projects / old-projects.git / blame_incremental
? search:
summary | shortlog | log | commit | commitdiff | tree
blob | blame (non-incremental) | history | HEAD
handle require_once and include_once, allow single quotes
[old-projects.git] / ekit / com / swabunga / spell / event / SpellChecker.java
... / ...
CommitLineData
1package com.swabunga.spell.event;\r\r
2\r\r
3import com.swabunga.spell.engine.*;\r\r
4import java.util.*;\r\r
5\r\r
6/**\r\r
7 * This is the main class for spell checking (using the new event based spell\r\r
8 * checking).\r\r
9 *\r\r
10 * @author Jason Height (jheight@chariot.net.au)\r\r
11 * @created 19 June 2002\r\r
12 */\r\r
13public class SpellChecker {\r\r
14 /** Flag indicating that the Spell Check completed without any errors present*/\r\r
15 public static final int SPELLCHECK_OK=-1;\r\r
16 /** Flag indicating that the Spell Check completed due to user cancellation*/\r\r
17 public static final int SPELLCHECK_CANCEL=-2;\r\r
18\r\r
19 private List eventListeners = new ArrayList();\r\r
20 private SpellDictionary dictionary;\r\r
21 \r\r
22 private Configuration config = Configuration.getConfiguration();\r\r
23\r\r
24 /**This variable holds all of the words that are to be always ignored */\r\r
25 private Set ignoredWords = new HashSet();\r\r
26 private Map autoReplaceWords = new HashMap();\r\r
27\r\r
28\r\r
29 /**\r\r
30 * Constructs the SpellChecker. The default threshold is used\r\r
31 *\r\r
32 * @param dictionary Description of the Parameter\r\r
33 */\r\r
34 public SpellChecker(SpellDictionary dictionary) {\r\r
35 if (dictionary == null) {\r\r
36 throw new IllegalArgumentException("dictionary must non-null");\r\r
37 }\r\r
38 this.dictionary = dictionary;\r\r
39 }\r\r
40\r\r
41\r\r
42 /**\r\r
43 * Constructs the SpellChecker with a threshold\r\r
44 *\r\r
45 * @param dictionary Description of the Parameter\r\r
46 * @param threshold Description of the Parameter\r\r
47 */\r\r
48 public SpellChecker(SpellDictionary dictionary, int threshold) {\r\r
49 this(dictionary);\r\r
50 config.setInteger( Configuration.SPELL_THRESHOLD, threshold );\r\r
51 }\r\r
52\r\r
53\r\r
54 /**\r\r
55 *Adds a SpellCheckListener\r\r
56 *\r\r
57 * @param listener The feature to be added to the SpellCheckListener attribute\r\r
58 */\r\r
59 public void addSpellCheckListener(SpellCheckListener listener) {\r\r
60 eventListeners.add(listener);\r\r
61 }\r\r
62\r\r
63\r\r
64 /**\r\r
65 *Removes a SpellCheckListener\r\r
66 *\r\r
67 * @param listener Description of the Parameter\r\r
68 */\r\r
69 public void removeSpellCheckListener(SpellCheckListener listener) {\r\r
70 eventListeners.remove(listener);\r\r
71 }\r\r
72\r\r
73\r\r
74 /**\r\r
75 * Fires off a spell check event to the listeners.\r\r
76 *\r\r
77 * @param event Description of the Parameter\r\r
78 */\r\r
79 protected void fireSpellCheckEvent(SpellCheckEvent event) {\r\r
80 for (int i = eventListeners.size() - 1; i >= 0; i--) {\r\r
81 ((SpellCheckListener) eventListeners.get(i)).spellingError(event);\r\r
82 }\r\r
83 }\r\r
84\r\r
85\r\r
86 /**\r\r
87 * This method clears the words that are currently being remembered as\r\r
88 * Ignore All words and Replace All words.\r\r
89 */\r\r
90 public void reset() {\r\r
91 ignoredWords.clear();\r\r
92 autoReplaceWords.clear();\r\r
93 }\r\r
94\r\r
95\r\r
96 /**\r\r
97 * Checks the text string.\r\r
98 * <p>\r\r
99 * Returns the corrected string.\r\r
100 *\r\r
101 * @param text Description of the Parameter\r\r
102 * @return Description of the Return Value\r\r
103 * @deprecated use checkSpelling(WordTokenizer)\r\r
104 */\r\r
105 public String checkString(String text) {\r\r
106 StringWordTokenizer tokens = new StringWordTokenizer(text);\r\r
107 checkSpelling(tokens);\r\r
108 return tokens.getFinalText();\r\r
109 }\r\r
110\r\r
111\r\r
112 /**\r\r
113 * Returns true iif this word contains a digit\r\r
114 *\r\r
115 * @param word Description of the Parameter\r\r
116 * @return The digitWord value\r\r
117 */\r\r
118 private final static boolean isDigitWord(String word) {\r\r
119 for (int i = word.length() - 1; i >= 0; i--) {\r\r
120 if (Character.isDigit(word.charAt(i))) {\r\r
121 return true;\r\r
122 }\r\r
123 }\r\r
124 return false;\r\r
125 }\r\r
126\r\r
127\r\r
128 /**\r\r
129 * Returns true iif this word looks like an internet address\r\r
130 *\r\r
131 * @param word Description of the Parameter\r\r
132 * @return The iNETWord value\r\r
133 */\r\r
134 private final static boolean isINETWord(String word) {\r\r
135 //JMH TBD\r\r
136 return false;\r\r
137 }\r\r
138\r\r
139\r\r
140 /**\r\r
141 * Returns true iif this word contains all upper case characters\r\r
142 *\r\r
143 * @param word Description of the Parameter\r\r
144 * @return The upperCaseWord value\r\r
145 */\r\r
146 private final static boolean isUpperCaseWord(String word) {\r\r
147 for (int i = word.length() - 1; i >= 0; i--) {\r\r
148 if (Character.isLowerCase(word.charAt(i))) {\r\r
149 return false;\r\r
150 }\r\r
151 }\r\r
152 return true;\r\r
153 }\r\r
154\r\r
155\r\r
156 /**\r\r
157 * Returns true iif this word contains mixed case characters\r\r
158 *\r\r
159 * @param word Description of the Parameter\r\r
160 * @param startsSentance True if this word is at the start of a sentance\r\r
161 * @return The mixedCaseWord value\r\r
162 */\r\r
163 private final static boolean isMixedCaseWord(String word, boolean startsSentance) {\r\r
164 int strLen = word.length();\r\r
165 boolean isUpper = Character.isUpperCase(word.charAt(0));\r\r
166 //Ignore the first character if this word starts the sentance and the first\r\r
167 //character was upper cased, since this is normal behaviour\r\r
168 if ((startsSentance) && isUpper && (strLen > 1))\r\r
169 isUpper = Character.isUpperCase(word.charAt(1));\r\r
170 if (isUpper) {\r\r
171 for (int i = word.length() - 1; i > 0; i--) {\r\r
172 if (Character.isLowerCase(word.charAt(i))) {\r\r
173 return true;\r\r
174 }\r\r
175 }\r\r
176 } else {\r\r
177 for (int i = word.length() - 1; i > 0; i--) {\r\r
178 if (Character.isUpperCase(word.charAt(i))) {\r\r
179 return true;\r\r
180 }\r\r
181 }\r\r
182 }\r\r
183 return false;\r\r
184 }\r\r
185\r\r
186\r\r
187 /**\r\r
188 * This method will fire the spell check event and then handle the event\r\r
189 * action that has been selected by the user.\r\r
190 *\r\r
191 * @param tokenizer Description of the Parameter\r\r
192 * @param event Description of the Parameter\r\r
193 * @return Returns true if the event action is to cancel the current spell checking, false if the spell checking should continue\r\r
194 */\r\r
195 protected boolean fireAndHandleEvent(WordTokenizer tokenizer, SpellCheckEvent event) {\r\r
196 fireSpellCheckEvent(event);\r\r
197 String word = event.getInvalidWord();\r\r
198 //Work out what to do in response to the event.\r\r
199 switch (event.getAction()) {\r\r
200 case SpellCheckEvent.INITIAL:\r\r
201 break;\r\r
202 case SpellCheckEvent.IGNORE:\r\r
203 break;\r\r
204 case SpellCheckEvent.IGNOREALL:\r\r
205 if (!ignoredWords.contains(word)) {\r\r
206 ignoredWords.add(word);\r\r
207 }\r\r
208 break;\r\r
209 case SpellCheckEvent.REPLACE:\r\r
210 tokenizer.replaceWord(event.getReplaceWord());\r\r
211 break;\r\r
212 case SpellCheckEvent.REPLACEALL:\r\r
213 String replaceAllWord = event.getReplaceWord();\r\r
214 if (!autoReplaceWords.containsKey(word)) {\r\r
215 autoReplaceWords.put(word, replaceAllWord);\r\r
216 }\r\r
217 tokenizer.replaceWord(replaceAllWord);\r\r
218 break;\r\r
219 case SpellCheckEvent.ADDTODICT:\r\r
220 String addWord = event.getReplaceWord();\r\r
221 tokenizer.replaceWord(addWord);\r\r
222 dictionary.addWord(addWord);\r\r
223 break;\r\r
224 case SpellCheckEvent.CANCEL:\r\r
225 return true;\r\r
226 default:\r\r
227 throw new IllegalArgumentException("Unhandled case.");\r\r
228 }\r\r
229 return false;\r\r
230 }\r\r
231\r\r
232\r\r
233 /**\r\r
234 * This method is called to check the spelling of the words that are returned\r\r
235 * by the WordTokenizer.\r\r
236 * <p>For each invalid word the action listeners will be informed with a new SpellCheckEvent</p>\r\r
237 *\r\r
238 * @param tokenizer Description of the Parameter\r\r
239 * @return Either SPELLCHECK_OK, SPELLCHECK_CANCEL or the number of errors found. The number of errors are those that are found BEFORE and corretions are made.\r\r
240 */\r\r
241 public final int checkSpelling(WordTokenizer tokenizer) {\r\r
242 int errors = 0;\r\r
243 boolean terminated = false;\r\r
244 //Keep track of the previous word\r\r
245 String previousWord = null;\r\r
246 while (tokenizer.hasMoreWords() && !terminated) {\r\r
247 String word = tokenizer.nextWord();\r\r
248 //Check the spelling of the word\r\r
249 if (!dictionary.isCorrect(word)) {\r\r
250 if (\r\r
251 (config.getBoolean(Configuration.SPELL_IGNOREMIXEDCASE) && isMixedCaseWord(word, tokenizer.isNewSentance())) ||\r\r
252 (config.getBoolean(Configuration.SPELL_IGNOREUPPERCASE) && isUpperCaseWord(word)) ||\r\r
253 (config.getBoolean(Configuration.SPELL_IGNOREDIGITWORDS) && isDigitWord(word)) ||\r\r
254 (config.getBoolean(Configuration.SPELL_IGNOREINTERNETADDRESSES) && isINETWord(word))) {\r\r
255 //Null event. Since we are ignoring this word due\r\r
256 //to one of the above cases.\r\r
257 } else {\r\r
258 //We cant ignore this misspelt word\r\r
259 //For this invalid word are we ignoreing the misspelling?\r\r
260 if (!ignoredWords.contains(word)) {\r\r
261 errors++;\r\r
262 //Is this word being automagically replaced\r\r
263 if (autoReplaceWords.containsKey(word)) {\r\r
264 tokenizer.replaceWord((String) autoReplaceWords.get(word));\r\r
265 } else {\r\r
266 //JMH Need to somehow capitalise the suggestions if\r\r
267 //ignoreSentanceCapitalisation is not set to true\r\r
268 //Fire the event.\r\r
269 SpellCheckEvent event = new BasicSpellCheckEvent(word, dictionary.getSuggestions(word,\r\r
270 config.getInteger(Configuration.SPELL_THRESHOLD)), tokenizer);\r\r
271 terminated = fireAndHandleEvent(tokenizer, event);\r\r
272 }\r\r
273 }\r\r
274 }\r\r
275 } else {\r\r
276 //This is a correctly spelt word. However perform some extra checks\r\r
277 /*\r\r
278 * JMH TBD //Check for multiple words\r\r
279 * if (!ignoreMultipleWords &&) {\r\r
280 * }\r\r
281 */\r\r
282 //Check for capitalisation\r\r
283 if ((!config.getBoolean(Configuration.SPELL_IGNORESENTANCECAPITALIZATION)) && (tokenizer.isNewSentance())\r\r
284 && (Character.isLowerCase(word.charAt(0)))) {\r\r
285 errors++;\r\r
286 StringBuffer buf = new StringBuffer(word);\r\r
287 buf.setCharAt(0, Character.toUpperCase(word.charAt(0)));\r\r
288 List suggestion = new LinkedList();\r\r
289 suggestion.add(new Word(buf.toString(), 0));\r\r
290 SpellCheckEvent event = new BasicSpellCheckEvent(word, suggestion,\r\r
291 tokenizer);\r\r
292 terminated = fireAndHandleEvent(tokenizer, event);\r\r
293 }\r\r
294 }\r\r
295 }\r\r
296 if (terminated)\r\r
297 return SPELLCHECK_CANCEL;\r\r
298 else if (errors == 0)\r\r
299 return SPELLCHECK_OK;\r\r
300 else return errors;\r\r
301 }\r\r
302}\r\r
303\r\r
304\r\r
Unused projects