1 // Copyright 2012 Google Inc. All Rights Reserved.
3 // Licensed under the Apache License, Version 2.0 (the "License");
4 // you may not use this file except in compliance with the License.
5 // You may obtain a copy of the License at
7 // http://www.apache.org/licenses/LICENSE-2.0
9 // Unless required by applicable law or agreed to in writing, software
10 // distributed under the License is distributed on an "AS IS" BASIS,
11 // WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
12 // See the License for the specific language governing permissions and
13 // limitations under the License.
15 package com.hughes.android.dictionary.parser.wiktionary;
17 import java.util.ArrayList;
18 import java.util.Arrays;
19 import java.util.Collection;
20 import java.util.LinkedHashMap;
21 import java.util.LinkedHashSet;
22 import java.util.List;
25 import java.util.concurrent.atomic.AtomicInteger;
26 import java.util.regex.Pattern;
28 import com.hughes.android.dictionary.engine.EntryTypeName;
29 import com.hughes.android.dictionary.engine.IndexBuilder;
30 import com.hughes.android.dictionary.engine.IndexedEntry;
31 import com.hughes.android.dictionary.engine.PairEntry;
32 import com.hughes.android.dictionary.engine.PairEntry.Pair;
33 import com.hughes.android.dictionary.parser.WikiTokenizer;
34 import com.hughes.util.ListUtil;
36 public abstract class EnParser extends AbstractWiktionaryParser {
38 // TODO: process {{ttbc}} lines
40 static final Pattern partOfSpeechHeader = Pattern.compile(
41 "Noun|Verb|Adjective|Adverb|Pronoun|Conjunction|Interjection|" +
42 "Preposition|Proper noun|Article|Prepositional phrase|Acronym|" +
43 "Abbreviation|Initialism|Contraction|Prefix|Suffix|Symbol|Letter|" +
44 "Ligature|Idiom|Phrase|\\{\\{acronym\\}\\}|\\{\\{initialism\\}\\}|" +
45 "\\{\\{abbreviation\\}\\}|" +
46 // These are @deprecated:
47 "Noun form|Verb form|Adjective form|Nominal phrase|Noun phrase|" +
48 "Verb phrase|Transitive verb|Intransitive verb|Reflexive verb|" +
49 // These are extras I found:
50 "Determiner|Numeral|Number|Cardinal number|Ordinal number|Proverb|" +
51 "Particle|Interjection|Pronominal adverb" +
52 "Han character|Hanzi|Hanja|Kanji|Katakana character|Syllable");
54 // Might only want to remove "lang" if it's equal to "zh", for example.
55 static final Set<String> USELESS_WIKI_ARGS = new LinkedHashSet<String>(
65 static boolean isIgnorableTitle(final String title) {
66 return title.startsWith("Wiktionary:") ||
67 title.startsWith("Template:") ||
68 title.startsWith("Appendix:") ||
69 title.startsWith("Category:") ||
70 title.startsWith("Index:") ||
71 title.startsWith("MediaWiki:") ||
72 title.startsWith("TransWiki:") ||
73 title.startsWith("Citations:") ||
74 title.startsWith("Concordance:") ||
75 title.startsWith("Help:");
78 final IndexBuilder enIndexBuilder;
79 final IndexBuilder foreignIndexBuilder;
80 final Pattern langPattern;
81 final Pattern langCodePattern;
84 // State used while parsing.
87 ENGLISH_DEF_OF_FOREIGN,
93 public boolean entryIsFormOfSomething = false;
94 final Collection<String> wordForms = new ArrayList<String>();
95 boolean titleAppended = false;
98 final AppendAndIndexWikiCallback<EnParser> appendAndIndexWikiCallback = new AppendAndIndexWikiCallback<EnParser>(this);
100 appendAndIndexWikiCallback.functionCallbacks.putAll(FunctionCallbacks.DEFAULT);
103 EnParser(final IndexBuilder enIndexBuilder, final IndexBuilder otherIndexBuilder, final Pattern langPattern, final Pattern langCodePattern, final boolean swap) {
104 this.enIndexBuilder = enIndexBuilder;
105 this.foreignIndexBuilder = otherIndexBuilder;
106 this.langPattern = langPattern;
107 this.langCodePattern = langCodePattern;
111 // --------------------------------------------------------------------
113 static final class EnToTranslationParser extends EnParser {
115 EnToTranslationParser(final IndexBuilder enIndexBuilder,
116 final IndexBuilder otherIndexBuilder, final Pattern langPattern,
117 final Pattern langCodePattern, final boolean swap) {
118 super(enIndexBuilder, otherIndexBuilder, langPattern, langCodePattern, swap);
122 void parseSection(String heading, String text) {
123 if (isIgnorableTitle(title)) {
126 heading = heading.replaceAll("=", "").trim();
127 if (!heading.contains("English")) {
134 final WikiTokenizer wikiTokenizer = new WikiTokenizer(text);
135 while (wikiTokenizer.nextToken() != null) {
137 if (wikiTokenizer.isHeading()) {
138 final String headerName = wikiTokenizer.headingWikiText();
140 if (wikiTokenizer.headingDepth() <= posDepth) {
145 if (partOfSpeechHeader.matcher(headerName).matches()) {
146 posDepth = wikiTokenizer.headingDepth();
147 pos = wikiTokenizer.headingWikiText();
148 // TODO: if we're inside the POS section, we should handle the first title line...
150 } else if (headerName.equals("Translations")) {
152 LOG.info("Translations without POS (but using anyway): " + title);
154 doTranslations(wikiTokenizer, pos);
155 } else if (headerName.equals("Pronunciation")) {
156 //doPronunciation(wikiLineReader);
158 } else if (wikiTokenizer.isFunction()) {
159 final String name = wikiTokenizer.functionName();
160 if (name.equals("head") && pos == null) {
161 LOG.warning("{{head}} without POS: " + title);
167 private void doTranslations(final WikiTokenizer wikiTokenizer, final String pos) {
168 if (title.equals("absolutely")) {
169 //System.out.println();
172 String topLevelLang = null;
174 boolean done = false;
175 while (wikiTokenizer.nextToken() != null) {
176 if (wikiTokenizer.isHeading()) {
177 wikiTokenizer.returnToLineStart();
184 // Check whether we care about this line:
186 if (wikiTokenizer.isFunction()) {
187 final String functionName = wikiTokenizer.functionName();
188 final List<String> positionArgs = wikiTokenizer.functionPositionArgs();
190 if (functionName.equals("trans-top")) {
192 if (wikiTokenizer.functionPositionArgs().size() >= 1) {
193 sense = positionArgs.get(0);
194 // TODO: could emphasize words in [[brackets]] inside sense.
195 sense = WikiTokenizer.toPlainText(sense);
196 //LOG.info("Sense: " + sense);
198 } else if (functionName.equals("trans-bottom")) {
200 } else if (functionName.equals("trans-mid")) {
201 } else if (functionName.equals("trans-see")) {
202 incrementCount("WARNING:trans-see");
203 } else if (functionName.startsWith("picdic")) {
204 } else if (functionName.startsWith("checktrans")) {
206 } else if (functionName.startsWith("ttbc")) {
207 wikiTokenizer.nextLine();
208 // TODO: would be great to handle ttbc
209 // TODO: Check this: done = true;
211 LOG.warning("Unexpected translation wikifunction: " + wikiTokenizer.token() + ", title=" + title);
213 } else if (wikiTokenizer.isListItem()) {
214 final String line = wikiTokenizer.listItemWikiText();
215 // This line could produce an output...
217 // if (line.contains("ich hoan dich gear")) {
218 // //System.out.println();
221 // First strip the language and check whether it matches.
222 // And hold onto it for sub-lines.
223 final int colonIndex = line.indexOf(":");
224 if (colonIndex == -1) {
228 final String lang = trim(WikiTokenizer.toPlainText(line.substring(0, colonIndex)));
229 incrementCount("tCount:" + lang);
230 final boolean appendLang;
231 if (wikiTokenizer.listItemPrefix().length() == 1) {
233 final boolean thisFind = langPattern.matcher(lang).find();
237 appendLang = !langPattern.matcher(lang).matches();
238 } else if (topLevelLang == null) {
241 // Two-level -- the only way we won't append is if this second level matches exactly.
242 if (!langPattern.matcher(lang).matches() && !langPattern.matcher(topLevelLang).find()) {
245 appendLang = !langPattern.matcher(lang).matches();
248 String rest = line.substring(colonIndex + 1).trim();
249 if (rest.length() > 0) {
250 doTranslationLine(line, appendLang ? lang : null, pos, sense, rest);
253 } else if (wikiTokenizer.remainderStartsWith("''See''")) {
254 wikiTokenizer.nextLine();
255 incrementCount("WARNING: ''See''" );
256 LOG.fine("Skipping See line: " + wikiTokenizer.token());
257 } else if (wikiTokenizer.isWikiLink()) {
258 final String wikiLink = wikiTokenizer.wikiLinkText();
259 if (wikiLink.contains(":") && wikiLink.contains(title)) {
260 } else if (wikiLink.contains("Category:")) {
262 incrementCount("WARNING: Unexpected wikiLink" );
263 LOG.warning("Unexpected wikiLink: " + wikiTokenizer.token() + ", title=" + title);
265 } else if (wikiTokenizer.isNewline() || wikiTokenizer.isMarkup() || wikiTokenizer.isComment()) {
267 final String token = wikiTokenizer.token();
268 if (token.equals("----")) {
270 LOG.warning("Unexpected translation token: " + wikiTokenizer.token() + ", title=" + title);
271 incrementCount("WARNING: Unexpected translation token" );
278 private void doTranslationLine(final String line, final String lang, final String pos, final String sense, final String rest) {
279 state = State.TRANSLATION_LINE;
280 // Good chance we'll actually file this one...
281 final PairEntry pairEntry = new PairEntry(entrySource);
282 final IndexedEntry indexedEntry = new IndexedEntry(pairEntry);
284 final StringBuilder foreignText = new StringBuilder();
285 appendAndIndexWikiCallback.reset(foreignText, indexedEntry);
286 appendAndIndexWikiCallback.dispatch(rest, foreignIndexBuilder, EntryTypeName.WIKTIONARY_TRANSLATION_OTHER_TEXT);
288 if (foreignText.length() == 0) {
289 LOG.warning("Empty foreignText: " + line);
290 incrementCount("WARNING: Empty foreignText" );
295 foreignText.insert(0, String.format("(%s) ", lang));
298 StringBuilder englishText = new StringBuilder();
300 englishText.append(title);
302 englishText.append(" (").append(sense).append(")");
303 enIndexBuilder.addEntryWithString(indexedEntry, sense, EntryTypeName.WIKTIONARY_TRANSLATION_SENSE);
306 englishText.append(" (").append(pos.toLowerCase()).append(")");
308 enIndexBuilder.addEntryWithString(indexedEntry, title, EntryTypeName.WIKTIONARY_TITLE_MULTI);
310 final Pair pair = new Pair(trim(englishText.toString()), trim(foreignText.toString()), swap);
311 pairEntry.pairs.add(pair);
312 if (!pairsAdded.add(pair.toString())) {
313 LOG.warning("Duplicate pair: " + pair.toString());
314 incrementCount("WARNING: Duplicate pair" );
317 } // EnToTranslationParser
319 // -----------------------------------------------------------------------
322 static final class ForeignParser extends EnParser {
324 ForeignParser(final IndexBuilder enIndexBuilder,
325 final IndexBuilder otherIndexBuilder, final Pattern langPattern,
326 final Pattern langCodePattern, final boolean swap) {
327 super(enIndexBuilder, otherIndexBuilder, langPattern, langCodePattern, swap);
331 void parseSection(String heading, String text) {
332 if (isIgnorableTitle(title)) {
335 final String lang = heading.replaceAll("=", "").trim();
336 if (!langPattern.matcher(lang).find()){
340 final WikiTokenizer wikiTokenizer = new WikiTokenizer(text);
341 while (wikiTokenizer.nextToken() != null) {
342 if (wikiTokenizer.isHeading()) {
343 final String headingName = wikiTokenizer.headingWikiText();
344 if (headingName.equals("Translations")) {
345 LOG.warning("Translations not in English section: " + title);
346 } else if (headingName.equals("Pronunciation")) {
347 //doPronunciation(wikiLineReader);
348 } else if (partOfSpeechHeader.matcher(headingName).matches()) {
349 doForeignPartOfSpeech(lang, headingName, wikiTokenizer.headingDepth(), wikiTokenizer);
352 // It's not a heading.
353 // TODO: optimization: skip to next heading.
358 static final class ListSection {
359 final String firstPrefix;
360 final String firstLine;
361 final List<String> nextPrefixes = new ArrayList<String>();
362 final List<String> nextLines = new ArrayList<String>();
364 public ListSection(String firstPrefix, String firstLine) {
365 this.firstPrefix = firstPrefix;
366 this.firstLine = firstLine;
370 public String toString() {
371 return firstPrefix + firstLine + "{ " + nextPrefixes + "}";
375 int foreignCount = 0;
376 private void doForeignPartOfSpeech(final String lang, String posHeading, final int posDepth, WikiTokenizer wikiTokenizer) {
377 if (++foreignCount % 1000 == 0) {
378 LOG.info("***" + lang + ", " + title + ", pos=" + posHeading + ", foreignCount=" + foreignCount);
380 if (title.equals("6")) {
381 System.out.println();
384 final StringBuilder foreignBuilder = new StringBuilder();
385 final List<ListSection> listSections = new ArrayList<ListSection>();
387 appendAndIndexWikiCallback.reset(foreignBuilder, null);
388 this.state = State.ENGLISH_DEF_OF_FOREIGN; // TODO: this is wrong, need new category....
389 titleAppended = false;
394 ListSection lastListSection = null;
396 int currentHeadingDepth = posDepth;
397 while (wikiTokenizer.nextToken() != null) {
398 if (wikiTokenizer.isHeading()) {
399 currentHeadingDepth = wikiTokenizer.headingDepth();
401 if (currentHeadingDepth <= posDepth) {
402 wikiTokenizer.returnToLineStart();
407 if (currentHeadingDepth > posDepth) {
408 // TODO: deal with other neat info sections
412 if (wikiTokenizer.isFunction()) {
413 final String name = wikiTokenizer.functionName();
414 final List<String> args = wikiTokenizer.functionPositionArgs();
415 final Map<String,String> namedArgs = wikiTokenizer.functionNamedArgs();
416 // First line is generally a repeat of the title with some extra information.
417 // We need to build up the left side (foreign text, tokens) separately from the
418 // right side (English). The left-side may get paired with multiple right sides.
419 // The left side should get filed under every form of the word in question (singular, plural).
421 // For verbs, the conjugation comes later on in a deeper section.
422 // Ideally, we'd want to file every English entry with the verb
423 // under every verb form coming from the conjugation.
424 // Ie. under "fa": see: "make :: fare" and "do :: fare"
425 // But then where should we put the conjugation table?
426 // 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!)
427 // for the conjugation table from "fa".
428 // Would like to be able to link to a lang#token.
430 appendAndIndexWikiCallback.onFunction(wikiTokenizer, name, args, namedArgs);
432 } else if (wikiTokenizer.isListItem()) {
433 final String prefix = wikiTokenizer.listItemPrefix();
434 if (lastListSection != null &&
435 prefix.startsWith(lastListSection.firstPrefix) &&
436 prefix.length() > lastListSection.firstPrefix.length()) {
437 lastListSection.nextPrefixes.add(prefix);
438 lastListSection.nextLines.add(wikiTokenizer.listItemWikiText());
440 lastListSection = new ListSection(prefix, wikiTokenizer.listItemWikiText());
441 listSections.add(lastListSection);
443 } else if (lastListSection != null) {
444 // Don't append anything after the lists, because there's crap.
445 } else if (wikiTokenizer.isWikiLink()) {
447 foreignBuilder.append(wikiTokenizer.wikiLinkText());
449 } else if (wikiTokenizer.isPlainText()) {
451 foreignBuilder.append(wikiTokenizer.token());
453 } else if (wikiTokenizer.isMarkup() || wikiTokenizer.isNewline() || wikiTokenizer.isComment()) {
456 LOG.warning("Unexpected token: " + wikiTokenizer.token());
461 // Here's where we exit.
462 // Should we make an entry even if there are no foreign list items?
463 String foreign = foreignBuilder.toString().trim();
464 if (!titleAppended && !foreign.toLowerCase().startsWith(title.toLowerCase())) {
465 foreign = String.format("%s %s", title, foreign);
467 if (!langPattern.matcher(lang).matches()) {
468 foreign = String.format("(%s) %s", lang, foreign);
470 for (final ListSection listSection : listSections) {
471 doForeignListSection(foreign, title, wordForms, listSection);
478 private void doForeignListSection(final String foreignText, String title, final Collection<String> forms, final ListSection listSection) {
479 state = State.ENGLISH_DEF_OF_FOREIGN;
480 final String prefix = listSection.firstPrefix;
481 if (prefix.length() > 1) {
482 // Could just get looser and say that any prefix longer than first is a sublist.
483 LOG.warning("Prefix too long: " + listSection);
487 final PairEntry pairEntry = new PairEntry(entrySource);
488 final IndexedEntry indexedEntry = new IndexedEntry(pairEntry);
490 entryIsFormOfSomething = false;
491 final StringBuilder englishBuilder = new StringBuilder();
492 final String mainLine = listSection.firstLine;
493 appendAndIndexWikiCallback.reset(englishBuilder, indexedEntry);
494 appendAndIndexWikiCallback.dispatch(mainLine, enIndexBuilder, EntryTypeName.WIKTIONARY_ENGLISH_DEF);
496 final String english = trim(englishBuilder.toString());
497 if (english.length() > 0) {
498 final Pair pair = new Pair(english, trim(foreignText), this.swap);
499 pairEntry.pairs.add(pair);
500 foreignIndexBuilder.addEntryWithString(indexedEntry, title, entryIsFormOfSomething ? EntryTypeName.WIKTIONARY_IS_FORM_OF_SOMETHING_ELSE : EntryTypeName.WIKTIONARY_TITLE_MULTI);
501 for (final String form : forms) {
502 foreignIndexBuilder.addEntryWithString(indexedEntry, form, EntryTypeName.WIKTIONARY_INFLECTED_FORM_MULTI);
507 String lastForeign = null;
508 for (int i = 0; i < listSection.nextPrefixes.size(); ++i) {
509 final String nextPrefix = listSection.nextPrefixes.get(i);
510 final String nextLine = listSection.nextLines.get(i);
512 // TODO: This splitting is not sensitive to wiki code.
513 int dash = nextLine.indexOf("—");
516 dash = nextLine.indexOf("—");
520 dash = nextLine.indexOf(" - ");
524 if ((nextPrefix.equals("#:") || nextPrefix.equals("##:")) && dash != -1) {
525 final String foreignEx = nextLine.substring(0, dash);
526 final String englishEx = nextLine.substring(dash + mdashLen);
527 final Pair pair = new Pair(formatAndIndexExampleString(englishEx, enIndexBuilder, indexedEntry), formatAndIndexExampleString(foreignEx, foreignIndexBuilder, indexedEntry), swap);
528 if (pair.lang1 != "--" && pair.lang1 != "--") {
529 pairEntry.pairs.add(pair);
532 } else if (nextPrefix.equals("#:") || nextPrefix.equals("##:")){
533 final Pair pair = new Pair("--", formatAndIndexExampleString(nextLine, null, indexedEntry), swap);
534 lastForeign = nextLine;
535 if (pair.lang1 != "--" && pair.lang1 != "--") {
536 pairEntry.pairs.add(pair);
538 } else if (nextPrefix.equals("#::") || nextPrefix.equals("#**")) {
539 if (lastForeign != null && pairEntry.pairs.size() > 0) {
540 pairEntry.pairs.remove(pairEntry.pairs.size() - 1);
541 final Pair pair = new Pair(formatAndIndexExampleString(nextLine, enIndexBuilder, indexedEntry), formatAndIndexExampleString(lastForeign, foreignIndexBuilder, indexedEntry), swap);
542 if (pair.lang1 != "--" || pair.lang2 != "--") {
543 pairEntry.pairs.add(pair);
547 LOG.warning("TODO: English example with no foreign: " + title + ", " + nextLine);
548 final Pair pair = new Pair("--", formatAndIndexExampleString(nextLine, null, indexedEntry), swap);
549 if (pair.lang1 != "--" || pair.lang2 != "--") {
550 pairEntry.pairs.add(pair);
553 } else if (nextPrefix.equals("#*")) {
554 // Can't really index these.
555 final Pair pair = new Pair("--", formatAndIndexExampleString(nextLine, null, indexedEntry), swap);
556 lastForeign = nextLine;
557 if (pair.lang1 != "--" || pair.lang2 != "--") {
558 pairEntry.pairs.add(pair);
560 } else if (nextPrefix.equals("#::*") || nextPrefix.equals("##") || nextPrefix.equals("#*:") || nextPrefix.equals("#:*") || true) {
561 final Pair pair = new Pair("--", formatAndIndexExampleString(nextLine, null, indexedEntry), swap);
562 if (pair.lang1 != "--" || pair.lang2 != "--") {
563 pairEntry.pairs.add(pair);
571 private String formatAndIndexExampleString(final String example, final IndexBuilder indexBuilder, final IndexedEntry indexedEntry) {
573 // if (wikiTokenizer.token().equals("'''")) {
574 // insideTripleQuotes = !insideTripleQuotes;
576 final StringBuilder builder = new StringBuilder();
577 appendAndIndexWikiCallback.reset(builder, indexedEntry);
578 appendAndIndexWikiCallback.entryTypeName = EntryTypeName.WIKTIONARY_EXAMPLE;
579 appendAndIndexWikiCallback.entryTypeNameSticks = true;
581 // TODO: this is a hack needed because we don't safely split on the dash.
582 appendAndIndexWikiCallback.dispatch(example, indexBuilder, EntryTypeName.WIKTIONARY_EXAMPLE);
583 } catch (AssertionError e) {
586 final String result = trim(builder.toString());
587 return result.length() > 0 ? result : "--";
591 private void itConjAre(List<String> args, Map<String, String> namedArgs) {
592 final String base = args.get(0);
593 final String aux = args.get(1);
595 putIfMissing(namedArgs, "inf", base + "are");
596 putIfMissing(namedArgs, "aux", aux);
597 putIfMissing(namedArgs, "ger", base + "ando");
598 putIfMissing(namedArgs, "presp", base + "ante");
599 putIfMissing(namedArgs, "pastp", base + "ato");
601 putIfMissing(namedArgs, "pres1s", base + "o");
602 putIfMissing(namedArgs, "pres2s", base + "i");
603 putIfMissing(namedArgs, "pres3s", base + "a");
604 putIfMissing(namedArgs, "pres1p", base + "iamo");
605 putIfMissing(namedArgs, "pres2p", base + "ate");
606 putIfMissing(namedArgs, "pres3p", base + "ano");
608 putIfMissing(namedArgs, "imperf1s", base + "avo");
609 putIfMissing(namedArgs, "imperf2s", base + "avi");
610 putIfMissing(namedArgs, "imperf3s", base + "ava");
611 putIfMissing(namedArgs, "imperf1p", base + "avamo");
612 putIfMissing(namedArgs, "imperf2p", base + "avate");
613 putIfMissing(namedArgs, "imperf3p", base + "avano");
615 putIfMissing(namedArgs, "prem1s", base + "ai");
616 putIfMissing(namedArgs, "prem2s", base + "asti");
617 putIfMissing(namedArgs, "prem3s", base + "ò");
618 putIfMissing(namedArgs, "prem1p", base + "ammo");
619 putIfMissing(namedArgs, "prem2p", base + "aste");
620 putIfMissing(namedArgs, "prem3p", base + "arono");
622 putIfMissing(namedArgs, "fut1s", base + "erò");
623 putIfMissing(namedArgs, "fut2s", base + "erai");
624 putIfMissing(namedArgs, "fut3s", base + "erà ");
625 putIfMissing(namedArgs, "fut1p", base + "eremo");
626 putIfMissing(namedArgs, "fut2p", base + "erete");
627 putIfMissing(namedArgs, "fut3p", base + "eranno");
629 putIfMissing(namedArgs, "cond1s", base + "erei");
630 putIfMissing(namedArgs, "cond2s", base + "eresti");
631 putIfMissing(namedArgs, "cond3s", base + "erebbe");
632 putIfMissing(namedArgs, "cond1p", base + "eremmo");
633 putIfMissing(namedArgs, "cond2p", base + "ereste");
634 putIfMissing(namedArgs, "cond3p", base + "erebbero");
635 // Subjunctive / congiuntivo
636 putIfMissing(namedArgs, "sub123s", base + "i");
637 putIfMissing(namedArgs, "sub1p", base + "iamo");
638 putIfMissing(namedArgs, "sub2p", base + "iate");
639 putIfMissing(namedArgs, "sub3p", base + "ino");
640 // Imperfect subjunctive
641 putIfMissing(namedArgs, "impsub12s", base + "assi");
642 putIfMissing(namedArgs, "impsub3s", base + "asse");
643 putIfMissing(namedArgs, "impsub1p", base + "assimo");
644 putIfMissing(namedArgs, "impsub2p", base + "aste");
645 putIfMissing(namedArgs, "impsub3p", base + "assero");
647 putIfMissing(namedArgs, "imp2s", base + "a");
648 putIfMissing(namedArgs, "imp3s", base + "i");
649 putIfMissing(namedArgs, "imp1p", base + "iamo");
650 putIfMissing(namedArgs, "imp2p", base + "ate");
651 putIfMissing(namedArgs, "imp3p", base + "ino");
654 itConj(args, namedArgs);
658 private void itConj(List<String> args, Map<String, String> namedArgs) {
659 // TODO Auto-generated method stub
664 private static void putIfMissing(final Map<String, String> namedArgs, final String key,
665 final String value) {
666 final String oldValue = namedArgs.get(key);
667 if (oldValue == null || oldValue.length() == 0) {
668 namedArgs.put(key, value);
672 // TODO: check how ='' and =| are manifested....
673 // TODO: get this right in -are
674 private static void putOrNullify(final Map<String, String> namedArgs, final String key,
675 final String value) {
676 final String oldValue = namedArgs.get(key);
677 if (oldValue == null/* || oldValue.length() == 0*/) {
678 namedArgs.put(key, value);
680 if (oldValue.equals("''")) {
681 namedArgs.put(key, "");
688 // -----------------------------------------------------------------------
690 static class FunctionCallbacks {
692 static final Map<String,FunctionCallback<EnParser>> DEFAULT = new LinkedHashMap<String, FunctionCallback<EnParser>>();
695 FunctionCallback<EnParser> callback = new TranslationCallback();
696 DEFAULT.put("t", callback);
697 DEFAULT.put("t+", callback);
698 DEFAULT.put("t-", callback);
699 DEFAULT.put("tø", callback);
700 DEFAULT.put("apdx-t", callback);
702 callback = new EncodingCallback();
703 Set<String> encodings = new LinkedHashSet<String>(Arrays.asList(
705 "sd-Arab", "ku-Arab", "Arab", "unicode", "Laoo", "ur-Arab", "Thai",
706 "fa-Arab", "Khmr", "Cyrl", "IPAchar", "ug-Arab", "ko-inline",
707 "Jpan", "Kore", "Hebr", "rfscript", "Beng", "Mong", "Knda", "Cyrs",
708 "yue-tsj", "Mlym", "Tfng", "Grek", "yue-yue-j"));
709 for (final String encoding : encodings) {
710 DEFAULT.put(encoding, callback);
713 callback = new l_term();
714 DEFAULT.put("l", callback);
715 DEFAULT.put("term", callback);
717 callback = new Gender();
718 DEFAULT.put("m", callback);
719 DEFAULT.put("f", callback);
720 DEFAULT.put("n", callback);
721 DEFAULT.put("p", callback);
722 DEFAULT.put("g", callback);
724 callback = new AppendArg0();
726 callback = new Ignore();
727 DEFAULT.put("trreq", callback);
728 DEFAULT.put("t-image", callback);
729 DEFAULT.put("defn", callback);
730 DEFAULT.put("rfdef", callback);
731 DEFAULT.put("rfdate", callback);
732 DEFAULT.put("rfex", callback);
733 DEFAULT.put("rfquote", callback);
734 DEFAULT.put("attention", callback);
735 DEFAULT.put("zh-attention", callback);
738 callback = new FormOf();
739 DEFAULT.put("form of", callback);
740 DEFAULT.put("conjugation of", callback);
741 DEFAULT.put("participle of", callback);
742 DEFAULT.put("present participle of", callback);
743 DEFAULT.put("past participle of", callback);
744 DEFAULT.put("feminine past participle of", callback);
745 DEFAULT.put("gerund of", callback);
746 DEFAULT.put("feminine of", callback);
747 DEFAULT.put("plural of", callback);
748 DEFAULT.put("feminine plural of", callback);
749 DEFAULT.put("inflected form of", callback);
750 DEFAULT.put("alternative form of", callback);
751 DEFAULT.put("dated form of", callback);
752 DEFAULT.put("apocopic form of", callback);
754 callback = new InflOrHead();
755 DEFAULT.put("infl", callback);
756 DEFAULT.put("head", callback);
758 callback = new AppendName();
759 DEFAULT.put("...", callback);
761 DEFAULT.put("qualifier", new QualifierCallback());
762 DEFAULT.put("italbrac", new italbrac());
763 DEFAULT.put("gloss", new gloss());
764 DEFAULT.put("not used", new not_used());
765 DEFAULT.put("wikipedia", new wikipedia());
768 static final NameAndArgs<EnParser> NAME_AND_ARGS = new NameAndArgs<EnParser>();
770 // ------------------------------------------------------------------
772 static final class TranslationCallback implements FunctionCallback<EnParser> {
774 public boolean onWikiFunction(final WikiTokenizer wikiTokenizer, final String name, final List<String> args,
775 final Map<String, String> namedArgs, final EnParser parser,
776 final AppendAndIndexWikiCallback<EnParser> appendAndIndexWikiCallback) {
778 final String transliteration = namedArgs.remove("tr");
779 final String alt = namedArgs.remove("alt");
780 namedArgs.keySet().removeAll(USELESS_WIKI_ARGS);
781 if (args.size() < 2) {
782 LOG.warning("{{t...}} with wrong args: title=" + parser.title);
785 final String langCode = ListUtil.get(args, 0);
786 if (!appendAndIndexWikiCallback.langCodeToTCount.containsKey(langCode)) {
787 appendAndIndexWikiCallback.langCodeToTCount.put(langCode, new AtomicInteger());
789 appendAndIndexWikiCallback.langCodeToTCount.get(langCode).incrementAndGet();
790 final String word = ListUtil.get(args, 1);
791 appendAndIndexWikiCallback.dispatch(alt != null ? alt : word, EntryTypeName.WIKTIONARY_TITLE_MULTI);
794 if (args.size() > 2) {
795 appendAndIndexWikiCallback.builder.append(" {");
796 for (int i = 2; i < args.size(); ++i) {
798 appendAndIndexWikiCallback.builder.append("|");
800 appendAndIndexWikiCallback.builder.append(args.get(i));
802 appendAndIndexWikiCallback.builder.append("}");
805 if (transliteration != null) {
806 appendAndIndexWikiCallback.builder.append(" (");
807 appendAndIndexWikiCallback.dispatch(transliteration, EntryTypeName.WIKTIONARY_TRANSLITERATION);
808 appendAndIndexWikiCallback.builder.append(")");
812 // If alt wasn't null, we appended alt instead of the actual word
813 // we're filing under..
814 appendAndIndexWikiCallback.builder.append(" (");
815 appendAndIndexWikiCallback.dispatch(word, EntryTypeName.WIKTIONARY_TITLE_MULTI);
816 appendAndIndexWikiCallback.builder.append(")");
819 // Catch-all for anything else...
820 if (!namedArgs.isEmpty()) {
821 appendAndIndexWikiCallback.builder.append(" {");
822 appendNamedArgs(namedArgs, appendAndIndexWikiCallback);
823 appendAndIndexWikiCallback.builder.append("}");
830 // ------------------------------------------------------------------
832 static final class QualifierCallback implements FunctionCallback<EnParser> {
834 public boolean onWikiFunction(final WikiTokenizer wikiTokenizer, final String name, final List<String> args,
835 final Map<String, String> namedArgs,
836 final EnParser parser,
837 final AppendAndIndexWikiCallback<EnParser> appendAndIndexWikiCallback) {
838 if (args.size() != 1 || !namedArgs.isEmpty()) {
839 LOG.warning("weird qualifier: ");
842 String qualifier = args.get(0);
843 appendAndIndexWikiCallback.builder.append("(");
844 appendAndIndexWikiCallback.dispatch(qualifier, null);
845 appendAndIndexWikiCallback.builder.append(")");
850 // ------------------------------------------------------------------
852 static final class EncodingCallback implements FunctionCallback<EnParser> {
854 public boolean onWikiFunction(final WikiTokenizer wikiTokenizer, final String name, final List<String> args,
855 final Map<String, String> namedArgs,
856 final EnParser parser,
857 final AppendAndIndexWikiCallback<EnParser> appendAndIndexWikiCallback) {
858 if (!namedArgs.isEmpty()) {
859 LOG.warning("weird encoding: " + wikiTokenizer.token());
861 if (args.size() == 0) {
862 // Things like "{{Jpan}}" exist.
866 for (int i = 0; i < args.size(); ++i) {
868 appendAndIndexWikiCallback.builder.append(", ");
870 final String arg = args.get(i);
871 // if (arg.equals(parser.title)) {
872 // parser.titleAppended = true;
874 appendAndIndexWikiCallback.dispatch(arg, appendAndIndexWikiCallback.entryTypeName);
881 // ------------------------------------------------------------------
883 static final class Gender implements FunctionCallback<EnParser> {
885 public boolean onWikiFunction(final WikiTokenizer wikiTokenizer, final String name, final List<String> args,
886 final Map<String, String> namedArgs,
887 final EnParser parser,
888 final AppendAndIndexWikiCallback<EnParser> appendAndIndexWikiCallback) {
889 if (!namedArgs.isEmpty()) {
892 appendAndIndexWikiCallback.builder.append("{");
893 appendAndIndexWikiCallback.builder.append(name);
894 for (int i = 0; i < args.size(); ++i) {
895 appendAndIndexWikiCallback.builder.append("|").append(args.get(i));
897 appendAndIndexWikiCallback.builder.append("}");
902 // ------------------------------------------------------------------
904 static final class l_term implements FunctionCallback<EnParser> {
906 public boolean onWikiFunction(final WikiTokenizer wikiTokenizer, final String name, final List<String> args,
907 final Map<String, String> namedArgs,
908 final EnParser parser,
909 final AppendAndIndexWikiCallback<EnParser> appendAndIndexWikiCallback) {
911 // for {{l}}, lang is arg 0, but not for {{term}}
912 if (name.equals("term")) {
916 final EntryTypeName entryTypeName;
917 switch (parser.state) {
918 case TRANSLATION_LINE: entryTypeName = EntryTypeName.WIKTIONARY_TRANSLATION_OTHER_TEXT; break;
919 case ENGLISH_DEF_OF_FOREIGN: entryTypeName = EntryTypeName.WIKTIONARY_ENGLISH_DEF_WIKI_LINK; break;
920 default: throw new IllegalStateException("Invalid enum value: " + parser.state);
923 final String langCode = args.get(0);
924 final IndexBuilder indexBuilder;
925 if ("".equals(langCode)) {
926 indexBuilder = parser.foreignIndexBuilder;
927 } else if ("en".equals(langCode)) {
928 indexBuilder = parser.enIndexBuilder;
930 indexBuilder = parser.foreignIndexBuilder;
933 String displayText = ListUtil.get(args, 2, "");
934 if (displayText.equals("")) {
935 displayText = ListUtil.get(args, 1, null);
938 if (displayText != null) {
939 appendAndIndexWikiCallback.dispatch(displayText, indexBuilder, entryTypeName);
941 LOG.warning("no display text: " + wikiTokenizer.token());
944 final String tr = namedArgs.remove("tr");
946 appendAndIndexWikiCallback.builder.append(" (");
947 appendAndIndexWikiCallback.dispatch(tr, indexBuilder, EntryTypeName.WIKTIONARY_TRANSLITERATION);
948 appendAndIndexWikiCallback.builder.append(")");
951 final String gloss = ListUtil.get(args, 3, "");
952 if (!gloss.equals("")) {
953 appendAndIndexWikiCallback.builder.append(" (");
954 appendAndIndexWikiCallback.dispatch(gloss, parser.enIndexBuilder, EntryTypeName.WIKTIONARY_ENGLISH_DEF);
955 appendAndIndexWikiCallback.builder.append(")");
958 namedArgs.keySet().removeAll(USELESS_WIKI_ARGS);
959 if (!namedArgs.isEmpty()) {
960 appendAndIndexWikiCallback.builder.append(" {").append(name);
961 appendNamedArgs(namedArgs, appendAndIndexWikiCallback);
962 appendAndIndexWikiCallback.builder.append("}");
969 // ------------------------------------------------------------------
971 static final class AppendArg0 implements FunctionCallback<EnParser> {
973 public boolean onWikiFunction(final WikiTokenizer wikiTokenizer, final String name, final List<String> args,
974 final Map<String, String> namedArgs,
975 final EnParser parser,
976 final AppendAndIndexWikiCallback<EnParser> appendAndIndexWikiCallback) {
977 if (args.size() != 1 || !namedArgs.isEmpty()) {
980 appendAndIndexWikiCallback.dispatch(args.get(0), EntryTypeName.WIKTIONARY_TRANSLATION_OTHER_TEXT);
981 // TODO: transliteration
986 // ------------------------------------------------------------------
988 static final class italbrac implements FunctionCallback<EnParser> {
990 public boolean onWikiFunction(final WikiTokenizer wikiTokenizer, final String name, final List<String> args,
991 final Map<String, String> namedArgs,
992 final EnParser parser,
993 final AppendAndIndexWikiCallback<EnParser> appendAndIndexWikiCallback) {
994 if (args.size() != 1 || !namedArgs.isEmpty()) {
997 appendAndIndexWikiCallback.builder.append("(");
998 appendAndIndexWikiCallback.dispatch(args.get(0), EntryTypeName.WIKTIONARY_TRANSLATION_OTHER_TEXT);
999 appendAndIndexWikiCallback.builder.append(")");
1004 // ------------------------------------------------------------------
1006 static final class gloss implements FunctionCallback<EnParser> {
1008 public boolean onWikiFunction(final WikiTokenizer wikiTokenizer, final String name, final List<String> args,
1009 final Map<String, String> namedArgs,
1010 final EnParser parser,
1011 final AppendAndIndexWikiCallback<EnParser> appendAndIndexWikiCallback) {
1012 if (args.size() != 1 || !namedArgs.isEmpty()) {
1015 appendAndIndexWikiCallback.builder.append("(");
1016 appendAndIndexWikiCallback.dispatch(args.get(0), EntryTypeName.WIKTIONARY_TRANSLATION_OTHER_TEXT);
1017 appendAndIndexWikiCallback.builder.append(")");
1022 // ------------------------------------------------------------------
1024 static final class Ignore implements FunctionCallback<EnParser> {
1026 public boolean onWikiFunction(final WikiTokenizer wikiTokenizer, final String name, final List<String> args,
1027 final Map<String, String> namedArgs,
1028 final EnParser parser,
1029 final AppendAndIndexWikiCallback<EnParser> appendAndIndexWikiCallback) {
1034 // ------------------------------------------------------------------
1036 static final class not_used implements FunctionCallback<EnParser> {
1038 public boolean onWikiFunction(final WikiTokenizer wikiTokenizer, final String name, final List<String> args,
1039 final Map<String, String> namedArgs,
1040 final EnParser parser,
1041 final AppendAndIndexWikiCallback<EnParser> appendAndIndexWikiCallback) {
1042 appendAndIndexWikiCallback.builder.append("(not used)");
1048 // ------------------------------------------------------------------
1050 static final class AppendName implements FunctionCallback<EnParser> {
1052 public boolean onWikiFunction(final WikiTokenizer wikiTokenizer, final String name, final List<String> args,
1053 final Map<String, String> namedArgs,
1054 final EnParser parser,
1055 final AppendAndIndexWikiCallback<EnParser> appendAndIndexWikiCallback) {
1056 if (!args.isEmpty() || !namedArgs.isEmpty()) {
1059 appendAndIndexWikiCallback.builder.append(name);
1064 // --------------------------------------------------------------------
1065 // --------------------------------------------------------------------
1068 static final class FormOf implements FunctionCallback<EnParser> {
1070 public boolean onWikiFunction(final WikiTokenizer wikiTokenizer, final String name, final List<String> args,
1071 final Map<String, String> namedArgs,
1072 final EnParser parser,
1073 final AppendAndIndexWikiCallback<EnParser> appendAndIndexWikiCallback) {
1074 parser.entryIsFormOfSomething = true;
1075 String formName = name;
1076 if (name.equals("form of")) {
1077 formName = ListUtil.remove(args, 0, null);
1079 if (formName == null) {
1080 LOG.warning("Missing form name: " + parser.title);
1081 formName = "form of";
1083 String baseForm = ListUtil.get(args, 1, "");
1084 if ("".equals(baseForm)) {
1085 baseForm = ListUtil.get(args, 0, null);
1086 ListUtil.remove(args, 1, "");
1088 ListUtil.remove(args, 0, null);
1090 namedArgs.keySet().removeAll(USELESS_WIKI_ARGS);
1092 appendAndIndexWikiCallback.builder.append("{");
1093 NAME_AND_ARGS.onWikiFunction(wikiTokenizer, formName, args, namedArgs, parser, appendAndIndexWikiCallback);
1094 appendAndIndexWikiCallback.builder.append("}");
1095 if (baseForm != null && appendAndIndexWikiCallback.indexedEntry != null) {
1096 parser.foreignIndexBuilder.addEntryWithString(appendAndIndexWikiCallback.indexedEntry, baseForm, EntryTypeName.WIKTIONARY_BASE_FORM_MULTI);
1098 // null baseForm happens in Danish.
1099 LOG.warning("Null baseform: " + parser.title);
1105 static final FormOf FORM_OF = new FormOf();
1108 // --------------------------------------------------------------------
1109 // --------------------------------------------------------------------
1111 static final class wikipedia implements FunctionCallback<EnParser> {
1113 public boolean onWikiFunction(final WikiTokenizer wikiTokenizer, final String name, final List<String> args,
1114 final Map<String, String> namedArgs,
1115 final EnParser parser,
1116 final AppendAndIndexWikiCallback<EnParser> appendAndIndexWikiCallback) {
1117 namedArgs.remove("lang");
1118 if (args.size() > 1 || !namedArgs.isEmpty()) {
1121 } else if (args.size() == 1) {
1129 static final class InflOrHead implements FunctionCallback<EnParser> {
1131 public boolean onWikiFunction(final WikiTokenizer wikiTokenizer, final String name, final List<String> args,
1132 final Map<String, String> namedArgs,
1133 final EnParser parser,
1134 final AppendAndIndexWikiCallback<EnParser> appendAndIndexWikiCallback) {
1135 // See: http://en.wiktionary.org/wiki/Template:infl
1136 final String langCode = ListUtil.get(args, 0);
1137 String head = namedArgs.remove("head");
1139 head = namedArgs.remove("title"); // Bug
1142 head = parser.title;
1144 parser.titleAppended = true;
1146 namedArgs.keySet().removeAll(USELESS_WIKI_ARGS);
1148 final String tr = namedArgs.remove("tr");
1149 String g = namedArgs.remove("g");
1151 g = namedArgs.remove("gender");
1153 final String g2 = namedArgs.remove("g2");
1154 final String g3 = namedArgs.remove("g3");
1156 appendAndIndexWikiCallback.dispatch(head, EntryTypeName.WIKTIONARY_TITLE_MULTI);
1159 appendAndIndexWikiCallback.builder.append(" {").append(g);
1161 appendAndIndexWikiCallback.builder.append("|").append(g2);
1164 appendAndIndexWikiCallback.builder.append("|").append(g3);
1166 appendAndIndexWikiCallback.builder.append("}");
1170 appendAndIndexWikiCallback.builder.append(" (");
1171 appendAndIndexWikiCallback.dispatch(tr, EntryTypeName.WIKTIONARY_TITLE_MULTI);
1172 appendAndIndexWikiCallback.builder.append(")");
1173 parser.wordForms.add(tr);
1176 final String pos = ListUtil.get(args, 1);
1178 appendAndIndexWikiCallback.builder.append(" (").append(pos).append(")");
1180 for (int i = 2; i < args.size(); i += 2) {
1181 final String inflName = ListUtil.get(args, i);
1182 final String inflValue = ListUtil.get(args, i + 1);
1183 appendAndIndexWikiCallback.builder.append(", ");
1184 appendAndIndexWikiCallback.dispatch(inflName, null, null);
1185 if (inflValue != null && inflValue.length() > 0) {
1186 appendAndIndexWikiCallback.builder.append(": ");
1187 appendAndIndexWikiCallback.dispatch(inflValue, null, null);
1188 parser.wordForms.add(inflValue);
1191 for (final String key : namedArgs.keySet()) {
1192 final String value = WikiTokenizer.toPlainText(namedArgs.get(key));
1193 appendAndIndexWikiCallback.builder.append(" ");
1194 appendAndIndexWikiCallback.dispatch(key, null, null);
1195 appendAndIndexWikiCallback.builder.append("=");
1196 appendAndIndexWikiCallback.dispatch(value, null, null);
1197 parser.wordForms.add(value);
1205 DEFAULT.put("it-noun", new it_noun());
1207 static final class it_noun implements FunctionCallback<EnParser> {
1209 public boolean onWikiFunction(final WikiTokenizer wikiTokenizer, final String name, final List<String> args,
1210 final Map<String, String> namedArgs,
1211 final EnParser parser,
1212 final AppendAndIndexWikiCallback<EnParser> appendAndIndexWikiCallback) {
1213 parser.titleAppended = true;
1214 final String base = ListUtil.get(args, 0);
1215 final String gender = ListUtil.get(args, 1);
1216 final String singular = base + ListUtil.get(args, 2, null);
1217 final String plural = base + ListUtil.get(args, 3, null);
1218 appendAndIndexWikiCallback.builder.append(" ");
1219 appendAndIndexWikiCallback.dispatch(singular, null, null);
1220 appendAndIndexWikiCallback.builder.append(" {").append(gender).append("}, ");
1221 appendAndIndexWikiCallback.dispatch(plural, null, null);
1222 appendAndIndexWikiCallback.builder.append(" {pl}");
1223 parser.wordForms.add(singular);
1224 parser.wordForms.add(plural);
1225 if (!namedArgs.isEmpty() || args.size() > 4) {
1226 LOG.warning("Invalid it-noun: " + wikiTokenizer.token());
1233 DEFAULT.put("it-proper noun", new it_proper_noun());
1235 static final class it_proper_noun implements FunctionCallback<EnParser> {
1237 public boolean onWikiFunction(final WikiTokenizer wikiTokenizer, final String name, final List<String> args,
1238 final Map<String, String> namedArgs,
1239 final EnParser parser,
1240 final AppendAndIndexWikiCallback<EnParser> appendAndIndexWikiCallback) {