AbstractTextAnnotation (illinois-cogcomp-nlp 3.1.29 API)

java.lang.Object
- edu.illinois.cs.cogcomp.core.datastructures.textannotation.AbstractTextAnnotation

All Implemented Interfaces:

Serializable

Direct Known Subclasses:

Sentence, TextAnnotation
```
public abstract class AbstractTextAnnotation
extends Object
implements Serializable
```
This is the abstract class that represents annotation for text -- this could represent a paragraph or just a sentence.
This class stores the raw text, the tokens and the list of views available for the text and provides methods for adding new views.

Author:

Vivek Srikumar

See Also:

TextAnnotation, Sentence, Serialized Form

Field Summary

Fields
Modifier and Type	Field and Description
`String`	`text` The raw text
`protected IntPair[]`	`tokenCharacterOffsets` The character offsets of the tokens.
`protected String`	`tokenizedText` The tokenized text
`protected String[]`	`tokens` The list of tokens in the text.
`protected Map<String,List<View>>`	`views` The list of views

Constructor Summary

Constructors
Modifier Constructor and Description

protected AbstractTextAnnotation()

Constructors
Modifier	Constructor and Description
`protected`	`AbstractTextAnnotation()`

Method Summary

All Methods Instance Methods Concrete Methods Deprecated Methods
Modifier and Type	Method and Description
`void`	`addTopKView(String viewName, List<View> view)` Adds a new view identified by `viewName` and sets the top K values for this view.
`void`	`addView(String viewName, View view)` Adds the top scoring value for the view identified by `viewName`.
`void`	`addViews(String[] viewNames, View[] views)`
`Set<String>`	`getAvailableViews()` Gets the set of views that are available
`String`	`getText()` Gets the raw text
`String`	`getToken(int position)` Get the token at position from this text
`IntPair`	`getTokenCharacterOffset(int position)` Get the character offset of the token at position in the raw text.
`String`	`getTokenizedText()` Gets the tokenized text.
`String[]`	`getTokens()` Gets the tokens in this text.
`String[]`	`getTokensInSpan(int start, int end)` Gets the tokens in the span.
`List<View>`	`getTopKViews(String viewName)` Get all the values available for a given view, identified by `viewName`
`View`	`getView(String viewName)` Gets the highest scoring value for the view identified by `viewName`.
`boolean`	`hasView(String viewName)` Checks if this text annotation has a view identified by `viewName`
`void`	`removeAllViews()`
`void`	`removeView(String viewName)` Remove a given view
`IQueryable<Constituent>`	`select(String viewName)` Gets a queryable list of all constituents from the view, represented by `viewName`.
`void`	`setTokens(String[] tokens, IntPair[] tokenCharacterOffsets)` Deprecated. There is now a TOKENS view that should be used to access the tokens
`int`	`size()` Get the number of tokens in this text.

Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

- Field Detail
  - text
```
public String text
```
    The raw text
  - tokens
```
protected String[] tokens
```
    The list of tokens in the text. This might be a contentious issue because different annotators might use different tokenization schemes. One way to fix that would be to use some sort of common subsequence finding algorithm to align the tokenization to the canonical one.
  - tokenCharacterOffsets
```
protected IntPair[] tokenCharacterOffsets
```
    The character offsets of the tokens.
  - views
```
protected Map<String,List<View>> views
```
    The list of views
  - tokenizedText
```
protected String tokenizedText
```
    The tokenized text
- Constructor Detail
  - AbstractTextAnnotation
```
protected AbstractTextAnnotation()
```
- Method Detail
  - addTopKView
```
public void addTopKView(String viewName,
                        List<View> view)
```
    Adds a new view identified by viewName and sets the top K values for this view.
    
    Parameters:
    
    viewName - The name of the new view to be added.
    
    view - The top K views
  - addViews
```
public void addViews(String[] viewNames,
                     View[] views)
```
  - addView
```
public void addView(String viewName,
                    View view)
```
    Adds the top scoring value for the view identified by viewName.
  - removeView
```
public void removeView(String viewName)
```
    Remove a given view
  - removeAllViews
```
public void removeAllViews()
```
  - getAvailableViews
```
public Set<String> getAvailableViews()
```
    Gets the set of views that are available
    
    Returns:
    
    A Set<String> of views that are available for this text annotation.
  - getView
```
public View getView(String viewName)
```
    Gets the highest scoring value for the view identified by viewName.
  - getTopKViews
```
public List<View> getTopKViews(String viewName)
```
    Get all the values available for a given view, identified by viewName
  - getText
```
public String getText()
```
    Gets the raw text
    
    Returns:
    
    The raw text
  - getTokenizedText
```
public String getTokenizedText()
```
    Gets the tokenized text.
    
    Returns:
    
    The tokenized text, as a string where each token is separated by a space.
  - getTokens
```
public String[] getTokens()
```
    Gets the tokens in this text.
  - getToken
```
public String getToken(int position)
```
    Get the token at position from this text
  - getTokenCharacterOffset
```
public IntPair getTokenCharacterOffset(int position)
```
    Get the character offset of the token at position in the raw text.
    
    Returns:
    
    An IntPair representing the pair (start, end+1)
  - getTokensInSpan
```
public String[] getTokensInSpan(int start,
                                int end)
```
    Gets the tokens in the span.
    
    Parameters:
    
    start - The start of the span
    
    end - The end of the span
  - hasView
```
public boolean hasView(String viewName)
```
    Checks if this text annotation has a view identified by viewName
    
    Returns:
    
    true if the text annotation contains a view called viewName and false otherwise
  - setTokens
```
public void setTokens(String[] tokens,
                      IntPair[] tokenCharacterOffsets)
```
    Deprecated. There is now a TOKENS view that should be used to access the tokens
    
    Set the tokens for this text.
    NOTE: This function should not be called by any method that is not a Tokenizer and will throw an exception if tokens are set more than once. So do not call this function unless you know what you are doing.
    
    Parameters:
    
    tokens - An array of tokens
    
    tokenCharacterOffsets - An array, whose length is the same as tokens. Each element indicates the character offset of this token in the raw text.
    
    Throws:
    
    IllegalArgumentException - if the tokens are set more than once.
  - size
```
public int size()
```
    Get the number of tokens in this text.
    
    Returns:
    
    the number of tokens
  - select
```
public IQueryable<Constituent> select(String viewName)
```
    Gets a queryable list of all constituents from the view, represented by viewName. This function can be used for SQL-like queries on the text.
    Examples: In all the examples that follow, we assume that text is a variable of type AbstractTextAnnotation.
    1. Given a constituent srlConstituent from the SRL_VERB view, find all constituents in the parse view that cover exactly this constituent.
      
      List<Constituent> parseConstituent = text.select(ViewNames.PARSE).where( Queries.sameSpanAsConstituent(srlConstituent));
    See Also:
    
    QueryableList, ViewNames

Class AbstractTextAnnotation

Field Summary

Constructor Summary

Method Summary

Methods inherited from class java.lang.Object

Field Detail

text

tokens

tokenCharacterOffsets

views

tokenizedText

Constructor Detail

AbstractTextAnnotation

Method Detail

addTopKView

addViews

addView

removeView

removeAllViews

getAvailableViews

getView

getTopKViews

getText

getTokenizedText

getTokens

getToken

getTokenCharacterOffset

getTokensInSpan

hasView

setTokens

size

select