RGBCube/serenity
1
Fork 0
mirror of https://github.com/RGBCube/serenity synced 2025-05-31 04:38:11 +00:00
Code Issues Activity Actions

LibJS: Pass Interpreter& to Value::to_number() et al.

Browse source 
This patch is unfortunately rather large and might make some things feel
bloated, but it is necessary to fix a few flaws in LibJS, primarily
blindly coercing values to numbers without exception checks - i.e.

interpreter.argument(0).to_i32();  // can fail!!!

Some examples where the interpreter would actually crash:

var o = { toString: () => { throw Error() } };
+o;
o - 1;
"foo".charAt(o);
"bar".repeat(o);

To fix this, we now have the following...

to_double(Interpreter&)
to_i32()
to_i32(Interpreter&)
to_size_t()
to_size_t(Interpreter&)

...and a whole lot of exception checking.

There's intentionally no to_double(), use as_double() directly instead.

This way we still can use these convenient utility functions but don't
need to check for exceptions if we are sure the value already is a
number.

Fixes #2267.
This commit is contained in:
Linus Groh 2020-05-18 00:28:00 +01:00 • committed by Andreas Kling
parent 1a1394f7a2
commit 476094922b
17 changed files with 491 additions and 187 deletions
Download patch file Download diff file Expand all files Collapse all files

59
Libraries/LibJS/Runtime/MathObject.cpp
View file

@ -67,7 +67,9 @@ MathObject::~MathObject()
Value MathObject::abs(Interpreter& interpreter)
{
auto number = interpreter.argument(0).to_number();
auto number = interpreter.argument(0).to_number(interpreter);
if (interpreter.exception())
return {};
if (number.is_nan())
return js_nan();
return Value(number.as_double() >= 0 ? number.as_double() : -number.as_double());
@ -85,7 +87,9 @@ Value MathObject::random(Interpreter&)
Value MathObject::sqrt(Interpreter& interpreter)
{
auto number = interpreter.argument(0).to_number();
auto number = interpreter.argument(0).to_number(interpreter);
if (interpreter.exception())
return {};
if (number.is_nan())
return js_nan();
return Value(::sqrt(number.as_double()));
@ -93,7 +97,9 @@ Value MathObject::sqrt(Interpreter& interpreter)
Value MathObject::floor(Interpreter& interpreter)
{
auto number = interpreter.argument(0).to_number();
auto number = interpreter.argument(0).to_number(interpreter);
if (interpreter.exception())
return {};
if (number.is_nan())
return js_nan();
return Value(::floor(number.as_double()));
@ -101,7 +107,9 @@ Value MathObject::floor(Interpreter& interpreter)
Value MathObject::ceil(Interpreter& interpreter)
{
auto number = interpreter.argument(0).to_number();
auto number = interpreter.argument(0).to_number(interpreter);
if (interpreter.exception())
return {};
if (number.is_nan())
return js_nan();
return Value(::ceil(number.as_double()));
@ -109,7 +117,9 @@ Value MathObject::ceil(Interpreter& interpreter)
Value MathObject::round(Interpreter& interpreter)
{
auto number = interpreter.argument(0).to_number();
auto number = interpreter.argument(0).to_number(interpreter);
if (interpreter.exception())
return {};
if (number.is_nan())
return js_nan();
return Value(::round(number.as_double()));
@ -120,12 +130,13 @@ Value MathObject::max(Interpreter& interpreter)
if (!interpreter.argument_count())
return js_negative_infinity();
if (interpreter.argument_count() == 1)
return interpreter.argument(0).to_number();
Value max = interpreter.argument(0).to_number();
auto max = interpreter.argument(0).to_number(interpreter);
if (interpreter.exception())
return {};
for (size_t i = 1; i < interpreter.argument_count(); ++i) {
Value cur = interpreter.argument(i).to_number();
auto cur = interpreter.argument(i).to_number(interpreter);
if (interpreter.exception())
return {};
max = Value(cur.as_double() > max.as_double() ? cur : max);
}
return max;
@ -136,12 +147,13 @@ Value MathObject::min(Interpreter& interpreter)
if (!interpreter.argument_count())
return js_infinity();
if (interpreter.argument_count() == 1)
return interpreter.argument(0).to_number();
Value min = interpreter.argument(0).to_number();
auto min = interpreter.argument(0).to_number(interpreter);
if (interpreter.exception())
return {};
for (size_t i = 1; i < interpreter.argument_count(); ++i) {
Value cur = interpreter.argument(i).to_number();
auto cur = interpreter.argument(i).to_number(interpreter);
if (interpreter.exception())
return {};
min = Value(cur.as_double() < min.as_double() ? cur : min);
}
return min;
@ -149,10 +161,11 @@ Value MathObject::min(Interpreter& interpreter)
Value MathObject::trunc(Interpreter& interpreter)
{
auto number = interpreter.argument(0).to_number();
auto number = interpreter.argument(0).to_number(interpreter);
if (interpreter.exception())
return {};
if (number.is_nan())
return js_nan();
if (number.as_double() < 0)
return MathObject::ceil(interpreter);
return MathObject::floor(interpreter);
@ -160,7 +173,9 @@ Value MathObject::trunc(Interpreter& interpreter)
Value MathObject::sin(Interpreter& interpreter)
{
auto number = interpreter.argument(0).to_number();
auto number = interpreter.argument(0).to_number(interpreter);
if (interpreter.exception())
return {};
if (number.is_nan())
return js_nan();
return Value(::sin(number.as_double()));
@ -168,7 +183,9 @@ Value MathObject::sin(Interpreter& interpreter)
Value MathObject::cos(Interpreter& interpreter)
{
auto number = interpreter.argument(0).to_number();
auto number = interpreter.argument(0).to_number(interpreter);
if (interpreter.exception())
return {};
if (number.is_nan())
return js_nan();
return Value(::cos(number.as_double()));
@ -176,7 +193,9 @@ Value MathObject::cos(Interpreter& interpreter)
Value MathObject::tan(Interpreter& interpreter)
{
auto number = interpreter.argument(0).to_number();
auto number = interpreter.argument(0).to_number(interpreter);
if (interpreter.exception())
return {};
if (number.is_nan())
return js_nan();
return Value(::tan(number.as_double()));