Description

Updates parsing on non-css attribute values that don’t need tokenisation with csssparser.

Motivation and Context

Further improves performance of parsing documents and reduces the risk of data-loss.

Given we usually know the exact type and length of an input, we can pull from the value as-is.

How Has This Been Tested?

• Unit tests
• w3c & oxygen

---
config:
    xyChart:
        width: 622
        height: 181

---
xychart horizontal
    title "Performance change (ms)"
    x-axis [Oxygen, w3c]
    y-axis "Time (ms)" 0 --> 32
    %% Before
    bar [31.8, 0.9]
    %% After
    bar [19.5, 0.1]

Types of changes

Tests

• more coverage of attribute parsing

Features

• added basic parser to oxvg_parse
• put lightningcss parsing behind a feature flag
• rewrite path parser to use oxvg_parse parser

Breaking changes

• path api updated slightly to reduce clones

Checklist:

• My code follows the code style of this project.
• My change requires a change to the documentation.
• I have updated the documentation accordingly.
• I have added tests to cover my changes.
• All new and existing tests passed.