linguist-js/dist/analyser/classifiers/byShebang.js

Analyse the programming languages used in a folder or from raw content, using the same rules that GitHub Linguist does.

const escapeRegExp = (value) => value.replace(/[.*+?^${}()|[\]\\]/g, '\\$&');
const categoryPriority = {
    programming: 0,
    markup: 1,
    prose: 2,
    data: 3,
};
export default function byShebang(file, langData, opts) {
    if (!opts.checkShebang || opts.quick)
        return [];
    const firstLine = file.firstLine;
    if (!firstLine || !firstLine.startsWith('#!'))
        return [];
    const matches = Object.entries(langData).flatMap(([lang, data]) => {
        const interpreters = data.interpreters ?? [];
        // Check for interpreter match
        const matchesInterpretor = interpreters.some((interpreter) => new RegExp(`\\b${escapeRegExp(interpreter)}\\b`).test(firstLine));
        return matchesInterpretor ? [lang] : [];
    });
    return [...new Set(matches)].sort((a, b) => {
        const aType = langData[a]?.type;
        const bType = langData[b]?.type;
        const aPriority = categoryPriority[aType ?? ''] ?? 4;
        const bPriority = categoryPriority[bType ?? ''] ?? 4;
        return aPriority - bPriority;
    });
}