Even if you are using a GUI tool to connect to your MySQL servers, one day or another, you will have to deal with the command line. So it is nice to know a few tips that can really make your work easier.
Note: The commands below are only available for Unix/Linux.
Most of the graphical tools paginate results, which is very handy. But this is not the way the command line client works: it just outputs all results. It can be annoying but it is easily solved by using the pager command:
|
1 |
<br>mysql> pager more<br>PAGER set to 'more'<br>mysql> select title from sakila.film;<br>+-----------------------------+<br>| title |<br>+-----------------------------+<br>| ACADEMY DINOSAUR |<br>| ACE GOLDFINGER |<br>| ADAPTATION HOLES |<br>| AFFAIR PREJUDICE |<br>| AFRICAN EGG |<br>| AGENT TRUMAN |<br>| AIRPLANE SIERRA |<br>| AIRPORT POLLOCK |<br>| ALABAMA DEVIL |<br>| ALADDIN CALENDAR |<br>| ALAMO VIDEOTAPE |<br>| ALASKA PHANTOM |<br>| ALI FOREVER |<br>| ALICE FANTASIA |<br>| ALIEN CENTER |<br>| ALLEY EVOLUTION |<br>| ALONE TRIP |<br>| ALTER VICTORY |<br>| AMADEUS HOLY |<br>--Plus--<br> |
Another example of the pager command is when you want to estimate a good size for you InnoDB redo logs: the estimation is based on the variation of the Log Sequence Number during a given period of time. Instead of manually looking for the right line in the output of SHOW ENGINE INNODB STATUS (which can be huge), you can call pager to the rescue:
|
1 |
<br>mysql> pager grep sequence<br>PAGER set to 'grep sequence'<br>mysql> show engine innodb statusGselect sleep(60);show engine innodb statusG<br>Log sequence number 380166807992<br>1 row in set (0.41 sec)<br><br>1 row in set (1 min 0.00 sec)<br><br>Log sequence number 380170274979<br>1 row in set (0.00 sec)<br> |
When you are done and you want to disable paging, you can simply run:
|
1 |
<br>mysql> pager<br>Default pager wasn't set, using stdout.<br> |
When you try to optimize a query, it often involves manipulating the text of the query, and sometimes it would be great to have a text editor inside the client. Well, this can be achieved by using the edit command.
Let’s say you have the following query:
|
1 |
<br>mysql> select count(*) from film left join film_category using(film_id) left join category using(category_id) where name='Music';<br> |
and let’s say you want to change the left joins to inner joins and use capital letters for reserved SQL words. Instead of manually editing the statement, which will be boring, you simply call edit:
|
1 |
<br>mysql> edit<br> |
and it will open your default text editor with the text of the last query. The default text editor is vi, so you now have the power of vi inside the mysql client!
Once you have made your changes, save and exit the editor: you are back in the mysql client where you can type ; or G to execute the query.
In some situations, like when you are testing a set of commands to write documentation or when you are in the middle of an emergency, you want to be able to record the queries that you have executed. The command line client offers the tee command, which will log to a file the statements you typed and their output, pretty much like the Unix tee command:
|
1 |
<br>mysql> tee queries.log<br>Logging to file 'queries.log'<br>mysql> use sakila<br>Reading table information for completion of table and column names<br>You can turn off this feature to get a quicker startup with -A<br><br>Database changed<br>mysql> select count(*) from sakila;<br>ERROR 1146 (42S02): Table 'sakila.sakila' doesn't exist<br>mysql> select count(*) from film;<br>+----------+<br>| count(*) |<br>+----------+<br>| 1000 |<br>+----------+<br>1 row in set (0.00 sec)<br><br>mysql> exit<br> |
And now if you look at the content of the queries.log file, you will see a copy of your session.
The mysql command line client is not as glossy as most of the graphical tools, but if you know some of its hidden features, it can be very powerful. If you enjoyed these tips, I will write another post with other useful but overlooked features.
Resources
RELATED POSTS
