Create mysql table directly from CSV file using the CSV Storage engine?

I just learned that MySQL has a native CSV storage engine which stores data in a Comma-Separated-Value file per table.

Is it possible to create a table directly from a uploaded CSV file, something like:

CREATE TABLE USERS < PATH/USERS.CSV

where users.csv is uploaded by the user?

I just discovered csvkit, which is a set of Unix command-line tools for CSV files. I installed it on my Mac with pip install csvkit. The command was:

csvsql --dialect mysql --snifflimit 100000 bigdatafile.csv > maketable.sql

You can alternatively provide a DB connection string and it can load the table directly.

This is not possible. To create a table you need a table schema. What you have is a data file. A schema cannot be created with it.

What you can do is check if your file has a header row, and, in that case, you can manually create a table using that header row.

However, there is a way to generate a create table statement using a batch file as described by John Swapceinski in the comment section of the MySQL manual.

> Posted by John Swapceinski on September 5 2011 5:33am.
Create a table using the .csv file's header:

#!/bin/sh
# pass in the file name as an argument: ./mktable filename.csv
echo "create table $1 ( "
head -1 $1 | sed -e 's/,/ varchar(255),\n/g'
echo " varchar(255) );"

I'm recommended use MySQL Workbench where is import data. Workbench allows the user to create a new table from a file in CSV or JSON format. It handles table schema and data import in just a few clicks through the wizard.

In MySQL Workbench, use the context menu on table list and click Table Data Import Wizard.

More from the MySQL Workbench 6.5.1 Table Data Export and Import Wizard documentation. Download MySQL Workbench here.

"Convert CSV to SQL" helped me. Add your CSV file and you are good to go.

There is an easier way if you are using phpMyAdmin as your MySQL front end:

1. Create a database with the default settings.
2. Select the database.
3. Click "Import" at the top of the screen.
4. Select "CSV" under "Format".
5. Choose the options appropriate to your CSV file, open the CSV file in a text editor and reference it to get the "appropriate" options.

If you have problems, no problem, simply drop the database and try again.

In addition to the other solutions mentioned Mac users may want to note that SQL Pro has a CSV import option which works fairly well and is flexible - you can change column names, and field types on import. Choose new table otherwise the initial dialogue can appear somewhat disheartening.

Sequel Pro - database management application for working with MySQL databases.

If someone is looking for a PHP solution see "PHP_MySQL_wrapper":

$db = new MySQL_wrapper(MySQL_HOST, MySQL_USER, MySQL_PASS, MySQL_DB);
$db->connect(); 

// this sample gets column names from first row of file
//$db->createTableFromCSV('test_files/countrylist.csv', 'csv_to_table_test');

// this sample generates column names 
$db->createTableFromCSV('test_files/countrylist1.csv', 'csv_to_table_test_no_column_names', ',', '"', '\\', 0, array(), 'generate', '\r\n');

/** Create table from CSV file and imports CSV data to Table with possibility to update rows while import.
 * @param 	string		$file			- CSV File path
 * @param 	string 		$table 			- Table name
 * @param	string		$delimiter		- COLUMNS TERMINATED BY (Default: ',')
 * @param	string 		$enclosure		- OPTIONALLY ENCLOSED BY (Default: '"')
 * @param 	string		$escape 		- ESCAPED BY (Default: '\')
 * @param 	integer 	$ignore 		- Number of ignored rows (Default: 1)
 * @param 	array		$update 		- If row fields needed to be updated eg date format or increment (SQL format only @FIELD is variable with content of that field in CSV row) $update = array('SOME_DATE' => 'STR_TO_DATE(@SOME_DATE, "%d/%m/%Y")', 'SOME_INCREMENT' => '@SOME_INCREMENT + 1')
 * @param 	string 		$getColumnsFrom	- Get Columns Names from (file or generate) - this is important if there is update while inserting (Default: file)
 * @param 	string 		$newLine		- New line delimiter (Default: \n)
 * @return 	number of inserted rows or false
 */
// function createTableFromCSV($file, $table, $delimiter = ',', $enclosure = '"', $escape = '\\', $ignore = 1, $update = array(), $getColumnsFrom = 'file', $newLine = '\r\n')

$db->close();

you can use this bash script

convert.sh

and run

./convert.sh -f example/mycsvfile.csv

I adopted the script from shiplu.mokadd.im to fit my needs. Whom it interests:

#!/bin/bash
if [ "$#" -lt 2 ]; then
    if [ "$#" -lt 1 ]; then	
	    echo "usage: $0 [path to csv file] <table name> > [sql filename]"
		exit 1
	fi
	TABLENAME=$1
else
	TABLENAME=$2
fi
echo "CREATE TABLE $TABLENAME ( "
FIRSTLINE=$(head -1 $1)
# convert lowercase characters to uppercase
FIRSTLINE=$(echo $FIRSTLINE | tr '[:lower:]' '[:upper:]')
# remove spaces
FIRSTLINE=$(echo $FIRSTLINE | sed -e 's/ /_/g')
# add tab char to the beginning of line
FIRSTLINE=$(echo "\t$FIRSTLINE")
# add tabs and newline characters
FIRSTLINE=$(echo $FIRSTLINE | sed -e 's/,/,\\n\\t/g')
# add VARCHAR
FIRSTLINE=$(echo $FIRSTLINE | sed -e 's/,/ VARCHAR(255),/g')
# print out result
echo -e $FIRSTLINE" VARCHAR(255));"

This is not possible, you can however overwrite an existing table file. But be sure, that the line endings in your file are unix style (ending only with \n), not windows style (ending with \r\n), whether you are working under windows or not.

I have made a Windows command line tool that do just that.

You can download it here: http://commandline.dk/csv2ddl.htm</strike>

C:\Temp>csv2ddl.exe mysql test.csv test.sql

C:\Temp>csv2ddl.exe mysql advanced doublequote comma test.csv test.sql

If you're ok with using Python, Pandas worked great for me (csvsql hanged forever for my case). Something like:

from sqlalchemy import create_engine
import pandas as pd

df = pd.read_csv('/PATH/TO/FILE.csv')
# Optional, set your indexes to get Primary Keys
df = df.set_index(['COL A', 'COL B'])

engine = create_engine('mysql://user:pass@host/db', echo=False)

df.to_sql(table_name, dwh_engine, index=False)

Also this doesn't solve the "using CSV engine" part which was part of the question but might me useful as well.

MySQL for excel plugin can help you.

http://dev.mysql.com/doc/refman/5.6/en/mysql-for-excel.html

Open your CSV file in excel. You can use this plugin to export excel data into a new table of remote or local mysql server. It will analyze your data (top 100 to 1000 rows) and create a corresponding table schema.