bom-test.js

"use strict";

const assert = require('assert'),
    Buffer = require('safer-buffer').Buffer,
    iconv = require(__dirname+'/../');

const sampleStr = '<?xml version="1.0" encoding="UTF-8"?>\n<俄语>данные</俄语>',
    strBOM = '\ufeff',
    utf8BOM = Buffer.from([0xEF, 0xBB, 0xBF]),
    utf16beBOM = Buffer.from([0xFE, 0xFF]),
    utf16leBOM = Buffer.from([0xFF, 0xFE]);

describe("BOM Handling", function() {
    it("strips UTF-8 BOM", function() {
        const body = Buffer.concat([utf8BOM, Buffer.from(sampleStr)]);
        assert.equal(iconv.decode(body, 'utf8'), sampleStr);
    });

    it("strips UTF-16 BOM", function() {
        const body = Buffer.concat([utf16leBOM, iconv.encode(sampleStr, 'utf16le')]);
        assert.equal(iconv.decode(body, 'utf16'), sampleStr);
        assert.equal(iconv.decode(body, 'utf16le'), sampleStr);

        const body2 = Buffer.concat([utf16beBOM, iconv.encode(sampleStr, 'utf16be')]);
        assert.equal(iconv.decode(body2, 'utf16'), sampleStr);
        assert.equal(iconv.decode(body2, 'utf16be'), sampleStr);
    });

    it("doesn't strip BOMs when stripBOM=false", function() {
        const body = Buffer.concat([utf8BOM, Buffer.from(sampleStr)]);
        assert.equal(iconv.decode(body, 'utf8', {stripBOM: false}), strBOM + sampleStr);

        const body2 = Buffer.concat([utf16leBOM, iconv.encode(sampleStr, 'utf16le')]);
        assert.equal(iconv.decode(body2, 'utf16', {stripBOM: false}), strBOM + sampleStr);
        assert.equal(iconv.decode(body2, 'utf16le', {stripBOM: false}), strBOM + sampleStr);

        const body3 = Buffer.concat([utf16beBOM, iconv.encode(sampleStr, 'utf16be')]);
        assert.equal(iconv.decode(body3, 'utf16', {stripBOM: false}), strBOM + sampleStr);
        assert.equal(iconv.decode(body3, 'utf16be', {stripBOM: false}), strBOM + sampleStr);
    });

    it("adds/strips UTF-7 BOM", function() {
        const bodyWithBOM = iconv.encode(sampleStr, 'utf7', {addBOM: true});
        const body = iconv.encode(sampleStr, 'utf7');
        assert.notEqual(body.toString('hex'), bodyWithBOM.toString('hex'));
        assert.equal(iconv.decode(body, 'utf7'), sampleStr);
    });

    it("adds UTF-8 BOM when addBOM=true", function() {
        const body = Buffer.concat([utf8BOM, Buffer.from(sampleStr)]).toString('hex');
        assert.equal(iconv.encode(sampleStr, 'utf8', {addBOM: true}).toString('hex'), body);
    });

    it("adds UTF-16 BOMs when addBOM=true", function() {
        const body = Buffer.concat([utf16leBOM, iconv.encode(sampleStr, 'utf16le')]).toString('hex');
        assert.equal(iconv.encode(sampleStr, 'utf16le', {addBOM: true}).toString('hex'), body);

        const body2 = Buffer.concat([utf16beBOM, iconv.encode(sampleStr, 'utf16be')]).toString('hex');
        assert.equal(iconv.encode(sampleStr, 'utf16be', {addBOM: true}).toString('hex'), body2);
    });

    it("'UTF-16' encoding adds BOM by default, but can be overridden with addBOM=false", function() {
        const body = Buffer.concat([utf16leBOM, iconv.encode(sampleStr, 'utf16le')]).toString('hex');
        assert.equal(iconv.encode(sampleStr, 'utf16').toString('hex'), body);

        const body2 = Buffer.concat([iconv.encode(sampleStr, 'utf16le')]).toString('hex');
        assert.equal(iconv.encode(sampleStr, 'utf16', {addBOM: false}).toString('hex'), body2);
    });


    it("when stripping BOM, calls callback 'stripBOM' if provided", function() {
        let bomStripped = false;
        const stripBOM = function() { bomStripped = true; }

        const body = Buffer.concat([utf8BOM, Buffer.from(sampleStr)]);
        assert.equal(iconv.decode(body, 'utf8', {stripBOM: stripBOM}), sampleStr);
        assert(bomStripped);

        bomStripped = false;

        const body2 = Buffer.from(sampleStr);
        assert.equal(iconv.decode(body2, 'utf8', {stripBOM: stripBOM}), sampleStr);
        assert(!bomStripped);
    });

});