CharacterSplitter Stage

Splits tokens on specified characters, typically punctuation. Multiple split characters in a row will create a single split (not multiple splits).

Operates On:  Lexical Items with TOKEN

Configuration

• splitChars (string, optional) - List of characters which should be used to split tokens.
  • If not present, then tokens are split on any sequence of punctuation. 
• dontSplitChars (string, optional) - List of characters which will NOT be used to split tokens.
  
  • This is typically used to identify exceptions (characters which are not used to split tokens) when splitChars is missing.
  • These characters are included in the produced tokens.
• splitFlag (string, optional) - The flag to be put on the vertex between the two tokens.
  
  • If missing, defaults to ALL_PUNCTUATION.

Examples

{
 "type":"CharacterSplitter",
 "dontSplitChars":"."
}

Splits on all punctuation, except periods.

For example, the token:  "SagaToolkit-1.0" will produce the following graph:

V-------[SagaToolkit-1.0]-------V

 ^---[SagaToolkit]--V--[1.0]----^

{
  "type":"CharacterSplitter",
  "splitChars":"-",
  "splitFlag":"DASH_SPLIT"
}

(splits tokens dashes)

Output Flags

Lex-Item Flags:

• TOKEN - All tokens produced are tagged as TOKEN

Vertex Flags:

• ALL_PUNCTUATION - Identifies the vertex as all token
  • The default flag if no "splitFlag" is present.
• <splitFlag> - Defines an alternative flag to ALL_PUNCTUATION, if desired (see above)