84 lines
2 KiB
Markdown
84 lines
2 KiB
Markdown
|
# One INI
|
||
|
|
|
||
|
|
The core implementation of an AST based, idiomatic INI parser which aims to provide an easy to implement and consistent INI-standard.
|
||
|
|
|
||
|
|
This reference implementation is provided as Rust-library and WASM-package.
|
||
|
|
|
||
|
|
<!-- markdownlint-disable -->
|
||
|
|
|
||
|
|
[](https://github.com/jedmao/editorconfig-ini/actions)
|
||
|
|
|
||
|
|
<!-- markdownlint-restore -->
|
||
|
|
|
||
|
|
<!-- markdownlint-disable commands-show-output -->
|
||
|
|
|
||
|
|
The work on this project started with the search for an universal parser for the [EditorConfig INI file format specification](https://editorconfig-specification.readthedocs.io/en/latest/#id3).
|
||
|
|
|
||
|
|
## WASM
|
||
|
|
|
||
|
|
To use from [Web Assembly](https://webassembly.org/), compile with:
|
||
|
|
|
||
|
|
```sh
|
||
|
|
wasm-pack build --release --target nodejs
|
||
|
|
```
|
||
|
|
|
||
|
|
and run the (limited) WASM tests with:
|
||
|
|
|
||
|
|
```sh
|
||
|
|
wasm-pack test --node
|
||
|
|
```
|
||
|
|
|
||
|
|
You can call the genereted JS wrapper with either:
|
||
|
|
|
||
|
|
```js
|
||
|
|
import { parse_to_json } from './pkg/editorconfig_ini.js'
|
||
|
|
|
||
|
|
const results = parse_to_json(`
|
||
|
|
root = true
|
||
|
|
|
||
|
|
[*]
|
||
|
|
# always use unix line endings
|
||
|
|
end_of_line = lf
|
||
|
|
`)
|
||
|
|
|
||
|
|
// {
|
||
|
|
// "version": "0.1.0",
|
||
|
|
// "body": [
|
||
|
|
// { "type": "Pair", "key": "root", "value": "true" },
|
||
|
|
// {
|
||
|
|
// "type": "Section",
|
||
|
|
// "name": "*",
|
||
|
|
// "body": [
|
||
|
|
// { "type": "Comment", "indicator": "#", "value": "always use unix line endings" },
|
||
|
|
// { "type": "Pair", "key": "end_of_line", "value": "lf" }
|
||
|
|
// ]
|
||
|
|
// }
|
||
|
|
// ]
|
||
|
|
// }
|
||
|
|
```
|
||
|
|
|
||
|
|
or:
|
||
|
|
|
||
|
|
```js
|
||
|
|
import { parse_to_uint32array, TokenTypes } from './pkg/editorconfig_ini.js'
|
||
|
|
const buf = Buffer.from(`
|
||
|
|
root = true
|
||
|
|
|
||
|
|
[*]
|
||
|
|
# always use unix line endings
|
||
|
|
end_of_line = lf
|
||
|
|
`, 'utf8')
|
||
|
|
const ary = parse_to_uint32array(buf)
|
||
|
|
|
||
|
|
// Array with token type, start byte offset, end byte offset for each token
|
||
|
|
// Uint32Array(21) [
|
||
|
|
// TokenTypes.Key, 1, 5,
|
||
|
|
// TokenTypes.Value, 8, 12,
|
||
|
|
// TokenTypes.Section, 15, 16,
|
||
|
|
// TokenTypes.CommentIndicator, 18, 19,
|
||
|
|
// TokenTypes.CommentValue, 20, 48,
|
||
|
|
// TokenTypes.Key, 49, 60,
|
||
|
|
// TokenTypes.Value, 63, 65
|
||
|
|
// ]
|
||
|
|
```
|