Class: SemanticInterpreter

Requires

• module:require

Methods

addGrammar(id, grammarImpl, fileFormatNo)

semantic/semanticInterpreter.js, line 825

Add/register grammar for use with interpret NOTE: if no other grammar is available yet, currentGrammarId will be set to id. NOTE: if currently disabled, calling this function automatically enables ( setEnabled(TRUE) ), the semantic interpreter.

Name	Type	Description
id	String	ID for the grammar (e.g. an ISO-639 language code)
grammarImpl	mmir.grammar.GrammarConverter | function	the executable JavaScript grammar implementation IF {mmir.grammar.GrammarConverter}: the impl. with valid member {Function} mmir.grammar.GrammarConverter.executeGrammar() IF {Function}: the {Function} mmir.grammar.GrammarConverter#executeGrammar() - In this case, if no GrammarConverter instance fo id is present, a new one will be created; The stopwords must already be set, be part of the options-argument (see doc for fileFormatNo), or must additionally be set for the GrammarConverter instance (e.g. using mmir.SemanticInterpreter.setStopwords)
fileFormatNo	Number | PlainObject	optional OPTIONAL If Number and the number given does not match GRAMMAR_FILE_FORMAT_VERSION the file format is assumed to be out-dated and an Error will be thrown. If PlainObject, i.e. an options object, the following properties are evaluated (all properties are optional): fileFormat: NUMBER, default: undefined (desc. see above) execMode: 'sync' | 'async', default: 'sync' if 'async' then the grammar is executed asynchronously, i.e. interpret() must be invoked with a callback function in order to retrieve the result stopwords: Array, default: null if given, the grammar (GrammarConverter) will be set with this stopword list, i.e. grammar.setStopwords(stopwords)

Throws:

Error if fileFormatNo is given, but does not match GRAMMAR_FILE_FORMAT_VERSION.

addProcessing(langCode, proc, isPrepend, callback)

semantic/semanticInterpreter.js, line 883

Shortcut for mmir.GrammarConverter#addProc: add pre-/post-processing step for running before/after interpret

Name	Type	Description
langCode	String	the language code, for which to add the (pre- and/or post-) processing step
proc	ProcessingStep	the processing step: { //the name of the processing step name: string, //OPTIONAL pre-processing function: pre(input: string | Positions, isCalcPos: boolean) pre: Function, //OPTIONAL post-processing function: post(result: any, pos: Positions) post: Function }
isPrepend	Boolean | Number	optional OPTIONAL if omitted (or FALSY): appended proc to processing steps if number: insert proc at this index into the processing steps-list if TRUE: prepend proc to processing steps
callback	function	optional OPTIONAL callback, in case of asnychronous initalization, i.e. if grammar is not loaded/compiled yet, and grammar.json is available. If omitted, an error is thrown, if the grammar has not been loaded/compiled yet.

See:

• mmir.GrammarConverter#addProc

Example

//poitionUtils:
var posUtil = mmir.require('mmirf/positionUtils');
//stemming function
var stemFunc = ...;
//add stemming function for pre-processing for "de" as first step
mmir.semantic.addProcessing('de', {
 name: 'stem',
 pre: posUtil.createWordPosPreProc(stem, this)
}, true);

applyPreProcessing(thePhrase, lang, processingSteps)

semantic/semanticInterpreter.js, line 767

Applies pre-processing for the corresponding parser/grammar of lang (e.g. removes stopwords using the stopword-list etc). NOTE: interpret automatically applies pre-processing i.e. there is no need to manually do this when using interpret). IMPORTANT: this helper function actually invokes mmir.grammar.GrammarConverter#preproc which by default removes stopwords; if the corresponding GrammarConverter instance has been set with a non-default pre-processing chain, results may be differ (i.e. may not remove stopwords).

Name	Type	Description
thePhrase	String	the Phrase for which stopwords should be removed
lang	String	optional the language code (identifier) for the parser/grammar (if omitted the currently set grammar is used)
processingSteps	Array.<ProcessingStep>	optional OPTIONAL if given, use processingSteps instead of the GrammarConverter's configured pre-processing chain. NOTE positional argument (i.e. must specify pos too)

See:

• mmir.grammar.GrammarConverter#preproc

createGrammar(rawGrammarSrc, id, doRecompile, generatedParserLanguageCode, callback){SemanticInterpreter.prototype}

semantic/semanticInterpreter.js, line 810

Name	Type	Description
rawGrammarSrc	String | JSONObject
id	String
doRecompile	String | JSONObject	IF {String}: the String's contents will be used as a String-representation of the JSON grammar IF {Object}: the Object will be used as JSON representation for the grammar
generatedParserLanguageCode	String	optional OPTIONAL if param doRecompile is used, this String specifies the language for the generated grammatic-parser. If omitted, the default "de" (German) will be used. NOTE: this must be a valid ISO language code!
callback	function	optional

Returns:

Type	Description
SemanticInterpreter.prototype

getCurrentGrammar()

semantic/semanticInterpreter.js, line 922

id (i.e. the key for map grammarImplMap) for currently used grammar. If for invocations of interpret(..) etc. function the ID/languageCode argument is missing/omitted, then this id will be used. NOTE: if not NULL, the grammar must be available, either as compiled JS file (which must be already loaded, i.e. already present in grammarImplMap), or as JSON grammar file (which must be available at /config/languages/[ID]/grammar.json

getEngineCompileMode(){Boolean}

semantic/semanticInterpreter.js, line 1014

Get compile-mode (sychronous or asynchronous) for the grammar engine, i.e. if the compiler engine for the JSON grammar should run synchronously or asynchronously.

Returns:

Type	Description
Boolean	the compile mode (sychronous or asynchronous) when generating new parsers with the grammar-engine.

getEngineCompileStrictMode(){Boolean}

semantic/semanticInterpreter.js, line 1024

Get JavaScript strict mode compile-setting for the grammar engine, i.e. if the compiler engine should generate code with strict-mode setting.

Returns:

Type	Description
Boolean	the strict mode setting

getGrammarConverter(id){mmir.grammar.GrammarConverter}

semantic/semanticInterpreter.js, line 799

Get the grammar converter instance (of registered grammar)

Name	Type	Description
id	String	optional the ID (identifier) / language code for grammar if omitted: the currently active grammar

See:

• addGrammar
• setCurrentGrammar

Returns:

Type	Description
mmir.grammar.GrammarConverter	the grammar converter

getGrammarDefinitionText(id)

semantic/semanticInterpreter.js, line 776

NOTE: the grammar must be compiled/registered first

Name	Type	Description
id	String	the ID (identifier) / language code for grammar

See:

• mmir.grammar.GrammarConverter#getGrammarDef

getGrammarEngine(){String}

semantic/semanticInterpreter.js, line 949

Get the ID of the current grammar engine / compiler.

Default Value:

• "jcss"

Returns:

Type	Description
String	the ID of the current grammar engine

getGrammarParserText(id)

semantic/semanticInterpreter.js, line 785

NOTE: the grammar must be compiled/registered first

Name	Type	Description
id	String	the ID (identifier) / language code for grammar

See:

• mmir.grammar.GrammarConverter#getGrammarSource

hasGrammar(id){Boolean}

semantic/semanticInterpreter.js, line 838

Check if grammar is register

Name	Type	Description
id	string	the grammar ID

Returns:

Type	Description
Boolean	true, if grammar with ID is registered

interpret(phrase, langCode, callback){Object}

semantic/semanticInterpreter.js, line 718

Name	Type	Description
phrase	String	the phrase that will be parsed
langCode	String	the language code (identifier) for the parser/grammar
callback	function | ParseOptions	optional OPTIONAL parsing-options or a callback: options.callback: FUNCTION the callback function (see below) options.debug: BOOLEAN enabling debug output (by default the logger's log-level <= 'debug' is used) options.trace: BOOLEAN enabling verbose/tracing output; may not be supported by all grammar engines (by default the logger's log-level <= 'verbose' is used) NOTE: some grammar engines may support additional parsing options If a callback function: receives the return value (instead of receiving the result as return value from this function directly). The signature for the callback is: callback(result: Object) (i.e. the result that would be returned by this function itself is passed as argument into the callback function; see also documentation for returns). NOTE: in case, the grammar for the requested langCode is not compiled yet (i.e. not present as executable JavaScript), the corresponding JSON definition of the grammar needs to be compiled first, before processing the ASR's semantics is possible. In this case, a callback function MUST be supplied in order to receive a result (since compilation of the grammar may be asynchronous).

Returns:

Type	Description
Object	the parsing result (as processed by the parser / grammar; usually a JSON-like object). WARNING: if a callback function was provided, then there is no return object.

isEnabled()

semantic/semanticInterpreter.js, line 937

Flag for enabling/disabling processing of SemanticInterpreter. If disabled, interpret(), applyPreProcessing() will return null values. NOTE: if no grammar for any language is available, the SemanticInterpreter should be disabled. Setting a language, automatically enables the the SemanticInterpreter.

isPreProcessPositionsEnabled(){Boolean}

semantic/semanticInterpreter.js, line 1064

If true, pre-processing input-phrase (before running interpretion) will include meta-data for changed positions (due to pre-processing) in input-string E.g. can be used to map semantic-results (matched tokens/utterances where e.g. stopwords would be removed) on the raw input-string. The meta-information will be included in field preproc of the interpretation result.

See:

• setPreProcessPositionsEnabled
• _isCalcProcPos

Returns:

Type	Description
Boolean	if calculation of modified positions during pre-processing is enabled

removeGrammar(id)

semantic/semanticInterpreter.js, line 844

Remove a registered grammar

Name	Type	Description
id	string	the grammar ID to remove

removeStopwords(thePhrase, lang)

semantic/semanticInterpreter.js, line 738

Removes stopwords using the stopword-list from the parser/grammar for lang.

Name	Type	Description
thePhrase	String	the Phrase for which stopwords should be removed
lang	String	the language code (identifier) for the parser/grammar

Deprecated

• use applyPreProcessing instead

See:

• applyPreProcessing

setCurrentGrammar(id)

semantic/semanticInterpreter.js, line 910

Sets the current grammar. If in invocations of interpret the grammar ID (e.g. language code) is missing, then this grammar that is set here is used. The id must reference either a grammar that was compiled (i.e. generated JavaScript file) for this id, or there must exists JSON-grammar file for which the language-dir matches the id parameter, e.g. config/languages/[id]/grammar.json.

Name	Type	Description
id	String	the ID for the grammar, e.g. an ISO language code

setEnabled()

semantic/semanticInterpreter.js, line 930

See:

• isEnabled

setEngineCompileMode(asyncCompileMode, disableStrictMode)

semantic/semanticInterpreter.js, line 999

Set compile-mode (sychronous or asynchronous) for the grammar engine, i.e. if the compiler engine for the JSON grammar should run synchronously or asynchronously. NOTE: if there is no asynchronous implementation available for the grammar engine, the sync-impl. is used by default. NOTE: asynchronous compile mode requires WebWorkers

Name	Type	Description
asyncCompileMode	Boolean	sets the compile mode (sychronous or asynchronous) when generating new parsers with the grammar-engine.
disableStrictMode	Boolean	optional OPTIONAL disable JavaScript strict mode when generating grammar code

Default Value:

• false (i.e. synchronous compile mode)

setGrammarEngine(egnineId, asyncCompileMode, disableStrictMode)

semantic/semanticInterpreter.js, line 975

Set the grammar engine, i.e. the compiler engine for the JSON grammar NOTE: implementations of the grammar engines are located at env/grammar/ The file-name for an implementation should follow the convention: ID+"Generator.js" and should be registered with requirejs with the module-ID: ID+"Gen"

Name	Type	Description
egnineId	String	the ID for the engine. Possible values: "jscc", "jison", "pegjs"
asyncCompileMode	Boolean	optional OPITIONAL sets the compile mode (sychronous or asynchronous) when generating new parsers with the grammar-engine. DEFAULT: VOID (i.e. leave current set compile-mode setting unchanged)
disableStrictMode	Boolean	optional OPTIONAL disable JavaScript strict mode when generating grammar code NOTE: this argument is positional, i.e. asyncCompileMode must also be given when using this argument

setPreProcessPositionsEnabled(isEnabled)

semantic/semanticInterpreter.js, line 1045

Enable / disable calculation of modified positions during pre-processing

Name	Type	Description
isEnabled	Boolean	if calculation of modified positions during pre-processing should be enabled

See:

• isPreProcessPositionsEnabled
• _isCalcProcPos

setStopwords(id, grammarImpl, fileFormatNo)

semantic/semanticInterpreter.js, line 831

Add/register grammar for use with interpret NOTE: if no other grammar is available yet, currentGrammarId will be set to id. NOTE: if currently disabled, calling this function automatically enables ( setEnabled(TRUE) ), the semantic interpreter.

Name	Type	Description
id	String	ID for the grammar (e.g. an ISO-639 language code)
grammarImpl	mmir.grammar.GrammarConverter | function	the executable JavaScript grammar implementation IF {mmir.grammar.GrammarConverter}: the impl. with valid member {Function} mmir.grammar.GrammarConverter.executeGrammar() IF {Function}: the {Function} mmir.grammar.GrammarConverter#executeGrammar() - In this case, if no GrammarConverter instance fo id is present, a new one will be created; The stopwords must already be set, be part of the options-argument (see doc for fileFormatNo), or must additionally be set for the GrammarConverter instance (e.g. using mmir.SemanticInterpreter.setStopwords)
fileFormatNo	Number | PlainObject	optional OPTIONAL If Number and the number given does not match GRAMMAR_FILE_FORMAT_VERSION the file format is assumed to be out-dated and an Error will be thrown. If PlainObject, i.e. an options object, the following properties are evaluated (all properties are optional): fileFormat: NUMBER, default: undefined (desc. see above) execMode: 'sync' | 'async', default: 'sync' if 'async' then the grammar is executed asynchronously, i.e. interpret() must be invoked with a callback function in order to retrieve the result stopwords: Array, default: null if given, the grammar (GrammarConverter) will be set with this stopword list, i.e. grammar.setStopwords(stopwords)

Throws:

Error if fileFormatNo is given, but does not match GRAMMAR_FILE_FORMAT_VERSION.