/**
* Reads an array of strings from the TIFF file.
*
* @param count Number of strings to read
* @param value Offset from which to read
*/
protected String[] readASCIIArray(long count, long value) throws IOException {
_raf.seek(value);
int nstrs = 0;
List list = new LinkedList();
byte[] buf = new byte[(int) count];
_raf.read(buf);
StringBuffer strbuf = new StringBuffer();
for (int i = 0; i < count; i++) {
int b = buf[i];
if (b == 0) {
list.add(strbuf.toString());
strbuf.setLength(0);
} else {
strbuf.append((char) b);
}
}
/* We can't use ArrayList.toArray because that returns an
Object[], not a String[] ... sigh. */
String[] strs = new String[nstrs];
ListIterator iter = list.listIterator();
for (int i = 0; i < nstrs; i++) {
strs[i] = (String) iter.next();
}
return strs;
}
/**
* Reads a chunk and puts an Instrument property into the RepInfo object.
*
* @return <code>false</code> if the chunk is structurally invalid, otherwise <code>true</code>
*/
public boolean readChunk(RepInfo info) throws IOException {
AiffModule module = (AiffModule) _module;
int baseNote = ModuleBase.readUnsignedByte(_dstream, module);
int detune = ModuleBase.readSignedByte(_dstream, module);
int lowNote = ModuleBase.readUnsignedByte(_dstream, module);
int highNote = ModuleBase.readUnsignedByte(_dstream, module);
int lowVelocity = ModuleBase.readUnsignedByte(_dstream, module);
int highVelocity = ModuleBase.readUnsignedByte(_dstream, module);
int gain = module.readSignedShort(_dstream);
Loop sustainLoop = readLoop(module);
Loop releaseLoop = readLoop(module);
List propList = new ArrayList(9);
propList.add(new Property("BaseNote", PropertyType.INTEGER, new Integer(baseNote)));
propList.add(new Property("Detune", PropertyType.INTEGER, new Integer(detune)));
propList.add(new Property("LowNote", PropertyType.INTEGER, new Integer(lowNote)));
propList.add(new Property("HighNote", PropertyType.INTEGER, new Integer(highNote)));
propList.add(new Property("LowVelocity", PropertyType.INTEGER, new Integer(lowVelocity)));
propList.add(new Property("HighVelocity", PropertyType.INTEGER, new Integer(highVelocity)));
propList.add(new Property("Gain", PropertyType.INTEGER, new Integer(gain)));
propList.add(sustainLoop.loopProp("SustainLoop"));
propList.add(releaseLoop.loopProp("ReleaseLoop"));
module.addAiffProperty(
new Property("Instrument", PropertyType.PROPERTY, PropertyArity.LIST, propList));
return true;
}
/* See if this document, even if it lacks a doctype, is most likely
* XHTML. The test is that the document starts with an XML declaration
* and has "html" for its first tag.
*
* Returns:
* 0 if there's no XML declaration
* 1 if there's an XML declaration but no html tag; in this
* case it's probably some other kind of XML
* 2 if there's an XML declaration and an html tag
*
*/
protected int seemsToBeXHTML(List elements) {
JHElement elem;
try {
elem = (JHElement) elements.get(0);
if (!(elem instanceof JHXmlDecl)) {
return 0;
}
Iterator iter = elements.iterator();
while (iter.hasNext()) {
elem = (JHElement) iter.next();
if (elem instanceof JHOpenTag) {
JHOpenTag tag = (JHOpenTag) elem;
return ("html".equals(tag.getName()) ? 2 : 1);
}
}
} catch (Exception e) {
return 0; // document must be really empty
}
return 1;
}
/**
* Returns a Property representing a bitmask. If <code>rawOutput</code> is true, returns a LIST
* property whose elements are STRING properties. The string values of these STRING properties are
* the elements of <code>labels</code> whose indices correspond to 1 bits in the bitmask, counting
* the low-order bit as bit 0. if <code>rawOutput</code> is false, returns a LONG property whose
* numeric value is <code>value</code>.
*/
protected Property addBitmaskProperty(
String name, long value, String[] labels, boolean rawOutput) {
Property prop = null;
if (!rawOutput) {
List list = new LinkedList();
try {
for (int i = 0; i < labels.length; i++) {
if ((value & (1 << i)) != 0) {
list.add(labels[i]);
}
}
} catch (Exception e) {
_errors.add(name + " value out of range: " + value);
}
prop = new Property(name, PropertyType.STRING, PropertyArity.LIST, list);
}
if (prop == null) {
prop = new Property(name, PropertyType.LONG, new Long(value));
}
return prop;
}
/* Check if there is a DOCTYPE at the start of the elements
* list. If there is, return the appropriate version string.
* If the DOCTYPE says it isn't HTML, trust it and call this
* document ill-formed by returning -1.
* If there is no DOCTYPE, or an unrecognized one, return 0.
*/
protected int checkDoctype(List elements) {
JHElement firstElem = (JHElement) elements.get(0);
if (firstElem instanceof JHXmlDecl && elements.size() >= 2) {
firstElem = (JHElement) elements.get(1);
}
if (!(firstElem instanceof JHDoctype)) {
return 0; // no DOCTYPE found
}
List dt = ((JHDoctype) firstElem).getDoctypeElements();
if (dt.size() < 3) {
return 0;
}
try {
// Is DOCTYPE case sensitive? Assume not.
String str = ((String) dt.get(0)).toUpperCase();
if (!"HTML".equals(str)) {
// It's not HTML
return -1;
}
str = ((String) dt.get(1)).toUpperCase();
if (!"PUBLIC".equals(str)) {
return 0;
}
str = stripQuotes(((String) dt.get(2)).toUpperCase());
_doctype = str;
if ("-//W3C//DTD HTML 3.2 FINAL//EN".equals(str) || "-//W3C//DTD HTML 3.2//EN".equals(str)) {
return HTML_3_2;
} else if ("-//W3C//DTD HTML 4.0//EN".equals(str)) {
return HTML_4_0_STRICT;
} else if ("-//W3C//DTD HTML 4.0 TRANSITIONAL//EN".equals(str)) {
return HTML_4_0_TRANSITIONAL;
} else if ("-//W3C//DTD HTML 4.0 FRAMESET//EN".equals(str)) {
return HTML_4_0_FRAMESET;
} else if ("-//W3C//DTD HTML 4.01//EN".equals(str)) {
return HTML_4_01_STRICT;
} else if ("-//W3C//DTD HTML 4.01 TRANSITIONAL//EN".equals(str)) {
return HTML_4_01_TRANSITIONAL;
} else if ("-//W3C//DTD HTML 4.01 FRAMESET//EN".equals(str)) {
return HTML_4_01_FRAMESET;
}
} catch (Exception e) {
// Really shouldn't happen, but if it does we've got
// a bad doctype
return 0;
}
return 0;
}
/**
* General function for adding a property with a 32-bit value, with two arrays of Strings to
* interpret 0 and 1 values as a bitmask.
*
* @param val The bitmask
* @param name The name for the Property
* @param oneValueNames Array of names to use for '1' values
* @param zeroValueNames Array of names to use for '0' values
*/
public Property buildBitmaskProperty(
int val, String name, String[] oneValueNames, String[] zeroValueNames) {
if (_je != null && _je.getShowRawFlag()) {
return new Property(name, PropertyType.INTEGER, new Integer(val));
} else {
List slist = new LinkedList();
try {
for (int i = 0; i < oneValueNames.length; i++) {
String s = null;
if ((val & (1 << i)) != 0) {
s = oneValueNames[i];
} else {
s = zeroValueNames[i];
}
if (s != null && s.length() > 0) {
slist.add(s);
}
}
} catch (Exception e) {
return null;
}
return new Property(name, PropertyType.STRING, PropertyArity.LIST, slist);
}
}
/**
* Returns an Property representing an integer value. If <code>rawOutput</code> is true, returns
* an INTEGER property, and <code>labels</code> and <code>index</code> are unused. Otherwise,
* returns a STRING property, with the string being the element of <code>labels</code> whose index
* is <code>value</code>.
*/
protected Property addIntegerProperty(
String name, int value, String[] labels, boolean rawOutput) {
Property prop = null;
if (!rawOutput) {
try {
prop = new Property(name, PropertyType.STRING, labels[value]);
} catch (Exception e) {
_errors.add(name + " value out of range: " + value);
}
}
if (prop == null) {
prop = new Property(name, PropertyType.INTEGER, new Integer(value));
}
return prop;
}
/**
* Returns an ARRAY Property representing an integer array. If <code>rawOutput</code> is true, the
* elements of the property array are INTEGER properties, and <code>labels</code> is unused.
* Otherwise, the elements of the array are STRING properties, with the elements of <code>value
* </code> used as indices into <code>labels</code>.
*/
protected Property addIntegerArrayProperty(
String name, int[] value, String[] labels, boolean rawOutput) {
Property prop = null;
if (!rawOutput) {
String[] s = new String[value.length];
for (int i = 0; i < value.length; i++) {
try {
s[i] = labels[value[i]];
} catch (Exception e) {
_errors.add(name + " value out of range: " + value[i]);
}
}
prop = new Property(name, PropertyType.STRING, PropertyArity.ARRAY, s);
}
if (prop == null) {
prop = new Property(name, PropertyType.INTEGER, PropertyArity.ARRAY, value);
}
return prop;
}
/** Initializes the state of the module for parsing. */
protected void initParse() {
super.initParse();
_propList = new LinkedList();
_notes = new LinkedList();
_labels = new LinkedList();
_labeledText = new LinkedList();
_samples = new LinkedList();
firstSampleOffsetMarked = false;
numSamples = 0;
_metadata = new Property("WAVEMetadata", PropertyType.PROPERTY, PropertyArity.LIST, _propList);
_aesMetadata = new AESAudioMetadata();
_aesMetadata.setByteOrder(AESAudioMetadata.LITTLE_ENDIAN);
_aesMetadata.setAnalogDigitalFlag("FILE_DIGITAL");
_aesMetadata.setFormat("WAVE");
_aesMetadata.setUse("OTHER", "JHOVE_validation");
_aesMetadata.setDirection("NONE");
_propList.add(new Property("AESAudioMetadata", PropertyType.AESAUDIOMETADATA, _aesMetadata));
// Most chunk types are allowed to occur only once,
// and a few must occur exactly once.
// Clear flags for whether they have been seen.
formatChunkSeen = false;
dataChunkSeen = false;
instrumentChunkSeen = false;
cartChunkSeen = false;
mpegChunkSeen = false;
broadcastExtChunkSeen = false;
peakChunkSeen = false;
linkChunkSeen = false;
cueChunkSeen = false;
// Initialize profile flags
flagPCMWaveFormat = false;
flagWaveFormatEx = false;
flagWaveFormatExtensible = false;
flagBroadcastWave = false;
}
/**
* Returns an Property representing an integer value. If <code>rawOutput</code> is true, returns
* an INTEGER property, and <code>labels</code> and <code>index</code> are unused. Otherwise,
* returns a STRING property, with the string being the element of <code>labels</code> whose index
* is the index of <code>value</code> in <code>index</code>.
*/
protected Property addIntegerProperty(
String name, int value, String[] labels, int[] index, boolean rawOutput) {
Property prop = null;
if (!rawOutput) {
int n = -1;
for (int i = 0; i < index.length; i++) {
if (value == index[i]) {
n = i;
break;
}
}
if (n > -1) {
prop = new Property(name, PropertyType.STRING, labels[n]);
} else {
_errors.add(name + " value out of range: " + value);
}
}
if (prop == null) {
prop = new Property(name, PropertyType.INTEGER, new Integer(value));
}
return prop;
}
/**
* Parses the content of a purported WAVE digital object and stores the results in RepInfo.
*
* @param stream An InputStream, positioned at its beginning, which is generated from the object
* to be parsed
* @param info A fresh RepInfo object which will be modified to reflect the results of the parsing
* @param parseIndex Must be 0 in first call to <code>parse</code>. If <code>parse</code> returns
* a nonzero value, it must be called again with <code>parseIndex</code> equal to that return
* value.
*/
public int parse(InputStream stream, RepInfo info, int parseIndex) throws IOException {
initParse();
info.setFormat(_format[0]);
info.setMimeType(_mimeType[0]);
info.setModule(this);
_aesMetadata.setPrimaryIdentifier(info.getUri());
if (info.getURLFlag()) {
_aesMetadata.setOtherPrimaryIdentifierType("URI");
} else {
_aesMetadata.setPrimaryIdentifierType(AESAudioMetadata.FILE_NAME);
}
/* We may have already done the checksums while converting a
temporary file. */
_ckSummer = null;
if (_je != null && _je.getChecksumFlag() && info.getChecksum().size() == 0) {
_ckSummer = new Checksummer();
_cstream = new ChecksumInputStream(stream, _ckSummer);
_dstream = getBufferedDataStream(_cstream, _je != null ? _je.getBufferSize() : 0);
} else {
_dstream = getBufferedDataStream(stream, _je != null ? _je.getBufferSize() : 0);
}
try {
// Check the start of the file for the right opening bytes
for (int i = 0; i < 4; i++) {
int ch = readUnsignedByte(_dstream, this);
if (ch != sigByte[i]) {
info.setMessage(new ErrorMessage("Document does not start with RIFF chunk", 0));
info.setWellFormed(false);
return 0;
}
}
/* If we got this far, take note that the signature is OK. */
info.setSigMatch(_name);
// Get the length of the Form chunk. This includes all
// the subsequent chunks in the file, but excludes the
// header ("FORM" and the length itself).
bytesRemaining = readUnsignedInt(_dstream);
// Read the file type.
String typ = read4Chars(_dstream);
bytesRemaining -= 4;
if (!"WAVE".equals(typ)) {
info.setMessage(new ErrorMessage("File type in RIFF header is not WAVE", _nByte));
info.setWellFormed(false);
return 0;
}
while (bytesRemaining > 0) {
if (!readChunk(info)) {
break;
}
}
} catch (EOFException e) {
info.setWellFormed(false);
info.setMessage(new ErrorMessage("Unexpected end of file", _nByte));
return 0;
}
// Set duration from number of samples and rate.
if (numSamples > 0) {
// _aesMetadata.setDuration((double) numSamples / sampleRate);
_aesMetadata.setDuration(numSamples);
}
// Add note and label properties, if there's anything
// to report.
if (!_labels.isEmpty()) {
_propList.add(new Property("Labels", PropertyType.PROPERTY, PropertyArity.LIST, _labels));
}
if (!_labeledText.isEmpty()) {
_propList.add(
new Property("LabeledText", PropertyType.PROPERTY, PropertyArity.LIST, _labeledText));
}
if (!_notes.isEmpty()) {
_propList.add(new Property("Notes", PropertyType.PROPERTY, PropertyArity.LIST, _notes));
}
if (!_samples.isEmpty()) {
_propList.add(new Property("Samples", PropertyType.PROPERTY, PropertyArity.LIST, _samples));
}
if (_exifInfo != null) {
_propList.add(_exifInfo.buildProperty());
}
if (!formatChunkSeen) {
info.setMessage(new ErrorMessage("No Format Chunk"));
info.setWellFormed(false);
return 0;
}
/* This file looks OK. */
if (_ckSummer != null) {
/* We may not have actually hit the end of file. If we're calculating
* checksums on the fly, we have to read and discard whatever is
* left, so it will get checksummed. */
for (; ; ) {
try {
int n = skipBytes(_dstream, 2048, this);
if (n == 0) {
break;
}
} catch (Exception e) {
break;
}
}
info.setSize(_cstream.getNBytes());
info.setChecksum(new Checksum(_ckSummer.getCRC32(), ChecksumType.CRC32));
String value = _ckSummer.getMD5();
if (value != null) {
info.setChecksum(new Checksum(value, ChecksumType.MD5));
}
if ((value = _ckSummer.getSHA1()) != null) {
info.setChecksum(new Checksum(value, ChecksumType.SHA1));
}
}
info.setProperty(_metadata);
// Indicate satisfied profiles.
if (flagPCMWaveFormat) {
info.setProfile("PCMWAVEFORMAT");
}
if (flagWaveFormatEx) {
info.setProfile("WAVEFORMATEX");
}
if (flagWaveFormatExtensible) {
info.setProfile("WAVEFORMATEXTENSIBLE");
}
if (flagBroadcastWave) {
// Need to do some additional checks.
if (!broadcastExtChunkSeen) {
flagBroadcastWave = false;
}
if (compressionCode == FormatChunk.WAVE_FORMAT_MPEG) {
if (!broadcastExtChunkSeen || !factChunkSeen) {
flagBroadcastWave = false;
}
}
if (flagBroadcastWave) {
String prof = null;
switch (broadcastVersion) {
case 0:
prof = "Broadcast Wave Version 0";
break;
case 1:
prof = "Broadcast Wave Version 1";
break;
// Other versions are unknown at this time
}
if (prof != null) {
info.setProfile(prof);
}
}
}
return 0;
}
/**
* Reads a chunk and puts a BroadcastAudioExtension Property into the RepInfo object.
*
* @return <code>false</code> if the chunk is structurally invalid, otherwise <code>true</code>
*/
public boolean readChunk(RepInfo info) throws IOException {
WaveModule module = (WaveModule) _module;
byte[] buf256 = new byte[256];
ModuleBase.readByteBuf(_dstream, buf256, module);
String description = byteBufString(buf256);
byte[] buf32 = new byte[32];
ModuleBase.readByteBuf(_dstream, buf32, module);
String originator = byteBufString(buf32);
ModuleBase.readByteBuf(_dstream, buf32, module);
String originatorRef = byteBufString(buf32);
byte[] buf10 = new byte[10];
ModuleBase.readByteBuf(_dstream, buf10, module);
String originationDate = byteBufString(buf10);
byte[] buf8 = new byte[8];
ModuleBase.readByteBuf(_dstream, buf8, module);
String originationTime = byteBufString(buf8);
// TimeReference is stored as a 64-bit little-endian
// number -- I think
long timeReference = module.readSignedLong(_dstream);
int version = module.readUnsignedShort(_dstream);
module.setBroadcastVersion(version);
byte[] smtpe_umid = new byte[64];
ModuleBase.readByteBuf(_dstream, smtpe_umid, module);
module.skipBytes(_dstream, 190, module);
String codingHistory = "";
if (bytesLeft > 602) {
byte[] bufCodingHistory = new byte[(int) bytesLeft - 602];
ModuleBase.readByteBuf(_dstream, bufCodingHistory, module);
codingHistory = byteBufString(bufCodingHistory);
}
// Whew -- we've read the whole thing. Now make that into a
// list of Properties.
List plist = new ArrayList(20);
if (description.length() > 0) {
plist.add(new Property("Description", PropertyType.STRING, description));
}
if (originator.length() > 0) {
plist.add(new Property("Originator", PropertyType.STRING, originator));
}
if (originationDate.length() > 0) {
plist.add(new Property("OriginationDate", PropertyType.STRING, originationDate));
}
if (originationTime.length() > 0) {
plist.add(new Property("OriginationTime", PropertyType.STRING, originationTime));
}
plist.add(new Property("TimeReference", PropertyType.LONG, new Long(timeReference)));
plist.add(new Property("Version", PropertyType.INTEGER, new Integer(version)));
plist.add(new Property("UMID", PropertyType.BYTE, PropertyArity.ARRAY, smtpe_umid));
if (codingHistory.length() > 0) {
plist.add(new Property("CodingHistory", PropertyType.STRING, codingHistory));
}
module.addWaveProperty(
new Property("BroadcastAudioExtension", PropertyType.PROPERTY, PropertyArity.LIST, plist));
// set time reference in AES metadata set @author David Ackerman
AESAudioMetadata aes = module.getAESMetadata();
aes.setStartTime(timeReference);
return true;
}
/** Adds a Label property */
public void addLabel(Property p) {
_labels.add(p);
}
/**
* Parse the content of a purported HTML stream digital object and store the results in RepInfo.
*
* @param stream An InputStream, positioned at its beginning, which is generated from the object
* to be parsed. If multiple calls to <code>parse</code> are made on the basis of a nonzero
* value being returned, a new InputStream must be provided each time.
* @param info A fresh (on the first call) RepInfo object which will be modified to reflect the
* results of the parsing If multiple calls to <code>parse</code> are made on the basis of a
* nonzero value being returned, the same RepInfo object should be passed with each call.
* @param parseIndex Must be 0 in first call to <code>parse</code>. If <code>parse</code> returns
* a nonzero value, it must be called again with <code>parseIndex</code> equal to that return
* value.
*/
public int parse(InputStream stream, RepInfo info, int parseIndex) throws IOException {
if (parseIndex != 0) {
// Coming in with parseIndex = 1 indicates that we've determined
// this is XHTML; so we invoke the XML module to parse it.
// If parseIndex is 100, this is the first invocation of the
// XML module, so we call it with 0; otherwise we call it with
// the value of parseIndex.
if (isXmlAvailable()) {
edu.harvard.hul.ois.jhove.module.XmlModule xmlMod =
new edu.harvard.hul.ois.jhove.module.XmlModule();
if (parseIndex == 100) {
parseIndex = 0;
}
xmlMod.setApp(_app);
xmlMod.setBase(_je);
xmlMod.setDefaultParams(_defaultParams);
try {
xmlMod.applyDefaultParams();
} catch (Exception e) {
// really shouldn't happen
}
xmlMod.setXhtmlDoctype(_doctype);
return xmlMod.parse(stream, info, parseIndex);
} else {
// The XML module shouldn't be missing from any installation,
// but someone who really wanted to could remove it. In
// that case, you deserve what you get.
info.setMessage(new ErrorMessage("XML-HUL module required to validate XHTML documents"));
info.setWellFormed(false); // Treat it as completely wrong
return 0;
}
} else {
/* parseIndex = 0, first call only */
_doctype = null;
}
// Test if textMD is to be generated
if (_defaultParams != null) {
Iterator iter = _defaultParams.iterator();
while (iter.hasNext()) {
String param = (String) iter.next();
if (param.toLowerCase().equals("withtextmd=true")) {
_withTextMD = true;
}
}
}
initParse();
info.setFormat(_format[0]);
info.setMimeType(_mimeType[0]);
info.setModule(this);
if (_textMD == null || parseIndex == 0) {
_textMD = new TextMDMetadata();
}
/* We may have already done the checksums while converting a
temporary file. */
Checksummer ckSummer = null;
if (_je != null && _je.getChecksumFlag() && info.getChecksum().size() == 0) {
ckSummer = new Checksummer();
_cstream = new ChecksumInputStream(stream, ckSummer);
_dstream = getBufferedDataStream(_cstream, _je != null ? _je.getBufferSize() : 0);
} else {
_dstream = getBufferedDataStream(stream, _je != null ? _je.getBufferSize() : 0);
}
ParseHtml parser = null;
HtmlMetadata metadata = null;
HtmlCharStream cstream = null;
try {
cstream = new HtmlCharStream(_dstream, "ISO-8859-1");
parser = new ParseHtml(cstream);
} catch (UnsupportedEncodingException e) {
info.setMessage(new ErrorMessage("Internal error: " + e.getMessage()));
info.setWellFormed(false);
return 0; // shouldn't happen!
}
int type = 0;
try {
List elements = parser.HtmlDoc();
if (elements.isEmpty()) {
// Consider an empty document bad
info.setWellFormed(false);
info.setMessage(new ErrorMessage("Document is empty"));
return 0;
}
type = checkDoctype(elements);
if (type < 0) {
info.setWellFormed(false);
info.setMessage(new ErrorMessage("DOCTYPE is not HTML"));
return 0;
}
/* Check if there is at least one html, head, body or title tag.
* A plain text document
* might be interpreted as a single PCDATA, which is in some
* ethereal sense well-formed HTML, but it's pointless to consider
* it such. It might also use angle brackets as a text delimiter,
* and that shouldn't count as HTML either. */
boolean hasElements = false;
Iterator iter = elements.iterator();
while (iter.hasNext()) {
Object o = iter.next();
if (o instanceof JHOpenTag) {
String name = ((JHOpenTag) o).getName();
if ("html".equals(name)
|| "head".equals(name)
|| "body".equals(name)
|| "title".equals(name)) {
hasElements = true;
}
break;
}
}
if (!hasElements) {
info.setMessage(new ErrorMessage("Document contains no html, head, body or title tags"));
info.setWellFormed(false);
return 0;
}
// CRLF from HtmlCharStream ...
String lineEnd = cstream.getKindOfLineEnd();
if (lineEnd == null) {
info.setMessage(new InfoMessage("Not able to determine type of end of line"));
_textMD.setLinebreak(TextMDMetadata.NILL);
} else if (lineEnd.equalsIgnoreCase("CR")) {
_textMD.setLinebreak(TextMDMetadata.LINEBREAK_CR);
} else if (lineEnd.equalsIgnoreCase("LF")) {
_textMD.setLinebreak(TextMDMetadata.LINEBREAK_LF);
} else if (lineEnd.equalsIgnoreCase("CRLF")) {
_textMD.setLinebreak(TextMDMetadata.LINEBREAK_CRLF);
}
if (type == 0) {
/* If we can't find a doctype, it still might be XHTML
* if the elements start with an XML declaration and
* the root element is "html" */
switch (seemsToBeXHTML(elements)) {
case 0: // Not XML
break; // fall through
case 1: // XML but not HTML
info.setMessage(
new ErrorMessage(
"Document has XML declaration but no DOCTYPE; "
+ "probably XML rather than HTML"));
info.setWellFormed(false);
return 0;
case 2: // probably XHTML
return 100;
}
info.setMessage(
new ErrorMessage(
"Unrecognized or missing DOCTYPE declaration; "
+ "validation continuing as HTML 3.2"));
info.setValid(false);
// But keep going
}
HtmlDocDesc docDesc = null;
switch (type) {
case HTML_3_2:
default:
docDesc = new Html3_2DocDesc();
_textMD.setMarkup_basis("HTML");
_textMD.setMarkup_basis_version("3.2");
break;
case HTML_4_0_FRAMESET:
docDesc = new Html4_0FrameDocDesc();
_textMD.setMarkup_basis("HTML");
_textMD.setMarkup_basis_version("4.0");
break;
case HTML_4_0_TRANSITIONAL:
docDesc = new Html4_0TransDocDesc();
_textMD.setMarkup_basis("HTML");
_textMD.setMarkup_basis_version("4.0");
break;
case HTML_4_0_STRICT:
docDesc = new Html4_0StrictDocDesc();
_textMD.setMarkup_basis("HTML");
_textMD.setMarkup_basis_version("4.0");
break;
case HTML_4_01_FRAMESET:
docDesc = new Html4_01FrameDocDesc();
_textMD.setMarkup_basis("HTML");
_textMD.setMarkup_basis_version("4.01");
break;
case HTML_4_01_TRANSITIONAL:
docDesc = new Html4_01TransDocDesc();
_textMD.setMarkup_basis("HTML");
_textMD.setMarkup_basis_version("4.01");
break;
case HTML_4_01_STRICT:
docDesc = new Html4_01StrictDocDesc();
_textMD.setMarkup_basis("HTML");
_textMD.setMarkup_basis_version("4.01");
break;
case XHTML_1_0_STRICT:
case XHTML_1_0_TRANSITIONAL:
case XHTML_1_0_FRAMESET:
case XHTML_1_1:
// Force a second call to parse as XML. 100 is a
// magic code for the first XML call.
return 100;
}
_textMD.setMarkup_language(_doctype);
if (docDesc == null) {
info.setMessage(
new InfoMessage(
"Code for appropriate HTML version not available yet:" + "substituting HTML 3.2"));
docDesc = new Html3_2DocDesc();
}
docDesc.validate(elements, info);
metadata = docDesc.getMetadata();
// Try to get the charset from the meta Content
if (metadata.getCharset() != null) {
_textMD.setCharset(metadata.getCharset());
} else {
_textMD.setCharset(TextMDMetadata.CHARSET_ISO8859_1);
}
String textMDEncoding = _textMD.getCharset();
if (textMDEncoding.indexOf("UTF") != -1) {
_textMD.setByte_order(
_bigEndian ? TextMDMetadata.BYTE_ORDER_BIG : TextMDMetadata.BYTE_ORDER_LITTLE);
_textMD.setByte_size("8");
_textMD.setCharacter_size("variable");
} else {
_textMD.setByte_order(
_bigEndian ? TextMDMetadata.BYTE_ORDER_BIG : TextMDMetadata.BYTE_ORDER_LITTLE);
_textMD.setByte_size("8");
_textMD.setCharacter_size("1");
}
} catch (ParseException e) {
Token t = e.currentToken;
info.setMessage(
new ErrorMessage("Parse error", "Line = " + t.beginLine + ", column = " + t.beginColumn));
info.setWellFormed(false);
} catch (TokenMgrError f) {
info.setMessage(new ErrorMessage("TokenMgrError: " + f.getLocalizedMessage()));
info.setWellFormed(false);
}
if (info.getWellFormed() == RepInfo.FALSE) {
return 0;
}
if (type != 0) {
if (profileNames[type] != null) {
info.setProfile(profileNames[type]);
}
info.setVersion(versionNames[type]);
}
if (metadata != null) {
Property property = metadata.toProperty(_withTextMD ? _textMD : null);
if (property != null) {
info.setProperty(property);
}
}
if (ckSummer != null) {
info.setSize(_cstream.getNBytes());
info.setChecksum(new Checksum(ckSummer.getCRC32(), ChecksumType.CRC32));
String value = ckSummer.getMD5();
if (value != null) {
info.setChecksum(new Checksum(value, ChecksumType.MD5));
}
if ((value = ckSummer.getSHA1()) != null) {
info.setChecksum(new Checksum(value, ChecksumType.SHA1));
}
}
return 0;
}
/** Adds a LabeledText property */
public void addLabeledText(Property p) {
_labeledText.add(p);
}
/** Adds a Property to the WAVE metadata. */
public void addWaveProperty(Property prop) {
_propList.add(prop);
}
/** Adds the ListInfo property, which is a List of String Properties. */
public void addListInfo(List l) {
_propList.add(new Property("ListInfo", PropertyType.PROPERTY, PropertyArity.LIST, l));
}
/** Adds a Note string */
public void addNote(Property p) {
_notes.add(p);
}
/** Adds a Sample property */
public void addSample(Property p) {
_samples.add(p);
}
