MimeBase64Decoder Class

Definition

Namespace: DataStandardizer.Communication.Mime

Decodes base64 data as it arrives, buffer by buffer.

Syntax

public sealed class MimeBase64Decoder

Remarks

The decoder works like System.Buffers.Text.Base64.DecodeFromUtf8, but keeps its state between calls. It holds an incomplete quantum until the next call, so it consumes all of a block that isn’t final, and the result is the same however the data is split.

The padding of the last quantum, = or ==, ends the data: after it, a character of the alphabet or another = is invalid. In Mime mode, characters outside the alphabet are ignored, before and after the padding, unless RejectNonAlphabetCharacters is set. The pad bits of the last quantum aren’t checked (RFC 4648 §3.5).

After a call with isFinalBlock set to true returns Done, the decoder is ready for new data. After InvalidData, call Reset first. An instance isn’t thread-safe.

Constructors

Constructor Notes
MimeBase64Decoder(MimeBase64Mode mode) With no options.
MimeBase64Decoder(MimeBase64Mode mode, MimeBase64DecoderOptions options) Throws ArgumentOutOfRangeException for a value that isn’t defined.

Properties

Property Type Notes
Mode MimeBase64Mode The form of base64 the decoder reads.
Options MimeBase64DecoderOptions The checks the decoder makes beyond those of its mode.

Methods

Method Notes
OperationStatus Decode(ReadOnlySpan<byte> source, Span<byte> destination, out int bytesConsumed, out int bytesWritten, bool isFinalBlock) Decodes from UTF-8, which for base64 is the same as ASCII.
OperationStatus Decode(ReadOnlySpan<char> source, Span<byte> destination, out int charsConsumed, out int bytesWritten, bool isFinalBlock) Decodes from characters.
void Reset() Discards any part of a quantum held, to start on new data.

Decode returns one of these:

Status Meaning
Done All of source was consumed, including anything the decoder holds for the next call.
DestinationTooSmall destination is full. Call again with the rest of source.
InvalidData The data isn’t valid. The consumed count is the position of the invalid character, or the length of source if the data ends inside a quantum.

This site uses Just the Docs, a documentation theme for Jekyll.