linguist-js/dist/analyser/pipeline/classify.js

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

import byAttributes from '../classifiers/byAttributes.js';
import byExtension from '../classifiers/byExtension.js';
import byFilename from '../classifiers/byFilename.js';
import byModeline from '../classifiers/byModeline.js';
import byShebang from '../classifiers/byShebang.js';
function dedupeClassifications(classifications) {
    const seen = new Set();
    const ordered = [];
    for (const classification of classifications) {
        if (!seen.has(classification)) {
            seen.add(classification);
            ordered.push(classification);
        }
    }
    return ordered;
}
export function classifyFiles(files, langData, opts) {
    const classifications = {};
    for (const file of files) {
        // Search each language
        // Note: order here is important; the first match will be the one chosen
        const candidates = [
            ...byAttributes(file, langData),
            ...byModeline(file, langData, opts),
            ...byFilename(file, langData),
            ...byShebang(file, langData, opts),
            ...byExtension(file, langData),
        ];
        classifications[file.path] = dedupeClassifications(candidates);
    }
    return classifications;
}