gate.lib.basicdocument.BdocDocumentBuilder Maven / Gradle / Ivy
/*
* Copyright (c) 2019 The University of Sheffield.
*
* This file is part of gateplugin-Format_Bdoc
* (see https://github.com/GateNLP/gateplugin-Format_Bdoc).
*
* This program is free software: you can redistribute it and/or modify
* it under the terms of the GNU Lesser General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public License
* along with this program. If not, see > knownSets = new HashMap<>();
Set includedSetNames = new HashSet<>();
HashMap nextAnnotationIds = new HashMap<>();
HashMap includedFeatures = new HashMap<>();
boolean includePlaceholderSets = false;
String offset_type = "j";
String name = "";
int nextAnnId = 1;
/**
* Tell the builder to create the JsonDocument from a GATE document.
* By default, this will include all features and all annotation sets,
* but this can be changed later.
* Alternately, the JsonDocument can be constructed by adding the parts
* (text, document features, annotation sets) individually.
*
* @param doc the Gate document to build the BdocDocument/JSON from
* @return modified BdocDocumentBuilder
*/
public BdocDocumentBuilder fromGate(Document doc) {
// TODO: for now check that Document is a SimpleDocument
this.text = doc.getContent().toString();
this.name = doc.getName();
if(! (doc instanceof DocumentImpl)) {
throw new GateRuntimeException("Cannot build Bdoc document from something that is not a gate.corpora.DocumentImpl");
}
nextAnnId = ((DocumentImpl)doc).getNextAnnotationId();
knownSets.put("", doc.getAnnotations());
includedSetNames.add("");
for (String name : doc.getNamedAnnotationSets().keySet()) {
knownSets.put(name, doc.getAnnotations(name));
includedSetNames.add(name);
}
BdocUtils.featureMap2Map(doc.getFeatures(), includedFeatures);
return this;
}
/**
* Add an annotation set.
* Same as addSet(name, annset) but the annotation set uses the given
* nextAnnotationId for new annotations.
* NOTE: if the maximum annotation id in the set is bigger than the
* given nextAnnotationId, then the bigger Id is used instead!
*
* @param name the name of the annotation set (this can differ from the
* original name if an annotation set is passed. Must not be null, the
* "default" set uses the empty string as name.
* @param annset a set of annotations, could be an AnnotationSet or a set
* of annotations.
* @param nextAnnotationId the annotation id to start from when new
* annotations are added to this set in the Bdoc document.
* @return modified BdocDocumentBuilder
*/
public BdocDocumentBuilder addSet(String name, Set annset,
int nextAnnotationId) {
knownSets.put(name, annset);
includedSetNames.add(name);
nextAnnotationIds.put(name, nextAnnotationId);
return this;
}
/**
* Add all features from the given (feature) map as document features.
*
* The map can have keys and values of any type, but when adding features,
* the following conversions are carried out: null keys are removed,
* any key that is not a String is converted to String. The value of all
* features should be something that is directly JSON-serialisable, but this
* is not checked. Note that some types can get serialised to JSON but
* will get converted to a different type when read back from JSON!
*
* The user is responsible for making sure that value types work with the
* JSON default serialisation.
*
* If a feature has already been added previously, its old value is
* replaced.
*
* @param fm a map to interpret as a feature map
* @return modified BdocDocumentBuilder
*/
public BdocDocumentBuilder addFeatures(Map fm) {
BdocUtils.featureMap2Map(fm, includedFeatures);
return this;
}
/**
* Set/update the text of the document.
*
* @param text the new document text
* @return modified BdocDocumentBuilder
*/
public BdocDocumentBuilder setText(String text) {
this.text = text;
return this;
}
/**
* Set/update the name of the bdoc document.
*
* @param name the new document name
* @return modified BdocDocumentBuilder
*/
public BdocDocumentBuilder setName(String name) {
this.name = name;
return this;
}
/**
* Set the list of included AnnotationSet names.
*
* This allows you to select the annotation sets to actually use from
* a GATE document.
*
* If set names are specified which are not in initial or currently known list of sets,
* this is silently ignored. That way the same selection can be used for many documents,
* even if some of them do not actually contain some of the sets.
*
* @param names a collection of names to choose and use for the BdocDocument
* @return modified BdocDocumentBuilder
*/
public BdocDocumentBuilder setAnnotationSetNames(Collection names) {
includedSetNames.clear();
includedSetNames.addAll(names);
return this;
}
/**
* Set if placeholders sets should get included.
*
* This is only relevant if a subset of all annotation sets is selected to get added.
* @param flag
* @return
*/
public BdocDocumentBuilder setIncludePlaceholderSets(boolean flag) {
includePlaceholderSets = flag;
return this;
}
/**
* Set the list of included features.
*
* This allows you to select the features to actually use from a GATE document.
* If features are selected which are not known, this is silently ignored.
*
*
* @param featurenames a collection of feature names to choose and use for
* the BdocDocument
* @return modified BdocDocumentBuilder
*/
public BdocDocumentBuilder setFeatureNames(Collection featurenames) {
HashMap newFeatures = new HashMap<>();
for (String tmpname : featurenames) {
if(includedFeatures.containsKey(tmpname)) {
newFeatures.put(tmpname, includedFeatures.get(tmpname));
}
}
includedFeatures = newFeatures;
return this;
}
/**
* Add/set a single feature as a document feature.
*
* @param name Feature name
* @param value Feature value
* @return modified BdocDocumentBuilder
*/
public BdocDocumentBuilder addFeature(String name, Object value) {
includedFeatures.put(name, value);
return this;
}
/**
* Make the JsonDocument use java offsets (the default).
* @return modified BdocDocumentBuilder
*/
public BdocDocumentBuilder javaOffsets() {
offset_type = "j";
return this;
}
/**
* Make the JsonDocument use python/unicode codepoint offsets.
* This wil fix all the annotation offsets so they refer to unicode
* code points instead of java utf16 code units. This will only work
* if the text of the json document is set and compatible with the
* offsets of the added annotations, otherwise an exception is thrown.
* @return modified BdocDocumentBuilder
*/
public BdocDocumentBuilder pythonOffsets() {
offset_type = "p";
return this;
}
/**
* Given all the info accumulated, build a JsonDocument and return it.
*
* This adds a BdocDocument instance with all the information added
* so far.
*
* @return the BdocDocument containing all the information added so far
*/
public BdocDocument buildBdoc() {
BdocDocument ret = new BdocDocument();
ret.text = text;
ret.name = name;
if(includedFeatures.size() > 0) {
ret.features = includedFeatures;
}
if(knownSets.size() > 0) {
HashMap annotation_sets = new HashMap<>();
for(String name : knownSets.keySet()) {
if(!includedSetNames.contains(name)) {
// if a set should not get added, we either completely ignore it, or if
// includePlaceholderSets is true, we add an empty set with that name which
// has the next annotation id set to the one from the gate document
if (includePlaceholderSets) {
BdocAnnotationSet annset = new BdocAnnotationSet();
annset.name = name;
annset.annotations = new ArrayList<>();
annset.next_annid = nextAnnId;
annotation_sets.put(annset.name, annset);
}
continue;
}
BdocAnnotationSet annset = new BdocAnnotationSet();
annset.name = name;
annset.annotations = new ArrayList<>();
int next_annid = 0;
for (Annotation ann : knownSets.get(name)) {
BdocAnnotation bdocann = BdocAnnotation.fromGateAnnotation(ann);
if(bdocann.id >= next_annid) {
next_annid = bdocann.id + 1;
}
annset.annotations.add(bdocann);
}
if(nextAnnotationIds.containsKey(name)) {
annset.next_annid = Math.max(next_annid, nextAnnotationIds.get(name));
} else {
annset.next_annid = next_annid;
}
annotation_sets.put(annset.name, annset);
}
ret.annotation_sets = annotation_sets;
}
// do any offset fixup, if necessary
ret.fixupOffsets(offset_type);
return ret;
}
/**
* Store the BdocDocument to a file.
*
* @param path where to store
*/
public void toJson(File path) {
BdocDocument bdoc = buildBdoc();
new Saver().format(Format.JSON_MAP).to(path).save(bdoc);
}
/**
* Store the BdocDocument to a file.
*
* @param path where to store
*/
public void toJson(String path) {
BdocDocument bdoc = buildBdoc();
new Saver().format(Format.JSON_MAP).to(new File(path)).save(bdoc);
}
/**
* Convert the BdocDocument to a JSON string.
*
*
* @return JSON String
*/
public String toJsonString() {
BdocDocument bdoc = buildBdoc();
return new Saver().format(Format.JSON_MAP).asString().save(bdoc);
}
/**
* Write the BdocDocument to the output stream.
*
* @param os stream to write to
*/
public void toJson(OutputStream os) {
BdocDocument bdoc = buildBdoc();
new Saver().format(Format.JSON_MAP).to(os).save(bdoc);
}
}
© 2015 - 2025 Weber Informatics LLC | Privacy Policy