public TokensRegexNERAnnotator(String name, Properties properties) {
String prefix = (name != null && !name.isEmpty()) ? name + '.' : "";
String backgroundSymbol =
properties.getProperty(prefix + "backgroundSymbol", DEFAULT_BACKGROUND_SYMBOL);
String[] backgroundSymbols = backgroundSymbol.split("\\s*,\\s*");
String mappingFiles =
properties.getProperty(prefix + "mapping", DefaultPaths.DEFAULT_REGEXNER_RULES);
String[] mappings = mappingFiles.split("\\s*[,;]\\s*");
String validPosRegex = properties.getProperty(prefix + "validpospattern");
this.posMatchType =
PosMatchType.valueOf(
properties.getProperty(prefix + "posmatchtype", DEFAULT_POS_MATCH_TYPE.name()));
String noDefaultOverwriteLabelsProp =
properties.getProperty(prefix + "noDefaultOverwriteLabels");
this.noDefaultOverwriteLabels =
(noDefaultOverwriteLabelsProp != null)
? Collections.unmodifiableSet(
CollectionUtils.asSet(noDefaultOverwriteLabelsProp.split("\\s*,\\s*")))
: Collections.unmodifiableSet(new HashSet<>());
this.ignoreCase = PropertiesUtils.getBool(properties, prefix + "ignorecase", false);
this.verbose = PropertiesUtils.getBool(properties, prefix + "verbose", false);
if (validPosRegex != null && !validPosRegex.isEmpty()) {
validPosPattern = Pattern.compile(validPosRegex);
} else {
validPosPattern = null;
}
entries =
Collections.unmodifiableList(
readEntries(name, noDefaultOverwriteLabels, ignoreCase, verbose, mappings));
IdentityHashMap<SequencePattern<CoreMap>, Entry> patternToEntry = new IdentityHashMap<>();
multiPatternMatcher = createPatternMatcher(patternToEntry);
this.patternToEntry = Collections.unmodifiableMap(patternToEntry);
Set<String> myLabels = Generics.newHashSet();
// Can always override background or none.
Collections.addAll(myLabels, backgroundSymbols);
myLabels.add(null);
// Always overwrite labels
for (Entry entry : entries) myLabels.add(entry.type);
this.myLabels = Collections.unmodifiableSet(myLabels);
}
