PDFBox 2.0.4 : XFA to text error

问题

I am getting the following errors while trying to convert PDF(XFA) to string. These errors started coming when I switched from PDFBox 1.8.12 to PDFBox 2.0.4

Here is the log

Mar 09, 2017 7:16:07 AM org.apache.pdfbox.pdfparser.BaseParser parseCOSArray
WARNING: Corrupt object reference at offset 779916
Mar 09, 2017 7:16:07 AM org.apache.pdfbox.pdfparser.BaseParser parseCOSArray
WARNING: Corrupt object reference at offset 780049
Mar 09, 2017 7:16:07 AM org.apache.pdfbox.pdfparser.BaseParser parseCOSArray
WARNING: Corrupt object reference at offset 780074
java.io.IOException: Unknown dir object c='>' cInt=62 peek='>' peekInt=62 at offset 780074
    at org.apache.pdfbox.pdfparser.BaseParser.parseDirObject(BaseParser.java:951)
    at org.apache.pdfbox.pdfparser.BaseParser.parseCOSArray(BaseParser.java:651)
    at org.apache.pdfbox.pdfparser.BaseParser.parseDirObject(BaseParser.java:866)
    at org.apache.pdfbox.pdfparser.BaseParser.parseCOSDictionaryValue(BaseParser.java:150)
    at org.apache.pdfbox.pdfparser.BaseParser.parseCOSDictionaryNameValuePair(BaseParser.java:274)
    at org.apache.pdfbox.pdfparser.BaseParser.parseCOSDictionary(BaseParser.java:207)
    at org.apache.pdfbox.pdfparser.BaseParser.parseDirObject(BaseParser.java:854)
    at org.apache.pdfbox.pdfparser.COSParser.parseFileObject(COSParser.java:772)
    at org.apache.pdfbox.pdfparser.COSParser.parseObjectDynamically(COSParser.java:741)
    at org.apache.pdfbox.pdfparser.COSParser.parseObjectDynamically(COSParser.java:672)
    at org.apache.pdfbox.pdfparser.COSParser.parseDictObjects(COSParser.java:632)
    at org.apache.pdfbox.pdfparser.PDFParser.initialParse(PDFParser.java:217)
    at org.apache.pdfbox.pdfparser.PDFParser.parse(PDFParser.java:252)

and

java.io.IOException: Wrong type of referenced length object COSObject{7, 0}: COSDictionary
    at org.apache.pdfbox.pdfparser.COSParser.getLength(COSParser.java:907)
    at org.apache.pdfbox.pdfparser.COSParser.parseCOSStream(COSParser.java:949)
    at org.apache.pdfbox.pdfparser.COSParser.parseFileObject(COSParser.java:780)
    at org.apache.pdfbox.pdfparser.COSParser.parseObjectDynamically(COSParser.java:741)
    at org.apache.pdfbox.pdfparser.COSParser.parseObjectDynamically(COSParser.java:672)
    at org.apache.pdfbox.pdfparser.COSParser.parseDictObjects(COSParser.java:632)
    at org.apache.pdfbox.pdfparser.PDFParser.initialParse(PDFParser.java:217)
    at org.apache.pdfbox.pdfparser.PDFParser.parse(PDFParser.java:252)
    at org.apache.pdfbox.pdmodel.PDDocument.load(PDDocument.java:966)
    at org.apache.pdfbox.pdmodel.PDDocument.load(PDDocument.java:922)
    at org.apache.pdfbox.pdmodel.PDDocument.load(PDDocument.java:870)

I read the migration and used load instead of loadNonSeq, because now PDFBox handles that internally.

Any suggestions on how to fix these errors.

EDIT Error#1 Error#2

EDIT#2 @TilmanHausherr I checked your theory. I opened the file in Sublime, removed the extra spaces in the starting and saved it. I got the following error

    org.apache.pdfbox.filter.FlateFilter decode
SEVERE: FlateFilter: stop reading corrupt stream due to a DataFormatException
java.io.IOException: java.util.zip.DataFormatException: invalid distance too far back
    at org.apache.pdfbox.filter.FlateFilter.decode(FlateFilter.java:82)
    at org.apache.pdfbox.cos.COSInputStream.create(COSInputStream.java:69)
    at org.apache.pdfbox.cos.COSStream.createInputStream(COSStream.java:162)
    at org.apache.pdfbox.pdfparser.PDFXrefStreamParser.<init>(PDFXrefStreamParser.java:56)
    at org.apache.pdfbox.pdfparser.COSParser.parseXrefStream(COSParser.java:2075)
    at org.apache.pdfbox.pdfparser.COSParser.parseXrefObjStream(COSParser.java:348)
    at org.apache.pdfbox.pdfparser.COSParser.parseXref(COSParser.java:303)
    at org.apache.pdfbox.pdfparser.PDFParser.initialParse(PDFParser.java:194)
    at org.apache.pdfbox.pdfparser.PDFParser.parse(PDFParser.java:252)
    at utils.PDFManager.PDFToText(PDFManager.java:280)
    at processing.charge.CertificateUtils.getCertificateTypeFromFile(CertificateUtils.java:56)
    at processing.charge.CertificateUtils.getCertificateType(CertificateUtils.java:48)
    at processing.Controller.getDocumentType(Controller.java:110)
    at processing.Controller.insertIntoDb(Controller.java:43)
    at Test.main(Test.java:203)
    at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
    at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
    at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
    at java.lang.reflect.Method.invoke(Method.java:498)
    at com.intellij.rt.execution.application.AppMain.main(AppMain.java:147)
Caused by: java.util.zip.DataFormatException: invalid distance too far back
    at java.util.zip.Inflater.inflateBytes(Native Method)
    at java.util.zip.Inflater.inflate(Inflater.java:259)
    at java.util.zip.Inflater.inflate(Inflater.java:280)
    at org.apache.pdfbox.filter.FlateFilter.decompress(FlateFilter.java:107)
    at org.apache.pdfbox.filter.FlateFilter.decode(FlateFilter.java:64)
    ... 19 more
Mar 09, 2017 11:07:22 PM org.apache.pdfbox.filter.FlateFilter decode
SEVERE: FlateFilter: stop reading corrupt stream due to a DataFormatException
java.io.IOException: java.util.zip.DataFormatException: invalid distance too far back
    at org.apache.pdfbox.filter.FlateFilter.decode(FlateFilter.java:82)
    at org.apache.pdfbox.cos.COSInputStream.create(COSInputStream.java:69)
    at org.apache.pdfbox.cos.COSStream.createInputStream(COSStream.java:162)
    at org.apache.pdfbox.pdfparser.PDFXrefStreamParser.<init>(PDFXrefStreamParser.java:56)
    at org.apache.pdfbox.pdfparser.COSParser.parseXrefStream(COSParser.java:2075)
    at org.apache.pdfbox.pdfparser.COSParser.parseXrefObjStream(COSParser.java:348)
    at org.apache.pdfbox.pdfparser.COSParser.parseXref(COSParser.java:303)
    at org.apache.pdfbox.pdfparser.PDFParser.initialParse(PDFParser.java:194)
    at org.apache.pdfbox.pdfparser.PDFParser.parse(PDFParser.java:252)
    at utils.PDFManager.PDFToText(PDFManager.java:280)
    at processing.charge.CertificateUtils.getCertificateTypeFromFile(CertificateUtils.java:56)
    at processing.charge.CertificateUtils.getCertificateType(CertificateUtils.java:49)
    at processing.Controller.getDocumentType(Controller.java:110)
    at processing.Controller.insertIntoDb(Controller.java:43)
    at Test.main(Test.java:203)
    at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
    at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
    at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
    at java.lang.reflect.Method.invoke(Method.java:498)
    at com.intellij.rt.execution.application.AppMain.main(AppMain.java:147)
Caused by: java.util.zip.DataFormatException: invalid distance too far back
    at java.util.zip.Inflater.inflateBytes(Native Method)
    at java.util.zip.Inflater.inflate(Inflater.java:259)
    at java.util.zip.Inflater.inflate(Inflater.java:280)
    at org.apache.pdfbox.filter.FlateFilter.decompress(FlateFilter.java:107)
    at org.apache.pdfbox.filter.FlateFilter.decode(FlateFilter.java:64)

Also to verify your theory, I opened another file(that was working correctly) in Sublime, and it had the same spaces, tabs and CRs.

Working File

回答1:

As discussed in the comments, the files have blanks (CRs and TABs) before the PDF header starts. You can remove them with NOTEPAD++ (or with any editor that can edit binary files), or (if all your files have that flaw) by writing a short code that opens an input stream, swallow bytes until you hit "%" and then copy all the rest from there to an output stream.

I've also opened issue PDFBOX-3714.

Update: This has been fixed in 2.0.5, now available.

来源：https://stackoverflow.com/questions/42685353/pdfbox-2-0-4-xfa-to-text-error

标签

pdf

pdfbox

xfa