Hi guys,

I have a customer using my app that's having issues with PDF conversion of tables.

It seems that the tables display fine in HTML, but when converted to PDF, they get all screwed up. The tables break apart, and go on multiple lines. It looks really bad.

Server is running Prince 6.0r8, and I've tested with Prince 7.0 (unlicensed) on my personal machine. The bug exists both places.

Is there a bugfix for this? Workaround?

1. estimation.pdf‎ 81.4 kB ‎
   Screwed up table PDF
2. pdf preview.png‎ 70.4 kB ‎
   How the table should look...

Strange, would you be able to post or email me (mikeday@yeslogic.com) a sample HTML document that demonstrates the problem?

...Emailed! Thanks.

The issue appears to be caused by some invisible ideographic space characters (U+3000) that were located in between two table heading elements. The HTML parser wrapped these up in a new table heading, resulting in an incorrect number of columns in the table. I'm still not sure if this counts as a bug in the parser or not, but an easy workaround is to remove the characters.

The HTML5 specification requires text between the table cell elements to be extracted and placed before the table, as the browser is doing. We will follow this recommendation when we switch to our new HTML parser, currently under development.

Our new HTML parser is finished, and included in Prince 9, available now. This fixes the problem with text content between table cells.