According to ES2015, functions created using the FunctionDeclaration
syntax may define "own" properties named "caller" and "arguments" as
long as this is done using the AddRestrictedFunctionProperties abstract
operation.
Remove the assertions that prohibit this extension.
Because implementations are free to select any valid NaN value during
GetValueFromBuffer, tests concerning semantics for consistent NaN value
encoding cannot rely on values returned from that abstract operation.
Update the test for `%TypedArray%.prototype.map` to set the same NaN
values via `map` as set in the "control" array.
Ensure that NaN values are canonicalized consistently by all invocations
of SetValueInBuffer. Also ensure that `%TypedArray%.prototype.set` and
`%TypedArray%.prototype.slice` preserve the bit-level encoding of the
source data.
Use a set of experimentally-derived expressions known to produce NaN
values with distinct bit patterns in various platforms.
The RegExpBuiltinExec internal operation was modified in March of 2016
[1]: instead of referencing the `global` and `sticky` properties of the
"this" value, the algorithm now infers those values from the object's
[[OriginalFlags]] internal slot.
This change invalidated a number of tests. In cases where the change
resulted in an observable behavior, update the tests to assert the
latest specification text. In cases where the change removed a
previously-observable behavior, remove the files completely.
Specification text change set:
> 1. Assert: Type(_S_) is String.
> 1. Let _length_ be the number of code units in _S_.
> 1. Let _lastIndex_ be ? ToLength(? Get(_R_, `"lastIndex"`)).
> - 1. Let _global_ be ToBoolean(? Get(_R_, `"global"`)).
> - 1. Let _sticky_ be ToBoolean(? Get(_R_, `"sticky"`)).
> + 1. Let _flags_ be the value of _R_'s [[OriginalFlags]] internal slot.
> + 1. If _flags_ contains `"g"`, let _global_ be *true*, else let _global_ be *false*.
> + 1. If _flags_ contains `"y"`, let _sticky_ be *true*, else let _sticky_ be *false*.
> 1. If _global_ is *false* and _sticky_ is *false*, let _lastIndex_ be 0.
> 1. Let _matcher_ be the value of _R_'s [[RegExpMatcher]] internal slot.
> - 1. Let _flags_ be the value of _R_'s [[OriginalFlags]] internal slot.
> 1. If _flags_ contains `"u"`, let _fullUnicode_ be *true*, else let _fullUnicode_ be *false*.
> 1. Let _matchSucceeded_ be *false*.
> 1. Repeat, while _matchSucceeded_ is *false*
[1] https://github.com/tc39/ecma262/pull/494
Assert the expected behavior of the ArraySpeciesCreate abstract
operation in the 5 Array instance methods from which it is invoked.
This change set does not include tests for ECMAScript realm
considerations because Test262 does not currently expose a mechanism for
interfacing with realms.
Limit tests in `language/expressions/call/` and `built-ins/eval/` to
only assert details that are directly related to the invocation pattern.
Re-organize all other tests within the `language/eval-code/` directory,
further categorizing each as `direct` or `indirect` as appropriate.
Remove the extraneous "executable" flag from those files which carried
it.
This harness function is not necessary in the majority of cases in which
it is used. Remove its usage to simplify tests and decrease the amount
of domain-specific knowledge necessary to contribute to the test suite.
Persist the harness function itself for use by future tests for ES2015
modules (such a helper is necessary for tests that are interpreted as
module code).
The millisecond representation of a given Date instance is dependent on
the local system's time zone settings. In order to pass consistently
across contexts, tests for this value must take the system configuration
into account.
Introduce a test harness utility function to encapsulate these concerns.
Re-use this function across all test files that assert the exact
millisecond representation of Date instances.
Closes gh-540
Because the specific value of each constant is specified as a
host-defined approximation, only the value type may be enforced by
Test262.
Except for a small set of expected input/output pairs, both ES5 and
ES2015 define the expected return value of these methods in terms of an
"implementation-dependent approximation." This makes it inappropriate to
enforce expectations for specific values, even if expressed imprecisely.
Because Test262 asserts the strict equality of
`Array.prototype.toString` and `TypedArray.prototype.toString`, tests
for properties of the object do not need to be reproduced in both
location.
Ref gh-526
Test specific behaviour for Integer Indexed exotic objects, WRT the
following internal methods:
- [[GetOwnProperty]]
- [[HasProperty]]
- [[DefineOwnProperty]]
- [[Get]]
- [[Set]]
- [[OwnPropertyKeys]]
As written, the test for `Math.random` would pass if the runtime
erroneously produced a non-numeric value. Add the necessary assertions
to guard against this case.
The harness file `Test262Error.js` has not contained executable code since it
was introduced in this project [1]. The definition of the `Test262Error`
function has consistently been located in the `sta.js` harness file which test
runners are expected to inject into the test environment.
Remove the file and all references to it.
[1] See commit c33bf0e043
For asynchronous tests, the contract between test file and test runner
is implicit: runners are expected to inspect the source code for
references to a global `$DONE` identifier.
Promote a more explicit contract between test file and test runner by
introducing a new frontmatter "tag", `async`. This brings asynchronous
test configuration in-line with other configuration mechanisms and also
provides a more natural means of test filtering.
The modifications to test files was made programatically using the
`grep` and `sed` utilities:
$ grep "\$DONE" test/ -r --files-with-match --null | \
xargs -0 sed -i 's/^\(flags:\s*\)\[/\1[async, /g'
$ grep "\$DONE" test/ -rl --null | \
xargs -0 grep -E '^flags:' --files-without-match --null | \
xargs -0 sed -i 's/^---\*\//flags: [async]\n---*\//'
Assert that the `constructor` property of the "this" value of
`Promise.prototype.then` is accessed exactly once. This guards against
implementations where repeated access is used instead of reference
passing (possibly motivated by convenience).
Repeated access of this kind was demonstrated by V8's implementation of
the specification:
https://bugs.chromium.org/p/v8/issues/detail?id=4539
Add tests that assert behavior when a Promise is resolved with another
Promise whose `then` method has been overridden. Because all objects
with a `then` method are treated equivalently, the presence of a
[[PromiseState]] internal slot should have no effect on program
behavior.
These tests guard against a faulty optimization originally implemented
in V8:
https://bugs.chromium.org/p/v8/issues/detail?id=3641
Replace a ES2015 test where calling the TypedArray constructor with
a floating number triggered a RangeError. Within the ES2016 specs,
the same call will trigger a TypeError, as the result for
`SameValue(NewTarget, here)` will be checked before.
Remove files that tested both PerformPromiseThen and
PromiseResolveFunction in favor of new tests that test
PromiseResolveFunction more directly and completely.
Because these tests concern the behavior of the PromiseReactionJob
abstract operation itself, they should avoid assumptions about the
correct implementation of that operation. Specifically: they should not
rely on the behavior of abupt completions returned from "reaction
handler" functions.
Re-implement tests to express control flow expectations using the
`$DONE` function only.
These tests concern the behavior of PerformPromiseThen for settled
Promises. That abstract operation behaves differently for pending
promises, so the file naming scheme should reflect this distinction in
order to support the future implementation of additional tests.