在上一次修复之后我现在有了一个新问题。建议问一个新问题,所以在这里! 它仍然强制关闭,但现在又开始了。
public void AI(String text) {
// initiate new lists to be on the safe side
List<String> indexer = new ArrayList<String>();
List<String> sentence = new ArrayList<String>();
List<String> explode = new ArrayList<String>();
List<String> ary = new ArrayList<String>();
explode = Arrays.asList(text.split(" "));
// initiate randint and trigger variable
Random rand = new Random();
int randint = rand.nextInt(explode.size());
// initiate the trigger variable
String trigger = explode.get(randint);
//check if word exists in database and add if it not.
int i = 0;
for(String word : explode) {
if(common.get(word) == null) {
words.add(word);
common.put(word, 1);
context.put(word, explode);
pos.put(word, i);
length.put(word, explode.size());
if(word.equals(trigger)) {
ary = new ArrayList<String>(explode);
}
}else{
common.put(word, common.get(word)+1 );
}
i++;
}
// fill the arraylist sentence with words to be used, some in context, some random from the database.
for(int i2 = 0; i2 < length.get(trigger); i2++ ) {
randint = rand.nextInt(length.get(trigger));
if(randint < length.get(trigger)/2) {
//
if(ary.get(i2)!=null) {
sentence.add(ary.get(i2));
}
}else{
sentence.add(words.get(rand.nextInt(words.size())));
}
}
// use to the pos-hashmap to check at which index the word was detected at, if not place it at the deliberate index.
for(int i3 = 0; i3 < sentence.size(); i3++) {
if(sentence.get(i3)!=null) {
indexer.add(pos.get(sentence.get(i3)), sentence.get(i3));
}
}
// compose the final string that is to be passed to the speak function
for(int i4 = 0; i4 < indexer.size(); i4++) {
say = say + indexer.get(i4)+" ";
}
// pass the string to the speak function
mTts.speak(say, TextToSpeech.QUEUE_FLUSH, null);
// removing final string to be ready for next iteration
say = "";
// end of AI stuff
}
变量:
// arraylists
public List<String> explode = new ArrayList<String>();
public List<String> words = new ArrayList<String>();
public List<String> sentence = new ArrayList<String>();
public List<String> indexer = new ArrayList<String>();
// hashmaps
public Map<String, Integer> common = new HashMap<String, Integer>();
public Map<String, List<String>> context = new HashMap<String, List<String>>();
public Map<String, Integer> pos = new HashMap<String, Integer>();
public Map<String, Integer> length = new HashMap<String, Integer>();
//strings
public String say;
和logcat:
05-26 19:24:42.926: W/dalvikvm(543): threadid=3: thread exiting with uncaught exception (group=0x4001b188)
05-26 19:24:42.926: E/AndroidRuntime(543): Uncaught handler: thread main exiting due to uncaught exception
05-26 19:24:42.956: E/AndroidRuntime(543): java.lang.IndexOutOfBoundsException: Invalid location 0, size is 0
05-26 19:24:42.956: E/AndroidRuntime(543): at java.util.ArrayList.get(ArrayList.java:341)
05-26 19:24:42.956: E/AndroidRuntime(543): at sarah.namespace.SarahActivity.AI(SarahActivity.java:186)
第二个logcat:
05-26 21:38:44.555: W/dalvikvm(571): threadid=3: thread exiting with uncaught exception (group=0x4001b188)
05-26 21:38:44.555: E/AndroidRuntime(571): Uncaught handler: thread main exiting due to uncaught exception
05-26 21:38:44.575: E/AndroidRuntime(571): java.lang.IndexOutOfBoundsException
05-26 21:38:44.575: E/AndroidRuntime(571): at java.util.ArrayList.add(ArrayList.java:145)
05-26 21:38:44.575: E/AndroidRuntime(571): at sarah.namespace.SarahActivity.AI(SarahActivity.java:188)
logcats在尖叫着:
首先:
sentence.add(words.get(rand.nextInt(words.size())));
第二
for(int i3 = 0; i3 < sentence.size(); i3++) {
if(sentence.get(i3)!=null) {
indexer.add(pos.get(sentence.get(i3)), sentence.get(i3));
}
}
如果有人能帮助我再次击败这个谜语,我会很高兴的!
答案 0 :(得分:0)
嗯,我不确定哪些行是186和188(两个抛出OutOfBounds异常)。但是看看这个循环,我有一些建议:
for(int i2 = 0; i2 < length.get(trigger); i2++ ) {
从上面我们注意到length.get(trigger)
始终设置为explode.size()
,我们可以看到explode.size()
永远不会改变。我们可以检查explode.size()
或者再次使用for-each循环。继续:
randint = rand.nextInt(length.get(trigger));
if(randint < length.get(trigger)/2) {
这看起来你想要50/50的机会:将原始单词添加到sentence
或
if(ary.get(i2)!=null) {
sentence.add(ary.get(i2));
}
}else{
sentence.add(words.get(rand.nextInt(words.size())));
}
}
... sentence
的随机词。
我也相信这可能会引发异常:
if(ary.get(i2)!=null) {
sentence.add(ary.get(i2));
因为您正在检查i2
对length.get(trigger)
而不是ary.size()
的有效性。现在请记住,ary
只是explode
的副本。
所以让我们将这个循环浓缩为:
for(String word : explode) {
if(rand.nextInt(1)) // 50/50 chance of being false or true (0 or 1)
sentence.add(word);
else
sentence.add(words.get(rand.nextInt(words.size())));
}
看看剩余的两个循环,如果你应用一个类似的for-each循环,你将能够压缩并加速这些循环。我不知道应用程序代码的其余部分,但您可以完全删除一些列表和地图。干杯!