Support Enhanced LRC format + additional QOL features (Part I)

[Nafeij]

Enhanced LRC is an extension of Simple LRC Format developed for the A2 Media Player. This supports adding an optional timestamp for words, syllables, or otherwise parts of a line. It takes the format: <mm:ss.xx>.

Enhanced LRC support

• Add functionality to parse Enhanced LRC
• Add additional interfaces and functions

20/4/23

• Add options:
  • enhanced: parse entire input as Simple or Enhanced LRC. default false.
  • strip: ignore empty lines. default false.
• Parser will now support repeating times for Enhanced LRC
  • Will pre-calculate timing offsets for syllables based on the line's start time.
• Added relevant types.

25/4/23

• Add functions:
  • toString: inverse of parse; returns valid lrc for given list of Line objects.
  • expand: alias for toString(parse(lrc: string))
    The use case for this is to 'unroll' or 'expand' lines with repeat-timestamps into separate lines. This is straightforward to implement as parse already does this when building the Lrc list. This is especially useful for Enhanced Lrc as repeating timestamps are not supported by the specification.
• Add playground options to reflect the features listed above.

Misc

• Added build command to prepare script to support sideloading from git repos.

This should be backwards-compatible with previous versions.

Note: This PR is required before merging Part II.

[mebtte]

@Nafeij Thanks for contributing.

Supporting enhanced format is a good idea, but i think it's usage may be less than simple format and compatibility with enhanced format will be slow down parse function, so adding a new function like parseEnhanced or adding a option like parse(lrc, { enhanced: true }) is a better choice?

[mebtte]

And there is a issue, [00:32.79][00:56.00]Little <00:33.58>jagged <00:34.14>edge<00:35.25> should be generate two items, but only one in the result.

[Nafeij]

@mebtte

20/4/23

• Add options:
  • enhanced: parse entire input as Enhanced LRC, instead of Simple LRC. default false.
  • strip: ignore empty lines. default false.
• Parser will now support repeating times for Enhanced LRC lines
  • Will pre-calculate timing offsets for syllables based on the line's start time.
• Added relevant types.

[mebtte]

I think strip is unnecessary and it will make developer confusing with empty line and empty content.

[mebtte]

expand is verbose and it's unnecessary.

[mebtte]

According to wiki, word is better than syllable .

[mebtte]

Rewrite parse(lrc, { enhanced: true }) to parseEnhanced(lrc), it can reuse parse and has better typescript return type.

[mebtte]

I think stringify is better than toString.

[mebtte]

From the wiki, enhanced lrc format is [time]<time>content<time>content, not [time]content<time>content<time>.

[mebtte]

It seems stringify doesn't do much, because array of Line is more complex than lrc string itself.