Conversion has problems with german Umlaute about email-to-pdf-converter HOT 9 CLOSED

I could not reproduce the issue (see attachment). Could you provide me with some details on your OS, Java version and the wkhtmltopdf version?

Bad.pdf

from email-to-pdf-converter.

I am using Fedora 26

Kernel: Linux localhost-uhlme 4.12.8-300.fc26.x86_64 #1 SMP Thu Aug 17 15:30:20 UTC 2017 x86_64 x86_64 x86_64 GNU/Linux

wkhtmltopdf: wkhtmltopdf 0.12.3

Which version are you using?

from email-to-pdf-converter.

Which version are you using?

from email-to-pdf-converter.

I produced the pdf on a machine running Windows 10, using wkhtmltopdf 0.12.4 (with patched qt).

from email-to-pdf-converter.

@patrickuhlmann Could you resolve your problem? And if, could you please share how?

from email-to-pdf-converter.

I forked the code and made a quick fix here: https://github.com/patrickuhlmann/eml-to-pdf-converter/commit/94597c23fd4fbfa02e11874ed474caff11e0b107

Basically the problem was the following: The encoding of the mail was iso-8859-1. The class ContentTypeCleaner needed to decode it as QuotedPrintable. However the method doing this was "forcing" UTF-8. I did not fix the detection but added a new parameter charset. That parameter allows me to properly convert my mails.

from email-to-pdf-converter.

Glad you could resolve your issue!

Both eml files you provided specify an UTF-8 encoding. Could you provide me with an exemplary iso-8859-1 encoded eml that caused the problem?

from email-to-pdf-converter.

https://drive.google.com/open?id=17zb586dYvNcrz415k5O-fPQkf4GOBrjx

You find in the archive an eml as well as the pdf file (- bad) that results when I convert it with the original version. I did also attach the pdf file (- correct) that results when I convert it with my patched version (passing the correct encoding).

from email-to-pdf-converter.

Can't reproduce

from email-to-pdf-converter.