private void InitEnv()
{
env = TokenSequencePattern.GetNewEnv();
env.SetDefaultTokensAnnotationKey(typeof(CoreAnnotations.NumerizedTokensAnnotation));
// Do case insensitive matching
env.SetDefaultStringMatchFlags(Pattern.CaseInsensitive | Pattern.UnicodeCase);
env.SetDefaultStringPatternFlags(Pattern.CaseInsensitive | Pattern.UnicodeCase);
try
{
Units.RegisterUnits(env, options.unitsFilename);
}
catch (IOException ex)
{
throw new Exception("Error loading units from " + options.unitsFilename, ex);
}
try
{
UnitPrefix.RegisterPrefixes(env, options.prefixFilename);
}
catch (IOException ex)
{
throw new Exception("Error loading prefixes from " + options.prefixFilename, ex);
}
env.Bind("options", options);
env.Bind("numcomptype", typeof(CoreAnnotations.NumericCompositeTypeAnnotation));
env.Bind("numcompvalue", typeof(CoreAnnotations.NumericCompositeValueAnnotation));
}
public KBPTokensregexExtractor(string tokensregexDir, bool verbose)
{
if (verbose)
{
logger.Log("Creating TokensRegexExtractor");
}
// Create extractors
foreach (KBPRelationExtractor.RelationType rel in KBPRelationExtractor.RelationType.Values())
{
string relFileNameComponent = rel.canonicalName.ReplaceAll(":", "_");
string path = tokensregexDir + File.separator + relFileNameComponent.ReplaceAll("/", "SLASH") + ".rules";
if (IOUtils.ExistsInClasspathOrFileSystem(path))
{
IList <string> listFiles = new List <string>();
listFiles.Add(tokensregexDir + File.separator + "defs.rules");
listFiles.Add(path);
if (verbose)
{
logger.Log("Rule files for relation " + rel + " is " + path);
}
Env env = TokenSequencePattern.GetNewEnv();
env.Bind("collapseExtractionRules", true);
env.Bind("verbose", verbose);
CoreMapExpressionExtractor extr = CoreMapExpressionExtractor.CreateExtractorFromFiles(env, listFiles).KeepTemporaryTags();
rules[rel] = extr;
}
}
}
public TokensRegexAnnotator(string name, Properties props)
{
string prefix = (name == null) ? string.Empty : name + '.';
string[] files = PropertiesUtils.GetStringArray(props, prefix + "rules");
env = TokenSequencePattern.GetNewEnv();
env.Bind("options", options);
if (PropertiesUtils.GetBool(props, prefix + "caseInsensitive"))
{
System.Console.Error.WriteLine("using case insensitive!");
env.SetDefaultStringMatchFlags(NodePattern.CaseInsensitive | Pattern.UnicodeCase);
env.SetDefaultStringPatternFlags(Pattern.CaseInsensitive | Pattern.UnicodeCase);
}
if (files.Length != 0)
{
extractor = CoreMapExpressionExtractor.CreateExtractorFromFiles(env, files);
}
else
{
extractor = null;
}
verbose = PropertiesUtils.GetBool(props, prefix + "verbose", false);
options.setTokenOffsets = PropertiesUtils.GetBool(props, prefix + "setTokenOffsets", options.setTokenOffsets);
options.extractWithTokens = PropertiesUtils.GetBool(props, prefix + "extractWithTokens", options.extractWithTokens);
options.flatten = PropertiesUtils.GetBool(props, prefix + "flatten", options.flatten);
string matchedExpressionsAnnotationKeyName = props.GetProperty(prefix + "matchedExpressionsAnnotationKey");
if (matchedExpressionsAnnotationKeyName != null)
{
options.matchedExpressionsAnnotationKey = EnvLookup.LookupAnnotationKeyWithClassname(env, matchedExpressionsAnnotationKeyName);
if (options.matchedExpressionsAnnotationKey == null)
{
string propName = prefix + "matchedExpressionsAnnotationKey";
throw new Exception("Cannot determine annotation key for " + propName + '=' + matchedExpressionsAnnotationKeyName);
}
}
}
public TokensRegexAnnotator(params string[] files)
{
env = TokenSequencePattern.GetNewEnv();
extractor = CoreMapExpressionExtractor.CreateExtractorFromFiles(env, files);
verbose = false;
}
/// <exception cref="System.IO.IOException"/>
public static void Main(string[] args)
{
string rules;
if (args.Length > 0)
{
rules = args[0];
}
else
{
rules = "edu/stanford/nlp/ling/tokensregex/demo/rules/expr.rules.txt";
}
PrintWriter @out;
if (args.Length > 2)
{
@out = new PrintWriter(args[2]);
}
else
{
@out = new PrintWriter(System.Console.Out);
}
CoreMapExpressionExtractor <MatchedExpression> extractor = CoreMapExpressionExtractor.CreateExtractorFromFiles(TokenSequencePattern.GetNewEnv(), rules);
StanfordCoreNLP pipeline = new StanfordCoreNLP(PropertiesUtils.AsProperties("annotators", "tokenize,ssplit,pos,lemma,ner"));
Annotation annotation;
if (args.Length > 1)
{
annotation = new Annotation(IOUtils.SlurpFileNoExceptions(args[1]));
}
else
{
annotation = new Annotation("( ( five plus three plus four ) * 2 ) divided by three");
}
pipeline.Annotate(annotation);
// An Annotation is a Map and you can get and use the various analyses individually.
@out.Println();
// The toString() method on an Annotation just prints the text of the Annotation
// But you can see what is in it with other methods like toShorterString()
@out.Println("The top level annotation");
@out.Println(annotation.ToShorterString());
IList <ICoreMap> sentences = annotation.Get(typeof(CoreAnnotations.SentencesAnnotation));
int i = 0;
foreach (ICoreMap sentence in sentences)
{
@out.Println("Sentence #" + ++i);
foreach (CoreLabel token in sentence.Get(typeof(CoreAnnotations.TokensAnnotation)))
{
@out.Println(" Token: " + "word=" + token.Get(typeof(CoreAnnotations.TextAnnotation)) + ", pos=" + token.Get(typeof(CoreAnnotations.PartOfSpeechAnnotation)) + ", ne=" + token.Get(typeof(CoreAnnotations.NamedEntityTagAnnotation)));
}
IList <MatchedExpression> matchedExpressions = extractor.ExtractExpressions(sentence);
foreach (MatchedExpression matched in matchedExpressions)
{
// Print out matched text and value
@out.Println("Matched expression: " + matched.GetText() + " with value " + matched.GetValue());
// Print out token information
ICoreMap cm = matched.GetAnnotation();
foreach (CoreLabel token_1 in cm.Get(typeof(CoreAnnotations.TokensAnnotation)))
{
string word = token_1.Get(typeof(CoreAnnotations.TextAnnotation));
string lemma = token_1.Get(typeof(CoreAnnotations.LemmaAnnotation));
string pos = token_1.Get(typeof(CoreAnnotations.PartOfSpeechAnnotation));
string ne = token_1.Get(typeof(CoreAnnotations.NamedEntityTagAnnotation));
@out.Println(" Matched token: " + "word=" + word + ", lemma=" + lemma + ", pos=" + pos + ", ne=" + ne);
}
}
}
@out.Flush();
}
private void InitEnv()
{
env = TokenSequencePattern.GetNewEnv();
env.SetDefaultResultsAnnotationExtractor(TimeExpression.TimeExpressionConverter);
env.SetDefaultTokensAnnotationKey(typeof(CoreAnnotations.NumerizedTokensAnnotation));
env.SetDefaultResultAnnotationKey(typeof(TimeExpression.Annotation));
env.SetDefaultNestedResultsAnnotationKey(typeof(TimeExpression.ChildrenAnnotation));
env.SetDefaultTokensAggregators(CoreMapAttributeAggregator.DefaultNumericTokensAggregators);
env.Bind("nested", typeof(TimeExpression.ChildrenAnnotation));
env.Bind("time", new TimeFormatter.TimePatternExtractRuleCreator());
// Do case insensitive matching
env.SetDefaultStringPatternFlags(Pattern.CaseInsensitive | Pattern.UnicodeCase);
env.Bind("options", options);
env.Bind("TIME_REF", SUTime.TimeRef);
env.Bind("TIME_REF_UNKNOWN", SUTime.TimeRefUnknown);
env.Bind("TIME_UNKNOWN", SUTime.TimeUnknown);
env.Bind("TIME_NONE", SUTime.TimeNone);
env.Bind("ERA_AD", SUTime.EraAd);
env.Bind("ERA_BC", SUTime.EraBc);
env.Bind("ERA_UNKNOWN", SUTime.EraUnknown);
env.Bind("HALFDAY_AM", SUTime.HalfdayAm);
env.Bind("HALFDAY_PM", SUTime.HalfdayPm);
env.Bind("HALFDAY_UNKNOWN", SUTime.HalfdayUnknown);
env.Bind("RESOLVE_TO_THIS", SUTime.ResolveToThis);
env.Bind("RESOLVE_TO_PAST", SUTime.ResolveToPast);
env.Bind("RESOLVE_TO_FUTURE", SUTime.ResolveToFuture);
env.Bind("RESOLVE_TO_CLOSEST", SUTime.ResolveToClosest);
env.Bind("numcomptype", typeof(CoreAnnotations.NumericCompositeTypeAnnotation));
env.Bind("numcompvalue", typeof(CoreAnnotations.NumericCompositeValueAnnotation));
env.Bind("temporal", typeof(TimeExpression.Annotation));
// env.bind("tags", SequenceMatchRules.Tags.TagsAnnotation.class);
env.Bind("::IS_TIMEX_DATE", new GenericTimeExpressionPatterns.TimexTypeMatchNodePattern(SUTime.TimexType.Date));
env.Bind("::IS_TIMEX_DURATION", new GenericTimeExpressionPatterns.TimexTypeMatchNodePattern(SUTime.TimexType.Duration));
env.Bind("::IS_TIMEX_TIME", new GenericTimeExpressionPatterns.TimexTypeMatchNodePattern(SUTime.TimexType.Time));
env.Bind("::IS_TIMEX_SET", new GenericTimeExpressionPatterns.TimexTypeMatchNodePattern(SUTime.TimexType.Set));
env.Bind("::IS_TIME_UNIT", new GenericTimeExpressionPatterns.MatchedExpressionValueTypeMatchNodePattern("TIMEUNIT"));
env.Bind("::MONTH", new GenericTimeExpressionPatterns.MatchedExpressionValueTypeMatchNodePattern("MONTH_OF_YEAR"));
env.Bind("::DAYOFWEEK", new GenericTimeExpressionPatterns.MatchedExpressionValueTypeMatchNodePattern("DAY_OF_WEEK"));
// BINDINGS for parsing from file!!!!!!!
foreach (SUTime.TemporalOp t in SUTime.TemporalOp.Values())
{
env.Bind(t.ToString(), new Expressions.PrimitiveValue <SUTime.TemporalOp>("TemporalOp", t));
}
foreach (SUTime.TimeUnit t_1 in SUTime.TimeUnit.Values())
{
if (!t_1.Equals(SUTime.TimeUnit.Unknown))
{
//env.bind(t.name(), new SequenceMatchRules.PrimitiveValue<SUTime.Temporal>("DURATION", t.getDuration(), "TIMEUNIT"));
env.Bind(t_1.ToString(), new Expressions.PrimitiveValue <SUTime.Temporal>("TIMEUNIT", t_1.GetDuration()));
}
}
foreach (SUTime.StandardTemporalType t_2 in SUTime.StandardTemporalType.Values())
{
env.Bind(t_2.ToString(), new Expressions.PrimitiveValue <SUTime.StandardTemporalType>("TemporalType", t_2));
}
env.Bind("Duration", new Expressions.PrimitiveValue <IValueFunction>(Expressions.TypeFunction, new _NamedValueFunction_124("Duration")));
// New so we get different time ids
// TODO: Check args
// TODO: Handle Strings...
// TODO: This should already be in durations....
//String durationUnitString = (durationUnitTokens != null)? durationUnitTokens.get(0).get(CoreAnnotations.TextAnnotation.class):null;
//SUTime.Duration durationUnit = getDuration(durationUnitString);
// TODO: Handle inexactness
// Create duration range...
// Add begin and end times
env.Bind("DayOfWeek", new Expressions.PrimitiveValue <IValueFunction>(Expressions.TypeFunction, new _NamedValueFunction_212("DayOfWeek")));
env.Bind("MonthOfYear", new Expressions.PrimitiveValue <IValueFunction>(Expressions.TypeFunction, new _NamedValueFunction_235("MonthOfYear")));
env.Bind("MakePeriodicTemporalSet", new Expressions.PrimitiveValue <IValueFunction>(Expressions.TypeFunction, new _NamedValueFunction_258("MakePeriodicTemporalSet")));
// First argument is the temporal acting as the base of the periodic set
// Second argument is the quantifier (string)
// Third argument is the multiple (how much to scale the natural period)
/*"P1X"*/
env.Bind("TemporalCompose", new Expressions.PrimitiveValue <IValueFunction>(Expressions.TypeFunction, new _NamedValueFunction_328("TemporalCompose")));
}