moparisthebest/filebot
1
0
Fork 0
mirror of https://github.com/mitb-archive/filebot synced 2024-12-24 08:48:51 -05:00
Code Issues Releases Wiki Activity

* detect locale from language suffix first, and then lookup via osdb language detection

Browse Source
This commit is contained in:
Reinhard Pointner 2012-01-02 16:59:37 +00:00
parent 55ada2f210
commit 4341922677
2 changed files with 36 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
source/net/sourceforge/filebot/format/MediaBindingBean.java
View File

@ -288,6 +288,13 @@ public class MediaBindingBean {
@Define("lang") @Define("lang")
public Locale detectSubtitleLanguage() throws Exception { public Locale detectSubtitleLanguage() throws Exception {
// make sure media file is defined
checkMediaFile();
Locale languageSuffix = new ReleaseInfo().getLanguageSuffix(FileUtilities.getName(mediaFile));
if (languageSuffix != null)
return languageSuffix;
// require subtitle file // require subtitle file
if (!SUBTITLE_FILES.accept(mediaFile)) if (!SUBTITLE_FILES.accept(mediaFile))
return null; return null;

41
source/net/sourceforge/filebot/media/ReleaseInfo.java
View File

@ -2,6 +2,7 @@
package net.sourceforge.filebot.media; package net.sourceforge.filebot.media;
import static java.util.Arrays.*;
import static java.util.ResourceBundle.*; import static java.util.ResourceBundle.*;
import static java.util.regex.Pattern.*; import static java.util.regex.Pattern.*;
import static net.sourceforge.filebot.similarity.Normalization.*; import static net.sourceforge.filebot.similarity.Normalization.*;
@ -12,13 +13,12 @@ import java.io.IOException;
import java.nio.ByteBuffer; import java.nio.ByteBuffer;
import java.nio.charset.Charset; import java.nio.charset.Charset;
import java.util.ArrayList; import java.util.ArrayList;
import java.util.Arrays;
import java.util.HashSet; import java.util.HashSet;
import java.util.List; import java.util.List;
import java.util.Locale; import java.util.Locale;
import java.util.Map;
import java.util.Scanner; import java.util.Scanner;
import java.util.Set; import java.util.TreeMap;
import java.util.TreeSet;
import java.util.regex.Matcher; import java.util.regex.Matcher;
import java.util.regex.Pattern; import java.util.regex.Pattern;
import java.util.zip.GZIPInputStream; import java.util.zip.GZIPInputStream;
@ -42,6 +42,16 @@ public class ReleaseInfo {
} }
public Locale getLanguageSuffix(String name) {
// match locale identifier and lookup Locale object
String lang = matchLast(getLanguageSuffixPattern(), null, name);
if (lang == null)
return null;
return getLanguageMap(Locale.ENGLISH, Locale.getDefault()).get(lang);
}
protected String matchLast(Pattern pattern, String[] standardValues, CharSequence... sequence) { protected String matchLast(Pattern pattern, String[] standardValues, CharSequence... sequence) {
String lastMatch = null; String lastMatch = null;
@ -57,7 +67,7 @@ public class ReleaseInfo {
} }
// prefer standard value over matched value // prefer standard value over matched value
if (lastMatch != null) { if (lastMatch != null && standardValues != null) {
for (String standard : standardValues) { for (String standard : standardValues) {
if (standard.equalsIgnoreCase(lastMatch)) { if (standard.equalsIgnoreCase(lastMatch)) {
return standard; return standard;
@ -102,22 +112,29 @@ public class ReleaseInfo {
public Pattern getLanguageSuffixPattern() { public Pattern getLanguageSuffixPattern() {
Set<String> tokens = new TreeSet<String>(); // .{language}[.srt]
return compile("(?<=\\p{Punct}|\\s)(" + join(getLanguageMap(Locale.ENGLISH, Locale.getDefault()).keySet(), "|") + ")(?=$)", CASE_INSENSITIVE | UNICODE_CASE | CANON_EQ);
}
public Map<String, Locale> getLanguageMap(Locale... supportedLanguageName) {
Map<String, Locale> languageMap = new TreeMap<String, Locale>(String.CASE_INSENSITIVE_ORDER);
for (String code : Locale.getISOLanguages()) { for (String code : Locale.getISOLanguages()) {
Locale locale = new Locale(code); Locale locale = new Locale(code);
tokens.add(locale.getLanguage()); languageMap.put(locale.getLanguage(), locale);
tokens.add(locale.getISO3Language()); languageMap.put(locale.getISO3Language(), locale);
for (Locale language : new HashSet<Locale>(Arrays.asList(Locale.ENGLISH, Locale.getDefault()))) {
tokens.add(locale.getDisplayLanguage(language)); // map display language names for given locales
for (Locale language : new HashSet<Locale>(asList(supportedLanguageName))) {
languageMap.put(locale.getDisplayLanguage(language), locale);
} }
} }
// remove illegal tokens // remove illegal tokens
tokens.remove(""); languageMap.remove("");
// .{language}[.srt] return languageMap;
return compile("(?<=\\p{Punct})(" + join(tokens, "|") + ")(?=$)", CASE_INSENSITIVE | UNICODE_CASE | CANON_EQ);
} }