Logo
  • PHP
    • HipHop / HHVM
    • Modern PHP
    • PHPStorm
    • LAMP
    • Laravel
    • Composer
    • PDO
  • JavaScript
    • node.js
    • AngularJS
  • CSS
    • SASS
    • “CSS4” (CSS level 4)
  • HTML
  • Git
  • LAMP
  • Vagrant
  • UI / UX
  • Architecture of …
  • Off-Topic
With ♥ from Berlin
December 20, 2014
Chris
Local Development, MySQL
Comments Off on How to get a single table out of a massive MySQL .sql database backup file (mysql dump splitter)

How to get a single table out of a massive MySQL .sql database backup file (mysql dump splitter)

PreviousNext

Imagine the following situation: Somebody backs up an entire MySQL database – a very large one – with common tools. This will usually result in a tar’ed and gzip’ed file, containing ALL the create/fill statements for ALL tables of the database. All tables in one file. This is also very common when third party clients deliver database backups.

This situation gets annoying when your backup is in the 100+ GB area and you just need ONE table (or at least not all). Re-deploying backups in these dimensions from scratch (for testing for example) is a huge time-killer (running these statements can take days!), and for sure using just that 2 GB table is by far the better option than running the whole script when you really just need that one table.

This was a common task at my former company, and unfortunately all professional tools were not able to read such a file or totally crashed while doing so. The final solution to solve this problem was a tiny self-written bash script I found in on Change is inevitable, a very nice dev blog. This awesome script is extremely easy to use and simply extracts one specific table out of the giant file (and puts it into another file) or grabs a set of tables, you can even use regex here. Fantastic!

Kedar of Change is inevitable has even made a little tutorial on Youtube, check the blog post for the video. Big thanks, man!

As the original blog post shows the script in a unformatted way, here’s the full code again:

#!/bin/sh
# http://kedar.nitty-witty.com
#SPLIT DUMP FILE INTO INDIVIDUAL TABLE DUMPS
# Text color variables
txtund=$(tput sgr 0 1)    # Underline
txtbld=$(tput bold)       # Bold
txtred=$(tput setaf 1)    # Red
txtgrn=$(tput setaf 2)    # Green
txtylw=$(tput setaf 3)    # Yellow
txtblu=$(tput setaf 4)    # Blue
txtpur=$(tput setaf 5)    # Purple
txtcyn=$(tput setaf 6)    # Cyan
txtwht=$(tput setaf 7)    # White
txtrst=$(tput sgr0)       # Text reset

TARGET_DIR="."
DUMP_FILE=$1
TABLE_COUNT=0

if [ $# = 0 ]; then
        echo "${txtbld}${txtred}Usage: sh MyDumpSplitter.sh DUMP-FILE-NAME${txtrst} -- Extract all tables as a separate file from dump."
        echo "${txtbld}${txtred}       sh MyDumpSplitter.sh DUMP-FILE-NAME TABLE-NAME ${txtrst} -- Extract single table from dump."
        echo "${txtbld}${txtred}       sh MyDumpSplitter.sh DUMP-FILE-NAME -S TABLE-NAME-REGEXP ${txtrst} -- Extract tables from dump for specified regular expression."
        exit;
elif [ $# = 1 ]; then
        #Loop for each tablename found in provided dumpfile
        for tablename in $(grep "Table structure for table " $1 | awk -F"\`" {'print $2'})
        do
                #Extract table specific dump to tablename.sql
                sed -n "/^-- Table structure for table \`$tablename\`/,/^-- Table structure for table/p" $1 > $TARGET_DIR/$tablename.sql
                TABLE_COUNT=$((TABLE_COUNT+1))
        done;
elif [ $# = 2  ]; then
        for tablename in $(grep -E "Table structure for table \`$2\`" $1| awk -F"\`" {'print $2'})
        do
                echo "Extracting $tablename..."
                #Extract table specific dump to tablename.sql
                sed -n "/^-- Table structure for table \`$tablename\`/,/^-- Table structure for table/p" $1 > $TARGET_DIR/$tablename.sql
                TABLE_COUNT=$((TABLE_COUNT+1))
        done;
elif [ $# = 3  ]; then

        if [ $2 = "-S" ]; then
                for tablename in $(grep -E "Table structure for table \`$3" $1| awk -F"\`" {'print $2'})
                do
                        echo "Extracting $tablename..."
                        #Extract table specific dump to tablename.sql
                        sed -n "/^-- Table structure for table \`$tablename\`/,/^-- Table structure for table/p" $1 > $TARGET_DIR/$tablename.sql
                        TABLE_COUNT=$((TABLE_COUNT+1))
                done;
        else
                echo "${txtbld}${txtred} Please provide proper parameters. ${txtrst}";
        fi
fi

#Summary
echo "${txtbld}$TABLE_COUNT Table extracted from $DUMP_FILE at $TARGET_DIR${txtrst}"

 

Usage

You can use the mysqldumpsplitter like this (assuming you have put the script inside a file called mysqldumpsplitter.sh):

Extract all tables of file XXX to lots of single sql files:

sh mysqldumpsplitter.sh XXX

Extract table ZZZ of file XXX to single file:

sh mysqldumpsplitter.sh XXX ZZZ

Extract table(s) via regex of file XXX:

sh mysqldumpsplitter.sh XXX -S REGEX
Share this
git-php-deployment

Extremely simple deployment with PHPloy

Let me ask you a question: How mad do you get when you have to remember which files you edited

css

How to center a div vertically and horizontally (modern methods, without fixed size!)

“How to center a div” (inside another div or inside the body itself) is one of the most discussed questions

[Link] Interesting: Designing a Nuclear Waste Warning Symbol That Will Still Make Sense in 10,000 Years

Interesting article for everybody with a faible for UI/UX: Slate – Designing a Nuclear Waste Warning Symbol That Will Still

phpstorm 7.0 php

How to setup and use XDEBUG with PHPStorm 6/7 (locally in Windows 7/8 and Mac OS X)

Real debugging is probably one of the most coolest things that are possible in software development: Remember the times where

mod-rewrite-ubuntu-14-04-lts

How to enable mod_rewrite in Ubuntu 12.04 LTS

A little note first: This is the most seen way to enable mod_rewrite in tutorials. However, it’s not the best

Microsoft enters post-password era with Hello (promo video)

Just a short promotion clip, but definitly interesting: Microsoft announces Hello, the authentication system inside Windows 10, using fingerprints, iris

php ide

Sitepoint asks for your favourite PHP IDE – take part!

Bruno Skvorc ask for your favourite PHP IDE in this new article on Sitepoint. If you use one, then take

organizing css

Jonathan Snook – CSS is a Mess – How to organize CSS in big projects (54min video talk)

vagrant

How to setup a local server (in a virtual machine) with Vagrant in PHPStorm

This is part 1 of a series on How to get a modern workflow in PHP development. Part 2 is

mod-rewrite-ubuntu-14-04-lts

How to enable mod_rewrite in Ubuntu 14.04 LTS

A little note first: This is the most basic way to enable mod_rewrite. However, it’s not the best way. The

1/4

Categories

Search

How to show the available version of a package (before doing apt-get install)
Useful basic linux stuff: Show kernel version, distribution name and distribution version on Ubuntu systems
gamescom 2014 trailers
GAMESCOM 2014: Awesome Next-Gen ingame graphics
october cms
[Link] How To Install October CMS on a VPS running Ubuntu 14.04
php-legacy-code
Interesting talk on modernizing a legacy PHP codebase
microsoft-windows-azure-cloud-hosting
Microsoft’s Azure platform gives away high money prizes for “testing out” their cloud services
battlefield-3-free
Electronic Arts / Origin offers Battlefield 3 for free (limited promo action) !
How Instagram.com works
vagrant
A preinstalled Vagrant box with PHP HipHop / HHVM and Ubuntu 12.04 (Precise Pangolin)
The New Era of JavaScript (28min conference talk, Jack Franklin, 2013)
[Link] Improving Smashing Magazine’s Performance: A Case Study
“Fuck you. Pay Me.” How to deal with clients, the professional way. An excellent talk with Mike Monteiro.
github-logo-octocat
GitHub rolls out .PSD diff and viewing
This picture shows the icon of blindness
How blind people use websites (video with Sina Bahram, blind accessibility researcher)
angularjs
Two excellent introductions into AngularJS by Todd Motto

Tags

apache bash centos composer conference coupon CSS debian fonts framework git GitHub hack HHVM HipHop HTML HTML5 IDE JavaScript JS LAMP laravel linux mod_rewrite MVC MySQL Nginx optimization PHP PHP 5.5 PHP 5.6 phpmyadmin PHPStorm security server SSD Ubuntu UI UX vagrant video virtual machine voucher VPS wordpress
Side-Project: Wordle-Solver:
www.wordle-helper.info

Pages

  • Privacy Policy