background

One day , I wrote the following line of comment in my code , What surprised me was that the code in the comment was executed .

According to the conventional idea , This line of code is annotated , The console should not have execution results , However, the following results are returned .

Come here , We can probably guess that Unicode Decoding occurs before any word decoding . and \u000d It's a newline , Therefore, the comment is terminated, resulting in the execution of the comment code after the newline character .

What is? Unicode The escape ？

I went to oracle Have a look at the official website Java language norm （JLS 3） Related explanation , The main idea is as follows ：Unicode Escape is used to indicate that only ASCII Character Unicode Symbol . When you need to insert characters that cannot be represented in the character set of the source file , It will come in handy .JLS 3.3 Relevant instructions in section ,Unicode The escape contains a backslash character （\）, One or more heels 'u' Characters and four hexadecimal digits .

therefore , In the example \u000d Will be treated as a newline .

The figure below shows the official JLS file .

The advantage of this mechanism is that it can be used in ASCII And any other code , And you don't need to figure out where the comments start and end ！

Analysis and verification

To confirm whether it's what we analyzed , I used it Java Built-in tools native2ascii To convert a file with any supported character encoding to a file with ASCII or Unicode Escaped file .

Execute the following command at a glance .

I can find that the converted code has been wrapped ！

Actually , I can also check class Bytecode to find the clue .

It seems to be the same as what we analyzed before .

summary

This operation also ensures Java The core idea —— Platform consistency .

Although the processing mechanism in this way seems elegant , however , It has side effects （ Interference semantics ）, Especially in the comments , We must pay attention to ！

Okay , Leave a question for you to think about , Create a new one Hi.java file , The following Unicode Copy code to file , See what the execution result will be ？