Class WordDelimiterGraphFilter

All Implemented Interfaces:
Closeable, AutoCloseable, Unwrappable<TokenStream>

public final class WordDelimiterGraphFilter extends TokenFilter
Splits words into subwords and performs optional transformations on subword groups, producing a correct token graph so that e.g. PhraseQuery can work correctly when this filter is used in the search-time analyzer. Unlike the deprecated WordDelimiterFilter, this token filter produces a correct token graph as output. However, it cannot consume an input token graph correctly. Processing is suppressed by KeywordAttribute.isKeyword()=true.

Words are split into subwords with the following rules:

  • split on intra-word delimiters (by default, all non alpha-numeric characters): "Wi-Fi""Wi", "Fi"
  • split on case transitions: "PowerShot""Power", "Shot"
  • split on letter-number transitions: "SD500""SD", "500"
  • leading and trailing intra-word delimiters on each subword are ignored: "//hello---there, 'dude'""hello", "there", "dude"
  • trailing "'s" are removed for each subword: "O'Neil's" "O", "Neil"
    • Note: this step isn't performed in a separate filter because of possible subword combinations.
The GENERATE... options affect how incoming tokens are broken into parts, and the various CATENATE_... parameters affect how those parts are combined.
  • If no CATENATE option is set, then no subword combinations are generated: "PowerShot" 0:"Power", 1:"Shot" (0 and 1 are the token positions)
  • CATENATE_WORDS means that in addition to the subwords, maximum runs of non-numeric subwords are catenated and produced at the same position of the last subword in the run:
    • "PowerShot"0:"Power", 1:"Shot" 1:"PowerShot"
    • "A's+B's&C's" > 0:"A", 1:"B", 2:"C", 2:"ABC"
    • "Super-Duper-XL500-42-AutoCoder!" 0:"Super", 1:"Duper", 2:"XL", 2:"SuperDuperXL", 3:"500" 4:"42", 5:"Auto", 6:"Coder", 6:"AutoCoder"
  • CATENATE_NUMBERS works like CATENATE_WORDS, but for adjacent digit sequences.
  • CATENATE_ALL smushes together all the token parts without distinguishing numbers and words.
One use for WordDelimiterGraphFilter is to help match words with different subword delimiters. For example, if the source text contained "wi-fi" one may want "wifi" "WiFi" "wi-fi" "wi+fi" queries to all match. One way of doing so is to specify CATENATE options in the analyzer used for indexing, and not in the analyzer used for querying. Given that the current StandardTokenizer immediately removes many intra-word delimiters, it is recommended that this filter be used after a tokenizer that does not do this (such as WhitespaceTokenizer).
  • Field Details

    • GENERATE_WORD_PARTS

      public static final int GENERATE_WORD_PARTS
      Causes parts of words to be generated:

      "PowerShot" => "Power" "Shot"

      See Also:
    • GENERATE_NUMBER_PARTS

      public static final int GENERATE_NUMBER_PARTS
      Causes number subwords to be generated:

      "500-42" => "500" "42"

      See Also:
    • CATENATE_WORDS

      public static final int CATENATE_WORDS
      Causes maximum runs of word parts to be catenated:

      "wi-fi" => "wifi"

      See Also:
    • CATENATE_NUMBERS

      public static final int CATENATE_NUMBERS
      Causes maximum runs of number parts to be catenated:

      "500-42" => "50042"

      See Also:
    • CATENATE_ALL

      public static final int CATENATE_ALL
      Causes all subword parts to be catenated:

      "wi-fi-4000" => "wifi4000"

      See Also:
    • PRESERVE_ORIGINAL

      public static final int PRESERVE_ORIGINAL
      Causes original words are preserved and added to the subword list (Defaults to false)

      "500-42" => "500" "42" "500-42"

      See Also:
    • SPLIT_ON_CASE_CHANGE

      public static final int SPLIT_ON_CASE_CHANGE
      Causes lowercase -> uppercase transition to start a new subword.
      See Also:
    • SPLIT_ON_NUMERICS

      public static final int SPLIT_ON_NUMERICS
      If not set, causes numeric changes to be ignored (subwords will only be generated given SUBWORD_DELIM tokens).
      See Also:
    • STEM_ENGLISH_POSSESSIVE

      public static final int STEM_ENGLISH_POSSESSIVE
      Causes trailing "'s" to be removed for each subword

      "O'Neil's" => "O", "Neil"

      See Also:
    • IGNORE_KEYWORDS

      public static final int IGNORE_KEYWORDS
      Suppresses processing terms with KeywordAttribute.isKeyword()=true.
      See Also:
    • protWords

      final CharArraySet protWords
      If not null is the set of tokens to protect from being delimited
    • flags

      private final int flags
    • bufferedParts

      private int[] bufferedParts
    • bufferedLen

      private int bufferedLen
    • bufferedPos

      private int bufferedPos
    • bufferedTermParts

      private char[][] bufferedTermParts
    • termAttribute

      private final CharTermAttribute termAttribute
    • keywordAttribute

      private final KeywordAttribute keywordAttribute
    • offsetAttribute

      private final OffsetAttribute offsetAttribute
    • posIncAttribute

      private final PositionIncrementAttribute posIncAttribute
    • posLenAttribute

      private final PositionLengthAttribute posLenAttribute
    • iterator

      private final WordDelimiterIterator iterator
    • concat

    • adjustInternalOffsets

      private final boolean adjustInternalOffsets
    • lastConcatCount

      private int lastConcatCount
    • concatAll

    • accumPosInc

      private int accumPosInc
    • savedTermBuffer

      private char[] savedTermBuffer
    • savedTermLength

      private int savedTermLength
    • savedStartOffset

      private int savedStartOffset
    • savedEndOffset

      private int savedEndOffset
    • savedState

      private AttributeSource.State savedState
    • lastStartOffset

      private int lastStartOffset
    • adjustingOffsets

      private boolean adjustingOffsets
    • wordPos

      private int wordPos
    • sorter

  • Constructor Details

    • WordDelimiterGraphFilter

      public WordDelimiterGraphFilter(TokenStream in, boolean adjustInternalOffsets, byte[] charTypeTable, int configurationFlags, CharArraySet protWords)
      Creates a new WordDelimiterGraphFilter
      Parameters:
      in - TokenStream to be filtered
      adjustInternalOffsets - if the offsets of partial terms should be adjusted
      charTypeTable - table containing character types
      configurationFlags - Flags configuring the filter
      protWords - If not null is the set of tokens to protect from being delimited
    • WordDelimiterGraphFilter

      public WordDelimiterGraphFilter(TokenStream in, int configurationFlags, CharArraySet protWords)
      Creates a new WordDelimiterGraphFilter using WordDelimiterIterator.DEFAULT_WORD_DELIM_TABLE as its charTypeTable
      Parameters:
      in - TokenStream to be filtered
      configurationFlags - Flags configuring the filter
      protWords - If not null is the set of tokens to protect from being delimited
  • Method Details

    • bufferWordParts

      private void bufferWordParts() throws IOException
      Iterates all words parts and concatenations, buffering up the term parts we should return.
      Throws:
      IOException
    • incrementToken

      public boolean incrementToken() throws IOException
      Description copied from class: TokenStream
      Consumers (i.e., IndexWriter) use this method to advance the stream to the next token. Implementing classes must implement this method and update the appropriate AttributeImpls with the attributes of the next token.

      The producer must make no assumptions about the attributes after the method has been returned: the caller may arbitrarily change it. If the producer needs to preserve the state for subsequent calls, it can use AttributeSource.captureState() to create a copy of the current attribute state.

      This method is called for every token of a document, so an efficient implementation is crucial for good performance. To avoid calls to AttributeSource.addAttribute(Class) and AttributeSource.getAttribute(Class), references to all AttributeImpls that this stream uses should be retrieved during instantiation.

      To ensure that filters and consumers know which attributes are available, the attributes must be added during instantiation. Filters and consumers are not required to check for availability of attributes in TokenStream.incrementToken().

      Specified by:
      incrementToken in class TokenStream
      Returns:
      false for end of stream; true otherwise
      Throws:
      IOException
    • reset

      public void reset() throws IOException
      Description copied from class: TokenFilter
      This method is called by a consumer before it begins consumption using TokenStream.incrementToken().

      Resets this stream to a clean state. Stateful implementations must implement this method so that they can be reused, just as if they had been created fresh.

      If you override this method, always call super.reset(), otherwise some internal state will not be correctly reset (e.g., Tokenizer will throw IllegalStateException on further usage).

      NOTE: The default implementation chains the call to the input TokenStream, so be sure to call super.reset() when overriding this method.

      Overrides:
      reset in class TokenFilter
      Throws:
      IOException
    • buffer

      void buffer(int startPos, int endPos, int startPart, int endPart)
      startPos, endPos -> graph start/end position startPart, endPart -> slice of the original term for this part
    • buffer

      void buffer(char[] termPart, int startPos, int endPos, int startPart, int endPart)
      a null termPart means it's a simple slice of the original term
    • saveState

      private void saveState()
      Saves the existing attribute states
    • flushConcatenation

      private void flushConcatenation(WordDelimiterGraphFilter.WordDelimiterConcatenation concat)
      Flushes the given WordDelimiterConcatenation by either writing its concat and then clearing, or just clearing.
      Parameters:
      concat - WordDelimiterConcatenation that will be flushed
    • shouldConcatenate

      private boolean shouldConcatenate(int wordType)
      Determines whether to concatenate a word or number if the current word is the given type
      Parameters:
      wordType - Type of the current word used to determine if it should be concatenated
      Returns:
      true if concatenation should occur, false otherwise
    • shouldGenerateParts

      private boolean shouldGenerateParts(int wordType)
      Determines whether a word/number part should be generated for a word of the given type
      Parameters:
      wordType - Type of the word used to determine if a word/number part should be generated
      Returns:
      true if a word/number part should be generated, false otherwise
    • concatenate

      private void concatenate(WordDelimiterGraphFilter.WordDelimiterConcatenation concatenation)
      Concatenates the saved buffer to the given WordDelimiterConcatenation
      Parameters:
      concatenation - WordDelimiterConcatenation to concatenate the buffer to
    • has

      private boolean has(int flag)
      Determines whether the given flag is set
      Parameters:
      flag - Flag to see if set
      Returns:
      true if flag is set
    • flagsToString

      public static String flagsToString(int flags)
      Returns string representation of configuration flags
    • toString

      public String toString()
      Description copied from class: AttributeSource
      Returns a string consisting of the class's simple name, the hex representation of the identity hash code, and the current reflection of all attributes.
      Overrides:
      toString in class AttributeSource
      See Also: