Renaming files according to a number of chars from the directory name

Question

I have a number of directories like these :

./lab-results.n3668.2017-01-16
./lab-results.n471.2017-02-24
./tests.n287.2017-01-28
./tests.n4501.2017-03-04

They contain unfortunately only one file but the name is the same.

./lab-results.n3668.2017-01-16/lab-results.pdf
./lab-results.n471.2017-02-24/lab-results.pdf
./tests.n287.2017-01-28/tests.pdf
./tests.n4501.2017-03-04/tests.pdf

I would like to move all these files after having renamed them like this :

./lab-results.n3668.pdf
./lab-results.n471.pdf
./tests.n287.pdf
./tests.n4501.pdf

So, it seems like some sh scripts I see there but I can't figure out how I can take a part of the directory's name characters to form the future name of the file.

lab-results.n9999 : max number available, will not change
tests.n9999 : max number

I also may have others directories but always followed by a bunch of numbers within their names :

./xxxxxxx.n9999
./yyyyyy.n9999
./zzz.n9999

Any tips or help very much appreciated. I use bash 4.3.11(1) on a linux distrib.

Answer 1

 for F in */* ; do echo mv "$F" "${F/.2017*/.pdf}" ; done

or as a script:

#! /bin/bash
for F in */*
do
    echo mv "$F" "${F/.2017*/.pdf}"
done

It just show what to it will do. If it is what you want then remove echo and run it.

The trick is with ${F/.2017*/.pdf}. It searches for .2017* in the filename+path and replace it with .pdf.

Answer 2

rename 's|^(\./[^.]*\.[^.]*).*|$1.pdf|' ./*.*.*/*.pdf

---

• We select directories that have atleast 2 dots and comprise pdf files.
• Then we keep info until the first 2 dots and construct the new filename.
• This assumes your rename is perl-rename. Debian-based systems have perl-rename as rename, others may have it as perl-rename or prename instead.