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.enwiktionary;
17 import java.util.LinkedHashMap;
18 import java.util.List;
20 import java.util.concurrent.atomic.AtomicInteger;
22 import com.hughes.android.dictionary.engine.EntryTypeName;
23 import com.hughes.android.dictionary.engine.IndexBuilder;
24 import com.hughes.android.dictionary.engine.IndexedEntry;
25 import com.hughes.android.dictionary.parser.WikiTokenizer;
26 import com.hughes.util.EnumUtil;
28 final class AppendAndIndexWikiCallback implements WikiTokenizer.Callback {
30 final EnWiktionaryXmlParser parser;
31 StringBuilder builder;
32 IndexedEntry indexedEntry;
33 IndexBuilder indexBuilder;
34 final Map<String,FunctionCallback> functionCallbacks = new LinkedHashMap<String, FunctionCallback>();
36 boolean entryTypeNameSticks = false;
37 EntryTypeName entryTypeName = null;
39 final Map<String,AtomicInteger> langCodeToTCount = new LinkedHashMap<String, AtomicInteger>();
41 public AppendAndIndexWikiCallback(final EnWiktionaryXmlParser parser) {
45 public void reset(final StringBuilder builder, final IndexedEntry indexedEntry) {
46 this.builder = builder;
47 this.indexedEntry = indexedEntry;
48 this.indexBuilder = null;
50 entryTypeNameSticks = false;
53 public void dispatch(final String wikiText, final IndexBuilder indexBuilder, final EntryTypeName entryTypeName) {
54 final IndexBuilder oldIndexBuilder = this.indexBuilder;
55 final EntryTypeName oldEntryTypeName = this.entryTypeName;
56 this.indexBuilder = indexBuilder;
57 if (!entryTypeNameSticks) {
58 this.entryTypeName = EnumUtil.min(entryTypeName, this.entryTypeName);
60 if (entryTypeName == null) this.entryTypeName = null;
61 WikiTokenizer.dispatch(wikiText, false, this);
62 this.indexBuilder = oldIndexBuilder;
63 this.entryTypeName = oldEntryTypeName;
66 public void dispatch(final String wikiText, final EntryTypeName entryTypeName) {
67 dispatch(wikiText, this.indexBuilder, entryTypeName);
72 public void onPlainText(final String plainText) {
73 // The only non-recursive callback. Just appends to the builder, and indexes.
74 builder.append(plainText);
75 if (indexBuilder != null && entryTypeName != null && indexedEntry != null) {
76 indexBuilder.addEntryWithString(indexedEntry, plainText, entryTypeName);
81 public void onWikiLink(WikiTokenizer wikiTokenizer) {
82 final String text = wikiTokenizer.wikiLinkText();
83 final String link = wikiTokenizer.wikiLinkDest();
85 if (link.contains("#English")) {
86 dispatch(text, parser.enIndexBuilder, EntryTypeName.WIKTIONARY_ENGLISH_DEF_WIKI_LINK);
87 } else if (link.contains("#") && parser.langPattern.matcher(link).find()) {
88 dispatch(text, parser.foreignIndexBuilder, EntryTypeName.WIKTIONARY_ENGLISH_DEF_OTHER_LANG);
89 } else if (link.equals("plural")) {
92 //LOG.warning("Special link: " + englishTokenizer.token());
93 dispatch(text, EntryTypeName.WIKTIONARY_ENGLISH_DEF_WIKI_LINK);
97 final EntryTypeName entryTypeName;
98 switch (parser.state) {
99 case TRANSLATION_LINE:
100 entryTypeName = EntryTypeName.WIKTIONARY_TRANSLATION_WIKI_TEXT;
102 case ENGLISH_DEF_OF_FOREIGN:
103 entryTypeName = EntryTypeName.WIKTIONARY_ENGLISH_DEF_WIKI_LINK;
106 throw new IllegalStateException("Invalid enum value: " + parser.state);
108 dispatch(text, entryTypeName);
113 public void onFunction(
114 final WikiTokenizer wikiTokenizer,
116 final List<String> args,
117 final Map<String, String> namedArgs) {
119 FunctionCallback functionCallback = functionCallbacks.get(name);
120 if (functionCallback == null || !functionCallback.onWikiFunction(wikiTokenizer, name, args, namedArgs, parser, this)) {
121 // Default function handling:
122 namedArgs.keySet().removeAll(EnWiktionaryXmlParser.USELESS_WIKI_ARGS);
123 final boolean single = args.isEmpty() && namedArgs.isEmpty();
124 builder.append(single ? "{" : "{{");
126 final IndexBuilder oldIndexBuilder = indexBuilder;
128 FunctionCallbacksDefault.NAME_AND_ARGS.onWikiFunction(wikiTokenizer, name, args, namedArgs, parser, this);
129 indexBuilder = oldIndexBuilder;
131 builder.append(single ? "}" : "}}");
136 public void onHtml(WikiTokenizer wikiTokenizer) {
137 // Unindexed for now.
138 builder.append(wikiTokenizer.token());
142 public void onMarkup(WikiTokenizer wikiTokenizer) {
147 public void onComment(WikiTokenizer wikiTokenizer) {
152 public void onNewline(WikiTokenizer wikiTokenizer) {
157 public void onHeading(WikiTokenizer wikiTokenizer) {
162 public void onListItem(WikiTokenizer wikiTokenizer) {