JDOM 2.0.6.1 · org.jdom2.input.stax.DTDParser

Parses out key information from a single String representing a DOCTYPE declaration. StAX parsers supply a single string representing the DOCTYPE and this needs to be processed to get items like the SystemID, etc. Additionally it needs to be reformatted to create a standardised representation.

The assumption is that the DTD is valid.

We need to pull out 4 elements of data:

The root element name
The SystemID (if available)
The PublicID (if available)
The internal subset (if available)

The internal-subset should be re-formatted to conform to the JDOM 'standard' where each declaration starts on a new line indented with 2 spaces. This 'standard' is defined by the way that JDOM formats the DTD declarations in the SAX parse process, which fires individual events for the content in the DTD.

We can do this all with a well-structured regular expression, which is actually simpler than trying to fish out all the components ourselves....

Field Summary

Modifier and Type	Field and Description
private static final String	metapattern This is the meta-pattern.
private static final Pattern	pattern The following Pattern is the final result after parsing/tokenizing/substituting the meta-pattern.

Constructor Summary

Access	Constructor and Description
private	DTDParser() Make instances 'impossible'.

Method Summary

Modifier and Type	Method and Description
private static final Pattern	Returns: The substituted pattern buildPattern(HashMap<String, String> The map containing substitution tokens/patterns map, String The meta-pattern to do the substitutions on. input) This method substitutes the simple tokens in the meta-pattern with the declared values in the map.
private static String	Returns: the formatted input. formatInternal(String the input DocType declaration as found in a StAX Reader. internal) Reformat an internal subset....
private static final String	Returns: The first found value. getGroup(final Matcher The match that has succeeded mat, final int... The groups to check for a value. groups) Looks in any number of matched groups for a value.
private static final boolean	Returns: true if it is a space, tab, newline, or carriage-return. isWhite(char The char to check ch) return true if the input character is one of the types recognized in the DTD spec.
public static DocType	Returns: The input string as a DocType. parse(final String the input DOCTYPE string to parse. Must be valid. input, final JDOMFactory The JDOM factory to use to build the JDOM DocType. factory) Parse out a DOCTYPE declaration as supplied by the standard StAX readers.
private static final HashMap<String, String>	Returns: The populated map. populatePatterns() This builds a substitution map containing the raw patterns for certain types of content we expect.

Inherited from java.lang.Object:: clone equals finalize getClass hashCode notify notifyAll toString wait wait wait

Field Detail

metapattern	back to summary
private static final String metapattern This is the meta-pattern. Where you see ' os ' there is optional space. Where you see ' name ' there is the element name. Where you see ' ms ' there is mandatory space. Where you see ' id ' there is some quoted identifier. Where you see ' internal ' there is the internal subset. Anything else will become part of the final regex. Space (' ') was chosen for the token delimiter because it makes the meta-pattern easy to read. There are a couple of places in this expression where there are two ' ' together, and it is critical that it does not change because there will be missed token matches then.

metapattern

back to summary

private static final String metapattern

This is the meta-pattern.

Where you see ' os ' there is optional space.
Where you see ' name ' there is the element name.
Where you see ' ms ' there is mandatory space.
Where you see ' id ' there is some quoted identifier.
Where you see ' internal ' there is the internal subset.

Anything else will become part of the final regex.

Space (' ') was chosen for the token delimiter because it makes the meta-pattern easy to read. There are a couple of places in this expression where there are two ' ' together, and it is critical that it does not change because there will be missed token matches then.

pattern	back to summary
private static final Pattern pattern The following Pattern is the final result after parsing/tokenizing/substituting the meta-pattern.

Constructor Detail

DTDParser	back to summary
private DTDParser() Make instances 'impossible'. Everything is static.

Method Detail

buildPattern	back to summary
private static final Pattern buildPattern(HashMap<String, String> map, String input) This method substitutes the simple tokens in the meta-pattern with the declared values in the map. Parameters map:HashMap<String, String> The map containing substitution tokens/patterns input:String The meta-pattern to do the substitutions on. Returns:Pattern The substituted pattern

buildPattern

back to summary

private static final Pattern buildPattern(HashMap<String, String> map, String input)

This method substitutes the simple tokens in the meta-pattern with the declared values in the map.

Parameters

map:HashMap<String, String>: The map containing substitution tokens/patterns
input:String: The meta-pattern to do the substitutions on.

Returns:Pattern

The substituted pattern

formatInternal	back to summary
private static String formatInternal(String internal) Reformat an internal subset.... Each declaration starts on an indented newline. Parameters internal:String the input DocType declaration as found in a StAX Reader. Returns:String the formatted input.

formatInternal

back to summary

private static String formatInternal(String internal)

Reformat an internal subset.... Each declaration starts on an indented newline.

Parameters

internal:String: the input DocType declaration as found in a StAX Reader.

Returns:String

the formatted input.

getGroup	back to summary
private static final String getGroup(final Matcher mat, final int... groups) Looks in any number of matched groups for a value. Returns the first set value. The assumption is that, depending on the pattern matches, the value could be in a few different locations. Parameters mat:Matcher The match that has succeeded groups:int[] The groups to check for a value. Returns:String The first found value.

getGroup

back to summary

private static final String getGroup(final Matcher mat, final int... groups)

Looks in any number of matched groups for a value. Returns the first set value. The assumption is that, depending on the pattern matches, the value could be in a few different locations.

Parameters

mat:Matcher: The match that has succeeded
groups:int[]: The groups to check for a value.

Returns:String

The first found value.

isWhite	back to summary
private static final boolean isWhite(char ch) return true if the input character is one of the types recognized in the DTD spec. Parameters ch:char The char to check Returns:boolean true if it is a space, tab, newline, or carriage-return.

isWhite

back to summary

private static final boolean isWhite(char ch)

return true if the input character is one of the types recognized in the DTD spec.

Parameters

ch:char: The char to check

Returns:boolean

true if it is a space, tab, newline, or carriage-return.

parse	back to summary
public static DocType parse(final String input, final JDOMFactory factory) throws JDOMException Parse out a DOCTYPE declaration as supplied by the standard StAX readers. Using 'XML' terminology, this method assumes that the input is both 'well-formed' and 'valid'. The assumptions that this class makes ensure that the 'right thing' is done for valid content, but invalid content may or may not fail with a JDOMException. The behaviour of this method with invalid input is 'undefined'. Parameters input:String the input DOCTYPE string to parse. Must be valid. factory:JDOMFactory The JDOM factory to use to build the JDOM DocType. Returns:DocType The input string as a DocType. Exceptions JDOMException: if the DocType is not generated.

parse

back to summary

public static DocType parse(final String input, final JDOMFactory factory) throws JDOMException

Parse out a DOCTYPE declaration as supplied by the standard StAX readers.

Using 'XML' terminology, this method assumes that the input is both 'well-formed' and 'valid'. The assumptions that this class makes ensure that the 'right thing' is done for valid content, but invalid content may or may not fail with a JDOMException. The behaviour of this method with invalid input is 'undefined'.

Parameters

input:String: the input DOCTYPE string to parse. Must be valid.
factory:JDOMFactory: The JDOM factory to use to build the JDOM DocType.

Returns:DocType

The input string as a DocType.

Exceptions

JDOMException:: if the DocType is not generated.

populatePatterns	back to summary
private static final HashMap<String, String> populatePatterns() This builds a substitution map containing the raw patterns for certain types of content we expect. Returns:HashMap<String, String> The populated map.

populatePatterns

back to summary

private static final HashMap<String, String> populatePatterns()

This builds a substitution map containing the raw patterns for certain types of content we expect.

Returns:HashMap<String, String>: The populated map.

public Class DTDParser

Field Summary

Constructor Summary

Method Summary

Field Detail

Constructor Detail

Method Detail