I have some SQL scripts that I'm trying to automate. In the past I have used SQL*Plus, and called the sqlplus binary manually, from a bash script.
However, I'm trying to figure out if there's a way to connect to the DB, and call the script from inside of the bash script... so that I can insert date and make the queries run relative to a certain number of days in the past.
I'm slightly confused. You should be able to call sqlplus from within the bash script. This may be what you were doing with your first statement
Try Executing the following within your bash script:
#!/bin/bash
echo Start Executing SQL commands
sqlplus <user>/<password> #file-with-sql-1.sql
sqlplus <user>/<password> #file-with-sql-2.sql
If you want to be able to pass data into your scripts you can do it via SQLPlus by passing arguments into the script:
Contents of file-with-sql-1.sql
select * from users where username='&1';
Then change the bash script to call sqlplus passing in the value
#!/bin/bash
MY_USER=bob
sqlplus <user>/<password> #file-with-sql-1.sql $MY_USER
You can also use a "here document" to do the same thing:
VARIABLE=SOMEVALUE
sqlplus connectioninfo << HERE
start file1.sql
start file2.sql $VARIABLE
quit
HERE
Here is a simple way of running MySQL queries in the bash shell
mysql -u [database_username] -p [database_password] -D [database_name] -e "SELECT * FROM [table_name]"
Maybe you can pipe SQL query to sqlplus. It works for mysql:
echo "SELECT * FROM table" | mysql --user=username database
I've used the jdbcsql project on Sourceforge.
On *nix systems, this will create a csv stream of results to standard out:
java -Djava.security.egd=file///dev/urandom -jar jdbcsql.jar -d oracledb_SID -h $host -p 1521 -U some_username -m oracle -P "$PW" -f excel -s "," "$1"
Note that adding the -Djava.security.egd=file///dev/urandom increases performance greatly
Windows commands are similar: see http://jdbcsql.sourceforge.net/
If you do not want to install sqlplus on your server/machine then the following command-line tool can be your friend. It is a simple Java application, only Java 8 that you need in order to you can execute this tool.
The tool can be used to run any SQL from the Linux bash or Windows command line.
Example:
java -jar sql-runner-0.2.0-with-dependencies.jar \
-j jdbc:oracle:thin:#//oracle-db:1521/ORCLPDB1.localdomain \
-U "SYS as SYSDBA" \
-P Oradoc_db1 \
"select 1 from dual"
Documentation is here.
You can download the binary file from here.
As Bash doesn't have built in sql database connectivity... you will need to use some sort of third party tool.
Related
I have a rather simple bash script.
#!/bin/bash
echo
echo What is the root statement for the new debate?
read body
echo
echo What is your mysql password?
read -s pass
echo
sql='INSERT
INTO `Debate` (`unblocked`, `debaterId`, `dirty` )
VALUES ('"'1'"', '"'15'"', '"'0'"' );
INSERT
INTO `Statement` (`body`, `debateId`, `debaterId` )
VALUES ('"'"${body}"'"', LAST_INSERT_ID(), '"'15'"' );'
#echo mysql -u resolution -p ${pass} -D resolution -e \"${sql//[^a-zA-Z0-9(),;\`\'_]/ }\"
mysql -u resolution -p ${pass} -D resolution -e \"${sql//[^a-zA-Z0-9(),;\`\'_]/ }\"
What this is meant to do should be pretty straight forward but it comes down to "ask user for input on terminal... form sql command with the provided input... insert into database in two tables." It's meant to be a quick hack for adding new debates into the database for testing while development is ongoing.
Instead, mysql prints the help as if I had used -?. No error... just the help text. Everything looks correct when the mysql command is echoed to the terminal. And if I copy and paste the echoed command it works just fine. I've searched google and stack overflow but found nothing about this.
U#H ~ mysql --version
mysql Ver 15.1 Distrib 10.3.28-MariaDB, for debian-linux-gnu (x86_64) using readline 5.2
I have a script that needs to connect to oracle db hosted on a different server .I am able to connect to this oracle db using sqldeveloper.But i am not able to configured it in my bash script .
SQLDEVELOPER 4.0 is the tool that i use to connect through gui .How can i use this in my script .Is there any other way to do it ?Do i need any other software (sqlplus)
try using following once:
sqlplus db_user_name/password_for_user#DB_schema < Input_file.sql > Output
You need sqlplus to achieve what you're trying to do. The syntax of the command you need to put in you shell script should be like:
sqlplus 'USER/PASSWORD#(DESCRIPTION=(ADDRESS_LIST=(ADDRESS=(PROTOCOL=TCP)(HOST=DB_HOST)(PORT=1521)))(CONNECT_DATA=(SERVER=DEDICATED)(SERVICE_NAME=SERVICE_NAME_YOU_USE_IN_SQLDEVELOPER)))'
On *nix systems, this will create a csv stream of results to standard out:
java -Djava.security.egd=file///dev/urandom -jar jdbcsql.jar -d oracledb_SID -h $host -p 1521 -U some_username -m oracle -P "$PW" -f excel -s "," "$1"
Note that adding the -Djava.security.egd=file///dev/urandom increases performance greatly
Windows commands are similar: see http://jdbcsql.sourceforge.net/
I need to execute SQL from batch file.
I am executing following to connect to Postgres and select data from table
C:/pgsql/bin/psql -h %DB_HOST% -p 5432 -U %DB_USER% -d %DB_NAME%
select * from test;
I am able to connect to database, however I'm getting the error
'select' is not recognized as an internal or external command,
operable program or batch file.
Has anyone faced such issue?
This is one of the query i am trying, something similar works in shell script, (please ignore syntax error in the query if there are any)
copy testdata (col1,col2,col3) from '%filepath%/%csv_file%' with csv;
You could pipe it into psql
(
echo select * from test;
) | C:/pgsql/bin/psql -h %DB_HOST% -p 5432 -U %DB_USER% -d %DB_NAME%
When closing parenthesis are part of the SQL query they have to be escaped with three carets.
(
echo insert into testconfig(testid,scenarioid,testname ^^^) values( 1,1,'asdf'^^^);
) | psql -h %DB_HOST% -p 5432 -U %DB_USER% -d %DB_NAME%
Use the -f parameter to pass the batch file name
C:/pgsql/bin/psql -h %DB_HOST% -p 5432 -U %DB_USER% -d %DB_NAME% -f 'sql_batch_file.sql'
http://www.postgresql.org/docs/current/static/app-psql.html
-f filename
--file=filename
Use the file filename as the source of commands instead of reading commands interactively. After the file is processed, psql terminates. This is in many ways equivalent to the meta-command \i.
If filename is - (hyphen), then standard input is read until an EOF indication or \q meta-command. Note however that Readline is not used in this case (much as if -n had been specified).
if running on Linux, this is what worked for me (need to update values below with your user, db name etc)
psql "host=YOUR_HOST port=YOUR_PORT dbname=YOUR_DB_NAME user=YOUR_USER_NAME password=YOUR_PASSWORD" -f "fully_qualified_path_to_your_script.sql"
You cannot put the query on separate line, batch interpreter will assume it's another command instead of a query for psql. I believe you will need to quote it as well.
I agree with Spidey:
1] if you are passing the file with the sql use -f or --file parameter
When you want to execute several commands the best way to do that is to add parameter -f, and after that just type path to your file without any " or ' marks (relative paths works also):
psql -h %host% -p 5432 -U %user% -d %dbname% -f ..\..\folder\Data.txt
It also works in .NET Core. I need it to add basic data to my database after migrations.
Kindly refer to the documentation
1] if you are passing the file with the sql use -f or --file parameter
2] if you are passing individual command use -c or --command parameter
If you are trying the shell script
psql postgresql://$username:$password#$host/$database < /app/sql_script/script.sql
I have a file that contains several SQL queries.
Can I somehow run them via isql (I'm doing the calls from Bash script, so no access to Perl DBI or JDBC)
I tried piping them into isql command via echo /my/file | isql -my-other-parameters but that didn't work.
Yes.
If you're running ISQL in interactive mode, you can load an entire contents of the file using :r my-filename command from > prompt.
From Bash script, it's also possible to do - but you need to carefully make sure that
The SQL file you are piping in has a go statement at the end. That is a VERY common cause of issues like the one you mentioned.
That statement has a newline after it.
From a script, you can do it 2 ways: Pass on STDIN via a pipe/redirect; OR, pass in the file name via isql's -i parameter
In my case it was isq -n for a piped query to work.
isql -U $DB_USR -P $DB_PWD -S $DB_PATH -D $DB_NAME -w 500 < $FILE
I'm looking to be able to run a single query on a remote server in a scripted task.
For example, intuitively, I would imagine it would go something like:
mysql -uroot -p -hslavedb.mydomain.com mydb_production "select * from users;"
mysql -u <user> -p -e 'select * from schema.table'
(Note the use of single quotes rather than double quotes, to avoid the shell expanding the * into filenames)
mysql -uroot -p -hslavedb.mydomain.com mydb_production -e "select * from users;"
From the usage printout:
-e, --execute=name
Execute command and quit. (Disables --force and history file)
here's how you can do it with a cool shell trick:
mysql -uroot -p -hslavedb.mydomain.com mydb_production <<< 'select * from users'
'<<<' instructs the shell to take whatever follows it as stdin, similar to piping from echo.
use the -t flag to enable table-format output
If it's a query you run often, you can store it in a file. Then any time you want to run it:
mysql < thefile
(with all the login and database flags of course)
echo "select * from users;" | mysql -uroot -p -hslavedb.mydomain.com mydb_production
As by the time of the question containerization wasn't that popular, this is how you pass a single query to a dockerized database cluster with Ansible, following #RC.'s answer:
ansible <host | group > -m shell -a "docker exec -it <container_name | container_id> mysql -u<your_user> -p<your_pass> <your_database> -e 'SELECT COUNT(*) FROM my_table;'"
If not using Ansible, just login to the server and use docker exec -it ... part.
MySQL will issue a warning that passing credentials in plain text may be insecure, so be aware of your risks.
From the mysql man page:
You can execute SQL statements in a script file (batch file) like this:
shell> mysql db_name < script.sql > output.tab
Put the query in script.sql and run it.