public String formatNoEmpty() {
StringBuilder sb = new StringBuilder("[(").append(dictionaryItem.lemma).append(") ");
for (InflectionalGroup ig : inflectionalGroups) {
sb.append(ig.formatNoEmpty());
}
sb.append("]");
return sb.toString();
}
public String formatOnlyIgs() {
StringBuilder sb = new StringBuilder();
for (InflectionalGroup ig : inflectionalGroups) {
sb.append(ig.formatNoSurface());
}
return sb.toString();
}
public MorphParse(StemNode stemNode, List<SuffixSurfaceNode> suffixSurfaceNodes) {
this.dictionaryItem = stemNode.getDictionaryItem();
this.root = stemNode.surfaceForm;
InflectionalGroup ig = new InflectionalGroup();
List<InflectionalGroup> igs = Lists.newArrayListWithCapacity(2);
ig.pos = dictionaryItem.primaryPos;
ig.spos = dictionaryItem.secondaryPos;
int j = 0;
for (SuffixSurfaceNode suffixNode : suffixSurfaceNodes) {
SuffixFormTemplate template = null;
if (suffixNode.getSuffixForm() instanceof NullSuffixForm) {
template = ((NullSuffixForm) suffixNode.getSuffixForm()).getTemplate();
}
// if node is a derivational node then we create a new ig and store the other
if (template != null && (template instanceof DerivationalSuffixTemplate || j == 0)) {
RootSuffix rootSuffix = (RootSuffix) template.getSuffix();
if (ig.pos == null) {
ig.pos = rootSuffix.pos;
}
if (j > 0) {
igs.add(ig);
ig = new InflectionalGroup();
ig.pos = rootSuffix.pos;
}
} else {
if (!(suffixNode.getSuffixForm().getSuffix() instanceof RootSuffix))
ig.suffixList.add(
new SuffixData(
suffixNode.getSuffixForm().suffix,
suffixNode.surfaceForm,
suffixNode.getSuffixForm().generation));
}
j++;
}
if (igs.isEmpty() || !ig.suffixList.isEmpty()) {
igs.add(ig);
}
this.inflectionalGroups = igs;
}
/**
* Returns surface forms list of all root and derivational roots of a parse. Examples: "kitaplar"
* ->["kitap"] "kitabım" ->["kitab"] "kitaplaşır"->["kitap", "kitaplaş"] "kavrulduk" ->
* ["kavr","kavrul"]
*/
public List<String> getStems() {
List<String> stems = Lists.newArrayListWithCapacity(2);
stems.add(root);
String previousStem = stems.get(0);
if (inflectionalGroups.size() > 1) {
previousStem = previousStem + inflectionalGroups.get(0).surfaceForm();
for (int i = 1; i < inflectionalGroups.size(); i++) {
InflectionalGroup ig = inflectionalGroups.get(i);
SuffixData suffixData = ig.suffixList.get(0);
if (suffixData.surface.length() > 0) {
String surface = suffixData.surface;
String stem = previousStem + surface;
if (!stems.contains(stem)) stems.add(stem);
}
previousStem = previousStem + ig.surfaceForm();
}
}
return stems;
}
/**
* Returns list of all lemmas of a parse. Examples: "kitaplar" ->["kitap"] "kitabım" ->["kitap"]
* "kitaplaşır"->["kitap", "kitaplaş"] "kitaplaş" ->["kitap", "kitaplaş"] "arattıragörür" ->
* ["ara","arat","arattır","arattıragör"]
*/
public List<String> getLemmas() {
List<String> lemmas = Lists.newArrayListWithCapacity(2);
lemmas.add(dictionaryItem.root);
String previousStem = root;
if (inflectionalGroups.size() > 1) {
previousStem = previousStem + inflectionalGroups.get(0).surfaceForm();
for (int i = 1; i < inflectionalGroups.size(); i++) {
InflectionalGroup ig = inflectionalGroups.get(i);
SuffixData suffixData = ig.suffixList.get(0);
if (suffixData.surface.length() > 0) {
String surface = suffixData.surface;
if (suffixData.lex.endsWith("~k")) {
surface = surface.substring(0, surface.length() - 1) + "k";
}
String stem = previousStem + surface;
if (!lemmas.contains(stem)) lemmas.add(stem);
}
previousStem = previousStem + ig.surfaceForm();
}
}
return lemmas;
}
public boolean containsSuffix(Suffix suffix) {
for (InflectionalGroup inflectionalGroup : inflectionalGroups) {
if (inflectionalGroup.containsSuffix(suffix)) return true;
}
return false;
}