I'm writing a script that takes a filename as an argument, which determines the "file type" of the file. I want to know if there is any command I can use to determine if a file is ASCII type, thanks all for giving a help.

In bash it would be something like this.

if test -f "$file"
then
echo "$file: Regular File"
fi


or you can use if [ -f $file ] ....

That test for -f will not help with an ASCII test. It determines if a filename is a regular file as opposed to a directory, a link, etc, and will return true even for binaries.

But I would start with that test, and if it is a regular file, then the "file" command will give a clue as to its content.

Popo,

You can try this :

for i in `find . -xdev -type f`
do
file $i | egrep -i "text|script" | awk -F\: '{ print $1 }'
done


Witt

Thanks all !

I've tried something like that, but i think it does not sound good

if file "$1" | grep ascii
then
echo It is an ASCII file
elif file "$1" |grep command
...


I'm now trying all the other method you mentioned. Thanks

I don't know how well it works in a script, but I have used the "file" command to tell a file type.

# file filename
filename :ascii text

# file script.sh
script.sh :commands text

It works from the command line very well.


Here is another suggestion...

Are all of your files in one directory? If so, you can do a for loop.

for name in `ls *`
do
test -f $name
some other command
some other command
done

__________________
My brain is your brain

The only language that I ever have encountered with a built-in test for this was perl. I dislike perl and seldom use it. But I did give this feature a try. It seemed broken because it allowed many non-ascii characters before it finally declared a file to be binary. Since I then had to code my own test, I returned to ksh. But I do prefer perl's terminology. It calls this "text files" and "binary files".

Unless you inspect every byte of the file, you are not going to get this 100%. And there is a big performance hit with inspecting every byte. But after some experiments, I settled on an algorithm that works for me. I examine the first line and declare the file to be binary if I encounter even one non-text byte. It seems a little slack, I know, but I seem to get away with it.

Here is a little script that demonstrates this. Note that where I have used (TAB) to indicate a place where you must actually type the tab character.