+ int currentHeadingDepth = posDepth;
+ while (wikiTokenizer.nextToken() != null) {
+ if (wikiTokenizer.isHeading()) {
+ currentHeadingDepth = wikiTokenizer.headingDepth();
+
+ if (currentHeadingDepth <= posDepth) {
+ wikiTokenizer.returnToLineStart();
+ return;
+ }
+ }
+
+ if (currentHeadingDepth > posDepth) {
+ // TODO: deal with other neat info sections
+ continue;
+ }
+
+ 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 (isGender(name)) {
+ appendGender(foreignBuilder, name, args);
+ } else if (name.equals("wikipedia")) {
+ namedArgs.remove("lang");
+ if (args.size() > 1 || !namedArgs.isEmpty()) {
+ // Unindexed!
+ foreignBuilder.append(wikiTokenizer.token());
+ } else if (args.size() == 1) {
+ foreignBuilder.append(wikiTokenizer.token());
+ } else {
+ //foreignBuilder.append(title);
+ }
+ } else if (name.equals("attention") || name.equals("zh-attention")) {
+ // See: http://en.wiktionary.org/wiki/Template:attention
+ // Ignore these.
+ } else if (name.equals("infl")) {
+ // See: http://en.wiktionary.org/wiki/Template:infl
+ final String langCode = get(args, 0);
+ String head = namedArgs.remove("head");
+ if (head == null) {
+ head = namedArgs.remove("title"); // Bug
+ }
+ if (head == null) {
+ head = title;
+ } else {
+ head = WikiTokenizer.toPlainText(head);
+ }
+ titleAppended = true;
+
+ namedArgs.remove("sc");
+ namedArgs.remove("lang");
+ namedArgs.remove("sort");
+ namedArgs.remove("cat");
+
+ final String tr = namedArgs.remove("tr");
+ String g = namedArgs.remove("g");
+ if (g == null) {
+ g = namedArgs.remove("gender");
+ }
+ final String g2 = namedArgs.remove("g2");
+ final String g3 = namedArgs.remove("g3");
+
+ foreignBuilder.append(head);
+
+ if (g != null) {
+ foreignBuilder.append(" {").append(g);
+ if (g2 != null) {
+ foreignBuilder.append("|").append(g2);
+ }
+ if (g3 != null) {
+ foreignBuilder.append("|").append(g3);
+ }
+ foreignBuilder.append("}");
+ }
+
+ if (tr != null) {
+ foreignBuilder.append(String.format(TRANSLITERATION_FORMAT, tr));
+ wordForms.add(tr);
+ }
+
+ final String pos = get(args, 1);
+ if (pos != null) {
+ foreignBuilder.append(" (").append(pos).append(")");
+ }
+ for (int i = 2; i < args.size(); i += 2) {
+ final String inflName = get(args, i);
+ final String inflValue = get(args, i + 1);
+ foreignBuilder.append(", ").append(WikiTokenizer.toPlainText(inflName));
+ if (inflValue != null && inflValue.length() > 0) {
+ foreignBuilder.append(": ").append(WikiTokenizer.toPlainText(inflValue));
+ wordForms.add(inflValue);
+ }
+ }
+ for (final String key : namedArgs.keySet()) {
+ final String value = WikiTokenizer.toPlainText(namedArgs.get(key));
+ foreignBuilder.append(" ").append(key).append("=").append(value);
+ wordForms.add(value);
+ }
+ } else if (name.equals("it-noun")) {
+ titleAppended = true;
+ 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);
+ foreignBuilder.append(String.format(" %s {%s}, %s {pl}", singular, gender, plural, plural));
+ wordForms.add(singular);
+ wordForms.add(plural);
+ } else if (name.equals("it-proper noun")) {
+ foreignBuilder.append(wikiTokenizer.token());
+ } else if (name.equals("it-adj")) {
+ foreignBuilder.append(wikiTokenizer.token());
+ } 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 {
+ LOG.warning("Unknown conjugation: " + wikiTokenizer.token());
+ }
+ } else {
+ // Unindexed!
+ foreignBuilder.append(wikiTokenizer.token());
+ // LOG.warning("Unknown function: " + wikiTokenizer.token());
+ }
+
+ } else if (wikiTokenizer.isListItem()) {
+ final String prefix = wikiTokenizer.listItemPrefix();
+ if (lastListSection != null &&
+ prefix.startsWith(lastListSection.firstPrefix) &&
+ prefix.length() > lastListSection.firstPrefix.length()) {
+ lastListSection.nextPrefixes.add(prefix);
+ lastListSection.nextLines.add(wikiTokenizer.listItemWikiText());
+ } else {
+ lastListSection = new ListSection(prefix, wikiTokenizer.listItemWikiText());
+ listSections.add(lastListSection);
+ }
+ } else if (lastListSection != null) {
+ // Don't append anything after the lists, because there's crap.
+ } else if (wikiTokenizer.isWikiLink()) {
+ // Unindexed!
+ foreignBuilder.append(wikiTokenizer.wikiLinkText());
+
+ } else if (wikiTokenizer.isPlainText()) {
+ // Unindexed!
+ foreignBuilder.append(wikiTokenizer.token());
+
+ } else if (wikiTokenizer.isMarkup() || wikiTokenizer.isNewline() || wikiTokenizer.isComment()) {
+ // Do nothing.
+ } else {
+ LOG.warning("Unexpected token: " + wikiTokenizer.token());