+ if (wikiTokenizer.isFunction()) {
+ final String name = wikiTokenizer.functionName();
+ final List<String> args = wikiTokenizer.functionPositionArgs();
+ final Map<String,String> namedArgs = wikiTokenizer.functionNamedArgs();
+ // First line is generally a repeat of the title with some extra information.
+ // We need to build up the left side (foreign text, tokens) separately from the
+ // right side (English). The left-side may get paired with multiple right sides.
+ // The left side should get filed under every form of the word in question (singular, plural).
+
+ // For verbs, the conjugation comes later on in a deeper section.
+ // Ideally, we'd want to file every English entry with the verb
+ // under every verb form coming from the conjugation.
+ // Ie. under "fa": see: "make :: fare" and "do :: fare"
+ // But then where should we put the conjugation table?
+ // I think just under fare. But then we need a way to link to the entry (actually the row, since entries doesn't show up!)
+ // for the conjugation table from "fa".
+ // Would like to be able to link to a lang#token.
+ if (name.equals("it-noun")) {
+ assert forms.isEmpty();
+ final String base = get(args, 0);
+ final String gender = get(args, 1);
+ final String singular = base + get(args, 2);
+ final String plural = base + get(args, 3);
+ side = String.format("%s {%s}, %s {pl}", singular, gender, plural, plural);
+ forms = Arrays.asList(singular, plural);
+ } else if (name.equals("it-proper noun")) {
+ // TODO
+ } else if (name.equals("it-adj")) {
+ // TODO
+ } else if (name.startsWith("it-conj")) {
+ if (name.equals("it-conj-are")) {
+ itConjAre(args, namedArgs);
+ } else if (name.equals("it-conj-ere")) {
+ } else if (name.equals("it-conj-ire")) {
+ } else {
+ System.err.println("Unknown conjugation: " + wikiTokenizer.token());
+ }
+
+ } else {
+ System.err.println("Unknown function: " + wikiTokenizer.token());
+ }
+
+ } else if (wikiTokenizer.isListItem()) {
+ handleForeignListItem(side != null ? side : title, title, forms, wikiTokenizer);
+
+ } else if (wikiTokenizer.isWikiLink()) {
+
+ } else {
+ }
+
+ }
+ }
+
+ private void itConjAre(List<String> args, Map<String, String> namedArgs) {
+ final String base = args.get(0);
+ final String aux = args.get(1);
+
+ putIfMissing(namedArgs, "inf", base + "are");
+ putIfMissing(namedArgs, "aux", aux);
+ putIfMissing(namedArgs, "ger", base + "ando");
+ putIfMissing(namedArgs, "presp", base + "ante");
+ putIfMissing(namedArgs, "pastp", base + "ato");
+ // Present
+ putIfMissing(namedArgs, "pres1s", base + "o");
+ putIfMissing(namedArgs, "pres2s", base + "i");
+ putIfMissing(namedArgs, "pres3s", base + "a");
+ putIfMissing(namedArgs, "pres1p", base + "iamo");
+ putIfMissing(namedArgs, "pres2p", base + "ate");
+ putIfMissing(namedArgs, "pres3p", base + "ano");
+ // Imperfect
+ putIfMissing(namedArgs, "imperf1s", base + "avo");
+ putIfMissing(namedArgs, "imperf2s", base + "avi");
+ putIfMissing(namedArgs, "imperf3s", base + "ava");
+ putIfMissing(namedArgs, "imperf1p", base + "avamo");
+ putIfMissing(namedArgs, "imperf2p", base + "avate");
+ putIfMissing(namedArgs, "imperf3p", base + "avano");
+ // Passato remoto
+ putIfMissing(namedArgs, "prem1s", base + "ai");
+ putIfMissing(namedArgs, "prem2s", base + "asti");
+ putIfMissing(namedArgs, "prem3s", base + "ò");
+ putIfMissing(namedArgs, "prem1p", base + "ammo");
+ putIfMissing(namedArgs, "prem2p", base + "aste");
+ putIfMissing(namedArgs, "prem3p", base + "arono");
+ // Future
+ putIfMissing(namedArgs, "fut1s", base + "erò");
+ putIfMissing(namedArgs, "fut2s", base + "erai");
+ putIfMissing(namedArgs, "fut3s", base + "erà");
+ putIfMissing(namedArgs, "fut1p", base + "eremo");
+ putIfMissing(namedArgs, "fut2p", base + "erete");
+ putIfMissing(namedArgs, "fut3p", base + "eranno");
+ // Conditional
+ putIfMissing(namedArgs, "cond1s", base + "erei");
+ putIfMissing(namedArgs, "cond2s", base + "eresti");
+ putIfMissing(namedArgs, "cond3s", base + "erebbe");
+ putIfMissing(namedArgs, "cond1p", base + "eremmo");
+ putIfMissing(namedArgs, "cond2p", base + "ereste");
+ putIfMissing(namedArgs, "cond3p", base + "erebbero");
+ // Subjunctive / congiuntivo
+ putIfMissing(namedArgs, "sub123s", base + "i");
+ putIfMissing(namedArgs, "sub1p", base + "iamo");
+ putIfMissing(namedArgs, "sub2p", base + "iate");
+ putIfMissing(namedArgs, "sub3p", base + "ino");
+ // Imperfect subjunctive
+ putIfMissing(namedArgs, "impsub12s", base + "assi");
+ putIfMissing(namedArgs, "impsub3s", base + "asse");
+ putIfMissing(namedArgs, "impsub1p", base + "assimo");
+ putIfMissing(namedArgs, "impsub2p", base + "aste");
+ putIfMissing(namedArgs, "impsub3p", base + "assero");
+ // Imperative
+ putIfMissing(namedArgs, "imp2s", base + "a");
+ putIfMissing(namedArgs, "imp3s", base + "i");
+ putIfMissing(namedArgs, "imp1p", base + "iamo");
+ putIfMissing(namedArgs, "imp2p", base + "ate");
+ putIfMissing(namedArgs, "imp3p", base + "ino");
+
+
+ itConj(args, namedArgs);
+ }
+
+
+ private void putIfMissing(final Map<String, String> namedArgs, final String key,
+ final String value) {
+ final String oldValue = namedArgs.get(key);
+ if (oldValue == null || oldValue.length() == 0) {
+ namedArgs.put(key, value);
+ }
+ }
+
+ // TODO: check how ='' and =| are manifested....
+
+ private void putOrNullify(final Map<String, String> namedArgs, final String key,
+ final String value) {
+ final String oldValue = namedArgs.get(key);
+ if (oldValue == null/* || oldValue.length() == 0*/) {
+ namedArgs.put(key, value);
+ } else {
+ if (oldValue.equals("''")) {
+ namedArgs.put(key, "");
+ }
+ }
+ }
+
+ final List<String> listPrefixes = new ArrayList<String>();
+ final List<String> listLines = new ArrayList<String>();
+
+static final Pattern UNINDEXED_WIKI_TEXT = Pattern.compile(
+ "(first|second|third)-person (singular|plural)|" +
+ "present tense|" +
+ "imperative"
+ );
+
+ private void handleForeignListItem(final String foreignText, String title, final Collection<String> forms, final WikiTokenizer wikiTokenizer) {
+
+ final String prefix = wikiTokenizer.listItemPrefix();
+ if (prefix.length() > 1) {
+ System.err.println("Prefix too long: " + wikiTokenizer.token());
+ return;
+ }
+
+ listPrefixes.clear();
+ listLines.clear();
+ listPrefixes.add(prefix);
+ listLines.add(wikiTokenizer.listItemWikiText());
+ while(wikiTokenizer.nextToken() != null &&
+ wikiTokenizer.isNewline() ||
+ wikiTokenizer.isComment() ||
+ (wikiTokenizer.isListItem() &&
+ wikiTokenizer.listItemPrefix().length() > prefix.length() &&
+ wikiTokenizer.listItemPrefix().startsWith(prefix))) {
+ if (wikiTokenizer.isListItem()) {
+ listPrefixes.add(wikiTokenizer.listItemPrefix());
+ listLines.add(wikiTokenizer.listItemWikiText());
+ }
+ }
+ if (wikiTokenizer.nextToken() != null) {
+ wikiTokenizer.returnToLineStart();
+ }
+ System.out.println("list lines: " + listLines);
+ System.out.println("list prefixes: " + listPrefixes);
+
+ final PairEntry pairEntry = new PairEntry();
+ final IndexedEntry indexedEntry = new IndexedEntry(pairEntry);
+
+ final String foreign = trim(title);
+
+ final StringBuilder englishBuilder = new StringBuilder();
+
+ final String mainLine = listLines.get(0);
+
+ final WikiTokenizer englishTokenizer = new WikiTokenizer(mainLine);
+ while (englishTokenizer.nextToken() != null) {
+ // TODO handle form of....
+ if (englishTokenizer.isPlainText()) {
+ englishBuilder.append(englishTokenizer.token());
+ enIndexBuilder.addEntryWithString(indexedEntry, englishTokenizer.token(), EntryTypeName.WIKTIONARY_ENGLISH_DEF);
+ } else if (englishTokenizer.isWikiLink()) {
+ final String text = englishTokenizer.wikiLinkText();
+ final String link = englishTokenizer.wikiLinkDest();
+ if (link != null) {
+ if (link.contains("#English")) {
+ englishBuilder.append(text);
+ enIndexBuilder.addEntryWithString(indexedEntry, text, EntryTypeName.WIKTIONARY_ENGLISH_DEF_WIKI_LINK);
+ } else if (link.contains("#") && this.langPattern.matcher(link).find()) {
+ englishBuilder.append(text);
+ otherIndexBuilder.addEntryWithString(indexedEntry, text, EntryTypeName.WIKTIONARY_ENGLISH_DEF_OTHER_LANG);
+ } else {
+ System.err.println("Special link: " + englishTokenizer.token());
+ // TODO: something here...
+ }
+ } else {
+ // link == null
+ englishBuilder.append(text);
+ if (!UNINDEXED_WIKI_TEXT.matcher(text).find()) {
+ enIndexBuilder.addEntryWithString(indexedEntry, text, EntryTypeName.WIKTIONARY_ENGLISH_DEF_WIKI_LINK);
+ }
+ }
+ } else if (englishTokenizer.isFunction()) {
+ final String name = englishTokenizer.functionName();
+ if (name.contains(" conjugation of ") ||
+ name.contains(" form of ") ||
+ name.contains(" feminine of ") ||
+ name.contains(" plural of ")) {
+ // Ignore these in the index, they're really annoying....
+ englishBuilder.append(englishTokenizer.token());
+ } else {
+ System.err.println("Unexpected function: " + englishTokenizer.token());
+ }
+ } else {
+ if (englishTokenizer.isComment() || englishTokenizer.isMarkup()) {
+ } else {
+ System.err.println("Unexpected definition text: " + englishTokenizer.token());
+ }
+ }
+ }
+ final String english = trim(englishBuilder.toString());
+ if (english.length() > 0) {
+ final Pair pair = new Pair(english, trim(foreignText), this.swap);
+ pairEntry.pairs.add(pair);
+ otherIndexBuilder.addEntryWithString(indexedEntry, title, EntryTypeName.WIKTIONARY_TITLE_SINGLE, EntryTypeName.WIKTIONARY_TITLE_MULTI);
+ for (final String form : forms) {
+ otherIndexBuilder.addEntryWithString(indexedEntry, form, EntryTypeName.WIKTIONARY_FORM_SINGLE, EntryTypeName.WIKTIONARY_FORM_MULTI);
+ }