I have a bunch of javascript files and I need to add a comment on a new line at the beginning of every file that contains the filename e.g.
/* app.js */
I'd like to do this recursively on a folder from the terminal on my mac.
Can anyone help?
Thanks!
First, we have to locate those files:
$ find $folder_name -type f -name "*.js" -print
This will locate all of the files with the suffix of *.js. The type -f means only look for files (just incase you have a directory name that ends in .js). I will assume that your JavaScript file names don't contain whitespace, or control characters like <NL> characters in the name. Otherwise, we'll have to do a bit more work.
We can use that find statement to locate all of those files, then we can munge those files. Once we find those files, we have to figure out how to munge the file to get the comment line on top.
We'll take three step approach:
For the first step, we'll use echo to create a new file with the old name. Then, we'll use the cat command to concatenate the old file on the new file. And, finally we'll use mv to rename the newly constructed file to the old name.
find $javascript_dir -type f -name "*.js" -print | while read file_name
do
basename=$(basename $file_name)
echo "/* $basename */" > "$basename.tmp"
cat "$file_name" >> "$basename.tmp"
mv "$basename.tmp" "$file_name"
done
Note that > redirects the output of the command to the named file, and that >> appends the output of the command to the named file. mv stands for move. In Unix, file are actually stored by inode identify. There's a table of contents that matches the inode to the full name of the file (including the directory). Thus, moving a file in Unix is the same as simply renaming it, so the directory name is changed. In Windows batch, there are separate move and rename commands. In Unix, they're both mv.
We could use sed to do an in place prepending of the line:
find . -name "*.js" -print | while read file_name
do
basename=$(basename $file_name)
sed -i "" "1i\\
/* $basename */
" "$file_name"
done
sed is a powerful line editor, and most people use only its substitution command mode. However, sed scripts can be way more complex and rich (and obscure which is why you rarely see them).
You can use cat - and a temp file:
# adds filename to top of each file in $DIR
DIR=. # set to whatever directory
FILES=$(find $DIR -type f)
for FILE in $FILES; do
FILE=$(echo $FILE | sed -e s/\.\\\///) # remove leading "./" if DIR is "."
echo "/* $FILE */" | cat - $FILE > tempfile.tmp # add the line using a temp file
mv tempfile.tmp $FILE # move the temp file back to the original file
done
Note that this will add the full path of the file relative to DIR if DIR is not '.'. To add only the filename itself you could use:
echo "/* $(basename $FILE) */" | cat - $FILE > tempfile.tmp # add the line using a temp file
You can use sed for this, try something like this:
#!/bin/bash
for i in $(find ./ -name *.js);
do
`sed -i 's/$i \*\//$i \*\/\n&/' "$i"`
done
Insert a newline with sed could be a problem since not all version works the same. To have an overview see this reference. You can for example replace the substitution part with G or {G;}. It depends on your sed version.
Similar Questions are already answered here and here
Another interesting solution:
If you want to append a newline after the first line you can also use this:
sed -i "1a\ " test.js
So change the script above into this:
#!/bin/bash
for i in $(find ./ -name *.js);
do
`sed -i "1a\ " $1`
done
