static final String INPUTS = "data/inputs/";
static final String STOPLISTS = "data/inputs/stoplists/";
static final String OUTPUTS = "data/outputs/";
+
+ // Build the non EN ones.
+ static final String[][] nonEnPairs = new String[][] {
+ {"EN"},
+ {"DE"},
+ {"IT"},
+ // This one takes a really long time, and the result is too big for code.google.com
+ //{"FR"},
+
+ // The 3 I use most:
+ {"IT", "EN" },
+ {"DE", "EN" },
+ {"DE", "IT" },
+
+ {"AR", "DE" },
+ {"AR", "ES" },
+ {"AR", "FR" },
+ {"AR", "HE" },
+ {"AR", "IT" },
+ {"AR", "JA" },
+ {"AR", "RU" },
+ {"AR", "TR" }, // Turkish
+ {"AR", "cmn" },
+
+ {"DE", "AR" },
+ {"DE", "FR" },
+ {"DE", "CA" }, // Catalan
+ {"DE", "CS" }, // Czech
+ {"DE", "EO" }, // Esperanto
+ {"DE", "ES" },
+ {"DE", "FR" },
+ {"DE", "HE" },
+ {"DE", "HU" }, // Hungarian
+ {"DE", "IT" },
+ {"DE", "JA" },
+ {"DE", "LA" }, // Latin
+ {"DE", "NL" }, // Dutch
+ {"DE", "PL" }, // Polish
+ {"DE", "RU" },
+ {"DE", "SV" }, // Swedish
+ {"DE", "TR" }, // Turkish
+ {"DE", "cmn" },
+ {"DE", "TA" }, // Tamil
+
+ {"ES", "RU" }, // Spanish-Russian
+
+ {"FR", "BG" }, // Bulgarian
+ {"FR", "CS" }, // Czech
+ {"FR", "DE" },
+ {"FR", "ES" },
+ {"FR", "IT" },
+ {"FR", "JA" },
+ {"FR", "LA" },
+ {"FR", "NL" }, // Dutch
+ {"FR", "RU" },
+ {"FR", "TR" }, // Turkish
+ {"FR", "cmn" },
+ {"FR", "EL" },
+
+ {"IT", "DE" },
+ {"IT", "EL" }, // Greek
+ {"IT", "ES" },
+ {"IT", "FR" },
+ {"IT", "HU" },
+ {"IT", "JA" },
+ {"IT", "LA" }, // Latin
+ {"IT", "LV" }, // Latvian
+ {"IT", "NL" },
+ {"IT", "PL" },
+ {"IT", "RU" },
+ {"IT", "SV" },
+ {"IT", "TR" }, // Turkish
+ {"IT", "cmn" },
+
+ {"JA", "cmn" },
+ {"JA", "AR" },
+ {"JA", "KO" },
+
+ {"cmn", "AR" },
+ {"cmn", "DE" },
+ {"cmn", "ES" },
+ {"cmn", "FR" },
+ {"cmn", "IT" },
+ {"cmn", "KO" },
+
+ {"NO", "SV" },
+ {"NO", "FI" },
+ {"FI", "SV" },
+
+ {"PL", "FR" }, // Polish
+ {"PL", "RU" }, // Polish
+ {"PL", "HU" }, // Polish
+ {"PL", "ES" }, // Polish
+
+ {"TR", "EL" }, // Turkish, Greek
+
+ {"FA", "HY" }, // Persian, Armenian, by request.
+ {"FA", "SV" }, // Persian, Swedish, by request.
+ {"NL", "PL" }, // Dutch, Polish, by request.
+
+ };
+
static final Map<String,String> isoToDedication = new LinkedHashMap<String, String>();
static {
- isoToDedication.put("AF", "Afrikaans dictionary dedicated to Heiko and Mariëtte Horn.");
- isoToDedication.put("HR", "Croatian dictionary dedicated to Ines Viskic and Miro Kresonja.");
- isoToDedication.put("NL", "Dutch dictionary dedicated to Mike LeBeau.");
- // German handled in file.
- isoToDedication.put("EL", "Greek dictionary dedicated to Noah Egge.");
- isoToDedication.put("IT", "Italian dictionary dedicated to Carolina Tropini, my favorite stardust in the whole universe! Ti amo!");
- isoToDedication.put("KO", "Korean dictionary dedicated to Ande Elwood--fall fashion und Fernsehturms!");
- isoToDedication.put("PT", "Portuguese dictionary dedicated to Carlos Melo, one Tough Mudder.");
- isoToDedication.put("RO", "Romanian dictionary dedicated to Radu Teodorescu.");
- isoToDedication.put("RU", "Russian dictionary dedicated to Maxim Aronin--best friend always!.");
- isoToDedication.put("SR", "Serbian dictionary dedicated to Filip Crnogorac--thanks for the honey.");
- isoToDedication.put("ES", "Spanish dictionary made especially for Carolina Tropini! <3 XoXoXXXXX!");
- isoToDedication.put("SV", "Swedish dictionary dedicated to Kajsa Palmblad--björn kramar!");
+ isoToDedication.put("AF", "Wiktionary-based Afrikaans dictionary dedicated to Heiko and Mariëtte Horn.");
+ isoToDedication.put("HR", "Wiktionary-based Croatian dictionary dedicated to Ines Viskic and Miro Kresonja.");
+ isoToDedication.put("NL", "Wiktionary-based Dutch dictionary dedicated to Mike LeBeau.");
+ isoToDedication.put("DE", "@data/inputs/de-en_dedication.txt");
+ isoToDedication.put("EL", "Wiktionary-based Greek dictionary dedicated to Noah Egge.");
+ isoToDedication.put("IT", "Wiktionary-based Italian dictionary dedicated to Carolina Tropini, my favorite stardust in the whole universe! Ti amo!");
+ isoToDedication.put("KO", "Wiktionary-based Korean dictionary dedicated to Ande Elwood--fall fashion und Fernsehturms!");
+ isoToDedication.put("PT", "Wiktionary-based Portuguese dictionary dedicated to Carlos Melo, one Tough Mudder.");
+ isoToDedication.put("RO", "Wiktionary-based Romanian dictionary dedicated to Radu Teodorescu.");
+ isoToDedication.put("RU", "Wiktionary-based Russian dictionary dedicated to Maxim Aronin--best friend always!.");
+ isoToDedication.put("SR", "Wiktionary-based Serbian dictionary dedicated to Filip Crnogorac--thanks for the honey!");
+ isoToDedication.put("ES", "Wiktionary-based Spanish dictionary made especially for Carolina Tropini! <3 XoXoXXXXX!");
+ isoToDedication.put("SV", "Wiktionary-based Swedish dictionary dedicated to Kajsa Palmblad--björn kramar!");
}
- private static String getDedication(String iso) {
- return isoToDedication.containsKey(iso) ? "\n\n" + isoToDedication.get(iso) : "";
+ private static String getEnDictionaryInfo(String iso) {
+ return isoToDedication.containsKey(iso) ? isoToDedication.get(iso) : String.format("Wiktionary-based %s dictionary.", iso);
}
static final Map<String,String> isoToStoplist = new LinkedHashMap<String, String>();
result.add(String.format("--input%dWiktionaryLang=%s", i, lang1));
result.add(String.format("--input%dSkipLang=%s", i, lang1));
result.add(String.format("--input%dWebUrlTemplate=http://%s.wiktionary.org/wiki/%%s", i, lang1.toLowerCase()));
- result.add(String.format("--input%dPageLimit=100", i));
+ //result.add(String.format("--input%dPageLimit=100", i));
++i;
} else {
System.err.println("Can't read file: " + wikiSplitFile);
}
- if (lang1.equals("EN")) {
+ if (lang1.equals("EN") && !lang1.equals("EN")) {
+ // Add a parser that tries to use the definitions. This is
+ // not very pretty yet.
result.add(String.format("--input%d=%s/wikiSplit/en/%s.data", i, INPUTS, lang1));
result.add(String.format("--input%dName=ENWiktionary.%s", i, lang1)) ;
result.add(String.format("--input%dFormat=enwiktionary", i));
result.add(String.format("--input%dLangPattern=%s", i, "English"));
result.add(String.format("--input%dLangCodePattern=%s", i, lang1.toLowerCase()));
result.add(String.format("--input%dEnIndex=%d", i, 1));
- result.add(String.format("--input%dPageLimit=100", i));
+ //result.add(String.format("--input%dPageLimit=100", i));
++i;
}
final String foreignIso = getOtherLang(pair, wikitionaryLang);
final String wikiSplitFile = String.format("%s/wikiSplit/%s/%s.data", INPUTS, wikitionaryLang.toLowerCase(), foreignIso);
if (!new File(wikiSplitFile).canRead()) {
- System.err.println("Can't read file: " + wikiSplitFile);
+ System.err.println("WARNING: Can't read file: " + wikiSplitFile);
continue;
}
result.add(String.format("--input%d=%s", i, wikiSplitFile));
// Deal with the pairs where one is English.
if (Arrays.asList(pair).contains("EN")) {
final String foreignIso = getOtherLang(pair, "EN");
-
String foreignRegex = WiktionaryLangs.isoCodeToEnWikiName.get(foreignIso);
- if (foreignIso.equals("ZH")) {
- // HACK: The missing "e" prevents a full match, causing "Cantonese" to be appended to the entries.
- foreignRegex = "Chinese|Mandarin|Cantones";
- }
result.add(String.format("--lang1=%s", lang1));
result.add(String.format("--lang2=%s", lang2));
- result.add(String.format("--dictInfo=(EN)Wikitionary-based EN-%s dictionary.%s", foreignIso, getDedication(foreignIso)));
+ result.add(String.format("--dictInfo=%s", getEnDictionaryInfo(foreignIso)));
+ // Foreign section.
result.add(String.format("--input%d=%s/wikiSplit/en/%s.data", i, INPUTS, foreignIso));
result.add(String.format("--input%dName=ENWiktionary.%s", i, foreignIso)) ;
result.add(String.format("--input%dFormat=enwiktionary", i));
result.add(String.format("--input%dEnIndex=%d", i, Arrays.asList(pair).indexOf("EN") + 1));
++i;
+ // Translation section.
result.add(String.format("--input%d=%swikiSplit/en/EN.data", i, INPUTS));
result.add(String.format("--input%dName=enwiktionary.english", i));
result.add(String.format("--input%dFormat=enwiktionary", i));
result.add(String.format("--input%dLangPattern1=%s", i, lang1));
result.add(String.format("--input%dLangPattern2=%s", i, lang2));
++i;
+
+ // TODO: Could use FR translation section here too.
}
return result;
final List<String[]> allPairs = new ArrayList<String[]>();
- // Build the non EN ones.
- final String[][] nonEnPairs = new String[][] {
- {"EN"},
- {"DE"},
- {"IT"},
- {"FR"},
-
- // The 3 I use most:
- {"IT", "EN" },
- {"DE", "EN" },
- {"DE", "IT" },
-
-
- {"AR", "DE" },
- {"AR", "ES" },
- {"AR", "FR" },
- {"AR", "HE" },
- {"AR", "IT" },
- {"AR", "JA" },
- {"AR", "RU" },
- {"AR", "TR" }, // Turkish
- {"AR", "ZH" },
-
- {"DE", "AR" },
- {"DE", "FR" },
- {"DE", "CA" }, // Catalan
- {"DE", "CS" }, // Czech
- {"DE", "EO" }, // Esperanto
- {"DE", "ES" },
- {"DE", "FR" },
- {"DE", "HE" },
- {"DE", "HU" }, // Hungarian
- {"DE", "IT" },
- {"DE", "JA" },
- {"DE", "LA" }, // Latin
- {"DE", "NL" }, // Dutch
- {"DE", "PL" }, // Polish
- {"DE", "RU" },
- {"DE", "SV" }, // Swedish
- {"DE", "TR" }, // Turkish
- {"DE", "ZH" },
- {"DE", "TA" }, // Tamil
-
- {"ES", "RU" }, // Spanish-Russian
-
- {"FR", "BG" }, // Bulgarian
- {"FR", "CS" }, // Czech
- {"FR", "DE" },
- {"FR", "ES" },
- {"FR", "IT" },
- {"FR", "JA" },
- {"FR", "LA" },
- {"FR", "NL" }, // Dutch
- {"FR", "RU" },
- {"FR", "TR" }, // Turkish
- {"FR", "ZH" },
-
- {"IT", "DE" },
- {"IT", "EL" }, // Greek
- {"IT", "ES" },
- {"IT", "FR" },
- {"IT", "HU" },
- {"IT", "JA" },
- {"IT", "LA" }, // Latin
- {"IT", "LV" }, // Latvian
- {"IT", "NL" },
- {"IT", "PL" },
- {"IT", "RU" },
- {"IT", "SV" },
- {"IT", "TR" }, // Turkish
- {"IT", "ZH" },
-
- {"JA", "ZH" },
- {"JA", "AR" },
- {"JA", "KO" },
-
- {"ZH", "AR" },
- {"ZH", "DE" },
- {"ZH", "ES" },
- {"ZH", "FR" },
- {"ZH", "IT" },
- {"ZH", "KO" },
-
-
- {"NO", "SV" },
- {"NO", "FI" },
- {"FI", "SV" },
-
- {"PL", "FR" }, // Polish
- {"PL", "RU" }, // Polish
- {"PL", "HU" }, // Polish
- {"PL", "ES" }, // Polish
-
- {"TR", "EL" }, // Turkish, Greek
-
- {"FA", "HY" }, // Persian, Armenian, by request.
- {"FA", "SV" }, // Persian, Swedish, by request.
-
- };
allPairs.addAll(Arrays.asList(nonEnPairs));
-
// Add all the EN-XX pairs.
for (final String isoCode : WiktionaryLangs.isoCodeToEnWikiName.keySet()) {
- if (isoCode.equals("EN") || isoCode.equals("DE")) {
- continue;
+ if (!isoCode.equals("EN")) {
+ allPairs.add(new String[] {"EN", isoCode});
}
- allPairs.add(new String[] {"EN", isoCode});
}
- allPairs.add(new String[] {"EN", "DE"});
final Set<List<String>> done = new LinkedHashSet<List<String>>();
+ boolean go = true;
for (final String[] pair : allPairs) {
Arrays.sort(pair);
final List<String> pairList = Arrays.asList(pair);
}
done.add(pairList);
- if (!pairList.contains("IT") || !pairList.contains("EN")) {
- //continue;
+ if (pairList.contains("EN") && pairList.contains("DE")) {
+ go = true;
+ } else {
+ go = false;
+ }
+
+ if (!go) {
+ continue;
}
DictionaryBuilder.main(getMainArgs(pair).toArray(new String[0]));