Renaming JSON files based on their contents

I have loads of geographical json files with date_geohash.json name format and I have to convert them to date_latitude-longotude.json format. For example I need to convert 2017-03-28_u13hhyn.json to 2017-03-28_N52.76-E1.62.json.   I’m trying to write a bash script to rename these json files using ‘rename’ and ‘jq’ (json query) utilities. Using jq I can extract latitude and longitude from the json files.

jq '"(.latitude)-(.longitude).json"' 2017-03-28_u13hhyn.json

gives me

“52.768471-1.623297.json"

  However, I don’t know how to truncate the latitude and longitude digits and rename the json files.  

 

Answer

It might be better to extract the latitude and longitude into variables, then manipulate into the filename format you want with printf. Then just rename the file. Here’s a basic example:

#!/bin/bash

declare    directory="/some/directory/path"
declare    pattern="*_*.json"

cd ${directory}
for file in ${pattern}; do
  prefix=${file%%_*}    # Remove first underscore and everything following.
  latitude=$(jq '.latitude' ${file})
  longitude=$(jq '.longitude' ${file})
  new_name=$(printf "%s_N%.2f-E%.2f.json" ${prefix} ${latitude} ${longitude})
  mv ${file} ${new_name}
done;

Note 1: printf does not truncate the numbers, it rounds them, which is going to give more accurate results.

Note 2: I have hard-coded ‘N’ and ‘E’ into the filename, on the assumption that latitude and longitude will be negative for S and W. If this is not the case, there is a little more work required, but the priciple stays the same.