UNIX for Beginners Questions & Answers

I'm trying to write a script that takes a file name in the form of Name_Num1_Num2.Extension and I want to separate the name portion and then use grep to see if that name part has any illegal characters in it.

I already have my grep command written and it works, I'm not sure how to separate the Name part from the two _Num1_Num2.Extension, is it possible to just look at just the one part of the file name and perform my command on it?

I also want to do the same with the _Num1_Num2 and the .Extension part with my script to check if they contain illegal characters as well, not sure how to approach that as well.

The reason I want to check each part separately is because each part of the file name has different illegal characters where the name cannot contain anything other than letters, digits and then the number portion should only contain digits.

So in short my question is: is there any way I can check portions of filenames? Or is there a better way to approach this? Thanks.

---------- Post updated at 04:22 PM ---------- Previous update was at 04:00 PM ----------

Never mind, I got a solution by doing the following:

Code:

inputFile="name_1_0.cc"
name="${inputFile%_*_*}"
echo $name
name

Thanks for sharing!
Lets see if others have a better suggestion...

I also found that the following works as well.

Code:

name=$(echo $inputFile | cut -d'_' -f1)
minor=$(echo $inputFile | cut -d'_' -f3 | cut -d'.' -f1)
major=$(echo $inputFile | cut -d'_' -f2)
ext=$(echo $inputFile | cut -d'.' -f2)

How about - provided you are using bash -

Code:

IFS="_." ARR=( $inputFile )
for IX in ${!ARR[@]}; do echo ${ARR[$IX]}; done
name
1
0
cc

General shell syntax, using unix pattern matching of the case statement to check the filename format, without separating the filename in different parts:

Code:

case $inputFile in
  (*[![:alnum:]]*_*_*.*|*_*[![:digit:]]*_*.*|*_*_*[![:digit:]]*.*|*_*_*.*[![:alnum:]]*)
    echo "filename is wrong"
  ;;
  (*_*_*.*)
    echo "filename is correct"
  ;;
  (*)
    echo "filename is wrong"
  ;;
esac

--
Or bash/ksh93/zsh, same thing, but this time using extended regular expression matching:

Code:

if [[ $inputFile =~ ^[[:alnum:]]+_[[:digit:]]+_[[:digit:]]+\.[[:alnum:]]+$ ]]; then
  echo "filename is correct"
else
  echo "filename is wrong"
fi

--
A quick general (POSIX) way to split the filename in different variables, using a here-document:

Code:

IFS='_.' read name major minor ext << EOF
$inputFile
EOF

Or bash/ksh93/zsh, using a here-string:

Code:

IFS='_.' read name major minor ext <<< "$inputFile"

---
Note: this will globally change IFS and not local to the array assignment. Also the order in which these two assignments will be executed is not defined, so it is best to use a semicolon or newline to separate them and to make sure the IFS assignment is done before the array assignment and save IFS and restore it later,

Code:

oldIFS=$IFS
IFS="_."
ARR=( $inputFile )
IFS=$oldIFS

or use (bash):

Code:

IFS="_." read -a ARR <<< "$inputFile"

In the latter case IFS does get set local to the read command.

Hi,

If the name format is same, then you can try something like this..

Code:

inputFile="name_1_0.cc"

Code:

echo $inputfile | cut -d "_" -f1 | xargs | grep -i "charecters"