QR Codes can encode text as bits in one of several modes, and can use multiple modes @@ -28,7 +32,7 @@ import java.io.UnsupportedEncodingException; * *
See ISO 18004:2006, 6.4.3 - 6.4.7
* - * @author srowen@google.com (Sean Owen) + * @author Sean Owen */ final class DecodedBitStreamParser { @@ -41,25 +45,17 @@ final class DecodedBitStreamParser { 'O', 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W', 'X', 'Y', 'Z', ' ', '$', '%', '*', '+', '-', '.', '/', ':' }; - private static final String SHIFT_JIS = "SJIS"; - private static final String EUC_JP = "EUC_JP"; - private static final boolean ASSUME_SHIFT_JIS; - private static final String UTF8 = "UTF8"; - private static final String ISO88591 = "ISO8859_1"; - - static { - String platformDefault = System.getProperty("file.encoding"); - ASSUME_SHIFT_JIS = SHIFT_JIS.equalsIgnoreCase(platformDefault) || EUC_JP.equalsIgnoreCase(platformDefault); - } private DecodedBitStreamParser() { } - static String decode(byte[] bytes, Version version) throws ReaderException { + static DecoderResult decode(byte[] bytes, Version version, ErrorCorrectionLevel ecLevel, Hashtable hints) + throws FormatException { BitSource bits = new BitSource(bytes); - StringBuffer result = new StringBuffer(); + StringBuffer result = new StringBuffer(50); CharacterSetECI currentCharacterSetECI = null; boolean fc1InEffect = false; + Vector byteSegments = new Vector(1); Mode mode; do { // While still another segment to read... @@ -67,53 +63,51 @@ final class DecodedBitStreamParser { // OK, assume we're done. Really, a TERMINATOR mode should have been recorded here mode = Mode.TERMINATOR; } else { - mode = Mode.forBits(bits.readBits(4)); // mode is encoded by 4 bits + try { + mode = Mode.forBits(bits.readBits(4)); // mode is encoded by 4 bits + } catch (IllegalArgumentException iae) { + throw FormatException.getFormatInstance(); + } } if (!mode.equals(Mode.TERMINATOR)) { if (mode.equals(Mode.FNC1_FIRST_POSITION) || mode.equals(Mode.FNC1_SECOND_POSITION)) { // We do little with FNC1 except alter the parsed result a bit according to the spec fc1InEffect = true; + } else if (mode.equals(Mode.STRUCTURED_APPEND)) { + // not really supported; all we do is ignore it + // Read next 8 bits (symbol sequence #) and 8 bits (parity data), then continue + bits.readBits(16); } else if (mode.equals(Mode.ECI)) { // Count doesn't apply to ECI int value = parseECIValue(bits); - try { - currentCharacterSetECI = CharacterSetECI.getCharacterSetECIByValue(value); - } catch (IllegalArgumentException iae) { - // unsupported... just continue? + currentCharacterSetECI = CharacterSetECI.getCharacterSetECIByValue(value); + if (currentCharacterSetECI == null) { + throw FormatException.getFormatInstance(); } } else { // How many characters will follow, encoded in this mode? - int count = bits.readBits(mode.getCharacterCountBits(version)); + int count = bits.readBits(mode.getCharacterCountBits(version)); if (mode.equals(Mode.NUMERIC)) { decodeNumericSegment(bits, result, count); } else if (mode.equals(Mode.ALPHANUMERIC)) { decodeAlphanumericSegment(bits, result, count, fc1InEffect); } else if (mode.equals(Mode.BYTE)) { - decodeByteSegment(bits, result, count, currentCharacterSetECI); + decodeByteSegment(bits, result, count, currentCharacterSetECI, byteSegments, hints); } else if (mode.equals(Mode.KANJI)) { decodeKanjiSegment(bits, result, count); } else { - throw new ReaderException("Unsupported mode indicator"); + throw FormatException.getFormatInstance(); } } } } while (!mode.equals(Mode.TERMINATOR)); - // I thought it wasn't allowed to leave extra bytes after the terminator but it happens - /* - int bitsLeft = bits.available(); - if (bitsLeft > 0) { - if (bitsLeft > 6 || bits.readBits(bitsLeft) != 0) { - throw new ReaderException("Excess bits or non-zero bits after terminator mode indicator"); - } - } - */ - return result.toString(); + return new DecoderResult(bytes, result.toString(), byteSegments.isEmpty() ? null : byteSegments, ecLevel); } private static void decodeKanjiSegment(BitSource bits, StringBuffer result, - int count) throws ReaderException { + int count) throws FormatException { // Each character will require 2 bytes. Read the characters as 2-byte pairs // and decode as Shift_JIS afterwards byte[] buffer = new byte[2 * count]; @@ -136,19 +130,21 @@ final class DecodedBitStreamParser { } // Shift_JIS may not be supported in some environments: try { - result.append(new String(buffer, SHIFT_JIS)); + result.append(new String(buffer, StringUtils.SHIFT_JIS)); } catch (UnsupportedEncodingException uee) { - throw new ReaderException(SHIFT_JIS + " encoding is not supported on this device"); + throw FormatException.getFormatInstance(); } } private static void decodeByteSegment(BitSource bits, StringBuffer result, int count, - CharacterSetECI currentCharacterSetECI) throws ReaderException { + CharacterSetECI currentCharacterSetECI, + Vector byteSegments, + Hashtable hints) throws FormatException { byte[] readBytes = new byte[count]; if (count << 3 > bits.available()) { - throw new ReaderException("Count too large: " + count); + throw FormatException.getFormatInstance(); } for (int i = 0; i < count; i++) { readBytes[i] = (byte) bits.readBits(8); @@ -160,15 +156,16 @@ final class DecodedBitStreamParser { // upon decoding. I have seen ISO-8859-1 used as well as // Shift_JIS -- without anything like an ECI designator to // give a hint. - encoding = guessEncoding(readBytes); + encoding = StringUtils.guessEncoding(readBytes, hints); } else { encoding = currentCharacterSetECI.getEncodingName(); } try { result.append(new String(readBytes, encoding)); } catch (UnsupportedEncodingException uce) { - throw new ReaderException(uce.toString()); + throw FormatException.getFormatInstance(); } + byteSegments.addElement(readBytes); } private static void decodeAlphanumericSegment(BitSource bits, @@ -206,13 +203,13 @@ final class DecodedBitStreamParser { private static void decodeNumericSegment(BitSource bits, StringBuffer result, - int count) throws ReaderException { + int count) throws FormatException { // Read three digits at a time while (count >= 3) { // Each 10 bits encodes three digits int threeDigitsBits = bits.readBits(10); if (threeDigitsBits >= 1000) { - throw new ReaderException("Illegal value for 3-digit unit: " + threeDigitsBits); + throw FormatException.getFormatInstance(); } result.append(ALPHANUMERIC_CHARS[threeDigitsBits / 100]); result.append(ALPHANUMERIC_CHARS[(threeDigitsBits / 10) % 10]); @@ -223,7 +220,7 @@ final class DecodedBitStreamParser { // Two digits left over to read, encoded in 7 bits int twoDigitsBits = bits.readBits(7); if (twoDigitsBits >= 100) { - throw new ReaderException("Illegal value for 2-digit unit: " + twoDigitsBits); + throw FormatException.getFormatInstance(); } result.append(ALPHANUMERIC_CHARS[twoDigitsBits / 10]); result.append(ALPHANUMERIC_CHARS[twoDigitsBits % 10]); @@ -231,66 +228,11 @@ final class DecodedBitStreamParser { // One digit left over to read int digitBits = bits.readBits(4); if (digitBits >= 10) { - throw new ReaderException("Illegal value for digit unit: " + digitBits); + throw FormatException.getFormatInstance(); } result.append(ALPHANUMERIC_CHARS[digitBits]); } } - - private static String guessEncoding(byte[] bytes) { - if (ASSUME_SHIFT_JIS) { - return SHIFT_JIS; - } - // Does it start with the UTF-8 byte order mark? then guess it's UTF-8 - if (bytes.length > 3 && bytes[0] == (byte) 0xEF && bytes[1] == (byte) 0xBB && bytes[2] == (byte) 0xBF) { - return UTF8; - } - // For now, merely tries to distinguish ISO-8859-1, UTF-8 and Shift_JIS, - // which should be by far the most common encodings. ISO-8859-1 - // should not have bytes in the 0x80 - 0x9F range, while Shift_JIS - // uses this as a first byte of a two-byte character. If we see this - // followed by a valid second byte in Shift_JIS, assume it is Shift_JIS. - // If we see something else in that second byte, we'll make the risky guess - // that it's UTF-8. - int length = bytes.length; - boolean canBeISO88591 = true; - boolean lastWasPossibleDoubleByteStart = false; - for (int i = 0; i < length; i++) { - int value = bytes[i] & 0xFF; - if (value >= 0x80 && value <= 0x9F && i < length - 1) { - canBeISO88591 = false; - // ISO-8859-1 shouldn't use this, but before we decide it is Shift_JIS, - // just double check that it is followed by a byte that's valid in - // the Shift_JIS encoding - if (lastWasPossibleDoubleByteStart) { - // If we just checked this and the last byte for being a valid double-byte - // char, don't check starting on this byte. If the this and the last byte - // formed a valid pair, then this shouldn't be checked to see if it starts - // a double byte pair of course. - lastWasPossibleDoubleByteStart = false; - } else { - // ... otherwise do check to see if this plus the next byte form a valid - // double byte pair encoding a character. - lastWasPossibleDoubleByteStart = true; - int nextValue = bytes[i + 1] & 0xFF; - if ((value & 0x1) == 0) { - // if even, next value should be in [0x9F,0xFC] - // if not, we'll guess UTF-8 - if (nextValue < 0x9F || nextValue > 0xFC) { - return UTF8; - } - } else { - // if odd, next value should be in [0x40,0x9E] - // if not, we'll guess UTF-8 - if (nextValue < 0x40 || nextValue > 0x9E) { - return UTF8; - } - } - } - } - } - return canBeISO88591 ? ISO88591 : SHIFT_JIS; - } private static int parseECIValue(BitSource bits) { int firstByte = bits.readBits(8);