public static IContainer GenerateContainer(ILog logger)
{
if (_testContainer != null)
{
return(_testContainer);
}
if (logger == null)
{
logger = new NoOpLog();
}
var headerParser = new FileHeaderParser(logger);
var trailerParser = new FileTrailerParser();
var filterProvider = new MemoryFilterProvider(new DecodeParameterResolver(logger), new PngPredictor(), logger);
var cmapParser = new CMapParser();
var afmParser = new AdobeFontMetricsParser();
var container = new Container();
container.Register(headerParser);
container.Register(trailerParser);
container.Register(filterProvider);
container.Register(cmapParser);
container.Register(afmParser);
container.Register(logger);
return(container);
}
public void NullScannerThrows()
{
var input = StringBytesTestConverter.Convert("11 0 obj", false);
Action action = () => FileTrailerParser.GetFirstCrossReferenceOffset(input.Bytes, null, false);
Assert.Throws <ArgumentNullException>(action);
}
public void CanReadStartXrefIfCommentsPresent()
{
var input = StringBytesTestConverter.Convert(@"
startxref %Commented here
57695
%%EOF", false);
var result = FileTrailerParser.GetFirstCrossReferenceOffset(input.Bytes, new CoreTokenScanner(input.Bytes), false);
Assert.Equal(57695, result);
}
/// <summary>
/// Merge the set of PDF documents.
/// </summary>
public static byte[] Merge(IReadOnlyList <byte[]> files, IReadOnlyList <IReadOnlyList <int> > pagesBundle = null)
{
if (files == null)
{
throw new ArgumentNullException(nameof(files));
}
const bool isLenientParsing = false;
var documentBuilder = new DocumentMerger();
foreach (var fileIndex in Enumerable.Range(0, files.Count))
{
var file = files[fileIndex];
IReadOnlyList <int> pages = null;
if (pagesBundle != null && fileIndex < pagesBundle.Count)
{
pages = pagesBundle[fileIndex];
}
var inputBytes = new ByteArrayInputBytes(file);
var coreScanner = new CoreTokenScanner(inputBytes);
var version = FileHeaderParser.Parse(coreScanner, isLenientParsing, Log);
var crossReferenceParser = new CrossReferenceParser(Log, new XrefOffsetValidator(Log),
new Parser.Parts.CrossReference.CrossReferenceStreamParser(FilterProvider));
CrossReferenceTable crossReference = null;
// ReSharper disable once AccessToModifiedClosure
var locationProvider = new ObjectLocationProvider(() => crossReference, inputBytes);
var pdfScanner = new PdfTokenScanner(inputBytes, locationProvider, FilterProvider, NoOpEncryptionHandler.Instance);
var crossReferenceOffset = FileTrailerParser.GetFirstCrossReferenceOffset(inputBytes, coreScanner, isLenientParsing);
crossReference = crossReferenceParser.Parse(inputBytes, isLenientParsing, crossReferenceOffset, version.OffsetInFile, pdfScanner, coreScanner);
var catalogDictionaryToken = ParseCatalog(crossReference, pdfScanner, out var encryptionDictionary);
if (encryptionDictionary != null)
{
throw new PdfDocumentEncryptedException("Unable to merge document with password");
}
var documentCatalog = CatalogFactory.Create(crossReference.Trailer.Root, catalogDictionaryToken, pdfScanner, isLenientParsing);
documentBuilder.AppendDocument(documentCatalog, version.Version, pdfScanner, pages);
}
return(documentBuilder.Build());
}
public void MissingNumericAfterStartXrefThrows()
{
var input = StringBytesTestConverter.Convert(@"11 0 obj
<< /Type/Font >>
endobj
startxref
", false);
Action action = () => FileTrailerParser.GetFirstCrossReferenceOffset(input.Bytes, new CoreTokenScanner(input.Bytes), false);
Assert.Throws <PdfDocumentFormatException>(action);
}
public static IContainer GenerateContainer(ILog logger)
{
if (_testContainer != null)
{
return(_testContainer);
}
if (logger == null)
{
logger = new NoOpLog();
}
var headerParser = new FileHeaderParser(logger);
var trailerParser = new FileTrailerParser();
var nameParser = new CosNameParser();
var dictionaryParser = new CosDictionaryParser(nameParser, logger);
var baseParser = new CosBaseParser(nameParser, new CosStringParser(), dictionaryParser, new CosArrayParser());
var streamParser = new CosStreamParser(logger);
var filterProvider = new MemoryFilterProvider(new DecodeParameterResolver(logger), new PngPredictor(), logger);
var crossReferenceParser = new CrossReferenceStreamParser(filterProvider);
var objectStreamParser = new ObjectStreamParser(logger, filterProvider, baseParser);
var dynamicParser = new DynamicParser(logger, baseParser, streamParser, objectStreamParser);
var crossReferenceTableParser = new CrossReferenceParser(logger, dictionaryParser, baseParser, streamParser, crossReferenceParser, new CrossReferenceTableParser(),
new OldCrossReferenceTableParser(logger, dictionaryParser, baseParser));
var cmapParser = new CMapParser();
var afmParser = new AdobeFontMetricsParser();
var container = new Container();
container.Register(headerParser);
container.Register(trailerParser);
container.Register(nameParser);
container.Register(dictionaryParser);
container.Register(baseParser);
container.Register(streamParser);
container.Register(crossReferenceParser);
container.Register(crossReferenceTableParser);
container.Register(dynamicParser);
container.Register(objectStreamParser);
container.Register(filterProvider);
container.Register(cmapParser);
container.Register(afmParser);
container.Register(logger);
return(container);
}
public void FindsCompliantStartXref()
{
var input = StringBytesTestConverter.Convert(@"sta455%r endstream
endobj
12 0 obj
1234 %eof
endobj
startxref
456
%%EOF", false);
var result = FileTrailerParser.GetFirstCrossReferenceOffset(input.Bytes, new CoreTokenScanner(input.Bytes), false);
Assert.Equal(456, result);
}
public void MissingStartXrefThrows()
{
var input = StringBytesTestConverter.Convert(@"11 0 obj
<< /Type/Something /W[12 0 5 6] >>
endobj
12 0 obj
1234 %eof
endobj
startref
1384733
%%EOF
% I decided to put some nonsense here:
% because I could hahaha
start_rexf
17", false);
Action action = () => FileTrailerParser.GetFirstCrossReferenceOffset(input.Bytes, new CoreTokenScanner(input.Bytes), false);
Assert.Throws <PdfDocumentFormatException>(action);
}
public void IncludesStartXrefFollowingEndOfFile()
{
var input = StringBytesTestConverter.Convert(@"11 0 obj
<< /Type/Something /W[12 0 5 6] >>
endobj
12 0 obj
1234 %eof
endobj
startxref
1384733
%%EOF
% I decided to put some nonsense here:
% because I could hahaha
startxref
17", false);
var result = FileTrailerParser.GetFirstCrossReferenceOffset(input.Bytes, new CoreTokenScanner(input.Bytes), false);
Assert.Equal(17, result);
}
public void TakesLastStartXrefPrecedingEndOfFile()
{
var input = StringBytesTestConverter.Convert(@"11 0 obj
<< /Type/Something /W[12 0 5 6] >>
endobj
12 0 obj
1234 %eof
endobj
startxref
1384733
%actually I changed my mind
startxref
1274665676543
%%EOF", false);
var result = FileTrailerParser.GetFirstCrossReferenceOffset(input.Bytes, new CoreTokenScanner(input.Bytes), false);
Assert.Equal(1274665676543, result);
}
private static PdfDocument OpenDocument(IInputBytes inputBytes, ISeekableTokenScanner scanner, ILog log, bool isLenientParsing,
IReadOnlyList <string> passwords, bool clipPaths)
{
var filterProvider = DefaultFilterProvider.Instance;
CrossReferenceTable crossReferenceTable = null;
var xrefValidator = new XrefOffsetValidator(log);
// We're ok with this since our intent is to lazily load the cross reference table.
// ReSharper disable once AccessToModifiedClosure
var locationProvider = new ObjectLocationProvider(() => crossReferenceTable, inputBytes);
var pdfScanner = new PdfTokenScanner(inputBytes, locationProvider, filterProvider, NoOpEncryptionHandler.Instance);
var crossReferenceStreamParser = new CrossReferenceStreamParser(filterProvider);
var crossReferenceParser = new CrossReferenceParser(log, xrefValidator, crossReferenceStreamParser);
var version = FileHeaderParser.Parse(scanner, isLenientParsing, log);
var crossReferenceOffset = FileTrailerParser.GetFirstCrossReferenceOffset(inputBytes, scanner,
isLenientParsing) + version.OffsetInFile;
// TODO: make this use the scanner.
var validator = new CrossReferenceOffsetValidator(xrefValidator);
crossReferenceOffset = validator.Validate(crossReferenceOffset, scanner, inputBytes, isLenientParsing);
crossReferenceTable = crossReferenceParser.Parse(inputBytes, isLenientParsing,
crossReferenceOffset,
version.OffsetInFile,
pdfScanner,
scanner);
var(rootReference, rootDictionary) = ParseTrailer(crossReferenceTable, isLenientParsing,
pdfScanner,
out var encryptionDictionary);
var encryptionHandler = encryptionDictionary != null ?
(IEncryptionHandler) new EncryptionHandler(encryptionDictionary, crossReferenceTable.Trailer, passwords)
: NoOpEncryptionHandler.Instance;
pdfScanner.UpdateEncryptionHandler(encryptionHandler);
var cidFontFactory = new CidFontFactory(pdfScanner, filterProvider);
var encodingReader = new EncodingReader(pdfScanner);
var type1Handler = new Type1FontHandler(pdfScanner, filterProvider, encodingReader);
var fontFactory = new FontFactory(log, new Type0FontHandler(cidFontFactory,
filterProvider, pdfScanner),
new TrueTypeFontHandler(log, pdfScanner, filterProvider, encodingReader, SystemFontFinder.Instance,
type1Handler),
type1Handler,
new Type3FontHandler(pdfScanner, filterProvider, encodingReader));
var resourceContainer = new ResourceStore(pdfScanner, fontFactory);
var information = DocumentInformationFactory.Create(pdfScanner, crossReferenceTable.Trailer);
var catalog = CatalogFactory.Create(rootReference, rootDictionary, pdfScanner, isLenientParsing);
var pageFactory = new PageFactory(pdfScanner, resourceContainer, filterProvider,
new PageContentParser(new ReflectionGraphicsStateOperationFactory()),
log);
var caching = new ParsingCachingProviders(resourceContainer);
var acroFormFactory = new AcroFormFactory(pdfScanner, filterProvider, crossReferenceTable);
var bookmarksProvider = new BookmarksProvider(log, pdfScanner);
return(new PdfDocument(log, inputBytes, version, crossReferenceTable, caching, pageFactory, catalog, information,
encryptionDictionary,
pdfScanner,
filterProvider,
acroFormFactory,
bookmarksProvider,
clipPaths));
}