Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

fix: AdaptSuggestionFilter only considered the first reading #9356

Open
wants to merge 1 commit into
base: master
Choose a base branch
from
Open

fix: AdaptSuggestionFilter only considered the first reading #9356

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
94 changes: 50 additions & 44 deletions ...ol-language-modules/de/src/main/java/org/languagetool/rules/de/AdaptSuggestionFilter.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -91,27 +91,29 @@ List<String> getAdaptedDet(AnalyzedTokenReadings detToken, String repl) {
String oldDetBaseform = getBaseform(detToken, "(ART|PRO):.*");
List<String> result = new ArrayList<>();
try {
String replGender = getNounGender(repl);
if (replGender == null || oldDetBaseform == null) {
List<String> replGenders = getNounGender(repl);
if (replGenders.size() == 0 || oldDetBaseform == null) {
return result;
}
for (AnalyzedToken reading : detToken.getReadings()) {
if (reading.getPOSTag() == null || !(reading.getPOSTag().startsWith("ART:") || reading.getPOSTag().startsWith("PRO:"))) {
continue;
}
String newDetPos = reading.getPOSTag().replaceAll("MAS|FEM|NEU", replGender).replaceFirst("BEG", "(BEG|B/S)").replaceFirst(":STV", "");
String[] replDet = GermanSynthesizer.INSTANCE.synthesize(new AnalyzedToken(oldDetBaseform, null, oldDetBaseform), newDetPos, true);
for (String s : replDet) {
if (StringTools.startsWithUppercase(detToken.getToken())) {
if (!s.toLowerCase().startsWith(detToken.getToken().substring(0, 1).toLowerCase())) {
continue; // see below
for (String replGender : replGenders) {
String newDetPos = reading.getPOSTag().replaceAll("MAS|FEM|NEU", replGender).replaceFirst("BEG", "(BEG|B/S)").replaceFirst(":STV", "");
String[] replDet = GermanSynthesizer.INSTANCE.synthesize(new AnalyzedToken(oldDetBaseform, null, oldDetBaseform), newDetPos, true);
for (String s : replDet) {
if (StringTools.startsWithUppercase(detToken.getToken())) {
if (!s.toLowerCase().startsWith(detToken.getToken().substring(0, 1).toLowerCase())) {
continue; // see below
}
result.add(StringTools.uppercaseFirstChar(s));
} else {
if (!s.startsWith(detToken.getToken().substring(0, 1))) {
continue; // mein, dein, sein etc. all share the same lemma ("mein"), but don't suggest "dein" for "mein"
}
result.add(s);
}
result.add(StringTools.uppercaseFirstChar(s));
} else {
if (!s.startsWith(detToken.getToken().substring(0, 1))) {
continue; // mein, dein, sein etc. all share the same lemma ("mein"), but don't suggest "dein" for "mein"
}
result.add(s);
}
}
}
Expand All @@ -126,37 +128,39 @@ List<String> getAdaptedDetAdj(AnalyzedTokenReadings detToken, AnalyzedTokenReadi
String oldAdjBaseform = getBaseform(adjToken, "ADJ:.*");
List<String> result = new ArrayList<>();
try {
String replGender = getNounGender(repl);
if (replGender == null || oldDetBaseform == null || oldAdjBaseform == null) {
List<String> replGenders = getNounGender(repl);
if (replGenders.size() == 0 || oldDetBaseform == null || oldAdjBaseform == null) {
return result;
}
for (AnalyzedToken reading : detToken.getReadings()) {
if (reading.getPOSTag() == null || !reading.getPOSTag().matches("(ART|PRO):.*")) {
continue;
}
String newDetPos = reading.getPOSTag().replaceAll("MAS|FEM|NEU", replGender).replaceFirst("BEG", "(BEG|B/S)");
String newAdjPos;
if (newDetPos.startsWith("ART:")) {
newAdjPos = reading.getPOSTag().replaceAll("MAS|FEM|NEU", replGender).replaceFirst("BEG", "(BEG|B/S)").replaceFirst(":STV", "");
} else if (newDetPos.startsWith("PRO:")) {
newAdjPos = newDetPos.replaceAll("PRO:POS:(NOM|AKK|GEN|DAT):(SIN|PLU):(MAS|FEM|NEU)", "ADJ:$1:$2:$3").replaceFirst(":(STV|BEG).*", ":GRU:IND");
} else {
throw new RuntimeException("Unexpected POS tag: " + newDetPos);
}
//System.out.println("newDetPos: " + newDetPos + " for " + oldDetBaseform);
//System.out.println("newAdjPos: " + newAdjPos + " for " + oldAdjBaseform);
String[] replDet = GermanSynthesizer.INSTANCE.synthesize(new AnalyzedToken(oldDetBaseform, null, oldDetBaseform), newDetPos, true);
String[] replAdj = GermanSynthesizer.INSTANCE.synthesize(new AnalyzedToken(oldAdjBaseform, null, oldAdjBaseform), newAdjPos, true);
//System.out.println("replDet: " + Arrays.toString(replDet));
//System.out.println("replAdj: " + Arrays.toString(replAdj));
for (String det : replDet) {
if (!det.startsWith(detToken.getToken().substring(0, 1))) {
continue; // mein, dein, sein etc. all share the same lemma ("mein"), but don't suggest "dein" for "mein"
for (String replGender : replGenders) {
String newDetPos = reading.getPOSTag().replaceAll("MAS|FEM|NEU", replGender).replaceFirst("BEG", "(BEG|B/S)");
String newAdjPos;
if (newDetPos.startsWith("ART:")) {
newAdjPos = reading.getPOSTag().replaceAll("MAS|FEM|NEU", replGender).replaceFirst("BEG", "(BEG|B/S)").replaceFirst(":STV", "");
} else if (newDetPos.startsWith("PRO:")) {
newAdjPos = newDetPos.replaceAll("PRO:POS:(NOM|AKK|GEN|DAT):(SIN|PLU):(MAS|FEM|NEU)", "ADJ:$1:$2:$3").replaceFirst(":(STV|BEG).*", ":GRU:IND");
} else {
throw new RuntimeException("Unexpected POS tag: " + newDetPos);
}
for (String adj : replAdj) {
String newDetAdj = det + " " + adj;
if (!result.contains(newDetAdj)) {
result.add(newDetAdj);
//System.out.println("newDetPos: " + newDetPos + " for " + oldDetBaseform);
//System.out.println("newAdjPos: " + newAdjPos + " for " + oldAdjBaseform);
String[] replDet = GermanSynthesizer.INSTANCE.synthesize(new AnalyzedToken(oldDetBaseform, null, oldDetBaseform), newDetPos, true);
String[] replAdj = GermanSynthesizer.INSTANCE.synthesize(new AnalyzedToken(oldAdjBaseform, null, oldAdjBaseform), newAdjPos, true);
//System.out.println("replDet: " + Arrays.toString(replDet));
//System.out.println("replAdj: " + Arrays.toString(replAdj));
for (String det : replDet) {
if (!det.startsWith(detToken.getToken().substring(0, 1))) {
continue; // mein, dein, sein etc. all share the same lemma ("mein"), but don't suggest "dein" for "mein"
}
for (String adj : replAdj) {
String newDetAdj = det + " " + adj;
if (!result.contains(newDetAdj)) {
result.add(newDetAdj);
}
}
}
}
Expand All @@ -178,20 +182,22 @@ private String getBaseform(AnalyzedTokenReadings token, String tagStartsWith) {
return baseform;
}

@Nullable
private String getNounGender(String word) throws IOException {
private List<String> getNounGender(String word) throws IOException {
List<AnalyzedTokenReadings> readings = GermanTagger.INSTANCE.tag(Collections.singletonList(word));
List<String> genders = new ArrayList<>();
for (AnalyzedTokenReadings atr : readings) {
if (atr.getReadings().size() > 0) {
String pos = atr.getReadings().get(0).getPOSTag();
for (AnalyzedToken reading : atr.getReadings()) {
String pos = reading.getPOSTag();
if (pos != null && pos.startsWith("SUB:")) {
String[] parts = pos.split(":");
if (parts.length >= 4) {
return parts[3];
if (!genders.contains(parts[3])) {
genders.add(parts[3]);
}
}
}
}
}
return null;
return genders;
}
}
2 changes: 2 additions & 0 deletions ...anguage-modules/de/src/test/java/org/languagetool/rules/de/AdaptSuggestionFilterTest.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -75,6 +75,8 @@ public void testAcceptRuleMatchWithDet() throws IOException {
runAcceptRuleMatch("Hier steht ihre Roadmap.", "Roadmap", "Verfahren", "[ihr Verfahren]");
runAcceptRuleMatch("Hier steht unsere Roadmap.", "Roadmap", "Verfahren", "[unser Verfahren]");
runAcceptRuleMatch("Hier steht eure Roadmap.", "Roadmap", "Verfahren", "[euer Verfahren]");

runAcceptRuleMatch("Ein Asylant.", "Asylant", "Asylberechtigter", "[Ein Asylberechtigter]");
}

@Ignore("WIP")
Expand Down