ink/build/sanitize-ansi.js

React for CLI

import { hasAnsiControlCharacters, tokenizeAnsi } from './ansi-tokenizer.js';
const sgrParametersRegex = /^[\d:;]*$/;
// Terminals print nothing for C0 controls and DEL, so strip them; tabs and newlines carry layout. ESC and C1 never reach text tokens, the tokenizer owns them.
const controlCharactersRegex = /(?![\t\n])\p{Control}/gu;
const leadingMarksRegex = /^\p{Mark}+/u;
// @alcalzone/ansi-tokenize only accepts `[0-9;]` SGR parameters and renders anything else as visible cells, so colon sub-parameters are rewritten before layout.
// 256-color and truecolor forms map losslessly to the semicolon form, underline styles (`4:n`) degrade to plain underline, everything else (underline color `58:...`, malformed color forms) is dropped.
const normalizeParameter = (parameter) => {
    const parts = parameter.split(':');
    if (parts.length === 1) {
        return parameter;
    }
    if (parts[0] === '4') {
        return parts[1] === '0' || parts[1] === '' ? '24' : '4';
    }
    if (parts[0] !== '38' && parts[0] !== '48') {
        return undefined;
    }
    if (parts[1] === '5' && parts.length === 3) {
        return parts.join(';');
    }
    if (parts[1] === '2') {
        // Drop an empty or default color space id.
        if (parts.length === 6 && (parts[2] === '' || parts[2] === '0')) {
            parts.splice(2, 1);
        }
        if (parts.length === 5) {
            return parts.join(';');
        }
    }
    return undefined;
};
// Strip ANSI escape sequences that would conflict with Ink's layout.
// Preserved: SGR sequences (colors, bold, etc. - end with 'm') and
// OSC sequences (hyperlinks, etc. - ESC ] or C1 OSC).
// Stripped: cursor movement, screen clearing, and other control sequences.
// Combining marks share their base character's cell, so SGR sequences are deferred until after the marks. Otherwise ANSI tokenization consumes the marks as part of the escape sequence.
const sanitizeAnsi = (text) => {
    if (!hasAnsiControlCharacters(text)) {
        return text.replaceAll(controlCharactersRegex, '');
    }
    let output = '';
    let pendingStyles = '';
    for (const token of tokenizeAnsi(text)) {
        if (token.type === 'text') {
            const value = token.value.replaceAll(controlCharactersRegex, '');
            const marks = leadingMarksRegex.exec(value)?.[0] ?? '';
            output += marks;
            if (value.length > marks.length) {
                output += pendingStyles + value.slice(marks.length);
                pendingStyles = '';
            }
            continue;
        }
        if (token.type === 'osc') {
            output += pendingStyles + token.value;
            pendingStyles = '';
            continue;
        }
        if (token.type !== 'csi' ||
            token.finalCharacter !== 'm' ||
            token.intermediateString !== '' ||
            !sgrParametersRegex.test(token.parameterString)) {
            continue;
        }
        const parameters = token.parameterString
            .split(';')
            .map(parameter => normalizeParameter(parameter))
            .filter(parameter => parameter !== undefined);
        if (parameters.length > 0) {
            pendingStyles += `\u{1B}[${parameters.join(';')}m`;
        }
    }
    return output + pendingStyles;
};
export default sanitizeAnsi;
//# sourceMappingURL=sanitize-ansi.js.map