图解云服务模型的演进
Java 8 stream特性是一个能快速降低开发人员工作量的语法糖,用起来很简单,用好了很难。这里就通过一系列的博客对几个常见的错误进行解释说明,并给出替代方法。这里先说明串行和终止操作。
首先,给出IBM官网给出的介绍,请仔细阅读。
https://www.ibm.com/developerworks/cn/java/j-lo-java8streamapi/
如何熟悉一个系统?(内含知识大图)
流的操作类型分为两种:
- Intermediate:一个流可以后面跟随零个或多个 intermediate 操作。其目的主要是打开流,做出某种程度的数据映射/过滤,然后返回一个新的流,交给下一个操作使用。这类操作都是惰性化的(lazy),就是说,仅仅调用到这类方法,并没有真正开始流的遍历。
- Terminal:一个流只能有一个 terminal 操作,当这个操作执行后,流就被使用“光”了,无法再被操作。所以这必定是流的最后一个操作。Terminal 操作的执行,才会真正开始流的遍历,并且会生成一个结果,或者一个 side effect。
在对于一个 Stream 进行多次转换操作 (Intermediate 操作),每次都对 Stream 的每个元素进行转换,而且是执行多次,这样时间复杂度就是 N(转换次数)个 for 循环里把所有操作都做掉的总和吗?其实不是这样的,转换操作都是 lazy 的,多个转换操作只会在 Terminal 操作的时候融合起来,一次循环完成。我们可以这样简单的理解,Stream 里有个操作函数的集合,每次转换操作就是把转换函数放入这个集合中,在 Terminal 操作的时候循环 Stream 对应的集合,然后对每个元素执行所有的函数。
这里给出一个例子:
- 创建一个整型stream,对1到10的流进行处理,分别将流中的数值乘以10、加上5,得到一个Stream 对象integerStream。
- 对integerStream进行collect,得到一个List 对象integerList。
- 输出integerList。
private static void testSerial() {
Integer[] integers = new Integer[]{1,2,3,4,5,6,7,8,9,10};
logger.info("1. create a integer stream, without terminal operation");
Stream<Integer> integerStream = Arrays.asList(integers).stream()
.map(integer->{
logger.info("multiply {} with 10", integer);
return integer * 10;
})
.map(integer -> {
logger.info("add {} with 5", integer);
return integer + 5;
});
logger.info("2. add a terminal operation to stream");
List<Integer> integerList = integerStream.collect(Collectors.toList());
logger.info("3. result {}", integerList);
}
根据下面的日志,可以确定,在第一步创建Stream对象时, intermediate 操作并没有被立即执行,在第二步执行 terminal 操作,Stream中的所有步骤才被执行。而且第一步中创建的两个intermediate 操作,会依次处理同一个数据,也就是说对于所有数据,先被第一个map操作处理,处理结果再被第二个map操作处理,然后被收集到List中。
[20:53:52:765] [INFO] - [main] - com.wangdan.practice.stream.PracticeMain.testSerial(PracticeMain.java:19) - 1. create a integer stream, without terminal operation
[20:53:52:821] [INFO] - [main] - com.wangdan.practice.stream.PracticeMain.testSerial(PracticeMain.java:29) - 2. add a terminal operation to stream
[20:53:52:827] [INFO] - [main] - com.wangdan.practice.stream.PracticeMain.lambda$testSerial$0(PracticeMain.java:22) - multiply 1 with 10
[20:53:52:829] [INFO] - [main] - com.wangdan.practice.stream.PracticeMain.lambda$testSerial$1(PracticeMain.java:26) - add 10 with 5
[20:53:52:829] [INFO] - [main] - com.wangdan.practice.stream.PracticeMain.lambda$testSerial$0(PracticeMain.java:22) - multiply 2 with 10
[20:53:52:829] [INFO] - [main] - com.wangdan.practice.stream.PracticeMain.lambda$testSerial$1(PracticeMain.java:26) - add 20 with 5
[20:53:52:829] [INFO] - [main] - com.wangdan.practice.stream.PracticeMain.lambda$testSerial$0(PracticeMain.java:22) - multiply 3 with 10
[20:53:52:829] [INFO] - [main] - com.wangdan.practice.stream.PracticeMain.lambda$testSerial$1(PracticeMain.java:26) - add 30 with 5
[20:53:52:830] [INFO] - [main] - com.wangdan.practice.stream.PracticeMain.lambda$testSerial$0(PracticeMain.java:22) - multiply 4 with 10
[20:53:52:830] [INFO] - [main] - com.wangdan.practice.stream.PracticeMain.lambda$testSerial$1(PracticeMain.java:26) - add 40 with 5
[20:53:52:830] [INFO] - [main] - com.wangdan.practice.stream.PracticeMain.lambda$testSerial$0(PracticeMain.java:22) - multiply 5 with 10
[20:53:52:830] [INFO] - [main] - com.wangdan.practice.stream.PracticeMain.lambda$testSerial$1(PracticeMain.java:26) - add 50 with 5
[20:53:52:830] [INFO] - [main] - com.wangdan.practice.stream.PracticeMain.lambda$testSerial$0(PracticeMain.java:22) - multiply 6 with 10
[20:53:52:830] [INFO] - [main] - com.wangdan.practice.stream.PracticeMain.lambda$testSerial$1(PracticeMain.java:26) - add 60 with 5
[20:53:52:830] [INFO] - [main] - com.wangdan.practice.stream.PracticeMain.lambda$testSerial$0(PracticeMain.java:22) - multiply 7 with 10
[20:53:52:831] [INFO] - [main] - com.wangdan.practice.stream.PracticeMain.lambda$testSerial$1(PracticeMain.java:26) - add 70 with 5
[20:53:52:832] [INFO] - [main] - com.wangdan.practice.stream.PracticeMain.lambda$testSerial$0(PracticeMain.java:22) - multiply 8 with 10
[20:53:52:832] [INFO] - [main] - com.wangdan.practice.stream.PracticeMain.lambda$testSerial$1(PracticeMain.java:26) - add 80 with 5
[20:53:52:833] [INFO] - [main] - com.wangdan.practice.stream.PracticeMain.lambda$testSerial$0(PracticeMain.java:22) - multiply 9 with 10
[20:53:52:833] [INFO] - [main] - com.wangdan.practice.stream.PracticeMain.lambda$testSerial$1(PracticeMain.java:26) - add 90 with 5
[20:53:52:833] [INFO] - [main] - com.wangdan.practice.stream.PracticeMain.lambda$testSerial$0(PracticeMain.java:22) - multiply 10 with 10
[20:53:52:833] [INFO] - [main] - com.wangdan.practice.stream.PracticeMain.lambda$testSerial$1(PracticeMain.java:26) - add 100 with 5
[20:53:52:833] [INFO] - [main] - com.wangdan.practice.stream.PracticeMain.testSerial(PracticeMain.java:31) - 3. result [15, 25, 35, 45, 55, 65, 75, 85, 95, 105]
Redis(二):命令集构建及关键属性解析