DocumentTaskSettings Class

[DataContractAttribute]
public class DocumentTaskSettings : ContentExtractionSettings

var taskSettings = new OpenDiscoverPlatform.DocumentTaskSettings();
taskSettings.CollectionId = "101";
taskSettings.TaskId       = Guid.NewGuid().ToString();

taskSettings.ProcessingTaskType = ProcessingType.SingleArchive; // Single (or multi-part) archive processing task
taskSettings.IsPartitioned   = true; // Archive will be partitioned
taskSettings.TotalPartitions = 4;    // Archive will be broken up and processed as 4 separate partitions (tasks).
taskSettings.PartitionTarget = 2;    // The partition # this task will work on (other DocumentTaskEngine instances can process the other partitions simultaneously)

var archivePath    = @"D:\InputDocuments\Archives\VeryLargeArchive.zip";
var outputRootPath = @"D:\Output\"; // Root path to store task output
var taskOutputPath = System.IO.Path.Combine(outputRootPath, string.Format(@"CollectionId_{0}\Task_{1}", taskSettings.CollectionId, taskSettings.TaskId));

// 
// For single archive or single mail store tasks, the input document(s) Document.FilePath and Document.FormatId properties should be set:
// 
var archiveDocument = new Document();
archiveDocument.FilePath = archivePath;
using (var docStream = System.IO.File.OpenRead(archivePath))
{
    archiveDocument.FormatId = OpenDiscoverSDK.DocumentIdentifier.Identify(docStream, archivePath);
}

//For a split (multi-part) archive, we would pass in a list of the split segment documents in order:
taskSettings.Documents      = new List<Document> () { archiveDocument };  
taskSettings.ProcessingMode = ProcessingMode.TextAndMetadata;
taskSettings.OutputMode     = OutputMode.IndividualFiles;  // Extracted attachments/ text will be saved as individual (flat) files.

// Set root path for processing output files:
taskSettings.DocumentArchiveRootPath = taskOutputPath;

taskSettings.Passwords        = null;  // No passwords to cycle through
taskSettings.PerformNistCheck = false; // No checking document binary hashes against NIST database.

taskSettings.EmbeddedObjectExtraction = EmbeddedExtractionType.EmbeddedDocumentsAndMedia;
taskSettings.ExcludeInlineEmailImages = true;

taskSettings.PdfDocument.ImageExtraction = PdfImageExtraction.OnlyFailedPdfPages;
taskSettings.PdfDocument.PageExtractedTextCriteria = 1;

taskSettings.TimeZoneAndEmail.CollectionTimeZone      = TimeZoneInfo.Utc;
taskSettings.TimeZoneAndEmail.ApplyTimeZoneToMetadata = false;
taskSettings.TimeZoneAndEmail.EmailDateTimeFormat     = EmailDateTimeFormat.MonthDayYearTime;
taskSettings.TimeZoneAndEmail.ShowUtcOffsetForTime    = true;

taskSettings.Hashing.HashingType = HashingType.BinaryAndContentHash;
taskSettings.Hashing.MaxBinaryHashLength = 10*1024*1024*1024;  // Hash up to a maximum of the first 10GB of a file
taskSettings.Hashing.IncludeBccRecipientsInEmailContentHash = false;

taskSettings.LanguageId.IdentifyLanguages = true;

taskSettings.UnsupportedFiltering.FilteringType = UnsupportedFilterType.Unsupported;
taskSettings.UnsupportedFiltering.LargeUnsupportedMaxFilteredChars = 1024 * 1024 * 1024; // Binary-to-text filter at max 1 billion chars

// 
// Create a document task engine instance to process the task:
// 
var documentTaskEngine = new DocumentTaskEngine(taskSettings);
documentTaskEngine.Completed      += _documentTaskEngine_Completed;
documentTaskEngine.FatalException += _documentTaskEngine_FatalException;
documentTaskEngine.LongProcessingDocumentWarning += _documentTaskEngine_LongProcessingDocumentWarning;

// 
// Run task synchronously (blocking):
// 
documentTaskEngine.RunTaskBlocking();

// TODO: do something with the output, like bulk insert into a document store or an eDiscovery document review system.