[PDFBOX-2449] Character missing in text extraction

• Type: Bug
• Status: Closed
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 1.8.8
• Fix Version/s: 1.8.8
• Assignee: Andreas Lehmkühler
• Reporter: Tilman Hausherr
• Created: 23/Oct/14 20:53
• Updated: 13/Dec/14 14:15
• Resolved: 29/Nov/14 18:35

Description

The attached file brings this text extraction:

1.8.6:
For safe! clean! abundant water?in our homes!
rivers! lakes! and streams?is one of our

1.8.7:
For safe! clean! abundant water?in our homes!
rivers! lakes! and streams?is one of our

1.8.8:
For safe! clean! abundant water?n our homes!
rivers! lakes! and streams?s one of our

2.0:
For safe! clean! abundant water–in our homes!
rivers! lakes! and streams–is one of our

AR:
For safe! clean! abundant water–in our homes!
rivers! lakes! and streams–is one of our

So the "i" has been lost in the 1.8.8 version. (2.0 and AR have a character that is invisible when viewing this issue, but can be seen when editing)

Issue Links

• is broken by: PDFBOX-2377 Apparent regression in character mapping in a few files from govdocs1
• relates to: TIKA-1442 Upgrade to PDFBox 1.8.8