我有一个txt文件,我需要将行的每个元素(字符串,空格)放到数组中。
示例:
1行"嘿,1天\ n"
2 /// etc
第一行的输出:[" H" " E" " Y" "" " " " 1" " d" "" " Y"]
这样做的最佳方式是什么?使用split和string还是可以在char类型上运行?
答案 0 :(得分:2)
为什么不在字符串上调用String#toCharArray
?
如果您使用nextLine
,读者将截断换行符。
类似的东西:
try(BufferedReader r = new BufferedReader(new FileReader("topSecretDocs.txt"))) {
for(String line = r.readLine(); line != null; line = r.readLine()) {
char[] chars = line.toCharArray();
System.out.println("Split line to " + Arrays.toString(chars));
}
} catch(IOException e) {
dealWithException();
}
编辑请注意,您可能希望使用String#charAt
来遍历字符串,因为无需复制字符数组,尤其是对于非常大的字符串。
编辑2:这是一个非常糟糕的基准。
static long time(Runnable r) {
long start = System.nanoTime();
r.run();
return System.nanoTime() - start;
}
static int cs = 0;
public static void main(String[] args) {
char[] bigString = new char[100000];
Arrays.fill(bigString, 0, bigString.length / 4, 'A');
Arrays.fill(bigString, bigString.length / 4, bigString.length / 2, 'B');
Arrays.fill(bigString, bigString.length / 2, bigString.length * 3 / 4, 'C');
Arrays.fill(bigString, bigString.length * 3 / 4, bigString.length, 'D');
String s = new String(bigString);
float avgCopy = 0, avgLoop = 0;
final int times = 100000;
for(int i = 0; i < times; i++) {
avgLoop += time(() -> {
cs = 0;
for(int j = 0; j < s.length(); j++)
cs += s.charAt(j) == 'C' ? 1 : 0;
});
} avgLoop /= times;
for(int i = 0; i < times; i++) {
avgCopy += time(() -> {
char[] chars = s.toCharArray();
cs = 0;
for(char c : chars)
cs += c == 'C' ? 1 : 0;
});
} avgCopy /= times;
System.out.println("copy: " + avgCopy + " ns");
System.out.println("loop: " + avgLoop + " ns");
System.out.println("There were (obviously) " + cs + " Cs in that string.");
}
我明白了:
copy: 70984.336 ns
loop: 40534.63 ns
所以,快一半左右。无论您是否愿意称这是一个重要的差异,我都会留给您。