How to rename a batch of files to a name that exists in the first line of each file

Question

I need to rename a batch of files in a single directory to the characters between the backticks on the first row of the file contents.

File name: schema.stored.procedure302
First line in the file reads:

/* Procedure structure for procedure `usp_stored_procedure_abc` */

Desired new filename: usp_stored_procedure_abc.sql

---

Here's the scenario in a little more detail:

I have a single schema.sprocs.sql file that contains only the stores procedures from a mysqldump.

I used csplit to split the file into 300+ separate sql files

csplit -f schema.stored.procedure schema.sprocs.sql '/Procedure structure for procedure/' {*}

This worked brilliantly, and now I have files like this:

schema.stored.procedure302
schema.stored.procedure303
schema.stored.procedure304
etc...

What I need help with is:
What commands can I use to read the contents of each file, find the pattern:

/* Procedure structure for procedure `

Then rename the file with the characters that follow that pattern until a backtick is found.

For example:

The file schema.stored.procedure302 Contains the following line on the first row:

/* Procedure structure for procedure `usp_stored_procedure_abc` */

I need that file to be renamed to: usp_stored_procedure_abc.sql (the contents between the backticks on the first row of the file contents).

I'm sure this is feasible, can someone help me?

Answer 1

Following string shows, how we will get our new filename:

head -n 1 file |awk -F'[`]' '{print $2".sql"}'

Complete script would be this:

#!/bin/bash
for f in *;
do
  d="$(head -n 1 $f |awk -F'[`]' '{print $2".sql"}')" ;
  if [ ! -f "$d" ];
  then
    mv "$f" "$d"
  else echo "File '$d' already exists! Skiped!"
  fi
done