I have an HTML file being generated that, for reasons, has some special characters in it. Probably stuff copied out of MS Word. I'm skipping details here, since I hope they're not relevant. What I'm seeing is that when I convert to PDF from the command line, it looks fine. The "e" with the accent, for example, renders perfectly. But when I use the Java wrapper, I get those classic diamond-question mark glyphs and other weirdness.

Anyone know if there's an option I ought to be using in the wrapper call? Or some sort of pre-processing before the call that needs to be done?

Which convert method are you calling in the Java wrapper?

convert(String inputPath, String outputPath)

Would you be able to attach or email me (mikeday@yeslogic.com) an example of the HTML and PDF? Is it using the same fonts when generated from the wrapper? Which version of Prince are you using? Are any options being set with the wrapper, such as input document type?