Even though a PDF with non-embedded fonts does not contain actual font data, the PDF format requires certain information from the fonts used to create the PDF file. For this reason, Qoppa’s Word to PDF conversion library, jWordConvert, needs to locate a Java system font corresponding to each Word document font for the required information. This font information is also used for Word document layout.

jWordConvert’s basic approach is:

  • Find a local Java “working font” by trying to match primary font name & style (loosely)
  • Perform Word document layout & write PDF font information with the working font

The best Word to PDF document conversion (i.e. most similar to Microsoft Word layout) will result when local Java system fonts are found that closely match the Word document fonts.

Font Not Found
If a matching local font is not found then jWordConvert uses the style of the font (ex. “Serif”, “SansSerif”, “Modern”, italic, monospace, etc…) to find a Java logical font to get character metrics for PDF creation. jWordConvert will still create each PDF font with the original font name from the Word document. In the case where your PDF viewing software is able to find the font by name on the system viewing the PDF file, some text layout issues could remain, due to differences in the non-matching font used for layout on the converting system. If the PDF viewing software cannot find the font, it will read information saved by jWordConvert in the PDF regarding the original font style and will try and find a substitute font based on this information.