I want to export a random entry of my database into a file with the command
SELECT * FROM my_table ORDER BY RANDOM() LIMIT 1 \g /path/file;
This query works if I enter it in my db terminal, but I want to us this query with a bash script but then I get the error: syntax error at or near "\g"
My bash script looks like this:
PGPASSWORD=*** psql -U user -d db_name -h localhost -p port -t -c "SELECT * FROM my_table ORDER BY RANDOM() LIMIT 1 \g /path/file"
Bash is interpreting the string and trying to interpolate it. Probably, escaping the backslash will solve your problem.
PGPASSWORD=*** psql -U user -d db_name -h localhost -p port -t -c "SELECT * FROM my_table ORDER BY RANDOM() LIMIT 1 \\g /path/file"
A SQL statement terminated by \g is not supported by the -c command switch. Per documentation of -c:
-c command
...
command must be either a command string that is completely parsable by the server (i.e., it contains no psql-specific features), or a single backslash command. Thus you cannot mix SQL
and psql meta-commands with this option
To redirect the results to a file, there are several options:
shell redirection: psql [other options] -Atc 'SELECT...' >/path/to/data.txt
-A is to switch to unaligned mode (no space fillers to align columns).
put the SQL part in a heredoc text instead of the command line:
psql [options] <<EOF
SELECT ... \g /path/to/file
EOF
This form has the advantage that multiline statements or multiple statements are supported directly.
\copy of the query. Be aware that COPY to a FILE is different: it creates the file on the server with the permissions of postgres and requires being a database superuser. COPY TO STDOUT works too but is not better than SELECT concerning the redirection.
I found a solution for my script, and now it works.
#!/bin/bash
RANDOM_NUMBER=0
while true
do
for i in `seq 1`
do
RANDOM_NUMBER=$(($RANDOM % 100000))
echo $RANDOM_NUMBER
PGPASSWORD=*** psql -U user_name -d db_name -h localhost -p PORT -c
"INSERT INTO numbers (number) VALUES ('$RANDOM_NUMBER');"
done
sleep 10
for i in `seq 1`
do
PGPASSWORD=*** psql -U user_name -d db_name -h localhost -p PORT -c
"DELETE FROM numbers WHERE id = (SELECT id FROM numbers ORDER BY RANDOM() LIMIT 1);"
done
done
Related
I have been tasked with replacing ISQL in a lot of our bash scripts with sqlcmd. ISQL allows piping a variable in it's execution.
An example would be:
SQL_STATEMENT="SELECT TOP 1 SYS_USER_NAME FROM SYS_USER"
echo $SQL_STATEMENT | isql -b -d, $DSN $DBUID $DBPWD >> setupdb_test.txt
From what I can tell this is not viable in sqlcmd. How can I do this? What flags does sqlcmd have to allow this to happen?
Here is what I have tried and have had a good result BUT I really do not want to create the file sql_command.sql every time a particular script runs:
echo "SELECT TOP 1 SYS_USER_NAME FROM SYS_USER" > sql_command.sql
sqlcmd -S $DB -U $DBUID -P $DBPWD -d $DSN -i sql_command.sql >> setupdb_test.txt
Programs originating on Windows can be picky about how they handle non-regular files and I don't have the opportunity to test, but you can try the typical Unix tricks for providing a "file" with data from an echo.
Either /dev/stdin:
echo "SELECT TOP 1 SYS_USER_NAME FROM SYS_USER" | sqlcmd -S "$DB" -U "$DBUID" -P "$DBPWD" -d "$DSN" -i /dev/stdin
or process substitution:
sqlcmd -S "$DB" -U "$DBUID" -P "$DBPWD" -d "$DSN" -i <(echo "SELECT TOP 1 SYS_USER_NAME FROM SYS_USER")
I have a bash file which executes a SQLCMD query to check for the existence of a database:
sqlcmd -S localhost -U sa -P $SA_PASSWORD -d master -Q "SET NOCOUNT ON SELECT count(*) FROM master.sys.databases WHERE name = N'MyDatabaseHere'")
I tried adding the SET NOCOUNT ON but it still outputs like this:
------
1
I am assigning the result to a variable and I want to be able to check it like this
myvariable == 1
Is there a way to do this?
Took me a while but I found a solution only using sqlcmd.
You need to use -W -h-1 -k
-W removes trailing space
-h-1 removes headers
-k Removes all control characters, such as tabs and new line characters from the output.
TABLENAME=$(/opt/mssql-tools/bin/sqlcmd -S localhost -U sa -P $SA_PASSWORD -d master -W -h-1 -k -Q "SET NOCOUNT ON SELECT name FROM master.sys.databases WHERE name = N'MyDatabaseNameHere'")
I'm trying to get a list of database names using the psql command. So far I have:
psql -h example.com -U backup -c '\l'
This however gives me the results in a table-like format. I ONLY want the table names (one on each line). How would I accomplish this?
This does it:
psql -h example.com -U backup -t -A -c "SELECT datname FROM pg_database WHERE datname <> ALL ('{template0,template1,postgres}')"
Using the system catalog pg_database.
Read the manual about psql.
I don't think you can use \l for that.
But the following should do it:
psql -h example.com -U backup -t -c "select datname from pg_database"
\t turns off the header and the other "noise". And the select statement only returns the database name.
Not sure if you need to use single quotes or double quotes for the SQL statement - on my Windows console I had to use double quotes.
What should I do for making it work?
#!/bin/bash
TABLENAMES="user_stats"
ssh -t railsapps#xxx.xxx.xxx.xx -p xxx bash -c "'
for TABLENAME in $TABLENAMES
do
psql -d mydb -P format=unaligned -P tuples_only -P fieldsep=\, -c "SELECT * FROM $TABLENAME" > /tmp/$TABLENAME
done
'"
General problem: how to periodically dump the database tables to a local machine from a psql database in a single bash script run on Mac OS X?
Firstly, you should test your SQL and bash scripts remotely (do SSH interactively).
I think your problem is caused by a bad mix of quote / double-quote. I think the star (*) and $TABLENAME are expensed before the SSH call, so too early. Try to put a backslash before the $ sign.
You should use the verbose or the debug option, to help to understand what is really executed:
ssh -t railsapps#xxx.xxx.xxx.xx -p xxx bash -vxc "'
for TABLENAME in \$TABLENAMES; do
psql -d mydb -P format=unaligned -P tuples_only -P fieldsep=\, -c "SELECT \* FROM \$TABLENAME" > /tmp/\$TABLENAME
done
'"
How to atore a scalar postgresql-value on a bash-variable like in script below?
dbname="testlauf"
username="postgres"
vartest='psql -c -d $dbname -U $username -h localhost -p 5432 "SELECT gid FROM testtable WHERE aid='1';"'
echo "$vartest"
I tried several different writings, but nothing seems to work. Thanks in advance.
Put the -c option just before its argument - the query. Mind also using the additional -t option to get just the tuple value. And of course, use the backticks (`) operator.
Using the -X option is also recommended, as sometimes a .psqlrc file might add some redundant output, as well as the -A option, which disables column aligning (whitespaces).
In order to skip NOTICE or other additional messages, include the -q flag.
vartest=`psql -d $db -U $user -AXqtc "SELECT gid FROM testtable WHERE aid='1'"`
Using -t option or --tuples-only will give you the rows only, so it will easier to store them in array variable (if the result from query more than one)
vartest =(`psql -t -d $dbname -U $username -c "SELECT gid FROM testtable WHERE aid='1';"`)
echo $vartest
example:
query result
ubuntu#ratnakri:~$ psql -h localhost -p 5432 -t -U postgres -d postgres -c "select slot_name from pg_replication_slots"
barman
barman2
make it into array variable
ubuntu#ratnakri:~$ RESULT=(`psql -h localhost -p 5432 -t -U postgres -d postgres -c "select slot_name from pg_replication_slots"`)
ubuntu#ratnakri:~$ echo ${RESULT[0]}
barman
ubuntu#ratnakri:~$ echo ${RESULT[1]}
barman2