/**
* <b>SAX:</b> Resolve the given entity into an input source. If the name can't be mapped to a
* preferred form of the entity, the URI is used. To resolve the entity, first a local catalog
* mapping names to URIs is consulted. If no mapping is found there, a catalog mapping names to
* java resources is consulted. Finally, if neither mapping found a copy of the entity, the
* specified URI is used.
*
* <p>
*
* <p>When a URI is used, <a href="#createInputSource">createInputSource</a> is used to correctly
* deduce the character encoding used by this entity. No MIME type checking is done.
*
* @param name Used to find alternate copies of the entity, when this value is non-null; this is
* the XML "public ID".
* @param uri Used when no alternate copy of the entity is found; this is the XML "system ID",
* normally a URI.
*/
public InputSource resolveEntity(String name, String uri) throws IOException {
InputSource retval;
String mappedURI = name2uri(name);
InputStream stream;
// prefer explicit URI mappings, then bundled resources...
if (mappedURI == null && (stream = mapResource(name)) != null) {
uri = "java:resource:" + (String) id2resource.get(name);
retval = new InputSource(XmlReader.createReader(stream));
// ...and treat all URIs the same (as URLs for now).
} else {
URL url;
URLConnection conn;
if (mappedURI != null) uri = mappedURI;
else if (uri == null) return null;
url = new URL(uri);
conn = url.openConnection();
uri = conn.getURL().toString();
// System.out.println ("++ URI: " + url);
if (ignoringMIME) retval = new InputSource(XmlReader.createReader(conn.getInputStream()));
else {
String contentType = conn.getContentType();
retval = createInputSource(contentType, conn.getInputStream(), false, url.getProtocol());
}
}
retval.setSystemId(uri);
retval.setPublicId(name);
return retval;
}
/**
* Creates an input source from a given URI.
*
* @param uri the URI (system ID) for the entity
* @param checkType if true, the MIME content type for the entity is checked for document type and
* character set encoding.
*/
public static InputSource createInputSource(URL uri, boolean checkType) throws IOException {
URLConnection conn = uri.openConnection();
InputSource retval;
if (checkType) {
String contentType = conn.getContentType();
retval = createInputSource(contentType, conn.getInputStream(), false, uri.getProtocol());
} else {
retval = new InputSource(XmlReader.createReader(conn.getInputStream()));
}
retval.setSystemId(conn.getURL().toString());
return retval;
}
/** Creates an input source from a given file, autodetecting the character encoding. */
public static InputSource createInputSource(File file) throws IOException {
InputSource retval;
String path;
retval = new InputSource(XmlReader.createReader(new FileInputStream(file)));
// On JDK 1.2 and later, simplify this:
// "path = file.toURL ().toString ()".
path = file.getAbsolutePath();
if (File.separatorChar != '/') path = path.replace(File.separatorChar, '/');
if (!path.startsWith("/")) path = "/" + path;
if (!path.endsWith("/") && file.isDirectory()) path = path + "/";
retval.setSystemId("file:" + path);
return retval;
}
/**
* Returns an input source, using the MIME type information and URL scheme to statically determine
* the correct character encoding if possible and otherwise autodetecting it. MIME carefully
* specifies the character encoding defaults, and how attributes of the content type can change
* it. XML further specifies two mandatory encodings (UTF-8 and UTF-16), and includes an XML
* declaration which can be used to internally label most documents encoded using US-ASCII
* supersets (such as Shift_JIS, EUC-JP, ISO-2022-*, ISO-8859-*, and more).
*
* <p>
*
* <p>This method can be used to access XML documents which do not have URIs (such as servlet
* input streams, or most JavaMail message entities) and to support access methods such as HTTP
* POST or PUT. (URLs normally return content using the GET method.)
*
* <p>
*
* <p><em> The caller should set the system ID in order for relative URIs found in this document
* to be interpreted correctly.</em> In some cases, a custom resolver will need to be used; for
* example, documents may be grouped in a single MIME "multipart/related" bundle, and relative
* URLs would refer to other documents in that bundle.
*
* @param contentType The MIME content type for the source for which an InputSource is desired,
* such as <em>text/xml;charset=utf-8</em>.
* @param stream The input byte stream for the input source.
* @param checkType If true, this verifies that the content type is known to support XML
* documents, such as <em>application/xml</em>.
* @param scheme Unless this is "file", unspecified MIME types default to US-ASCII. Files are
* always autodetected since most file systems discard character encoding information.
*/
public static InputSource createInputSource(
String contentType, InputStream stream, boolean checkType, String scheme) throws IOException {
InputSource retval;
String charset = null;
if (contentType != null) {
int index;
contentType = contentType.toLowerCase();
index = contentType.indexOf(';');
if (index != -1) {
String attributes;
attributes = contentType.substring(index + 1);
contentType = contentType.substring(0, index);
// use "charset=..." if it's available
index = attributes.indexOf("charset");
if (index != -1) {
attributes = attributes.substring(index + 7);
// strip out subsequent attributes
if ((index = attributes.indexOf(';')) != -1) attributes = attributes.substring(0, index);
// find start of value
if ((index = attributes.indexOf('=')) != -1) {
attributes = attributes.substring(index + 1);
// strip out rfc822 comments
if ((index = attributes.indexOf('(')) != -1)
attributes = attributes.substring(0, index);
// double quotes are optional
if ((index = attributes.indexOf('"')) != -1) {
attributes = attributes.substring(index + 1);
attributes = attributes.substring(0, attributes.indexOf('"'));
}
charset = attributes.trim();
// XXX "\;", "\)" etc were mishandled above
}
}
}
//
// Check MIME type.
//
if (checkType) {
boolean isOK = false;
for (int i = 0; i < types.length; i++)
if (types[i].equals(contentType)) {
isOK = true;
break;
}
if (!isOK) throw new IOException("Not XML: " + contentType);
}
//
// "text/*" MIME types have hard-wired character set
// defaults, as specified in the RFCs. For XML, we
// ignore the system "file.encoding" property since
// autodetection is more correct.
//
if (charset == null) {
contentType = contentType.trim();
if (contentType.startsWith("text/")) {
if (!"file".equalsIgnoreCase(scheme)) charset = "US-ASCII";
}
// "application/*" has no default
}
}
retval = new InputSource(XmlReader.createReader(stream, charset));
retval.setByteStream(stream);
retval.setEncoding(charset);
return retval;
}
