Beispiel #1
0
size_t noopEncoder::decode(utility::inputStream& in,
	utility::outputStream& out, utility::progressListener* progress)
{
	in.reset();  // may not work...

	// No decoding performed
	size_t res = 0;

	if (progress)
		res = utility::bufferedStreamCopy(in, out, 0, progress);
	else
		res = utility::bufferedStreamCopy(in, out);

	return res;
}
Beispiel #2
0
size_t qpEncoder::decode(
	utility::inputStream& in,
	utility::outputStream& out,
	utility::progressListener* progress
) {

	in.reset();  // may not work...

	// Process the data
	const bool rfc2047 = getProperties().getProperty <bool>("rfc2047", false);

	byte_t buffer[16384];
	size_t bufferLength = 0;
	size_t bufferPos = 0;

	byte_t outBuffer[16384];
	size_t outBufferPos = 0;

	size_t total = 0;
	size_t inTotal = 0;

	while (bufferPos < bufferLength || !in.eof()) {

		// Flush current output buffer
		if (outBufferPos >= sizeof(outBuffer)) {

			QP_WRITE(out, outBuffer, outBufferPos);

			total += outBufferPos;
			outBufferPos = 0;
		}

		// Need to get more data?
		if (bufferPos >= bufferLength) {

			bufferLength = in.read(buffer, sizeof(buffer));
			bufferPos = 0;

			// No more data
			if (bufferLength == 0) {
				break;
			}
		}

		// Decode the next sequence (hex-encoded byte or printable character)
		byte_t c = buffer[bufferPos++];

		++inTotal;

		switch (c) {

			case '=': {

				if (bufferPos >= bufferLength) {
					bufferLength = in.read(buffer, sizeof(buffer));
					bufferPos = 0;
				}

				if (bufferPos < bufferLength) {

					c = buffer[bufferPos++];

					++inTotal;

					switch (c) {

						// Ignore soft line break ("=\r\n" or "=\n")
						case '\r':

							// Read one byte more
							if (bufferPos >= bufferLength) {
								bufferLength = in.read(buffer, sizeof(buffer));
								bufferPos = 0;
							}

							if (bufferPos < bufferLength) {
								++bufferPos;
								++inTotal;
							}

							break;

						case '\n':

							break;

						// Hex-encoded char
						default:
						{
							// We need another byte...
							if (bufferPos >= bufferLength) {
								bufferLength = in.read(buffer, sizeof(buffer));
								bufferPos = 0;
							}

							if (bufferPos < bufferLength) {

								const byte_t next = buffer[bufferPos++];

								++inTotal;

								const byte_t value = static_cast <byte_t>(
									sm_hexDecodeTable[c] * 16 + sm_hexDecodeTable[next]
								);

								outBuffer[outBufferPos++] = value;

							} else {

								// Premature end-of-data
							}

							break;
						}

					}

				} else {

					// Premature end-of-data
				}

				break;
			}
			case '_': {

				if (rfc2047) {

					// RFC-2047, Page 5, 4.2. The "Q" encoding:
					// << Note that the "_" always represents hexadecimal 20, even if the SPACE
					// character occupies a different code position in the character set in use. >>
					outBuffer[outBufferPos++] = 0x20;
					break;
				}

				outBuffer[outBufferPos++] = c;
				break;
			}
			default: {

				outBuffer[outBufferPos++] = c;
				break;
			}

		}

		if (progress) {
			progress->progress(inTotal, inTotal);
		}
	}

	// Flush remaining output buffer
	if (outBufferPos != 0) {
		QP_WRITE(out, outBuffer, outBufferPos);
		total += outBufferPos;
	}

	if (progress) {
		progress->stop(inTotal);
	}

	return total;
}
Beispiel #3
0
size_t qpEncoder::encode(
	utility::inputStream& in,
	utility::outputStream& out,
	utility::progressListener* progress
) {

	in.reset();  // may not work...

	const size_t propMaxLineLength =
		getProperties().getProperty <size_t>("maxlinelength", static_cast <size_t>(-1));

	const bool rfc2047 = getProperties().getProperty <bool>("rfc2047", false);
	const bool text = getProperties().getProperty <bool>("text", false);  // binary mode by default

	const bool cutLines = (propMaxLineLength != static_cast <size_t>(-1));
	const size_t maxLineLength = std::min(propMaxLineLength, static_cast <size_t>(74));

	// Process the data
	byte_t buffer[16384];
	size_t bufferLength = 0;
	size_t bufferPos = 0;

	size_t curCol = 0;

	byte_t outBuffer[16384];
	size_t outBufferPos = 0;

	size_t total = 0;
	size_t inTotal = 0;

	if (progress) {
		progress->start(0);
	}

	while (bufferPos < bufferLength || !in.eof()) {

		// Flush current output buffer
		if (outBufferPos + 6 >= static_cast <int>(sizeof(outBuffer))) {

			QP_WRITE(out, outBuffer, outBufferPos);

			total += outBufferPos;
			outBufferPos = 0;
		}

		// Need to get more data?
		if (bufferPos >= bufferLength) {

			bufferLength = in.read(buffer, sizeof(buffer));
			bufferPos = 0;

			// No more data
			if (bufferLength == 0) {
				break;
			}
		}

		// Get the next char and encode it
		const byte_t c = buffer[bufferPos++];

		if (rfc2047) {

			if (c >= 128 || sm_RFC2047EncodeTable[c] != 0) {

				if (c == 32) {  // space

					// RFC-2047, Page 5, 4.2. The "Q" encoding:
					// << The 8-bit hexadecimal value 20 (e.g., ISO-8859-1 SPACE) may be
					// represented as "_" (underscore, ASCII 95.). >>
					outBuffer[outBufferPos++] = '_';
					++curCol;

				} else {

					// Other characters: '=' + hexadecimal encoding
					QP_ENCODE_HEX(c);
				}

			} else {

				// No encoding
				outBuffer[outBufferPos++] = c;
				++curCol;
			}

		} else {

			switch (c) {

				case 46: {  // .

					if (curCol == 0) {
						// If a '.' appears at the beginning of a line, we encode it to
						// to avoid problems with SMTP servers... ("\r\n.\r\n" means the
						// end of data transmission).
						QP_ENCODE_HEX('.');
						continue;
					}

					outBuffer[outBufferPos++] = '.';
					++curCol;
					break;
				}
				case 32: {  // space

					// Need to get more data?
					if (bufferPos >= bufferLength) {
						bufferLength = in.read(buffer, sizeof(buffer));
						bufferPos = 0;
					}

					// Spaces cannot appear at the end of a line. So, encode the space.
					if (bufferPos >= bufferLength ||
					    (buffer[bufferPos] == '\r' || buffer[bufferPos] == '\n')) {

						QP_ENCODE_HEX(' ');

					} else {

						outBuffer[outBufferPos++] = ' ';
						++curCol;
					}

					break;
				}
				case 9: {   // TAB

					QP_ENCODE_HEX(c);
					break;
				}
				case 13:    // CR
				case 10: {  // LF

					// RFC-2045/6.7(4)

					// Text data
					if (text && !rfc2047) {

						outBuffer[outBufferPos++] = c;
						++curCol;

						if (c == 10) {
							curCol = 0;  // reset current line length
						}

					// Binary data
					} else {

						QP_ENCODE_HEX(c);
					}

					break;
				}
				case 61: {  // =

					QP_ENCODE_HEX('=');
					break;
				}
				/*
					Rule #2: (Literal representation) Octets with decimal values of 33
					through 60 inclusive, and 62 through 126, inclusive, MAY be
					represented as the ASCII characters which correspond to those
					octets (EXCLAMATION POINT through LESS THAN, and GREATER THAN
					through TILDE, respectively).
				*/
				default:

					//if ((c >= 33 && c <= 60) || (c >= 62 && c <= 126))
					if (c >= 33 && c <= 126 && c != 61 && c != 63) {

						outBuffer[outBufferPos++] = c;
						++curCol;

					// Other characters: '=' + hexadecimal encoding
					} else {

						QP_ENCODE_HEX(c);
					}

					break;

			} // switch (c)

			// Soft line break : "=\r\n"
			if (cutLines && curCol >= maxLineLength - 1) {

				outBuffer[outBufferPos] = '=';
				outBuffer[outBufferPos + 1] = '\r';
				outBuffer[outBufferPos + 2] = '\n';

				outBufferPos += 3;
				curCol = 0;
			}

		} // !rfc2047

		++inTotal;

		if (progress) {
			progress->progress(inTotal, inTotal);
		}
	}

	// Flush remaining output buffer
	if (outBufferPos != 0) {
		QP_WRITE(out, outBuffer, outBufferPos);
		total += outBufferPos;
	}

	if (progress) {
		progress->stop(inTotal);
	}

	return total;
}