mycsv

mycsv

A command line tool for writing CSV files from MySQL queries. MySQL has a "select into outfile" feature to write CSV files but the files are written to the local file system and are owned by the MySQL user. This requires giving users access to the database machine, setting up an NFS mount or creating an automated process to retrieve the CSV files. mycsv allows users or a process to write CSV files directly to a destination.

Features:

• Customizable CSV format
• Write initial header line based on columns in query
• Cross platform (linux, osx & windows)*
• Supports SSL configured MySQL accounts
• Designed for interative and non-interactive use
• Write to a file destination or stdout for redirection or piping
• Stdin redirection to obtain query from pipe or file**

*64bit binaries provided by build script. Minor changes to the build script can provide 32bit binaries.

**Stdin redirection not supported on windows

Dependencies

Go
Git required for go get

Installation

$ go get -d github.com/joshuaprunier/mycsv

or

git clone https://github.com/joshuaprunier/mycsv.git

$ ./build.sh

Building mycsv version
1.0.30-421b387-20150810.155045


Building Linux
        mycsv - OK

Building Windows
        mycsv.exe - OK

Building Darwin
        mycsv - OK

Done!

$ ls -R bin
bin:
darwin  linux  windows

bin/darwin:
mycsv

bin/linux:
mycsv

bin/windows:
mycsv.exe

Usage

mycsv version 1.0.30-421b387-20150810.155045

USAGE:
mycsv DB_COMMANDS [CSV OUTPUT FLAGS] [DEBUG FLAGS] [CSV OUTFILE] query

EXAMPLES:
mycsv -user=jprunier -pass= -file=my.csv -charset=utf8 -query="select * from jjp.example_table where filter in ('1', 'test', 'another')"
echo "select * from mysql.plugin" | mycsv -user=jprunier -pass=mypass -host=remotedb > my.csv
mycsv -user=jprunier -pass= -file=my.csv -d="|" -q="'" < queryfile

DATABASE FLAGS
==============
-user: Database Username (required)
-pass: Database Password (interactive prompt if blank)
-host: Database Host (localhost assumed if blank)
-port: Database Port (3306 default)
-charset: Database character set (binary default)

CSV FLAGS
=========
-file: CSV output filename (Write to stdout if not supplied)
-query: MySQL query (required, can be sent via stdin redirection)
-header: Print initial column name header line (true default)
-d: CSV field delimiter ("," default)
-q: CSV quote character ("\"" default)
-e: CSV escape character ("\\" default)
-t: CSV line terminator ("\n" default)
-v: Print more information (false default)

DEBUG FLAGS
===========
-debug_cpu: CPU debugging filename
-debug_mem: Memory debugging filename
-version: Version information

Examples

mycsv -user=jprunier -pass= -host=db1 -file=my.csv \
-query="select * from test.table1 where filter in ('1', 'test', 'another')"

mycsv -user=jprunier -pass= -host=db1 -file=my.csv \
-query="select a as 'column1', b, c as 'column c' from test.table2"

mycsv -user=jprunier -pass= -host=db1 -file=my.csv -charset=utf8 -header=false\
-query="select * from test.table1 where filter in ('1', 'test', 'another')"

mycsv -user=jprunier -pass= -host=db1 -file=my.csv -d="|" -q="'" -t="\r\n"\
-query="select * from test.table1 where filter in ('1', 'test', 'another')"

echo "select * from test.table1 where filter in ('1', 'test', 'another')" |\
mycsv -user=jprunier -pass= -host=db1 -file=my.csv

or

echo "select * from test.table1 where filter in ('1', 'test', 'another')" > myquery
mycsv -user=jprunier -pass= -host=db1 -file=my.csv < myquery

mycsv -user=jprunier -pass=mypass -host=db1 \
-query="select * from test.table1 where filter in ('1', 'test', 'another')" | sed 's/\\N/NULL/g' > my.csv

mycsv -user=jprunier -pass=mypass -host=db1 -file=my.csv -query="select * from test.table3 limit 100000" -v
CSV output will be written to my.csv
A '.' will be shown for every 10,000 CSV rows written
..........
100001 rows written
Total runtime = 10.269565988s

License

[MIT] (LICENSE)