THotPDF.DecodeLoadedStreamToStream Method

Decodes a loaded PDF stream through its complete filter chain and writes the final stage directly to a caller-owned stream

Declaration

function DecodeLoadedStreamToStream(AStream: THPDFStreamObject;
  Output: TStream; out Info: THPDFDecodePipelineInfo): Boolean; overload;
function DecodeLoadedStreamToStream(AStream: THPDFStreamObject;
  Output: TStream; CancellationToken: THPDFCancellationToken;
  out Info: THPDFDecodePipelineInfo): Boolean; overload;

Unit

HPDFDoc

Parameters

Name	Purpose
`AStream`	Loaded `THPDFStreamObject` whose `/Filter` entry is decoded
`Output`	Caller-owned destination that receives decoded bytes at its current position
`CancellationToken`	Optional cooperative cancellation shared with custom filter stages
`Info`	Receives input and output byte counts, filter counts, depth, status, failing filter, and diagnostics

Return value

Returns True only when every requested stage succeeds

Usage notes

The final filter writes to Output without creating a second full-size final memory stream
Intermediate stages use THPDFAdaptiveTempStream and spill to owned temporary files after LoadedStreamMemoryThreshold
Every stage remains bounded by DecodeBudgetBytes and every decoded byte also consumes DocumentDecodeBudgetBytes
ASCII85 requires an exact ~> end marker and ASCIIHex requires >; malformed loaded pipelines fail instead of accepting truncated data
DecodeFilterLimit rejects oversized arrays before stage allocation
DecodePipelineDepthLimit, cyclic indirect-reference detection, and same-stream recursion detection bound hostile object graphs
The source stream position is restored after success or failure
Registered IHPDFCustomStreamFilter stages share the same aggregate budgets and pipeline depth
Custom push and pull output is isolated in an adaptive temporary stream until that custom stage succeeds
The cancellation overload returns dpsCancelled instead of leaking EHPDFOperationCancelled
A failed streaming destination can contain bytes already accepted before the failure; use a temporary destination when atomic output is required

Diagnostics

THPDFDecodePipelineInfo = record
  Status: THPDFDecodePipelineStatus;
  InputBytes: Int64;
  OutputBytes: Int64;
  FilterCount: Integer;
  AppliedFilterCount: Integer;
  CustomFilterCount: Integer;
  CustomFilterFailureCount: Integer;
  PipelineDepth: Integer;
  StopBeforeLast: Boolean;
  FailedFilter: AnsiString;
  Diagnostic: AnsiString;
end;

Status distinguishes invalid streams and filters, filter-count and depth limits, cyclic references, recursive streams, unsupported filters, budget failures, decoder failures, custom failures, cancellation, and seek-contract violations

Example

Output := TFileStream.Create('payload.bin', fmCreate);
try
  if not PDF.DecodeLoadedStreamToStream(StreamObject, Output, Info) then
    raise Exception.Create(String(Info.Diagnostic));
finally
  Output.Free;
end;