private MultiPatternMatcher<CoreMap> createPatternMatcher(
Map<SequencePattern<CoreMap>, Entry> patternToEntry) {
// Convert to tokensregex pattern
int patternFlags = ignoreCase ? Pattern.CASE_INSENSITIVE : 0;
int stringMatchFlags = ignoreCase ? NodePattern.CASE_INSENSITIVE : 0;
Env env = TokenSequencePattern.getNewEnv();
env.setDefaultStringPatternFlags(patternFlags);
env.setDefaultStringMatchFlags(stringMatchFlags);
NodePattern<String> posTagPattern =
(validPosPattern != null && PosMatchType.MATCH_ALL_TOKENS.equals(posMatchType))
? new CoreMapNodePattern.StringAnnotationRegexPattern(validPosPattern)
: null;
List<TokenSequencePattern> patterns = new ArrayList<>(entries.size());
for (Entry entry : entries) {
TokenSequencePattern pattern;
if (entry.tokensRegex != null) {
// TODO: posTagPatterns...
pattern = TokenSequencePattern.compile(env, entry.tokensRegex);
} else {
List<SequencePattern.PatternExpr> nodePatterns = new ArrayList<>();
for (String p : entry.regex) {
CoreMapNodePattern c = CoreMapNodePattern.valueOf(p, patternFlags);
if (posTagPattern != null) {
c.add(CoreAnnotations.PartOfSpeechAnnotation.class, posTagPattern);
}
nodePatterns.add(new SequencePattern.NodePatternExpr(c));
}
pattern =
TokenSequencePattern.compile(new SequencePattern.SequencePatternExpr(nodePatterns));
}
if (entry.annotateGroup < 0 || entry.annotateGroup > pattern.getTotalGroups()) {
throw new RuntimeException("Invalid match group for entry " + entry);
}
pattern.setPriority(entry.priority);
patterns.add(pattern);
patternToEntry.put(pattern, entry);
}
return TokenSequencePattern.getMultiPatternMatcher(patterns);
}
@Override
protected List<Attribute> init(
AbstractDefinition abstractDefinition,
ExpressionExecutor[] attributeExpressionExecutors,
ExecutionPlanContext executionPlanContext) {
if (logger.isDebugEnabled()) {
logger.debug("Initializing Query ...");
}
if (attributeExpressionLength < 2) {
throw new ExecutionPlanCreationException(
"Query expects at least two parameters. Received only "
+ attributeExpressionLength
+ ".\nUsage: #nlp.findTokensRegexPattern(regex:string, text:string-variable)");
}
String regex;
try {
if (attributeExpressionExecutors[0] instanceof ConstantExpressionExecutor) {
regex = (String) attributeExpressionExecutors[0].execute(null);
} else {
throw new ExecutionPlanCreationException(
"First parameter should be a constant."
+ ".\nUsage: #nlp.findTokensRegexPattern(regex:string, text:string-variable)");
}
} catch (ClassCastException e) {
throw new ExecutionPlanCreationException(
"First parameter should be of type string. Found "
+ attributeExpressionExecutors[0].getReturnType()
+ ".\nUsage: #nlp.findTokensRegexPattern(regex:string, text:string-variable)");
}
try {
regexPattern = TokenSequencePattern.compile(regex);
} catch (Exception e) {
throw new ExecutionPlanCreationException("Cannot parse given regex " + regex, e);
}
if (!(attributeExpressionExecutors[1] instanceof VariableExpressionExecutor)) {
throw new ExecutionPlanCreationException(
"Second parameter should be a variable."
+ ".\nUsage: #nlp.findTokensRegexPattern(regex:string, text:string-variable)");
}
if (logger.isDebugEnabled()) {
logger.debug(
String.format(
"Query parameters initialized. Regex: %s Stream Parameters: %s",
regex, abstractDefinition.getAttributeList()));
}
initPipeline();
ArrayList<Attribute> attributes = new ArrayList<Attribute>(1);
attributes.add(new Attribute("match", Attribute.Type.STRING));
attributeCount = regexPattern.getTotalGroups();
for (int i = 1; i < attributeCount; i++) {
attributes.add(new Attribute(groupPrefix + i, Attribute.Type.STRING));
}
return attributes;
}
