CODEWITHSUNDEEP
perlgitbash
David Allen
David Allen

Reputation: 760

Extract file names from git pull output

I am creating a script that will pull down the latest master branch of code, and check to see if any of the file names are SQL. After I have determined the filename, I am going to source the stored procedures. Currently I have it pulling the code and readying the output line by line, however my grep command doesn't seem to be able to find the sql filename. This is what I have

#!/bin/bash

# Pull all of the latest code for the master branch, and set the commit log to 
# a variable
commit_log=$(git pull origin master --quiet)

#read the commit log line by line
echo $commit_log | while read line
do
    if [ `echo "$line" | grep -E "(\.sql)$"` ]; then
        echo "SQL"
    fi      
    echo "LINE:" $line 
done

I am not stuck on bash, I could also do this in perl.

Upvotes: 1

Views: 641

Answers (1)

michas
michas

Reputation: 26515

After a pull the new state is stored in HEAD and the old one in HEAD@{1}. You can find the changed files between both with git diff --name-only. A small perl one-liner seems to be the easiest way to check the file names:

git pull
git diff --name-only HEAD@{1} | perl -wpe 'print "SQL:" if /\.sql$/'

Upvotes: 4

Related Questions