mobeigi
/
filebot
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

* make sure substring metric only matches word sequences, rather than simple substrings which may match half of a word which wouldn't never make sense and could only cause issues

This commit is contained in:
Reinhard Pointner 2014-03-24 20:32:27 +00:00
parent f61f30e862
commit 1ca8de3ab7
1 changed files with 16 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

17
source/net/sourceforge/filebot/similarity/SubstringMetric.java
View File

@ -26,7 +26,22 @@ public class SubstringMetric implements SimilarityMetric {
if (s2 == null || s2.isEmpty()) if (s2 == null || s2.isEmpty())
return 0; return 0;
return (o1c2 && s1.contains(s2)) || (o2c1 && s2.contains(s1)) ? 1 : 0; return (o1c2 && matches(s1, s2) || (o2c1 && matches(s2, s1))) ? 1 : 0;
}
protected boolean matches(String s1, String s2) {
int index = s1.lastIndexOf(s2);
if (index < 0)
return false;
// check before and after and make sure we're only matching between word boundries
if (index - 1 >= 0 && !Character.isLetterOrDigit(s1.charAt(index - 1)))
return false;
if (index + s2.length() < s1.length() && !Character.isLetterOrDigit(index + s2.length()))
return false;
return true;
} }
protected String normalize(Object object) { protected String normalize(Object object) {