lsm/PdfPig
1
0
Fork 0
mirror of https://github.com/UglyToad/PdfPig.git synced 2025-04-05 20:55:01 +08:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity

fix #176, allow startxref to appear earlier in the document

Browse Source
This commit is contained in:
Eliot Jones 2020-05-31 17:01:38 +01:00
parent 4312aa470e
commit bf45602ac5
2 changed files with 51 additions and 29 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
src/UglyToad.PdfPig.Tests/Integration/SinglePageSimpleOpenOfficeTests.cs
View File

@ -90,5 +90,21 @@
Assert.False(document.TryGetBookmarks(out _));
}
}
[Fact]
public void StartXRefNotNearEnd()
{
var bytes = File.ReadAllBytes(GetFilename());
var emptyTrailer = new byte[2026];
emptyTrailer[0] = 10;
bytes = bytes.Concat(emptyTrailer).ToArray();
using (var document = PdfDocument.Open(bytes, ParsingOptions.LenientParsingOff))
{
Assert.Equal(1, document.NumberOfPages);
}
}
}
}

64
src/UglyToad.PdfPig/Parser/FileStructure/FileTrailerParser.cs
View File

@ -38,7 +38,7 @@
(byte) 'e',
(byte) 'f'
};
public static long GetFirstCrossReferenceOffset(IInputBytes bytes, ISeekableTokenScanner scanner, bool isLenientParsing)
{
if (bytes == null)
@ -55,10 +55,6 @@
var offsetFromEnd = fileLength < EndOfFileSearchRange ? (int)fileLength : EndOfFileSearchRange;
var startPosition = fileLength - offsetFromEnd;
bytes.Seek(startPosition);
var startXrefPosition = GetStartXrefPosition(bytes, offsetFromEnd);
scanner.Seek(startXrefPosition);
@ -96,38 +92,48 @@
var startXrefs = new List<int>();
var index = 0;
var offset = 0;
// Starting scanning the last 1024 bytes.
while (bytes.MoveNext())
var fileLength = bytes.Length;
var multiple = 1;
var actualStartOffset = Math.Max(0, fileLength - (offsetFromEnd * multiple));
do
{
offset++;
if (bytes.CurrentByte == StartXRefBytes[index])
multiple *= 2;
bytes.Seek(actualStartOffset);
// Starting scanning the file bytes.
while (bytes.MoveNext())
{
// We might be reading "startxref".
index++;
}
else
{
index = 0;
if (bytes.CurrentByte == StartXRefBytes[index])
{
// We might be reading "startxref".
index++;
}
else
{
index = 0;
}
if (index == StartXRefBytes.Length)
{
// Add this "startxref" (position from the start of the document to the first 's').
startXrefs.Add((int)bytes.CurrentOffset - StartXRefBytes.Length);
// Continue scanning in case there are further "startxref"s. Not sure if this ever happens.
index = 0;
}
}
if (index == StartXRefBytes.Length)
{
// Add this "startxref" (position from the end of the document to the first 's').
startXrefs.Add(offsetFromEnd - (offset - StartXRefBytes.Length));
// Continue scanning in case there are further "startxref"s. Not sure if this ever happens.
index = 0;
}
}
actualStartOffset = Math.Max(0, fileLength - (offsetFromEnd * multiple));
} while (startXrefs.Count == 0 && actualStartOffset > 0);
if (startXrefs.Count == 0)
{
throw new PdfDocumentFormatException("Could not find the startxref within the last 1024 characters.");
throw new PdfDocumentFormatException($"Could not find the startxref within the last {offsetFromEnd} characters.");
}
return bytes.Length - startXrefs[startXrefs.Count - 1];
return startXrefs[startXrefs.Count - 1];
}
}
}