Question

First, I tried asking the same on askubuntu and I’m not getting far there…
https://askubuntu.com/questions/186114/how-to-diff-two-folders-to-multiple-patch-files/186121#comment231985_186121

So, i’ll copy/paste here in hopes SO gives me a working solution…

I’m not much of a unix guy so I’ll just go ahead and ask:

I have a following problem – two folders with plenty of subfolders. I need to diff them.

I tried this:

diff -drupN vanila_kernel_3.0.8 my_kernel_3.0.8 > kernel.patch

and that results in a 185mb file… and not really what I want.

I want the result of the diff to be many smaller patches, ideally one for every changed file with the contents of the change. That means I have t ochange the way I use diff, and I need to put it in some sort of a loop… So i tried running this little script

for file in original_308/*.*; do
    diff -dupN "$file" "my_308/${file##*/}" > "$file".patch
done

But it doesn’t work :/

Ideally, I want to have a .patch file for every change, but having patches for files that changed in original would do just fine (as I could filter the newly added files and just copy them over)

Can someone provide me with a decent way to do this please?

UPDATE:
I know there are 231 altered files between these folders and 4546 newly added files… I’m just interested in generating those 231 patches. I even have a file containing those 231 filenames so that can be used in a script as well.

I used this to generate the changelog

diff -qdr -x *.o -x *.cmd -x *.d original_308 my_308 | sed "s/^.* and \(.*\) differ/\1/" | sort > changes.txt

and that just prints the filenames that changed, trailed by “only in my_308” files… the 231 files are on the top of that file (the first 231 lines)

So let’s make it clear – I want .patch files for source files. If the patches can end up in patches//file_name.c.patch that would be really great…

If you give me PERL scripts please tell me how to best use them since I do not understand that garbage. Please don’t fail me SO. I need to get this thing ported, tested and running and I’m wasting time on unix commands and untangible scripts ffs.

Answer 1

Thanks to everyone for their contributions… Here’s the end result

#!/bin/bash                               
FOLDER_OLD="original"
FOLDER_NEW="changed"
FOLDER_PATCHES="./fresh_patches"

echo "Generating changelog for $FOLDER_OLD --> $FOLDER_NEW"
diff -qdr ${FOLDER_OLD} ${FOLDER_NEW} | sed "s/^.* and \(.*\) differ/\1/" | sort > newchanges.txt

mkdir -p ${FOLDER_PATCHES}                                                                                                           
for newfile in `find $FOLDER_NEW -type f`
do
sufix=${newfile#${FOLDER_NEW}/}
oldfile="${FOLDER_OLD}/${sufix}"

if [ -f $oldfile ]; then
    patchfile="${FOLDER_PATCHES}/`echo $sufix |sed 's/\//_/g'`.patch"
    if [ "`diff $oldfile $newfile`" != "" ]; then
        echo "Make patch ${patchfile}"
        diff -dupN "$oldfile" "$newfile" > "$patchfile"
    fi
else
patchfile="${FOLDER_PATCHES}/ADDONS/${newfile#$FOLDER_NEW}"
    echo "Copy $newfile --> $patchfile"
    mkdir -p "`dirname ${patchfile}`"
    cp $newfile $patchfile
fi
done

The script first generates a list of changed files and files unique only to the new project/folder; then it generates a patch file for every file and copies the newly added files to the ADDONS folder (unique to the new project)