I have a Java application that is generating JasperReports. It will create as many as three JasperPrints from a single report: one prints on the printer, one is serialized and saved to the database, and the third is exported to PDF using Jasper’s built-in export capability.
The problem is that when exporting to PDF, characters containing 8 or more bits (i.e. not 7-bit ASCII) are showing up as empty squares, meaning Acrobat Reader is not able to display that character. The print version is correct, and loading the database version and printing it shows up correctly. If I change the PDF exported version to a different format, e.g. XML, the character shows up fine in a web browser.
Based on the evidence, I believe the issue is something specific to font handling in PDFs, but I am not sure what.
The font used is Lucida Sans Typewriter, a Unicode monospaced font. The Windows “font” directory is listed in the Java classpath: without this step, PDF exporting fails miserably with zero text at all, so I know it is finding the font.
The specific characters not displayed are accented characters used in Spanish text: á, é, í, ó, and ú. I haven’t checked ñ but I am guessing that won’t work too.
Any ideas what the problem is, areas of the system to check, or maybe parameters I need to send to the export process?
The PDF encoding used for exporting was UTF-8, and apparently the font didn’t support that properly. When I changed it to ISO-8859-1, every character showed up correctly in the PDF output.