Class: Predictionary

Predictionary()

new Predictionary()

Constructs a Predictionary word prediction class. It's possible to manage multiple internal dictionaries, retrieve predictions (suggestions) for a given input and learn/refine the dictionaries from user input.
Source:

Members

DEFAULT_DICTIONARY_KEY :string

Default dictionary key, if no key is specified.
Type:
  • string
Source:

Methods

(static) instance() → {Predictionary}

Constructs a new instance of Predictionary
Source:
Returns:
Type
Predictionary

addDictionary(dictionaryKey, wordsopt)

Add a new internal dictionary for predictions.
Parameters:
Name Type Attributes Description
dictionaryKey string the key of the dictionary to add
words Array <optional>
 Optional array of words (string) that should be added to the new dictionary.
Source:

addWord(element, dictionaryKeyopt)

Add a single new word/element to a dictionary.
Parameters:
Name Type Attributes Default Description
element string | Object the element to add, can be either a plain word as a string or an object containing the properties object.word (word to add as string) and object.rank (number specifying the rank of the word, a lower rank causes the word to be ranked to front as a suggestion).
dictionaryKey string <optional>
 Predictionary#DEFAULT_DICTIONARY_KEY optional key of the dictionary to add the element.
Source:

addWords(elements, dictionaryKeyopt)

Add multiple new words/elements to a dictionary.
Parameters:
Name Type Attributes Default Description
elements string | Object the elements to add, can be an Array of either plain words (string) or of objects containing the properties object.word (word to add as string) and object.rank (number specifying the rank of the word, a lower rank causes the word to be ranked to front as a suggestion).
dictionaryKey string <optional>
 Predictionary#DEFAULT_DICTIONARY_KEY optional key of the dictionary to add the element.
Source:

applyPrediction(input, chosenPrediction, optionsopt) → {string}

Apply chosen suggestion to a given input, e.g. input = "this is an ap" and chosenPrediction = "Apple" results in "this is an Apple". Calling this function automatically refines the saved frequency of the chosen word making it more likely to be suggested in the future.
Parameters:
Name Type Attributes Description
input string the current input string (e.g. from textfield)
chosenPrediction string the chosen prediction which should be applied to the input string
options Object <optional>
 options object containing additional properties.
Properties
Name Type Attributes Description
addToDictionary string <optional>
 the key of the dictionary where new learned words should be added. If not set the dictionary to add is automatically determined.
shouldCompleteLastWord number <optional>
 if true the last word is completed, if false the chosen prediction is added as new word. If not set this decision is done automatically (last character is space -> new word).
dontLearn boolean <optional>
 if true the chosen predictions are not added or their frequencies updated
Source:
Returns:
the given input with the applied suggestion
Type
string

delete(inputOrWord, optionsopt)

Deletes a single word from one or all dictionaries.
Parameters:
Name Type Attributes Description
inputOrWord string a single word or longer string where the last word will be deleted in the dictionaries.
options Object <optional>
 Object for options
Properties
Name Type Attributes Description
dictionaryKey string <optional>
 the key of the dictionary where the word should be deleted. If not specified the word is deleted in all dictionaries.
ignoreCase string <optional>
 if false or undefined (default) only words with matching cases are deleted, otherwise also words with non-matching case.
Source:

dictionariesToJSON() → {string}

Exports all dictionaries to a JSON string. Subsequently the dictionaries can be imported using the resulting string and Predictionary#loadDictionaries.
Source:
Returns:
JSON string representing all currently loaded dictionaries.
Type
string

dictionaryToJSON(dictionaryKeyopt) → {string}

Exports a single dictionary to a JSON string. Subsequently the dictionary can be imported using the resulting string and Predictionary#loadDictionary.
Parameters:
Name Type Attributes Default Description
dictionaryKey string <optional>
 Predictionary#DEFAULT_DICTIONARY_KEY the key of the dictionary to export.
Source:
Returns:
JSON string representing the currently loaded dictionary with specified dictionaryKey.
Type
string

getDictionaryKeys(onlyEnabledopt) → {Array.<string>}

Returns a list of currently loaded dictionary keys.
Parameters:
Name Type Attributes Description
onlyEnabled boolean <optional>
 if true only keys of dictionaries that are enabled are returned. See e.g. Predictionary#useDictionaries
Source:
Returns:
Type
Array.<string>

getWords(dictionaryKeyopt) → {Array.<string>}

Retrieves saved words of a single or all loaded dictionaries.
Parameters:
Name Type Attributes Description
dictionaryKey string <optional>
 key of the dictionary from which the words should be retrieved, if not set all dictionaries are used.
Source:
Returns:
array of saved words (string) for the dictionary with the given key.
Type
Array.<string>

hasWord(word, dictionaryKeyopt, matchCaseopt) → {boolean}

Test if a given word exists a single or all loaded dictionaries.
Parameters:
Name Type Attributes Description
word string to test if existing
dictionaryKey string <optional>
 key of the dictionary to check, if not set all dictionaries are used
matchCase boolean <optional>
 if set (true) the word is searched case-sensitive, otherwise case-insensitive (default)
Source:
Returns:
true if the given word is existing
Type
boolean

isUsingOnlyDefaultDictionary() → {boolean}

Returns true if only the default dictionary is used (key Predictionary#DEFAULT_DICTIONARY_KEY).
Source:
Returns:
Type
boolean

learn(chosenWord, previousWordopt, addToDictionaryopt)

Updates the frequencies for given words, making them to be more likely suggested in the future.
Parameters:
Name Type Attributes Description
chosenWord string a suggestion/word the user has chosen
previousWord string <optional>
 the previous word of the chosen suggestion
addToDictionary string <optional>
 the key of the dictionary where new words should be added. Automatically determined, if not specified.
Source:

learnFromInput(input, dictionaryKeyopt) → {boolean}

Learns from input text while the user is typing. This method can be called with e.g. the value of a text input field for every character the user is typing.
Parameters:
Name Type Attributes Description
input string the text string to learn with. The second last and third last words are learned.
dictionaryKey string <optional>
 the key of the dictionary where new words should be added. Automatically determined, if not specified.
Source:
Returns:
true if something was learned, false if not
Type
boolean

learnFromText(text, dictionaryKeyopt)

Learns words and transitions from a given text/phrase.
Parameters:
Name Type Attributes Default Description
text string the text to learn from
dictionaryKey string <optional>
 Predictionary#DEFAULT_DICTIONARY_KEY the key of the dictionary where the words should be learned/added.
Source:

loadDictionaries(dictionariesJSON)

Loads all dictionaries from a JSON string that was previously exported by Predictionary#dictionariesToJSON. This method replaces/deletes all currently loaded dictionaries!
Parameters:
Name Type Description
dictionariesJSON string json string representing dictionaries, exported by Predictionary#dictionariesToJSON
Source:

loadDictionary(dictionaryJSON, dictionaryKeyopt)

Loads a single dictionary from a JSON string that was previously exported by Predictionary#dictionaryToJSON. If the given dictionaryKey already exists, the existing dictionary is replaced.
Parameters:
Name Type Attributes Default Description
dictionaryJSON string json string representing a dictionary, exported by Predictionary#dictionaryToJSON
dictionaryKey string <optional>
 Predictionary#DEFAULT_DICTIONARY_KEY the key for which the dictionary should be imported.
Source:

parseWords(importString, optionsopt)

Import words from a plain string (e.g. text file).
Parameters:
Name Type Attributes Description
importString string a plain text string (e.g. from a text file)
options Object <optional>
 options object containing additional properties. The default properties are suited for a plain text string in format: "word1;word2;word3;...", setting rankPosition=1 would be suited for a plain text in format: "word1 rank1;word2 rank2;word3 rank3;...".
Properties
Name Type Attributes Default Description
elementSeparator string <optional>
 ; separator to split the elements from the importString
rankSeparator string <optional>
 <space> separator to split a single element into word and rank
wordPosition string <optional>
 0 position of the word in the element (0-based)
rankPosition string <optional>
 position of the rank in the element (0-based)
addToDictionary string <optional>
 Predictionary#DEFAULT_DICTIONARY_KEY key of the dictionary where the words should be added.
Source:

predict(input, optionsopt) → {Array.<string>}

Returns word suggestions for a given input. Automatically detects if the last word should be completed (last character is not space) or if a next word should be suggested (last character is space).
Parameters:
Name Type Attributes Description
input string string for which the predictions should be calculated, e.g. the value of a text input where the user is typing.
options Object <optional>
 options object containing additional properties.
Properties
Name Type Attributes Default Description
maxPredictions number <optional>
 10 number of suggestions that should be retrieved maximally //TODO maxPredictions
applyToInput boolean <optional>
 if true the suggestions are applied to the original input before being returned
Source:
Returns:
list of words that are predictions/suggestions for the given input, ordered by relevance.
Type
Array.<string>

predictCompleteWord(input, optionsopt) → {Array.<string>}

Returns word suggestions for a given input. Last word is assumed to be incomplete and has to be completed.
Parameters:
Name Type Attributes Description
input string string for which the predictions should be calculated, e.g. the value of a text input where the user is typing.
options Object <optional>
 options object containing additional properties.
Properties
Name Type Attributes Default Description
maxPredictions number <optional>
 10 number of suggestions that should be retrieved maximally
applyToInput boolean <optional>
 if true the suggestions are applied to the original input before being returned
Source:
Returns:
list of words that are predictions/suggestions for the given input, ordered by relevance.
Type
Array.<string>

predictNextWord(input, optionsopt) → {Array.<string>}

Returns word suggestions for a given input. Last word is assumed to be complete and suggestions for the next words are calculated.
Parameters:
Name Type Attributes Description
input string string for which the predictions should be calculated, e.g. the value of a text input where the user is typing.
options Object <optional>
 options object containing additional properties.
Properties
Name Type Attributes Default Description
maxPredictions number <optional>
 10 number of suggestions that should be retrieved maximally
applyToInput boolean <optional>
 if true the suggestions are applied to the original input before being returned
Source:
Returns:
list of words that are predictions/suggestions for the given input, ordered by relevance.
Type
Array.<string>

useAllDictionaries()

Use all currently loaded dictionaries for predictions.
Source:

useDictionaries(dictionaryKeys)

Use a set of dictionaries for predictions, specified by an array of dictionaryKeys.
Parameters:
Name Type Description
dictionaryKeys Array an array of strings, specifying the dictionaryKeys to use
Source:

useDictionary(dictionaryKey)

Use only a singe loaded dictionary for predictions.
Parameters:
Name Type Description
dictionaryKey string the key of the dictionary to use
Source: