[PDFBOX-4944] Built-in fonts are reporting nbsp char as having zero width.

• Type: Bug
• Status: Closed
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 2.0.21
• Fix Version/s: 2.0.22, 3.0.0 PDFBox
• Labels: regression
• Assignee: Tilman Hausherr
• Reporter: Dan Fickling
• Created: 29/Aug/20 08:31
• Updated: 20/Dec/20 18:47
• Resolved: 17/Sep/20 03:41

Description

        PDFont font = PDType1Font.HELVETICA; // Also tested with Times.
        System.out.println("width = " + font.getStringWidth("\u00a0"));
 
Non-breaking space is reporting as zero width (standard 14 fonts) in 2.0.21 while being encoded as a (width taking) character. As you can imagine this causes serious errors in line and justification layout.
 
Related:
Prior to PDFBOX-4891 non-breaking space threw an IllegalArgumentException.
 
Originally reported at Github:
https://github.com/danfickle/openhtmltopdf/issues/543
 
Discovered by Github user lagar84.

Issue Links

• is caused by: PDFBOX-4891 nbspace missing in WinAnsiEncoding and MacRomanEncoding
• is related to: PDFBOX-4988 Space rendered as missing glyph (2)