Retrieve email subject from file via bash -

i've shell script downloads files servers mail folder nas devices client has copies made locally backup.

the files saved 11469448248.h15587p19346.smtp.x14.eu_2,s files. i've changed extension standard .eml format email clients can read them disc.

for f in *.smtp.x14.eu_2,s; #sed "9q;d" $f #tail -n+9 $f | head -n1 mv -- "$f" "${f%.smtp.x14.eu_2,s}.eml"; done 

as can see i've tried use sed , tail command 9th line file; problem subject isn't on 9th line , file names don't of content.

i'm trying files names in understandable format, figured subject helpful.

on nth line of email file line begins subject: pd: subject

im trying find line fet rid of subject: pd: , leave rest ne file name

the following wrong implements seem asking.

subj=$(sed -n '/^subject: pd *//!d;p;q;/^$/q' "$f") 

the problem succeeds in trivial case, fails when have mime rfc2047-encoded subject: header, , (more trivially) when subject: header spans more single line.

i approach more modern programming language. it's not quite one-liner, it's easy enough python.

subj=$(./emailsubj.py "$f") 

where emailsubj.py contains more or less like

#!/usr/bin/env python email.parser import parser email.header import header, decode_header sys import argv  filename in argv[1:]:     open(filename, 'rb') handle:  # handle file not found etc?         message = parser().parse(handle)     try:         subj = ''.join([frag.decode(enc) if enc else frag             frag, enc in decode_header(message['subject'])])     except headerparseerror, unicodedecodeerror:         subj = message['subject']   # maybe warn error?     print(subj) 

(remember chmod +x emailsubj.py, obviously.)