Subset font by private use codepoints?

[audunru]

Trying to create a subset of the Bootstrap3 glyphicons so I get a font file with only the few glyphs I actually use.

Most of the icons use private use codepoints in UTF-8. I've been able to subset them by copying the character from http://www.fileformat.info/info/unicode/char/e013/browsertest.htm

But it isn't very readable. Any option I've missed that could make this more readable?

From my gulpfile.js:

var Fontmin = require('fontmin');
gulp.task('fontmin', function () {
    var characters = [
        '+', // plus    \002b
        '', // ok      \e013   http://www.fileformat.info/info/unicode/char/e013/browsertest.htm
        '', // remove  \e014
        '', // cog     \e019
        '', // trash   \e020
        '', // lock    \e033
        '', // stats   \e185
    ];
    var fontmin = new Fontmin()
        .src('node_modules/bootstrap-sass/assets/fonts/bootstrap/glyphicons-halflings-regular.ttf')
        .dest('public/build/fonts/bootstrap/subset')
        .use(Fontmin.glyph({
            text: characters.join(),
            hinting: false         // keep ttf hint info (fpgm, prep, cvt). default = true
        }))
        .use(Fontmin.ttf2eot())
        .use(Fontmin.ttf2woff())
        .use(Fontmin.ttf2svg());
    fontmin.run(function (err, files) {
        if (err) {
            throw err;
        }
        console.log(files[0]);
        // => { contents: <Buffer 00 01 00 ...> }
    });

});

[junmer]

let me try