Question

I meet the following code in a book and still not understand why. Anyone can help to explain it please?

[root@wd00070319 test]# ls
[root@wd00070319 test]# touch file1
[root@wd00070319 test]# ls
file1
[root@wd00070319 test]# ls > file2
[root@wd00070319 test]# ls
file1  file2
[root@wd00070319 test]# cat file2
file1
file2
[root@wd00070319 test]#

Answer 1

[root@wd00070319 test]# ls            # list the files. (directory is empty!)
[root@wd00070319 test]# touch file1   # create an empty file called "file1"
[root@wd00070319 test]# ls            # list the files again (to see "file1")
file1                                 
[root@wd00070319 test]# ls > file2    # list the files, put the result in "file2"
[root@wd00070319 test]# ls            # list the files again,
file1  file2
[root@wd00070319 test]# cat file2     # show content of file2
file1
file2                                 # <-- Note that "file2" is present as well.
[root@wd00070319 test]#

The lesson here is probably that a command like ls > file2 creates the output-file (file2) before actually executing the ls command.

---

This behavior is confirmed by the bash reference manual:

3.6 Redirections

[…] Before a command is executed, its input and output may be redirected
using a special notation interpreted by the shell. […]