org.apache.http.message
Class BasicTokenIterator

java.lang.Object
  org.apache.http.message.BasicTokenIterator

All Implemented Interfaces:

Iterator<Object>, TokenIterator

public class BasicTokenIterator
extends Object
implements TokenIterator

Basic implementation of a TokenIterator. This implementation parses #token sequences as defined by RFC 2616, section 2. It extends that definition somewhat beyond US-ASCII.

Since:

4.0

Field Summary

protected String	currentHeader The value of the current header.
protected String	currentToken The token to be returned by the next call to nextToken().
protected HeaderIterator	headerIt The iterator from which to obtain the next header.
static String	HTTP_SEPARATORS The HTTP separator characters.
protected int	searchPos The position after currentToken in currentHeader.

Constructor Summary

BasicTokenIterator(HeaderIterator headerIterator)
Creates a new instance of BasicTokenIterator.

Method Summary

protected String	createToken(String value, int start, int end) Creates a new token to be returned.
protected int	findNext(int pos) Determines the next token.
protected int	findTokenEnd(int from) Determines the ending position of the current token.
protected int	findTokenSeparator(int pos) Determines the position of the next token separator.
protected int	findTokenStart(int pos) Determines the starting position of the next token.
boolean	hasNext() Indicates whether there is another token in this iteration.
protected boolean	isHttpSeparator(char ch) Checks whether a character is an HTTP separator.
protected boolean	isTokenChar(char ch) Checks whether a character is a valid token character.
protected boolean	isTokenSeparator(char ch) Checks whether a character is a token separator.
protected boolean	isWhitespace(char ch) Checks whether a character is a whitespace character.
Object	next() Returns the next token.
String	nextToken() Obtains the next token from this iteration.
void	remove() Removing tokens is not supported.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Detail

HTTP_SEPARATORS

public static final String HTTP_SEPARATORS

The HTTP separator characters. Defined in RFC 2616, section 2.2.

See Also:

Constant Field Values

headerIt

protected final HeaderIterator headerIt

The iterator from which to obtain the next header.

currentHeader

protected String currentHeader

The value of the current header. This is the header value that includes currentToken. Undefined if the iteration is over.

currentToken

protected String currentToken

The token to be returned by the next call to nextToken(). null if the iteration is over.

searchPos

protected int searchPos

The position after currentToken in currentHeader. Undefined if the iteration is over.

Constructor Detail

BasicTokenIterator

public BasicTokenIterator(HeaderIterator headerIterator)

Creates a new instance of BasicTokenIterator.

Parameters:

headerIterator - the iterator for the headers to tokenize

Method Detail

hasNext

public boolean hasNext()

Description copied from interface: TokenIterator

Indicates whether there is another token in this iteration.

Specified by:

hasNext in interface Iterator<Object>

Specified by:

hasNext in interface TokenIterator

Returns:

true if there is another token, false otherwise

nextToken

public String nextToken()
                 throws NoSuchElementException,
                        ParseException

Obtains the next token from this iteration.

Specified by:

nextToken in interface TokenIterator

Returns:

the next token in this iteration

Throws:

NoSuchElementException - if the iteration is already over

ParseException - if an invalid header value is encountered

next

public final Object next()
                  throws NoSuchElementException,
                         ParseException

Returns the next token. Same as nextToken(), but with generic return type.

Specified by:

next in interface Iterator<Object>

Returns:

the next token in this iteration

Throws:

NoSuchElementException - if there are no more tokens

ParseException - if an invalid header value is encountered

remove

public final void remove()
                  throws UnsupportedOperationException

Removing tokens is not supported.

Specified by:

remove in interface Iterator<Object>

Throws:

UnsupportedOperationException - always

findNext

protected int findNext(int pos)
                throws ParseException

Determines the next token. If found, the token is stored in currentToken. The return value indicates the position after the token in currentHeader. If necessary, the next header will be obtained from headerIt. If not found, currentToken is set to null.

Parameters:

pos - the position in the current header at which to start the search, -1 to search in the first header

Returns:

the position after the found token in the current header, or negative if there was no next token

Throws:

ParseException - if an invalid header value is encountered

createToken

protected String createToken(String value,
                             int start,
                             int end)

Creates a new token to be returned. Called from findNext after the token is identified. The default implementation simply calls String.substring.

If header values are significantly longer than tokens, and some tokens are permanently referenced by the application, there can be problems with garbage collection. A substring will hold a reference to the full characters of the original string and therefore occupies more memory than might be expected. To avoid this, override this method and create a new string instead of a substring.

Parameters:

value - the full header value from which to create a token

start - the index of the first token character

end - the index after the last token character

Returns:

a string representing the token identified by the arguments

findTokenStart

protected int findTokenStart(int pos)

Determines the starting position of the next token. This method will iterate over headers if necessary.

Parameters:

pos - the position in the current header at which to start the search

Returns:

the position of the token start in the current header, negative if no token start could be found

findTokenSeparator

protected int findTokenSeparator(int pos)

Determines the position of the next token separator. Because of multi-header joining rules, the end of a header value is a token separator. This method does therefore not need to iterate over headers.

Parameters:

pos - the position in the current header at which to start the search

Returns:

the position of a token separator in the current header, or at the end

Throws:

ParseException - if a new token is found before a token separator. RFC 2616, section 2.1 explicitly requires a comma between tokens for #.

findTokenEnd

protected int findTokenEnd(int from)

Determines the ending position of the current token. This method will not leave the current header value, since the end of the header value is a token boundary.

Parameters:

from - the position of the first character of the token

Returns:

the position after the last character of the token. The behavior is undefined if from does not point to a token character in the current header value.

isTokenSeparator

protected boolean isTokenSeparator(char ch)

Checks whether a character is a token separator. RFC 2616, section 2.1 defines comma as the separator for #token sequences. The end of a header value will also separate tokens, but that is not a character check.

Parameters:

ch - the character to check

Returns:

true if the character is a token separator, false otherwise

isWhitespace

protected boolean isWhitespace(char ch)

Checks whether a character is a whitespace character. RFC 2616, section 2.2 defines space and horizontal tab as whitespace. The optional preceeding line break is irrelevant, since header continuation is handled transparently when parsing messages.

Parameters:

ch - the character to check

Returns:

true if the character is whitespace, false otherwise

isTokenChar

protected boolean isTokenChar(char ch)

Checks whether a character is a valid token character. Whitespace, control characters, and HTTP separators are not valid token characters. The HTTP specification (RFC 2616, section 2.2) defines tokens only for the US-ASCII character set, this method extends the definition to other character sets.

Parameters:

ch - the character to check

Returns:

true if the character is a valid token start, false otherwise

isHttpSeparator

protected boolean isHttpSeparator(char ch)

Checks whether a character is an HTTP separator. The implementation in this class checks only for the HTTP separators defined in RFC 2616, section 2.2. If you need to detect other separators beyond the US-ASCII character set, override this method.

Parameters:

ch - the character to check

Returns:

true if the character is an HTTP separator