fune/parser/html/nsHtml5String.cpp
Henri Sivonen 3edc601325 Bug 1402247 - Use encoding_rs for XPCOM string encoding conversions. r=Nika,erahm,froydnj.
Correctness improvements:

 * UTF errors are handled safely per spec instead of dangerously truncating
   strings.

 * There are fewer converter implementations.

Performance improvements:

 * The old code did exact buffer length math, which meant doing UTF math twice
   on each input string (once for length calculation and another time for
   conversion). Exact length math is more complicated when handling errors
   properly, which the old code didn't do. The new code does UTF math on the
   string content only once (when converting) but risks allocating more than
   once. There are heuristics in place to lower the probability of
   reallocation in cases where the double math avoidance isn't enough of a
   saving to absorb an allocation and memcpy.

 * Previously, in UTF-16 <-> UTF-8 conversions, an ASCII prefix was optimized
   but a single non-ASCII code point pessimized the rest of the string. The
   new code tries to get back on the fast ASCII path.

 * UTF-16 to Latin1 conversion guarantees less about handling of out-of-range
   input to eliminate an operation from the inner loop on x86/x86_64.

 * When assigning to a pre-existing string, the new code tries to reuse the
   old buffer instead of first releasing the old buffer and then allocating a
   new one.

 * When reallocating from the new code, the memcpy covers only the data that
   is part of the logical length of the old string instead of memcpying the
   whole capacity. (For old callers old excess memcpy behavior is preserved
   due to bogus callers. See bug 1472113.)

 * UTF-8 strings in XPConnect that are in the Latin1 range are passed to
   SpiderMonkey as Latin1.

New features:

 * Conversion between UTF-8 and Latin1 is added in order to enable faster
   future interop between Rust code (or otherwise UTF-8-using code) and text
   node and SpiderMonkey code that uses Latin1.

MozReview-Commit-ID: JaJuExfILM9
2018-08-14 14:43:42 +03:00

215 lines
5.6 KiB
C++

/* This Source Code Form is subject to the terms of the Mozilla Public
* License, v. 2.0. If a copy of the MPL was not distributed with this
* file, You can obtain one at http://mozilla.org/MPL/2.0/. */
#include "nsHtml5String.h"
#include "nsCharTraits.h"
#include "nsHtml5TreeBuilder.h"
#include "nsUTF8Utils.h"
void
nsHtml5String::ToString(nsAString& aString)
{
switch (GetKind()) {
case eStringBuffer:
return AsStringBuffer()->ToString(Length(), aString);
case eAtom:
return AsAtom()->ToString(aString);
case eEmpty:
aString.Truncate();
return;
default:
aString.Truncate();
aString.SetIsVoid(true);
return;
}
}
void
nsHtml5String::CopyToBuffer(char16_t* aBuffer) const
{
memcpy(aBuffer, AsPtr(), Length() * sizeof(char16_t));
}
bool
nsHtml5String::LowerCaseEqualsASCII(const char* aLowerCaseLiteral) const
{
return !nsCharTraits<char16_t>::compareLowerCaseToASCIINullTerminated(
AsPtr(), Length(), aLowerCaseLiteral);
}
bool
nsHtml5String::EqualsASCII(const char* aLiteral) const
{
return !nsCharTraits<char16_t>::compareASCIINullTerminated(
AsPtr(), Length(), aLiteral);
}
bool
nsHtml5String::LowerCaseStartsWithASCII(const char* aLowerCaseLiteral) const
{
const char* litPtr = aLowerCaseLiteral;
const char16_t* strPtr = AsPtr();
const char16_t* end = strPtr + Length();
char16_t litChar;
while ((litChar = *litPtr)) {
MOZ_ASSERT(!(litChar >= 'A' && litChar <= 'Z'),
"Literal isn't in lower case.");
if (strPtr == end) {
return false;
}
char16_t strChar = *strPtr;
if (strChar >= 'A' && strChar <= 'Z') {
strChar += 0x20;
}
if (litChar != strChar) {
return false;
}
++litPtr;
++strPtr;
}
return true;
}
bool
nsHtml5String::Equals(nsHtml5String aOther) const
{
MOZ_ASSERT(operator bool());
MOZ_ASSERT(aOther);
if (Length() != aOther.Length()) {
return false;
}
return !memcmp(AsPtr(), aOther.AsPtr(), Length() * sizeof(char16_t));
}
nsHtml5String
nsHtml5String::Clone()
{
switch (GetKind()) {
case eStringBuffer:
AsStringBuffer()->AddRef();
break;
case eAtom:
AsAtom()->AddRef();
break;
default:
break;
}
return nsHtml5String(mBits);
}
void
nsHtml5String::Release()
{
switch (GetKind()) {
case eStringBuffer:
AsStringBuffer()->Release();
break;
case eAtom:
AsAtom()->Release();
break;
default:
break;
}
mBits = eNull;
}
// static
nsHtml5String
nsHtml5String::FromBuffer(char16_t* aBuffer,
int32_t aLength,
nsHtml5TreeBuilder* aTreeBuilder)
{
if (!aLength) {
return nsHtml5String(eEmpty);
}
// Work with nsStringBuffer directly to make sure that storage is actually
// nsStringBuffer and to make sure the allocation strategy matches
// nsAttrValue::GetStringBuffer, so that it doesn't need to reallocate and
// copy.
RefPtr<nsStringBuffer> buffer(
nsStringBuffer::Alloc((aLength + 1) * sizeof(char16_t)));
if (!buffer) {
if (!aTreeBuilder) {
MOZ_CRASH("Out of memory.");
}
aTreeBuilder->MarkAsBroken(NS_ERROR_OUT_OF_MEMORY);
buffer = nsStringBuffer::Alloc(2 * sizeof(char16_t));
if (!buffer) {
MOZ_CRASH(
"Out of memory so badly that couldn't even allocate placeholder.");
}
char16_t* data = reinterpret_cast<char16_t*>(buffer->Data());
data[0] = 0xFFFD;
data[1] = 0;
return nsHtml5String(reinterpret_cast<uintptr_t>(buffer.forget().take()) |
eStringBuffer);
}
char16_t* data = reinterpret_cast<char16_t*>(buffer->Data());
memcpy(data, aBuffer, aLength * sizeof(char16_t));
data[aLength] = 0;
return nsHtml5String(reinterpret_cast<uintptr_t>(buffer.forget().take()) |
eStringBuffer);
}
// static
nsHtml5String
nsHtml5String::FromLiteral(const char* aLiteral)
{
size_t length = std::strlen(aLiteral);
if (!length) {
return nsHtml5String(eEmpty);
}
// Work with nsStringBuffer directly to make sure that storage is actually
// nsStringBuffer and to make sure the allocation strategy matches
// nsAttrValue::GetStringBuffer, so that it doesn't need to reallocate and
// copy.
RefPtr<nsStringBuffer> buffer(
nsStringBuffer::Alloc((length + 1) * sizeof(char16_t)));
if (!buffer) {
MOZ_CRASH("Out of memory.");
}
char16_t* data = reinterpret_cast<char16_t*>(buffer->Data());
ConvertLatin1toUTF16(MakeSpan(aLiteral, length), MakeSpan(data, length));
data[length] = 0;
return nsHtml5String(reinterpret_cast<uintptr_t>(buffer.forget().take()) |
eStringBuffer);
}
// static
nsHtml5String
nsHtml5String::FromString(const nsAString& aString)
{
auto length = aString.Length();
if (!length) {
return nsHtml5String(eEmpty);
}
RefPtr<nsStringBuffer> buffer = nsStringBuffer::FromString(aString);
if (buffer && (length == buffer->StorageSize() / sizeof(char16_t) - 1)) {
return nsHtml5String(reinterpret_cast<uintptr_t>(buffer.forget().take()) |
eStringBuffer);
}
buffer = nsStringBuffer::Alloc((length + 1) * sizeof(char16_t));
if (!buffer) {
MOZ_CRASH("Out of memory.");
}
char16_t* data = reinterpret_cast<char16_t*>(buffer->Data());
memcpy(data, aString.BeginReading(), length * sizeof(char16_t));
data[length] = 0;
return nsHtml5String(reinterpret_cast<uintptr_t>(buffer.forget().take()) |
eStringBuffer);
}
// static
nsHtml5String
nsHtml5String::FromAtom(already_AddRefed<nsAtom> aAtom)
{
return nsHtml5String(reinterpret_cast<uintptr_t>(aAtom.take()) | eAtom);
}
// static
nsHtml5String
nsHtml5String::EmptyString()
{
return nsHtml5String(eEmpty);
}