URI malformed error during base64 encoding

[sanex3339]

var foo = '\ud801\ud804-';

when encodeUriComponent

[sanex3339]

Anyone have a idea how to fix this when this string is passed to `encodeUriComponent'?

[zamotkin]

I can review today.

[zamotkin]

@sanex3339 I can't understand how did you receive such a string, but you can try to strip invalid surrogate pairs by something like this:

function stripUnmatchedSurrogates (str) {
    return str.replace(/[\uD800-\uDBFF](?![\uDC00-\uDFFF])/g, '').split('').reverse().join('').replace(/[\uDC00-\uDFFF](?![\uD800-\uDBFF])/g, '').split('').reverse().join('');
}

[cliffhall]

I ran into a similar problem, and was getting URI Malformed error when doing a production build of my react app using webpack-obfusicator. I have two apps actually and one did not have the problem but the other did. In the app that did, I was basically redoing the configuration that I had on the one that worked.

When I hit the error, I just started commenting out likely suspects in the JavascriptObfusicator configuration. I found quickly that it was stringArrayEncoding: "base64" causing the issue. The 'fix' turned out to be setting splitStrings: false.

Cheers, and thanks for such an awesome tool!

[sanex3339]

Hi. @cliffhall can you show a small example here https://obfuscator.io/ with enabled stringArrayEncoding and splitStrings?

[cliffhall]

Unfortunately it’s a large app and I have no idea what specifically broke and caused the error.

[msheakoski]

Here is how to reproduce:

1. Visit https://obfuscator.io
2. Use the following code for the input: "12😃45"
3. Change String Array Threshold to: 1
4. Change String Array Encoding to: Base64
5. Enable: Split Strings
6. Change Split Strings Chunk Length to: 3
7. Click the Obfuscate button
8. Output should be: URIError: URI malformed

It looks like the combination of "Split Strings Chunk Length" and "Base64" is not correctly handling multibyte characters if they occur on a chunk boundary. 😃 is composed of 2 unicode sequences, \uD83D and \uDE03. With a chunk length of 3, this particular string ends up being split into 12\uD83D and \uDE0345

[sanex3339]

Ok. Will fix it using https://github.com/sallar/stringz

[sanex3339]

Fixed in 2.4.2