[PDFBOX-603] PDFBox performance issue: Encoding.java getCharacter() method tweak

• Type: Improvement
• Status: Closed
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 0.8.0-incubator
• Fix Version/s: 1.0.0
• Component/s: Text extraction
• Environment: All
• Assignee: Jukka Zitting
• Reporter: Mel Martinez
• Created: 14/Jan/10 21:44
• Updated: 22/Feb/10 18:28
• Resolved: 14/Jan/10 22:58

Description

During parsing / text extraction the Encoding.getCharacter(COSName) method is invoked repeatedly.

It includes a string test that is performed up front but should only occur rarely. The code should be restructured slightly to only perform that test later. I.E. it should succeed fast and fail slow.

I'll post an attachment that rewrites the method slightly. The performance gains is fairly significant.