find $path1 $path2 -regex '.*\(\.log\|\.err\|\.out\)' -mtime -10 -exec cp "{}" $destination \;
This is my command in script to find a filename with extension as shown above. It was success until I found out some of my file contain 'filename.extension.date' which I cannot retrieve it. So I try and error (also using -iname but Im dont understand) to find the file which I want find 'word/string' instead of extension so I can get the file (filename.extension.date & filename.extension).
If you want to also copy the .log/.err/.out files that have an extra .date suffix, and assuming those dates are in YYYY-MM-DD format, and assuming GNU find (as your regexp suggests you're using), and GNU cp, you'd do
LC_ALL=C find "$path1" "$path2" \
-regextype posix-extended \
-regex '.*\.(log|err|out)(\.[0-9]{4}-[0-9]{2}-[0-9]{2})?' \
-type f \
-exec cp -t "$distination" {} +
Replace [0-9]{4}-[0-9]{2}-[0-9]{2} with [0-9]{8} for YYYYMMDD dates, etc.
The .*\.(log|err|out)(\.[0-9]{4}-[0-9]{2}-[0-9]{2})? extended regular expression matches on:
.*: any number (*) of characters (.), which thanks to LC_ALL=C means any byte, those bytes not having to form valid characters in the user's locale. Followed by:\.: a literal . (dot / period). Followed by:(log|err|out): either log, err or out.(...)?: an optional (?) ..., ... being:\.[0-9]{4}-[0-9]{2}-[0-9]{2}, a literal . again followed by 4 characters in the 0-9 range which thanks to LC_ALL=C again is limited to 0123456789, followed by -, 2 more of those digits, - and 2 more digits.With BSD find, you'd use the -E option instead of the -regextype posix-extended predicate for the argument to -regex to be treated as an extended regexp.
? makes the date optional. See edit.
file.log.tar.gz,tar.gzbeing the date?