[PDFBOX-1304] Text extraction meets "Could not parse predefined CMAP" and returns just a small part of the content containing garbage chars.

• Type: Bug
• Status: Closed
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 1.6.0
• Fix Version/s: 2.0.0
• Component/s: Text extraction
• Environment: Win7 32bits
• Assignee: Unassigned
• Reporter: Huan LI
• Created: 06/May/12 11:01
• Updated: 06/Nov/14 06:10
• Resolved: 06/Nov/14 06:10

Description

i'm using pdfbox-1.6.0 for text extraction from a Chinese pdf file(see the attachment "fj.pdf").

the extraction code looks like below:
[code]
stripper = new PDFTextStripper(encoding);
txt = stripper.getText(_pdfDoc);
[/code]
when running getText(), the console says :
[console]
五月 06, 2012 4:13:51 下午 org.apache.pdfbox.pdmodel.font.PDCIDFont determineEncoding
严重: Error: Could not parse predefined CMAP file for 'Founder-PKU2-UCS2'
五月 06, 2012 4:13:51 下午 org.apache.pdfbox.pdmodel.font.PDCIDFont determineEncoding
严重: Error: Could not parse predefined CMAP file for 'Founder-PKU2-UCS2'
五月 06, 2012 4:13:51 下午 org.apache.pdfbox.pdmodel.font.PDCIDFont determineEncoding
严重: Error: Could not parse predefined CMAP file for 'Founder-PKU2-UCS2'
五月 06, 2012 4:13:51 下午 org.apache.pdfbox.pdmodel.font.PDCIDFont determineEncoding
严重: Error: Could not parse predefined CMAP file for 'Founder-PKUE1-UCS2'
五月 06, 2012 4:13:51 下午 org.apache.pdfbox.pdmodel.font.PDCIDFont determineEncoding
严重: Error: Could not parse predefined CMAP file for 'Founder-PKUO1-UCS2'
五月 06, 2012 4:13:51 下午 org.apache.pdfbox.pdmodel.font.PDCIDFont determineEncoding
严重: Error: Could not parse predefined CMAP file for 'Founder-PKUF1-UCS2'
五月 06, 2012 4:13:51 下午 org.apache.pdfbox.pdmodel.font.PDCIDFont determineEncoding
严重: Error: Could not parse predefined CMAP file for 'Founder-PKU2-UCS2'
五月 06, 2012 4:13:51 下午 org.apache.pdfbox.pdmodel.font.PDCIDFont determineEncoding
严重: Error: Could not parse predefined CMAP file for 'Founder-PKUF1-UCS2'
五月 06, 2012 4:13:51 下午 org.apache.pdfbox.pdmodel.font.PDCIDFont determineEncoding
严重: Error: Could not parse predefined CMAP file for 'Founder-PKU2-UCS2'
五月 06, 2012 4:13:51 下午 org.apache.pdfbox.pdmodel.font.PDCIDFont determineEncoding
严重: Error: Could not parse predefined CMAP file for 'Founder-PKUE1-UCS2'
五月 06, 2012 4:13:51 下午 org.apache.pdfbox.pdmodel.font.PDCIDFont determineEncoding
严重: Error: Could not parse predefined CMAP file for 'Founder-PKUE1-UCS2'
五月 06, 2012 4:13:51 下午 org.apache.pdfbox.pdmodel.font.PDCIDFont determineEncoding
严重: Error: Could not parse predefined CMAP file for 'Founder-PKUF1-UCS2'
五月 06, 2012 4:13:51 下午 org.apache.pdfbox.pdmodel.font.PDCIDFont determineEncoding
严重: Error: Could not parse predefined CMAP file for 'Founder-PKU2-UCS2'
[/console]
after getText() returns, the txt contains just a small part of the pdf content (lots are missing) and some garbage chars like "犖犑狌犣犎犗犝犔犻犺犅"(see attachment "fj.txt").

I've heard some said that the "org.apache.pdfbox.cos.COSString.java" has some errors when pdfbox-0.7.3. Has COSString.java been corrected in 1.6.0?