Escaped unicode code points being mangled

[Alan]

When I run this json through the json-loader:

https://github.com/fb55/htmlparser2/blob/master/lib/entities/entities.json

I get:

63: function(module) {
    module.exports = {
        quot: 34,
        amp: 38,
        apos: 39,
        lpar: 40,
        rpar: 41,
        lt: 60,
        gt: 62,
        nbsp: 160,
        iexcl: 161
        ...
    }
}

It looks like the escaped unicode code points are being lost. I'm not exactly sure what the resulting values are, but it is neither the unescaped unicode characters nor the escaped code point sequences.

It's interesting that quot became "34" instead of just """, which is technically the correct ASCII value, but the json-loader should not be doing such conversions.

[Tobias Koppers]

This doesn't occur for me:

function(module, exports, __webpack_require__) {
    module.exports = {
        "Aacute": "Á",
        "aacute": "á",
        "Abreve": "Ă",
        "abreve": "ă",
        "ac": "∾",
        "acd": "∿",
        "acE": "∾̳",
        ...
        "quot": "\"",
        ...

function(r){r.exports={Aacute:"Á",aacute:"á",Abreve:"Ă",abreve:"ă",ac:"∾",...,quot:'"',...

[Tobias Koppers]

any update to this?

[Alan]

Hmmm, I tried creating a minimal-repro of this issue and it just wouldn't reproduce. I'll close for now, and when I get back to working on my main project later I'll re-open the issue if I can figure out how to recreate it.

Thanks for taking a look.