-
- static final String INPUTS = "data/inputs/";
- static final String STOPLISTS = "data/inputs/stoplists/";
- static final String OUTPUTS = "data/outputs/";
-
- public static void main(final String[] args) throws Exception {
-
- // Builds all the dictionaries it can, outputs list to a text file.
-
- final Map<String,String> isoToWikiName = new LinkedHashMap<String, String>(EnWiktionaryLangs.isoCodeToWikiName);
- isoToWikiName.remove("EN");
- isoToWikiName.remove("DE");
-
- final Map<String,String> isoToDedication = new LinkedHashMap<String, String>();
- 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!");
-
- final Map<String,String> isoToStoplist = new LinkedHashMap<String, String>();
- isoToStoplist.put("DE", "de.txt");
- isoToStoplist.put("EN", "en.txt");
- isoToStoplist.put("ES", "es.txt");
- isoToStoplist.put("IT", "it.txt");
- isoToStoplist.put("FR", "fr.txt");
-
- final Map<String,String> isoToRegex = new LinkedHashMap<String, String>();
- // HACK: The missing "e" prevents a full match, causing "Cantonese" to be appended to the entries.
- isoToRegex.put("ZH", "Chinese|Mandarin|Cantones");
-
-// isoToWikiName.keySet().retainAll(Arrays.asList("UK", "HR", "FI"));
- //isoToWikiName.clear();
- boolean go = false;
- for (final String foreignIso : isoToWikiName.keySet()) {
- if (foreignIso.equals("blah")) {
- go = true;
- }
- if (!go) {
- continue;
- }
-
- final String dictFile = String.format("%s/EN-%s_enwiktionary.quickdic", OUTPUTS, foreignIso);
- System.out.println("building dictFile: " + dictFile);
-
- if (!isoToStoplist.containsKey(foreignIso)) {
- isoToStoplist.put(foreignIso, "empty.txt");
+
+ 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", "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 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>();
+ static {
+ isoToStoplist.put("DE", "de.txt");
+ isoToStoplist.put("EN", "en.txt");
+ isoToStoplist.put("ES", "es.txt");
+ isoToStoplist.put("IT", "it.txt");
+ isoToStoplist.put("FR", "fr.txt");
+ }
+ private static String getStoplist(String iso) {
+ return isoToStoplist.containsKey(iso) ? isoToStoplist.get(iso) : "empty.txt";
+ }
+
+ static String getOtherLang(final String[] pair, final String first) {
+ assert Arrays.asList(pair).contains(first);
+ assert pair.length == 2;
+ return pair[0].equals(first) ? pair[1] : pair[0];
+ }
+
+ static List<String> getMainArgs(final String[] pair) {
+ final List<String> result = new ArrayList<String>();
+
+ int i = 1;
+
+ if (pair.length == 1) {
+ final String lang1 = pair[0];
+ final String dictFile = String.format("%s/%s.quickdic", OUTPUTS, lang1);
+ result.add(String.format("--dictOut=%s", dictFile));
+ result.add(String.format("--lang1=%s", lang1));
+ result.add(String.format("--lang1Stoplist=%s", STOPLISTS + getStoplist(lang1)));
+ result.add(String.format("--dictInfo=Wikitionary-based %s dictionary.", lang1));
+
+
+ final String wikiSplitFile = String.format("%s/wikiSplit/%s/%s.data", INPUTS, lang1.toLowerCase(), lang1);
+ if (new File(wikiSplitFile).canRead()) {
+ result.add(String.format("--input%d=%s", i, wikiSplitFile));
+ result.add(String.format("--input%dName=%s.wiktionary.org", i, lang1.toLowerCase()));
+ result.add(String.format("--input%dFormat=%s", i, WholeSectionToHtmlParser.NAME));
+ result.add(String.format("--input%dTitleIndex=%d", i, 1));
+ 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));
+ ++i;
+ } else {
+ System.err.println("Can't read file: " + wikiSplitFile);
+ }
+
+ 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%dWiktionaryType=EnEnglish", 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));
+ ++i;
+ }
+
+ return result;
+ } // Single-lang dictionaries.
+
+ final String lang1 = pair[0];
+ final String lang2 = pair[1];
+
+ final String dictFile = String.format("%s/%s-%s.quickdic",
+ OUTPUTS, lang1, lang2);
+
+ result.add(String.format("--dictOut=%s", dictFile));
+ result.add(String.format("--lang1Stoplist=%s", STOPLISTS + getStoplist(lang1)));
+ result.add(String.format("--lang2Stoplist=%s", STOPLISTS + getStoplist(lang2)));
+
+ // For a few langs, put the defs of the other language in DE/IT/FR using WholeSection.
+ for (final String wikitionaryLang : Arrays.asList("EN", "DE", "IT", "FR")) {
+ if (!Arrays.asList(pair).contains(wikitionaryLang)) {
+ continue;
+ }
+ 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("WARNING: Can't read file: " + wikiSplitFile);
+ continue;
+ }
+ result.add(String.format("--input%d=%s", i, wikiSplitFile));
+ result.add(String.format("--input%dName=%s.wiktionary.org", i, wikitionaryLang.toLowerCase()));
+ result.add(String.format("--input%dFormat=%s", i, WholeSectionToHtmlParser.NAME));
+ result.add(String.format("--input%dTitleIndex=%d", i, Arrays.asList(pair).indexOf(foreignIso) + 1));
+ result.add(String.format("--input%dWiktionaryLang=%s", i, wikitionaryLang));
+ result.add(String.format("--input%dSkipLang=%s", i, foreignIso));
+ result.add(String.format("--input%dWebUrlTemplate=http://%s.wiktionary.org/wiki/%%s", i, wikitionaryLang.toLowerCase()));
+ ++i;