -1078,7 +1078,7 @@ CJS_Return CJS_PublicMethods::AFPercent_Format(
     return CJS_Return(JSMessage::kBadObjectError);
 
   WideString& Value = pEvent->Value();
-  ByteString strValue = StrTrim(ByteString::FromUnicode(Value));
+  ByteString strValue = StrTrim(Value.ToDefANSI());
   if (strValue.IsEmpty())
     return CJS_Return();
 
@@ -1683,7 +1683,7 @@ CJS_Return CJS_PublicMethods::AFRange_Validate(
   if (pEvent->Value().IsEmpty())
     return CJS_Return();
 
-  double dEentValue = atof(ByteString::FromUnicode(pEvent->Value()).c_str());
+  double dEentValue = atof(pEvent->Value().ToDefANSI().c_str());
   bool bGreaterThan = pRuntime->ToBoolean(params[0]);
   double dGreaterThan = pRuntime->ToDouble(params[1]);
   bool bLessThan = pRuntime->ToBoolean(params[2]);
diff --git a/fxjs/cjs_runtime.cpp b/fxjs/cjs_runtime.cpp
index 9329a483f4..6042538e30 100644
--- a/fxjs/cjs_runtime.cpp
+++ b/fxjs/cjs_runtime.cpp
@@ -234,7 +234,7 @@ v8::Local<v8::Value> CJS_Runtime::MaybeCoerceToNumber(
     v8::Local<v8::Value> value) {
   bool bAllowNaN = false;
   if (value->IsString()) {
-    ByteString bstr = ByteString::FromUnicode(ToWideString(value));
+    ByteString bstr = ToWideString(value).ToDefANSI();
     if (bstr.GetLength() == 0)
       return value;
     if (bstr == "NaN")