UTF-8 Support

[sati-bodhi]

Is it possible to add utf-8 support to the app? Trying to convert pdfs with CJK characters ended up with garbled text.

[jzillmann]

Hey @sati-bodhi , the site is using https://mozilla.github.io/pdf.js/ for the underlying parsing. Can you check if your PDF renders correctly with it ? You should be able to test through https://mozilla.github.io/pdf.js/web/viewer.html

[sati-bodhi]

Thanks, @jzillmann! I just cloned the code. How do I go about testing?

---

I am following the instructions from readme.md right now. Will update you how it goes.

---

I got to the server part, but I am not sure how to initiate the conversion program from there.

[jzillmann]

Hmm, i'm confused a bit.. You cloned pdf.js or pdf-markdown ? As a first step i would recommend to just upload your PDF to https://mozilla.github.io/pdf.js/web/viewer.html. If this online viewer displays your PDF correctly, we know that pdf.js ia able to parse your PDF and we can go a step further, if not, we probably reached a dead end...

[sati-bodhi]

I cloned pdf.js I did the upload and it showed correctly.

[sati-bodhi]

The http://pdf2md.morethan.io/ engine works for this sample:

sample_success.pdf

But not this one:

sample_fail.pdf

Both can be rendered correctly by the online viewer:

[jzillmann]

Hey @sati-bodhi thanks for the samples. I tried a few things but had no luck so far... Asked the pdf.js community for help: https://github.com/mozilla/pdf.js/issues/9692

[jzillmann]

Based on the response in https://github.com/mozilla/pdf.js/issues/9692 i'm closing this as won't fix. The PDF seems to have a incomplete ToUnicode data...

[sati-bodhi]

Thanks! The answer given by @timvandermeij was elaborate indeed.