temporal

temporal copied to clipboard

Investigate using `EpochNanoseconds` and implement wherever possible

1

In #116, the `EpochNanoseconds` new type was added. The introduction of `EpochNanoseconds` was fairly bare, as further mentioned in #115. We should look into using `EpochNanoseconds` more throughout the codebase...

nekevss

Wrap `IxdtfParser` in a `TemporalParser` and ensure parsing invariants in `temporal_rs`

2

Currently parsing in `temporal_rs` is handled by parsing functions that are private. This should eventually be updates to a `TemporalParser` that wraps the `IxdtfParser` and ensures the various invariants of...

nekevss

Update Duration.prototype.round to latest spec

8

This function appears to be written against an old version of the spec. It is hard to verify correctness and update it to use the newer AOs.

sffc

Audit return types on all built-in methods

3

So far routinely `TemporalResult` has been used as the default return type. But currently a `TemporalResult` is sometimes used even in cases where the operation is infallible (most cases tend...

nekevss

Bikeshed for method naming conventions

3

There has been various changes in naming for methods over the lifetime of the development of `temporal_rs`, specifically around the use of snake case in methods and file names. The...

nekevss

Migrate errors to static enums

10

Our current design for `TemporalError` is essentially a mirror of ECMAScript's error types: https://github.com/boa-dev/temporal/blob/345ad548db498a5fe596ebebfc7f0ea6214fb079/src/error.rs#L39-L43 This works well for engines that have thousands of errors, with each one having a custom...

jedel1043

Currently, some parsing logic had to be reimplemented in `temporal_rs` to handle the `TimeZone::from_str`. This should be exposed upstream in `ixdtf` if possible as smaller a small `OffsetParser` and `IanaIdentifierParser`,...

nekevss

Related to #217. We need to be able to return named Time Zone identifiers in a case insensitive manner. The operations steps can be viewed in ECMA402 [6.5.1](https://tc39.es/ecma402/#sup-availablenamedtimezoneidentifiers) and [6.5.2](https://tc39.es/ecma402/#sec-getavailablenamedtimezoneidentifier)

nekevss

@nekevss says: > Are you looking into the test being broken on windows? If so, it may be worthwhile to know that windows right now is back by a prepackaged...

sffc

Reinvestigate `TemporalUnit::Auto` and it's general handling

2

Currently, `TemporalUnit` has `Auto` as an option. It was raised early on in the implementation (circa Oct. 2023) about whether `TemporalUnit::Auto` could be removed in favor of `Option`. The blocker...

nekevss