use of org.omegat.core.data.IProject.DefaultTranslationsIterator in project omegat by omegat-org.
the class FindMatches method search.
public List<NearString> search(final String searchText, final boolean requiresTranslation, final boolean fillSimilarityData, final IStopped stop) throws StoppedException {
result = new ArrayList<>(OConsts.MAX_NEAR_STRINGS + 1);
srcText = searchText;
removedText = "";
// of the translatable text
if (removePattern != null) {
StringBuilder removedBuffer = new StringBuilder();
Matcher removeMatcher = removePattern.matcher(srcText);
while (removeMatcher.find()) {
removedBuffer.append(removeMatcher.group());
}
srcText = removeMatcher.replaceAll("");
removedText = removedBuffer.toString();
}
// get tokens for original string
strTokensStem = tokenizeStem(srcText);
strTokensNoStem = tokenizeNoStem(srcText);
strTokensAll = tokenizeAll(srcText);
// travel by project entries, including orphaned
if (project.getProjectProperties().isSupportDefaultTranslations()) {
project.iterateByDefaultTranslations(new DefaultTranslationsIterator() {
public void iterate(String source, TMXEntry trans) {
checkStopped(stop);
if (!searchExactlyTheSame && source.equals(searchText)) {
// skip original==original entry comparison
return;
}
if (requiresTranslation && trans.translation == null) {
return;
}
String fileName = project.isOrphaned(source) ? ORPHANED_FILE_NAME : null;
processEntry(null, source, trans.translation, NearString.MATCH_SOURCE.MEMORY, false, 0, fileName, trans.creator, trans.creationDate, trans.changer, trans.changeDate, null);
}
});
}
project.iterateByMultipleTranslations(new MultipleTranslationsIterator() {
public void iterate(EntryKey source, TMXEntry trans) {
checkStopped(stop);
if (!searchExactlyTheSame && source.sourceText.equals(searchText)) {
// skip original==original entry comparison
return;
}
if (requiresTranslation && trans.translation == null) {
return;
}
String fileName = project.isOrphaned(source) ? ORPHANED_FILE_NAME : null;
processEntry(source, source.sourceText, trans.translation, NearString.MATCH_SOURCE.MEMORY, false, 0, fileName, trans.creator, trans.creationDate, trans.changer, trans.changeDate, null);
}
});
// travel by translation memories
for (Map.Entry<String, ExternalTMX> en : project.getTransMemories().entrySet()) {
int penalty = 0;
Matcher matcher = SEARCH_FOR_PENALTY.matcher(en.getKey());
if (matcher.find()) {
penalty = Integer.parseInt(matcher.group(1));
}
for (PrepareTMXEntry tmen : en.getValue().getEntries()) {
checkStopped(stop);
if (tmen.source == null) {
// Not all TMX entries have a source; in that case there can be no meaningful match, so skip.
continue;
}
if (requiresTranslation && tmen.translation == null) {
continue;
}
processEntry(null, tmen.source, tmen.translation, NearString.MATCH_SOURCE.TM, false, penalty, en.getKey(), tmen.creator, tmen.creationDate, tmen.changer, tmen.changeDate, tmen.otherProperties);
}
}
// travel by all entries for check source file translations
for (SourceTextEntry ste : project.getAllEntries()) {
checkStopped(stop);
if (ste.getSourceTranslation() != null) {
processEntry(ste.getKey(), ste.getSrcText(), ste.getSourceTranslation(), NearString.MATCH_SOURCE.MEMORY, ste.isSourceTranslationFuzzy(), 0, ste.getKey().file, "", 0, "", 0, null);
}
}
if (separateSegmentMatcher != null) {
// split paragraph even when segmentation disabled, then find matches for every segment
List<StringBuilder> spaces = new ArrayList<StringBuilder>();
List<Rule> brules = new ArrayList<Rule>();
Language sourceLang = project.getProjectProperties().getSourceLanguage();
Language targetLang = project.getProjectProperties().getTargetLanguage();
List<String> segments = Core.getSegmenter().segment(sourceLang, srcText, spaces, brules);
if (segments.size() > 1) {
List<String> fsrc = new ArrayList<String>(segments.size());
List<String> ftrans = new ArrayList<String>(segments.size());
// multiple segments
for (short i = 0; i < segments.size(); i++) {
String onesrc = segments.get(i);
// find match for separate segment
List<NearString> segmentMatch = separateSegmentMatcher.search(onesrc, requiresTranslation, false, stop);
if (!segmentMatch.isEmpty() && segmentMatch.get(0).scores[0].score >= SUBSEGMENT_MATCH_THRESHOLD) {
fsrc.add(segmentMatch.get(0).source);
ftrans.add(segmentMatch.get(0).translation);
} else {
fsrc.add("");
ftrans.add("");
}
}
// glue found sources
String foundSrc = Core.getSegmenter().glue(sourceLang, sourceLang, fsrc, spaces, brules);
// glue found translations
String foundTrans = Core.getSegmenter().glue(sourceLang, targetLang, ftrans, spaces, brules);
processEntry(null, foundSrc, foundTrans, NearString.MATCH_SOURCE.TM, false, 0, "", "", 0, "", 0, null);
}
}
if (fillSimilarityData) {
// fill similarity data only for result
for (NearString near : result) {
// fix for bug 1586397
byte[] similarityData = FuzzyMatcher.buildSimilarityData(strTokensAll, tokenizeAll(near.source));
near.attr = similarityData;
}
}
return result;
}
use of org.omegat.core.data.IProject.DefaultTranslationsIterator in project omegat by omegat-org.
the class TestTeamIntegrationChild method checkAll.
/**
* Check in memory and in file.
*/
static void checkAll() throws Exception {
ProjectTMX tmx = new ProjectTMX(TestTeamIntegration.SRC_LANG, TestTeamIntegration.TRG_LANG, false, new File(dir + "/omegat/project_save.tmx"), TestTeamIntegration.checkOrphanedCallback);
for (int c = 0; c < segCount; c++) {
checkTranslation(c);
checkTranslationFromFile(tmx, c);
}
Core.getProject().iterateByDefaultTranslations(new DefaultTranslationsIterator() {
public void iterate(String source, TMXEntry trans) {
Long prev = values.get(source);
if (prev == null) {
prev = 0L;
}
long curr = Long.parseLong(trans.translation);
if (curr < prev) {
throw new RuntimeException(source + ": Wrong value in " + source + ": current(" + curr + ") less than previous(" + prev + ")");
}
}
});
}
Aggregations