* fix anime matching

source/net/sourceforge/filebot/Main.java

@@ -409,15 +409,15 @@ public class Main {
 					MediaDetection.getClutterFileFilter();
 					MediaDetection.getDiskFolderFilter();
 
+					Collection<File> empty = Collections.emptyList();
+					MediaDetection.matchSeriesByDirectMapping(empty);
+
 					// pre-load movie/series index
 					List<String> dummy = Collections.singletonList("");
 					MediaDetection.stripReleaseInfo(dummy, true);
-					MediaDetection.matchSeriesByName(dummy, -1);
+					MediaDetection.matchSeriesByName(dummy, -1, MediaDetection.getSeriesIndex());
+					MediaDetection.matchSeriesByName(dummy, -1, MediaDetection.getAnimeIndex());
 					MediaDetection.matchMovieName(dummy, true, -1);
-
-					Collection<File> empty = Collections.emptyList();
-					MediaDetection.matchSeriesByDirectMapping(empty);
-					WebServices.TheTVDB.getLocalIndex();
 				} catch (Exception e) {
 					Logger.getLogger(getClass().getName()).log(Level.WARNING, e.getMessage(), e);
 				}

source/net/sourceforge/filebot/media/MediaDetection.java

@@ -280,7 +280,7 @@ public class MediaDetection {
 		return detectSeriesNames(files, locale, index);
 	}
 
-	public static List<String> detectSeriesNames(Collection<File> files, Locale locale, List<IndexEntry<SearchResult>> seriesIndex) throws Exception {
+	public static List<String> detectSeriesNames(Collection<File> files, Locale locale, List<IndexEntry<SearchResult>> index) throws Exception {
 		List<String> names = new ArrayList<String>();
 
 		// try xattr metadata if enabled
@@ -327,12 +327,12 @@ public class MediaDetection {
 			}
 
 			// check foldernames first
-			List<String> matches = matchSeriesByName(folders, 0);
+			List<String> matches = matchSeriesByName(folders, 0, index);
 
 			// check all filenames if necessary
 			if (matches.isEmpty()) {
-				matches.addAll(matchSeriesByName(filenames, 0));
+				matches.addAll(matchSeriesByName(filenames, 0, index));
-				matches.addAll(matchSeriesByName(stripReleaseInfo(filenames, false), 0));
+				matches.addAll(matchSeriesByName(stripReleaseInfo(filenames, false), 0, index));
 			}
 
 			// use lenient sub sequence matching only as fallback and try name without spacing logic that may mess up any lookup
@@ -348,13 +348,13 @@ public class MediaDetection {
 						sns.set(i, sn);
 					}
 				}
-				for (SearchResult it : matchSeriesFromStringWithoutSpacing(stripReleaseInfo(sns, false), true)) {
+				for (SearchResult it : matchSeriesFromStringWithoutSpacing(stripReleaseInfo(sns, false), true, index)) {
 					matches.add(it.getName());
 				}
 
 				// less reliable CWS deep matching
-				matches.addAll(matchSeriesByName(folders, 2));
+				matches.addAll(matchSeriesByName(folders, 2, index));
-				matches.addAll(matchSeriesByName(filenames, 2));
+				matches.addAll(matchSeriesByName(filenames, 2, index));
 
 				// pass along only valid terms
 				names.addAll(stripBlacklistedTerms(matches));
@@ -454,7 +454,7 @@ public class MediaDetection {
 		}
 	}
 
-	public static List<String> matchSeriesByName(Collection<String> files, int maxStartIndex) throws Exception {
+	public static List<String> matchSeriesByName(Collection<String> files, int maxStartIndex, List<IndexEntry<SearchResult>> index) throws Exception {
 		HighPerformanceMatcher nameMatcher = new HighPerformanceMatcher(maxStartIndex);
 		List<String> matches = new ArrayList<String>();
 
@@ -462,7 +462,7 @@ public class MediaDetection {
 
 		for (CollationKey[] name : names) {
 			IndexEntry<SearchResult> bestMatch = null;
-			for (IndexEntry<SearchResult> it : getSeriesIndex()) {
+			for (IndexEntry<SearchResult> it : index) {
 				CollationKey[] commonName = nameMatcher.matchFirstCommonSequence(name, it.getLenientKey());
 				if (commonName != null && commonName.length >= it.getLenientKey().length && (bestMatch == null || commonName.length > bestMatch.getLenientKey().length)) {
 					bestMatch = it;
@@ -485,7 +485,7 @@ public class MediaDetection {
 		return matches;
 	}
 
-	public static List<SearchResult> matchSeriesFromStringWithoutSpacing(Collection<String> names, boolean strict) throws IOException {
+	public static List<SearchResult> matchSeriesFromStringWithoutSpacing(Collection<String> names, boolean strict, List<IndexEntry<SearchResult>> index) throws IOException {
 		// clear name of punctuation, spacing, and leading 'The' or 'A' that are common causes for word-lookup to fail
 		Pattern spacing = Pattern.compile("(^(?i)(The|A)\\b)|[\\p{Punct}\\p{Space}]+");
 
@@ -502,7 +502,7 @@ public class MediaDetection {
 		float similarityThreshold = strict ? 0.75f : 0.5f;
 
 		List<SearchResult> seriesList = new ArrayList<SearchResult>();
-		for (IndexEntry<SearchResult> it : getSeriesIndex()) {
+		for (IndexEntry<SearchResult> it : index) {
 			String name = spacing.matcher(it.getLenientName()).replaceAll("").toLowerCase();
 			for (String term : terms) {
 				if (term.contains(name)) {

source/net/sourceforge/filebot/web/AnidbClient.java

@@ -180,6 +180,7 @@ public class AnidbClient extends AbstractEpisodeListProvider {
 		typeOrder.add("1");
 		typeOrder.add("4");
 		typeOrder.add("2");
+		typeOrder.add("3");
 
 		// fetch data
 		Map<Integer, List<Object[]>> entriesByAnime = new HashMap<Integer, List<Object[]>>(65536);
@@ -196,15 +197,18 @@ public class AnidbClient extends AbstractEpisodeListProvider {
 					String title = matcher.group(4);
 
 					if (aid > 0 && title.length() > 0 && typeOrder.contains(type) && languageOrder.contains(language)) {
+						// resolve HTML entities
+						title = Jsoup.parse(title).text();
+
+						if (type.equals("3") && (title.length() < 5 || !Character.isUpperCase(title.charAt(0)) || Character.isUpperCase(title.charAt(title.length() - 1)))) {
+							continue;
+						}
+
 						List<Object[]> names = entriesByAnime.get(aid);
 						if (names == null) {
 							names = new ArrayList<Object[]>();
 							entriesByAnime.put(aid, names);
 						}
-
-						// resolve HTML entities
-						title = Jsoup.parse(title).text();
-
 						names.add(new Object[] { typeOrder.indexOf(type), languageOrder.indexOf(language), title });
 					}
 				}

website/data/query-blacklist.txt

@@ -111,6 +111,7 @@
 ^test$
 ^testdata$
 ^tmp$
+^to$
 ^Torrent$
 ^Torrents$
 ^Tracker$