UTF-8 NOT WORKING

[deshario]

Seems like UTF-8 (thai characters) not working well .. Is there anything missing from me

[killianhuyghe]

I had the same issue with diacritics on latin letters (é, à, ...).

FYI, the bug is not present in the repo it has been forked from, so you might as well do without the stream part until this is fixed.

[jasonphillips]

FYI, the bug is not present in the repo it has been forked from, so you might as well do without the stream part until this is fixed.

Hmm, I've never been able to resolve this, but partly because I do get the same problem when using the latest (0.0.11) version of the forked library.

Keeping strictly to their main simple example from the docs, and using some characters like (é, or text like मुख्यपृष्ठम्) those fields are either garbled or fail to write. Also, when I have attempted to use the underlying binary pdftk directly, I still run into errors at the moment. Would happily accept a PR or focused test that can target the problem more precisely.

[drnjha]

In fdf.js I made the following modification to make UTF-8 work:

var header = Buffer.concat([
    Buffer.from("%FDF-1.2\n"),
    headerChars,
    Buffer.from(
        "1 0 obj \n" +
        "<<\n" +
        "/Version /1.3 \n" +     // neu wg. Umlauten
        "/Encoding /utf_8 \n" +  // neu wg. Umlauten
        "/FDF \n" +
        "<<\n" +
        "/Fields [\n"
    )
]);

I hope it makes sense to s.o.
:-)