converting epoch in a file name to human readable date/time

[sinusoid]

All,

I have searched for a solution to this and tried numerous things, but I am limited on my abilities with grep/awk/shell scripting etc and spent too many hours already. Maybe someone smarter than I can easily help me with a quick one-liner, perl/python script, etc...?

I have log files that everyday are downloaded from my webserver in the format:

samplesite.com.xxxxxxxxxxx.gz

xxxxxxxxxx is a 10 digit epoch time.

I am trying to figure out a way in batch to:

1. find all of exisiting files containing the pattern (after the first run it will only be one a day)
2. Isolate the epoch string
3. convert the epoch string to human readable date/time
4. rename the original file as samplesite.com.mmddYYYY.gz

Thanks so much!

[fabtk]

Hi sinusoid

About step 1: Are you downloading these backup files manually or should this also happen automatically?

For step 2, 3 and 4 I would do something like this ...

#!/bin/bash

for i in *.gz
do
  # getting length of timestamp (if 10 = timestamp - if not, already converted file)
  TSLENGHT=$(echo $i | cut -d'.' -f3 | awk '{print length}')
  if [ $TSLENGHT == 10 ]; then
    # cutting timestamp out of filename
    FILETIMESTAMP=$(echo $i | cut -d'.' -f3)
    # getting correct date format
    FILEDATE=$(date -d "1970-01-01 $FILETIMESTAMP sec" "+%d%m%Y")
    # renaming
    mv $i "samplesite.com.$FILEDATE.gz"
  fi
done

Greetings from Switzerland

FabTK

//edit: tested and works

[sinusoid]

FabTK --- this is FANTASTIC!

Thank you SO much!