I am writing a bash script to loop through 2 identical directory trees in order to do run diffs on matching files. I just need to know what is correct syntax to setup 2 loops that will run in parallel. I'd also like to iterated through directories recursively, so if there's an additional step to accomplish that, can you mention it?
#!/bin/bash
FILES1=/path/to/one
FILES2=/path/to/two
for f1 in $FILES1; f2 in $FILES2
do
echo "Processing $f1 $f2 file..."
done
diff already has the builtin functionality to run recursively against directories. You can run diff -Naur /path/to/one /path/to/two.
-N will show the diff for new files, instead of just saying it exists in the second path-a treats all files as text, so you may or may not need this.-u uses the common unified diff format-r is the important, in this case, recursive flagThere is no guarantee that the files will be in the same order in the two directories.
Do it like this:
FILES1=/path/to/one
FILES2=/path/to/two
for f1 in $FILES1/*
do
f2=$FILES2/`basename f1`
# ...
done
And if you need to do it recursively:
cd $FILES1
for f1 in `find`; do
f2=$FILES2/$f1
# ...
done
The closest you can get is to nest the for loops, i.e.
for f in $FILES1 ; do
for f2 in $FILES2 ; do
if [ ${f##*/} = ${f2##*/} ]; then
diff $f $f2
break
fi
done
done
to iterate thru recursive dirs, you could rely on find to return a list of files for processing
PATH1=/path/to/one
PATH2=/path/to/two
for f in $(find $PATH1 -print ) ; do
for f2 in $(find $PATH2 -print ) ; do
if [ ${f##*/} = ${f2##*/} ]; then
diff $f $f2
break
fi
done
done
Sorry I don't have a way to test this right now, the ${f##*/} may not be exactly right. The intention is to create just the basename of a file (I guess you could use that too, $(basename $f ) at the expense of an 2 processes for each.
Both of these are very expensive processes, checking lots of non-matching filename pairs. The break inside loop2 helps a little.
An alternate solution is to rely on one path to create the driver list and then use the supplied names to generate alternate relative paths, i.e.
for f in $FILES1 ; do
f2=../two/$f{##*/}
if [ -f ${f2} ] ; the
diff $f $f2
else
printf "no alternate file found in ../two for f=${f}"
fi
done
If need be, you can duplicate that loop and reverse the location of f2 and f to check the 'other-side' of your system.
I hope this helps.
diff -Naur /path/to/one /path/to/two?