env: python\r: No such file or directory
Solution 1:
Open the file in vim
or vi
, and administer the following command:
:set ff=unix
Save and exit:
:wq
Done!
Explanation
ff
stands for file format, and can accept the values of unix
(\n
), dos
(\r\n
) and mac
(\r
) (only meant to be used on pre-intel macs, on modern macs use unix
).
To read more about the ff
command:
:help ff
:wq
stands for Write and Quit, a faster equivalent is Shift+zz (i.e. hold down Shift then press z
twice).
Both commands must be used in command mode.
Usage on multiple files
It is not necessary to actually open the file in vim. The modification can be made directly from the command line:
vi +':wq ++ff=unix' file_with_dos_linebreaks.py
To process multiple *.py
files (in bash
):
for file in *.py ; do
vi +':w ++ff=unix' +':q' "${file}"
done
😱 offtopic: if by chance you are stuck in vim and need to exit, here are some easy ways.
Removing the BOM mark
Sometimes even after setting unix line endings you might still get an error running the file, especially if the file is executable and has a shebang. The script might have a BOM marker (such as 0xEFBBBF
or other) which makes the shebang invalid and causes the shell to complain. In these cases python myscript.py
will work fine (since python can handle the BOM) but ./myscript.py
will fail even with the execution bit set because your shell (sh, bash, zsh, etc) can't handle the BOM mark. (It's usually windows editors such as Notepad which create files with a BOM mark.)
The BOM can be removed by opening the file in vim
and administering the following command:
:set nobomb
Solution 2:
The script contains CR characters. The shell interprets these CR characters as arguments.
Solution: Remove the CR characters from the script using the following script.
with open('beak', 'rb+') as f:
content = f.read()
f.seek(0)
f.write(content.replace(b'\r', b''))
f.truncate()
Solution 3:
You can convert the line ending into *nix-friendly ones with
dos2unix beak