moparisthebest/filebot
1
0
Fork 0
mirror of https://github.com/mitb-archive/filebot synced 2024-11-17 23:05:03 -05:00
Code Issues Releases Wiki Activity

* improved matchMovieFromStringWithoutSpacing method

Browse Source
This commit is contained in:
Reinhard Pointner 2012-07-08 09:47:13 +00:00
parent 34b01c63dc
commit 85d736bf75
1 changed files with 22 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

30
source/net/sourceforge/filebot/media/MediaDetection.java
View File

@ -23,6 +23,7 @@ import java.util.HashMap;
import java.util.HashSet; import java.util.HashSet;
import java.util.LinkedHashMap; import java.util.LinkedHashMap;
import java.util.LinkedHashSet; import java.util.LinkedHashSet;
import java.util.LinkedList;
import java.util.List; import java.util.List;
import java.util.Locale; import java.util.Locale;
import java.util.Map; import java.util.Map;
@ -334,7 +335,11 @@ public class MediaDetection {
// assume name without spacing will mess up any lookup // assume name without spacing will mess up any lookup
if (movieNameMatches.isEmpty()) { if (movieNameMatches.isEmpty()) {
movieNameMatches = matchMovieFromStringWithoutSpacing(terms, new NameSimilarityMetric(), strict ? 0.9f : 0.5f); movieNameMatches = matchMovieFromStringWithoutSpacing(terms, strict);
if (movieNameMatches.isEmpty() && !terms.equals(stripReleaseInfo(terms, true))) {
movieNameMatches = matchMovieFromStringWithoutSpacing(stripReleaseInfo(terms, true), strict);
}
} }
// query by file / folder name // query by file / folder name
@ -443,29 +448,38 @@ public class MediaDetection {
} }
public static List<Movie> matchMovieFromStringWithoutSpacing(List<String> files, SimilarityMetric metric, float similarityThreshold) throws IOException { public static List<Movie> matchMovieFromStringWithoutSpacing(List<String> names, boolean strict) throws IOException {
Pattern spacing = Pattern.compile("[\\p{Punct}\\p{Space}]+"); Pattern spacing = Pattern.compile("[\\p{Punct}\\p{Space}]+");
List<String> terms = new ArrayList<String>(files.size()); List<String> terms = new ArrayList<String>(names.size());
for (String it : files) { for (String it : names) {
String term = spacing.matcher(it).replaceAll("").toLowerCase(); String term = spacing.matcher(it).replaceAll("").toLowerCase();
if (term.length() >= 3) { if (term.length() >= 3) {
terms.add(term); // only consider words, not just random letters terms.add(term); // only consider words, not just random letters
} }
} }
List<Movie> movies = new ArrayList<Movie>(); // similarity threshold based on strict/non-strict
SimilarityMetric metric = new NameSimilarityMetric();
float similarityThreshold = strict ? 0.9f : 0.5f;
LinkedList<Movie> movies = new LinkedList<Movie>();
for (Entry<String, Movie> it : getMovieIndex()) { for (Entry<String, Movie> it : getMovieIndex()) {
String name = spacing.matcher(it.getKey()).replaceAll("").toLowerCase(); String name = spacing.matcher(it.getKey()).replaceAll("").toLowerCase();
for (String term : terms) { for (String term : terms) {
if (term.contains(name) && metric.getSimilarity(name, term) >= similarityThreshold) { if (term.contains(name)) {
movies.add(it.getValue()); String year = String.valueOf(it.getValue().getYear());
if (term.contains(year) && metric.getSimilarity(term, name + year) > similarityThreshold) {
movies.addFirst(it.getValue());
} else if (metric.getSimilarity(term, name) > similarityThreshold) {
movies.addLast(it.getValue());
}
break; break;
} }
} }
} }
return movies; return new ArrayList<Movie>(movies);
} }